The Bug is back [Archive] - The Art of Sound Forum

Bksabath

06-06-2018, 09:19

23643

UP and downs
Downs
Those capacitors need screws so I ordered same from RS
Part N 6086785 £22.00 for 100 bag
and 537151 £5.00 for 100 bag

RS put them as stock items delivery for next day not true the top one was sent 2 days later from France
I was expecting a much better product as it cost over 4 times as much and it is not
They are exactly the same
No need to say quick phone call and RS re founded the money but one think RS components used to be a good source for products and now ? One fraud prone scam ?
The web site catalog is a bit of a joke as well to try and find parts is not what it used to be

23644

Up
Rapid electronics
Now stock Kerafoll 86/82 as far as I know it is the best interface between heat sink and transistor unless

Down
one manage to get same CVD diamond

http://www.e6.com/en/home/materials+and+products/single+crystal+synthetic+diamond/cvd+synthetic+diamond

I have spoken whit a few peoples whit loads of alphabet after the name they do not know that it has ever been used for this kind of thing I told them that I was going to make a study and that it would have been nice to get a few samples

well they sent me a link to the shop that has not replied so far
I will look more in to this for the F6 build
In the mean time I am covering my bases
CVD Diamond as thermal Pad for transistors is my idea and you read of it here first :lol:

sq225917

06-06-2018, 12:47

Using copper instead of aluminium on the other side of the heatpad would make a big difference.

Bksabath

06-06-2018, 13:53

23645yes quite right
Dam one could never find the standard BIC Lighter gauge :scratch:
This is going to be the heat sink for the F6
Scale I have tip over 20Kg but take that whit pinch of salt
It came out of a 200 Kw Danfoss Inverter
Plan is to cut in half and force cool it I will post the simple circuit later on

23646

In the mean time I have started to reform the capacitors
10V for 24 hours then slowly up to the nominal voltage when I would let them settle for a few days

Nice to see you here Simon !

Bksabath

07-06-2018, 05:38

The boards have arrived one word … NICE!

23653

In the picture 2 sets out of 4
those will be 2 50W F5 Turbo mono bocks

Plan is to go for 2 mono blocks so we have 2 soft start and 2 speaker protection
I know very little about those I use CL60 thermistors for the soft start and UPC1237 (or NTE 7100) whit Amplimo speakers relay on mine.

The soft start, speaker protection, rectifiers and most probably ground lift + DC blocker will be on the tunnel under the sinks while the Input board will be mounted on the back plate
Me and Martin spent 4 hours messing about and making list of things to do, piking Sink sizes (planning on 300 X 82 X 170 (4U)) and so far the lay out may be like this

23654

Things to do
Fan speed controllers

Bksabath

07-06-2018, 17:39

Today I had beans !
sorry wrong tread

Lets keep it fifi

Today I got a drill

23667

Should be enough to drill any thing of any size
Pretty quiet to

sketch of fan controller

l 23668

one may like the amplifiers to warm up pretty fast, no need to call home before hand to get GF, Wife or whatever to turn FiFi on and keep the temperature stable as the lauder one play the cooler the sinks get (up to a point)

On sketch is just general Idea (may not work as drawn)

There may be a few LED to put here and there say yellow cold, green You can now stop doing whatever you doing whit GF, Wife and such and listen to same music, Red for 2 hot fan failure and switching power off

Bigman80

07-06-2018, 18:40

Bksabath

07-06-2018, 21:57

Empirical results

the soldering Iron is putting out 40W of heat
Moss thermistor and thermometer on the heat sink
By adjusting the pot I can set the temperature at any where I like

236692367023671

Snags
1)Moss start conducting and the fans start to run at 3.17 V 40 C at 3.22 V before that fans stall, how bad is that for the motors ?

use green led as on previous post to turn fans on simple!

2) fans make noise (especially the ones I have on the bench)

is it worth to spend money on the fans and complicate things when that money could be spent on bigger heat sinks ?

not my decision .

Still amazing how such a small heath sink can dissipate whit a bit of air flowing Yes?

Bksabath

08-06-2018, 13:18

23674

Amplimo relay's have landed

they do not fit on boards but I was going to home brew same whit NTE710 or uPC1283 that I like very much

Do you want to know more ? (could that movie get any worst?)
https://www.halfgaar.net/dc-protection-with-relays

PS white one is old stock I have, black one is new type same spec same foot print as confirmed by Amplimo

Bksabath

10-06-2018, 10:13

236842368523686
Those are same ideas on how to make things, more like finding 100 ways that do not work.

I made small rig whit a couple of fans and I have a spare heat sink that is more or less the size we are going to get

I have a few spare Mosfet so here is more or less the way they are going to be clamped on the sinks

Kerafoll 82/83 is the best pad I know (apart from Diamond :D ) the efficiency increase whit the pressure, on the manufacturer spec sheet graph stops at 60 Nm CM2, in pints that is 6Kg centimetre square, there are going to be bars on top of each "transistor" each clamped whit M5 captive studs and bolts, it is much betrerer than the single M3 screw as that either strip the tread or snap

Same basic about Mosfets as I understand them, they are like a tap that either sink or drain a tank, the higher the tank the higher the pressure (voltage) the wider the pipe the more current flow

The Mosfet let more water flow if the voltage to the gate go up an over a certain point.
This voltage is different for each type of Mosfet and has smaller variation between those dudes of the same type, it will also change whit the temperature of the mosfet.

As we are going to use pairs of those dudes they must be matched as if one conduct more easily it will steal current from the other (source resistor take care of that in part) maybe a bit more clamping force on the strongest can make things even better?

Mosfet cam in 2 type P and N one let a negative voltage flow one way the other a positive one, different type of those 2 basic flavors exist …

There is an area at which they work best this is dependent on dude temperature and current flow, get past a certain point and they will break, when they break the tap is fully open and this is not really horroshow for your speakers that will not only blow up but also give your " beloved" a further chance to say "Told you so" (that is why protection is always recommended especially if one have beloved)
As Nelson Pass said over and over again 50C on the sinks is ideal whit a bias current at about 1.3 A on each

Voltage on the rails is (Secondary AC x 1.471 Less drops due to the R in CRC) call it 35 V for now

W = V X A = 35X1.3 45.5 Watt This is a lot to ask , will need to look at the Dudas case temperature and speck again, but based on this for now the sinks need to get rid of nearly 200 W of heat when music is off

Class A bias
the 2 flavors of Mosfet work against each other, Let 2.6 A flow in one pair and 2.6 A flow on the other, when the 2 current's meet in the middle we have 0 volts, all that current is going to contribute to global warming

Put a bit of music on to twist the taps one way or the other, the taps have a more or less solid link in between and only move in the same direction and the voltage at the middle will go up and down and your speakers cones, that is if you have any of those (Jup here is another one), will go in and out (love the Burgess/Kubrick reference) and contribute to your and neighbours pleasure centres in many different ways.

Barry

10-06-2018, 16:40

Today I got a drill

23667

Should be enough to drill any thing of any size
Pretty quiet too

I'm looking for something like that. Who is the manufacturer, what is the model number and how much do they cost?

Thanks

sq225917

10-06-2018, 17:59

Forced air cooling, you're crazy. I love that.

Bksabath

10-06-2018, 21:41

Hi Barry
It is Axminster they make same very nice drills I was looking at the micro milling machine as well and they do same super duper micro drills ideal for PCB
I got that one because it could take big pieces.
Drill https://www.axminster.co.uk/axminster-hobby-series-ahrd16b-bench-radial-drill-101928
On my wish list https://www.axminster.co.uk/axminster-model-engineer-series-sx1-micro-mill-505098

As things go the drill quality is pretty good but still made in china (blue and green machines are much better IMO)

Hi Simon
You know me and know that I take that as a compliment, crazier still my the F5 that I am using at present have 300 X 80 X400

Forced air cooling has same advantages 1) amplifiers go up to temperature faster 2) sink can be smaller or bias can be higher

We are looking at same heat sinks not from Modushoop IMO those are way to small and , most important, the base plate to which the sink are attached is not thick enough.

The 50 W turbo may not need it (it is going to be partnered to same unusual speakers, 92 dB 4 Homs 18 transducers this is all I tell you for now), the 100 W may do as sinks will be in the middle most likely and having it going hot faster is one of the things Martin was asking for

Paul and Martin where here today to talk about the shopping list, things are going to be on hold for a couple of weeks as big money is going to be spent on the Traffos

For the F6, I will start that when F5 T are completed, forced cooling will have to be a necessity as , and it break my hart, that copper heat sink will need to be cut in half as I only have one

walpurgis

10-06-2018, 22:31

I bought a cross slide vice to go with my drill. It allows me to do some basic milling. My vertical angle vice is handy too.

Bksabath

20-06-2018, 14:40

Got prices for heat sinks

£35.26 each for 8 units
http://www.abl-heatsinks.co.uk/index.php?page=extrudedproduct&product=20

Bksabath

30-06-2018, 13:08

23762

Work in progress
Heat sink ordered and payed for
Jfet ordered and payed for
Bit and pieces Boom ready order out on Tuesday
Traffos ??????????????????????? WTF?

It is all going to happen pretty quick once we get all the parts ...

Barry

30-06-2018, 14:02

This project is turning out to be a little different to what I expected:

http://theartofsound.net/forum/attachment.php?attachmentid=23762&d=1530363877

:D

Bksabath

02-07-2018, 20:50

One got to eat No?
Just waiting for ingredients for the F5T's
J fets have been shipped today from Hong Kong :D got 12 pairs

Bksabath

02-07-2018, 20:51

PS. how do you get big pictures like that ?

Barry

02-07-2018, 23:56

I just clicked on your 'thumbnail' image, copied and pasted it into my post.

Bksabath

03-07-2018, 05:07

tanks going to try that as soon as I get same nice pictures or get bored

Bksabath

03-07-2018, 19:14

Well no chance to get bored

23789

We got heat sinks

23790

Landed this afternoon

23791

F5T 50 W left F5T 100 W right

tomorrow I will place the order for bits and pieces

Bksabath

12-07-2018, 17:54

Anather parchell arrived today
50 X 1R MPC 75 Futaba resistors
Transformers ?

Bigman80

12-07-2018, 18:20

Loving the thread!!!

Bksabath

17-07-2018, 11:33

23810

Loving the thread!!!
You must be pulling my leg :D You have PM

It is not going as quick as I was expecting, but then again I am making those for 2 "friends"

In the mean time I a think I am making progress here is a spread sheet whit resistor values I am working on

more as I carry on, hope it is of same use, later Part numbers cost and source

File size let me post picture only (one is not so good whit Puters) PM if you like the spread sheet when is completed
I am making this for mono blocks so 2 soft start and 2 speaker protection boards
For the 50 W mono-blocks use those QTY for the 100W double up on output boards

Bksabath

19-07-2018, 18:36

Got the MCP 71 1R today

Conrads only do the 2 W and blipping RS will tell you that they keep them but then they don't

RS is a joke not what it used to be...

Any way got 100 only :D one does not want to be out of stock on those ;)

Burklin seems to be the only source

Bksabath

23-07-2018, 11:07

238492385023851
Things should take off now at least for the build of the 50 W Paul's F5 Turbo
Other build has been suspended ;)

Parcel from Mouser has landed

Bksabath

24-07-2018, 15:24

Out put Boards

I have mounted the thermistor for the temperature compensation as close as possible to the centre pin
This is the hottest point, that is accessible , of the mosfets

May be to fast on picking up changes ? comments welcome

As far as I am concerned, this is the most fiddly part of the build

238522385323854

Bigman80

24-07-2018, 15:34

238492385023851
Things should take off now at least for the build of the 50 W Paul's F5 Turbo
Other build has been suspended ;)

Parcel from Mouser has landedOh dear, suspended a build! Must be bad,

Looks like a good build anyway, looking forward to seeing the finished article!!

Bksabath

24-07-2018, 16:57

Oh dear, suspended a build! Must be bad,

Looks like a good build anyway, looking forward to seeing the finished article!!

Hope it is only a delay, I am not that bothered about it, the spare bits I have will surely go in a snip.

And once Paul's is up and running …;):D

sq225917

24-07-2018, 19:02

I wouldn't turn my nose at an F5t 100w. Even though I've a brace of mod86p amps that need casing up and a preamp with no case...

Bksabath

30-07-2018, 11:11

things are moving along
here is how I will clamp the transistors

Sizes are
Sink base plate ,this is quite important as heat travels trough this to reach the fins, 13 mm (not 8mm like the Modu sinks ;))

drill holes tap M5 depth 10 mm (nice to have a drill whit a stop so one does not drill right trough)
cooper bar is 20mm X3 mm

total depth whit transistors from top of bar to hole bottom 17 mm add 2 mm for washers
Bolts M5 X16

I like this way to clamp the transistors
1) pressure is distributed evenly
2) taightening the screws does not lift the lip of the hole and pushes the transistors up
3) whit M5 bolts much more pressure can be applied
Look at the graph in the spec sheet of the Kerafoll 86/82 curve drop whit pressure and stop at 60 Nm

We are planning on a bias of 2 Amps maybe more at 35 V that will be 70 W for each sink or 35 W on each transistor 140 W idle

Once set up bias may be increased a fair bit so to get the sinks at a nice 50 C

23890 23891

This is the 50 W build
Meeting tonight as we have to talk about a few things

Bksabath

30-07-2018, 11:33

23892

23893

A bit more details about the boards

The thermistor is fitted to the middle leg of the Mosfets

this is the hottest accessible point as the leg is welded to the junction where all the up/down up/down happen

As the temperature go up the mosfets conduct more and the current increase and this increase the temperature

As the temperature go up the mosfets conduct more and the current increase and this increase the temperature

As the temperature go up the mosfets conduct more and the current increase and this increase the temperature.....
Boom :oops:

The thermistor is there to reduce the Bias
Much more about bias later

Bksabath

01-08-2018, 20:50

Well met did not happen :D not so fuss about it. I carry on regardless
Input boards are finished

2390423905

sq225917

01-08-2018, 20:57

Your progress makes me ashamed, I still have power amps to case up 16 months after I bought them. I ordered cases last week though.

Bksabath

02-08-2018, 07:37

Your progress makes me ashamed, I still have power amps to case up 16 months after I bought them. I ordered cases last week though.

what a load of ….

You been making one paradise after the other everybody knows
:D

Bksabath

19-08-2018, 20:49

Things have been moving along and the amplifiers are starting to take shape

I had my niece and her boy friend around and it has been a week of live music over here from them playing guitar in my living room to the Chester cathedral organ and quire and few live gigs in Liverpool pubs
File size is to large to post those here I may get around it same time

https://rikyandthekids.bandcamp.com/album/lets-rumble

This is a link to Riky (Riccardo) Latest CD hope you like it ( I certainly do)

23970Input boards ,speaker protection and soft start board

2397123972 First lay out of Paul's amplifiers

23973 Martin has been around as well and started soldering the components to his boards so the build of the 100W turbo has started

Bksabath

27-08-2018, 20:38

It is a slow process but we are learning the 100 ways that do not work
Copper bar is lovely (not) to work whit and take a rustic look if you touch it so I have been finding the wrong way to do electroplating at home there is a fair bit to do but I think I have cracked it. (Clean clean clean and clean)

Then there is the lay out to be considered

Any way same more porn for you in the pictures

100 W build sink bottom view whit fans

24054

One way to go may need different base plate
24055

Capacitors stick out this way
24056

and this is another way
24057

Bksabath

27-08-2018, 21:02

PS you may noticed the different tint of the cooper plating bars
The one whit no resistors is the last one I did and probably the best of the lot
that was after cleaning the lot whit same alkaline gigamagic one keep in the kitchen (oven cleaner) and new sponges.
Easy enough to start again and make the lot much better

+ we have found a source for better and more color-full potions (not royal WE this is DIY collaboration and Jobs are delegated around)
I may delegate the plating process once it has decent standard

at present I am using a Stainless steel wand whit 2 1.5 V batteries

As soon as we get the new staff I will post

About the resistors both on the boards (1R X2) and on the CRC (0.47 R X6) I am using Fucushima Futaba MPC 71and MPC 74

Capacitors: at the moment the white ones are PEH 200 10mF

On the 100 W build we are looking at getting a ridiculous amount of Beer Can sized dudas probably in CRCRC

Bksabath

02-09-2018, 12:24

While Martin is soldering it's own board's and doing same home work (read the Pass Write up willya)
I am making progress whit the lay out

24109

Front view Traffo not to scale

24110

Top view Caps are on the way so for now same small bear cans it will have 3 X 47mF PEH200 in CRCRC or CRCC or maybe CRCLC CCRC or CLCRC please vote for your fav! yup 12 total

24111

Back view whit input board on heat sink again maybe
2411224113

And the 2 sides

One may have noticed the space under the sinks ? more about that later

Bigman80

02-09-2018, 12:53

[emoji23][emoji23][emoji23] beer can markers, I love that.

Sent from my VKY-L09 using Tapatalk

Bksabath

02-09-2018, 13:46

Warning Maths In this post !!!:scratch:

Assume 100 W nominal
Assume 35V rails
Assume 1 A Bias on each output duda 35X1 + W each duda is loaded at 35 W max may be as hi as 50 W = 1.4 A

Class A idle current on each rail 4 A X 35 V =140 W in heath that has to be dissipated on each heath sink .

Total energy saving 560 W You can certainly switch off a decent sized radiator in the listening room and use your FiFi instead a much greener approach would be to turn off the central heating completely and just stay in the listening room all the time

To improve on the efficiency and give you even more savings lets puts same fans on the sinks so that cold air go in from the bottom and pop out from the top much warmer as the music (or the absence of it ) will do that for you free of charge (Actually is a bit more than that as Gas is more expensive)

Bottom and side view of the 2 cheap fans they spec at 25 dB noise and should be running at less than 1/2 voltage As fitted there will be a sheet of acrylic bolted to the bottom of the sinks whit holes for the fans and the beer cans, sorry meant capacitors, so to hold those in place

2411424115

As fitted there will be a sheet of acrylic bolted to the bottom of the sinks whit holes for the fans and the beer cans, sorry meant capacitors, so to hold those in place

24116

As fitted the 2 circuit boards whit the 4 dudas on the low side diodes on top

24117

The amplifier will be on Class A till when one side turns on so much that the other side turn off the.
To do that one need 8 A assuming a standard 8 Homes speaker 8 x 8=64W
assuming a standard 4 Homes speaker 8 x 4 =36W
assuming one has a pair of isobaric 8 X whatever one fill like = not enough data :D

As the F5 will clip at about 2 volts from the rails assume 32 V swing up or down or better in and out (that is how a speaker move yes?)

Remember this ?

V
A R
W = A X V or V/R X V = 128 W if we where talking about DC instead of AC and considering the out part instead of in out in out that the cones do plus the fact that FiFi manufacturer's have a certain standard to measure amplifier's output power so that they can be pretty vague....

Is any one able to give me a figure please ?
Double that for 4 Homes and take a pinch of salt if you have a pair of Isobaric

Bksabath

13-09-2018, 20:56

Big day tomorrow
smoke test

The 50 W may well be a 3 box duda whit transformers soft start and the first bunch of capacitors then 2 cables to carry the DC to the mono-blocks
For now all is open air and whit a spare transformer because the Toroidy have not arrived

24181
I will be keeping this switched on for a while and increase the voltage to re form the capacitors
Must say well pleased whit the soft start as it worked from Go NO SMOKE

One channel heath sinks whit the transistors bolted down

24182

More or less this is one of the 50 W mono blocks

24183

They will be as small as possible and will be placed behind the speakers
Each will have another 50000uF on each rail and speaker protection

The input / Bias board of the F5 Turbo
24184

Just a few wires to solder and then I can start setting the Bias
May get a few tunes out of it by evening.

sq225917

13-09-2018, 22:36

Good luck

Bksabath

14-09-2018, 16:54

Tanks Simon

We are listening to it as I type this
Hfidog and pc141263 are here :D

I am so pissed off …

Bias set at 250 mV across the 0.5 R bias resistors sinks at 32 C and it sound bloody better than mine.

Pictures later

sq225917

14-09-2018, 17:23

Lol

pc141263

15-09-2018, 04:48

BKsabath powered this amp up last night and it sounded amazing, even in its raw form, What an amazing build you've carried out on this f5t amp.

Thank you so much for all your hard work.

Bksabath

15-09-2018, 20:29

You are welcome I really enjoy doing this and as I said the bug is back, so tank you !
I most probably rebuild mine after
It will really start to sing when the Bias is set properly and the presentation, imaging and so on will be much better whit the 2 running

at the moment it was just 250 mV across the 0.5 homes resistors, could same one cek my maths please , that is about 500 mA from Papa notes it should go to around 350, this said, most of the builds out there do not have your heat sinks so there is plenty of head room to play whit

they where running at about 32 C and if the above is correct we where only pushing 35 W on each

I was going to post same pictures of the night, sorry I am to tired after a 12 hour shift

I most probably get same in for Monday

Bksabath

29-09-2018, 20:40

I was going to post same measurement of the Mule last week new job and other things in the way …
And now the Toroidy have landed
so I toke the mule apart and started to put things back again
Those from last week

24283Paul and Martin listening to the Mule

24284Martin having a play whit his capacitors (well You quite open minded here so I hope I am not going to get banned)

24285 and here the Mule in my system

24286And a close up of the Mule

Bias is about half of what it is meant to be and it sound better than mine already

Bksabath

29-09-2018, 20:45

This was yesterday Martin 100W Turbos
at last the Toroidy are here

24289

24288

24287This is an Idea for Martin lay out the traffo will be bolted to the sink

Bksabath

29-09-2018, 21:13

and this is today
Dressing up the cable
We got the base plate optional extras, IMO they are not optional at all trafo fit there nicely whit a M8 bolt

24290

This is Paul power supply
His requirement are that all should be pretty small so this lot is going inside a 300 mm deep U3 enclosure and then a length of cable to take the DC to the second bank of capacitors. we are using the cable as R in the CRC
At the moment I am thinking about 2 meters of 2.5

24291

Look like the box just fit whit a bit to spare
24292
I am making new plates to keep the capacitors together 40 mm wide X 3mm Brunel will be proud of it and we may cut down in width a bit

24293
Going for 50 mF or 40 mF for the second leg of the CRC ? we are stuck whit only 40 mF in the first as there is no space left
24294

sq225917

29-09-2018, 23:27

Jesus wept, what size VA are those?

Light Dependant Resistor

30-09-2018, 00:13

Hi
Large toroids can be quite problematic when first turning on - some good instruction here to follow to overcome that difficulty.
http://sound.whsites.net/project39.htm

Bksabath

30-09-2018, 09:36

Hi Simon

They are only 600 VA
Martin was going to get the 800 but then he got 600 as well
Size would have been the same from the speck

One wonder what discount he got for the 4 ?:D
Not told us jet:eyebrows:

Tanks for the link Chris

I always enjoy reading Rod pages he has a nice understandable way to put things across, not like in same forums where Learners are just a pain in peak hour traffic ;) and same gang up on them.
I can certainly say to many, if you have not been there you are probably missing out (the Rod Eliot site)

I am using one of such soft starters whit CL 60 I may double up on them but it seems ok on the Mule ( was using a spare 1000 VA traffo)
Paul and Martin got the official soft starters and speaker protection.

I use home brewed one on my F5 which IMO are better but being home brewed they look ugly I will post a bit more about that later ( I am thinking about building F5T for myself)

I forgot to mention the speaker protection circuit …

I got it working but whit a LM 317 for a 24 V supply they are fine this way but certainly do not like the 35 V, cant complain about the boards as they take a loot of punishment while swapping blown transistor

they work fine and switch off the speakers whit about 3 V DC but I will change the Blink rate and increase the switch on delay

They are there to try and save the speakers if one of the Mossfet blow up and go shorth circuit (soon followed by the others ) and stop rail voltage getting in the woofers

I much prefer the NTE7100 or UPC1237 (very similar beast's)

I have NTE7100 whit Amplimo speaker relay in my F5 and not a problem in the last 8 years.

As I mentioned earlier we had one of Pauls amplifiers on the Right channel of my system and I don't think we could do that later this afternoon as Paul is popping down

I live to him and Martin to post first impression …
Just worth a mention I had a 2.5 A fuse on the mains and it gave up soon after we put same pressure on the pedal

Light Dependant Resistor

01-10-2018, 04:30

LM317 should be OK with 35v and 24v out, as they rely on Vin/Vout difference which is only 11v , perhaps a heatsink is needed

I would in the circumstance arrange an initial 317 or LM338 TO3 as a current regulator, taking output from its adjustment terminal and
have resistance between output and adjustment as Vref/ r which is 1.25/R

Attached schematic shows a 24 v regulator using current regulation of 463ma ( hopefully that is enough ) then voltage regulation, and I added a IRF4905 mosfet
working nicely off the resistive divider with its gate from the 338. 24299

Bksabath

01-10-2018, 17:58

The 1 A from the 317 is plenty to switch on a relay and sinks are already there as I am going to bolt it on a corner of the 4U X300 but tanks for that.

It is certainly going to came handy sooner or later

Bksabath

07-10-2018, 09:36

A Power Supply
2435124352243532435424355

Bksabath

07-10-2018, 09:39

Hope this is not to boring
:D
2435624357243582435924360

Bigman80

07-10-2018, 09:39

Can't see the pictures

Sent from my VKY-L09 using Tapatalk

Bksabath

07-10-2018, 09:43

243612436224363yes strange they only open if you click on them got 3 on this post

Bksabath

07-10-2018, 09:48

Just in case

We will not comment, (:lol: Just to make it sound like you are on another forum) on the green LED it was only one I had of right size

It will be changed for a blue one so do not you worry about that

Apart from that fire away, one thing I must say: It was a thigh squeeze

pc141263

07-10-2018, 11:29

That looks amazing my friend.

Hifidog

07-10-2018, 14:10

This looks really nice. I'm tempted to go this route myself now:scratch::scratch:

Bksabath

07-10-2018, 14:43

This looks really nice. I'm tempted to go this route myself now:scratch::scratch:

Paul requirement was that he has same discrete small amplifiers to hide away

The wow up in your face is the other option

this is DIY so anything is possible

Inside U3 Box a can you can fit "Joda"

Bksabath

15-10-2018, 15:20

This is one of the mono blocks just waiting for the front plate
24408244092441024411

The amplifier is warming up and I am setting up the BIAS

Bksabath

15-10-2018, 15:32

I am using one meter to monitor the Off set
24414
The other 2 meter measure the voltage drop on the 0.5 Homes source resistors
24413
I am also using a timer to keep track of things
and a 4 Homes load

24415
I am letting things get nice and warm before stepping up the bias
So far I have not changed any thing from last week
Things are really stable and I am going to be patient for an hour or so before I take it up (sorry meant to say WE like in other forums)

Bksabath

15-10-2018, 15:57

It has now been about 45 minutes

The voltage drop on the bias resistors for Positive and negative rail (WE :eyebrows: may do same meter swapping later as one cost 10 times the price of the other and more than my scope)

24416

Main thing is that the offset is dropping as things get warm (again meter swap later as this one was born in 1992)

24417

For now sinks at about 28 C (way to go so far) and the offset below the mV mark as it was last week

Yes it is a very stable design whit out loads of components to keep thing stable and muddy up the sound

Bigman80

15-10-2018, 16:09

243612436224363yes strange they only open if you click on them got 3 on this postNice!!!!!!!!

pc141263

15-10-2018, 16:23

Wow!!!! mate that looks FANTASTIC. I better get on with the front ends. :):):):)

Bksabath

15-10-2018, 16:46

Yes You better do.

They are not going home whit out

Plus can't wait to see same one face when he see that surprise finish :lol:

BTW 350 mV do us a favour and read the manual bit about Bias whit sinks at 32 C one does not know when to stop …

pc141263

15-10-2018, 18:50

quality and reliability over max power every time for me .

Bksabath

15-10-2018, 19:53

funny You should say that

:oops:

I was testing things and must have slipped whit the volume I got smoke out of the final boards
Must say it was not on any condition you are likely to see at home ever
100 Khz whit a square wave volume unknown
No big deal, my fault, it will get sorted no worries about that

Bksabath

15-10-2018, 20:22

Any way same pictures before the smoke
this is getting to technical but other on this tread may appreciate and comment

24418

My scope is not that good one of the channels, top is pretty bad
24419

You may notice the shape of the curve on top channel

All the test have been taken whit a 4 homes load
2442024421

Bksabath

15-10-2018, 20:29

24424
24423

Start clipping
If you look at scope the 10 X switch multiplier is on the probe

same square wave and such
244252442624427

Bksabath

15-10-2018, 20:42

244282442924430

It start loosing shape at 60KHz
I was going to post same current results but I have blown up (not surprising) one of the Output boards

So more to follow ...

Bigman80

15-10-2018, 20:57

244282442924430

It start loosing shape at 60KHz
I was going to post same current results but I have blown up (not surprising) one of the Output boards

So more to follow ...Oh dear.

pc141263

16-10-2018, 07:05

Oh no, feeling really gutted for you. I know how much work you have put into these Amps. No worries I'm sure it's only a minor setback.

Bksabath

16-10-2018, 08:00

Well I got it apart already and the negative board has 2 blown Mosfets
I have already matched a god pair out of my stash

It looks like the J fet is gone as well but again I got plenty of spares, this is a bit more upsetting because it is going to be one less pair that I leave to my grand children's
It should be back on later on
Positive thing about it the speaker protection works :D

2443124432

Bksabath

23-10-2018, 10:50

24459244602446124462

Back on after changing the power dudas
Found cause of blow up as well, little strand of wire on the pot I was using to set the volumes went the wrong way

Any way
It has been cooking overnight
Offset is on the mV mark and the temperature on the sinks at a conservative 35 C
I have set the bias at 0.35 V on the resistors

Here is the output at 1Khz ~ :D

I am using 4 30W 1R in series and they do get hot

sq225917

23-10-2018, 11:47

Looking good.

pc141263

27-10-2018, 15:16

looking good, sounding amazing went round for a visit today, one of the mono amp needs a little more setting up but the sound is totally awesome.

Bksabath

29-10-2018, 00:08

Quite a few things to do bit of hum whit the XP10 at max

Earth lift or not to earth lift :scratch:

Light Dependant Resistor

29-10-2018, 08:04

Quite a few things to do bit of hum whit the XP10 at max

Earth lift or not to earth lift :scratch:

Just earth, however do so using star connection. If you have hum across audio signal, consider the benefits of
using encapsulated opto resistive attenuation ie LDRs, The NSL32SR3 in particular, as they provide black silence at all volume levels .

Bksabath

29-10-2018, 22:36

There is a 40 mm wide and 3 mm thick plate connecting the capacitors and all the grounds I will get same SY 2.5 CSA cable for the supply cables
Just ned to look at the earth lift and maybe a DC suppressor, 4 big caps and diode bridge.

When I say loads of hum this is compared whit my F5s' ear close to the speakers and the XP10 at max can't hear a thing from a meter away

Finishing the enclosures will also help

Just a few holes to drill for those

No much else to report apart that they have been on for a couple of days at idle to do same burn in and form the capacitors

They should be in the new home in the next few days

I will post pictures before they go

And then I can start on the other set

Bksabath

01-11-2018, 15:53

Paul F5 ready for delivery

2451824519245202452124522

Bigman80

01-11-2018, 17:50

Superb stuff mate.

Bksabath

01-11-2018, 19:11

Delivered and running in Paul's

I will start the other pair soon …

PS Oliver price for Klotz whit decent RCA phono plugs 1.5 mt ? can you do it ? or maybe spotfire

I was going to let Paul try mine but I have XLR on those

Hifidog

01-11-2018, 19:30

They look fantastic. I'm excited to hear them. Mine next, lots of interesting things to do on them.

Bksabath

07-11-2018, 20:55

So far this has been just a progress report
I would like to make this tread a bit more interesting, bear in mind that I am a novice in electronics, and useful
All the information I get is mostly from the Nelson Pass web sites

On Paul amplifiers I have set the Bias at 350 mV across the 1 R resistor as in the article, probably it could go way up considering that the temperature on the sinks is only around 35 C

On Martin amplifiers I have 2 times as many mosfets and the power dissipation ,heat generated, is going to be times as much
another thing that Martin will like is to get the temperature up quickly

So here is the fan speed controller

It just uses the same principles and parts I have used on the amplifiers
A voltage divider Potentiometer and 2 thermistors give a voltage output to the gate of the mosfet as this voltage go up due to an increase in temperature the mosfet start conducting more and the fan will run faster cooling things down.
By adjusting the pot there is a point at which things reach an equilibrium and the temperature remain constant

The rig is just a cooper plate bolted to a computer CPU fan, those fans are designed to get rid of about 100 W of heat
At first look a lot of effort for such a simple thing,:D there is a bit more to it later on ;)

2457824579245802458124582

Bksabath

09-11-2018, 09:01

To see how this work I need a heater to warm up the cooper plate

Same all same all

I am using a mosfet whit a 10 R resistor to warm up the plate
10 V drop across the resistor give me 1 A the supply voltage is 30 V so I am draining 30W
I can get more by increasing the voltage between the gate and the source (VGS)
For now I need to prove that the fan circuit is working and that it is keeping a constant temperature on the plate

24603

Next post may be usefull

Light Dependant Resistor

09-11-2018, 09:12

Looking good as a test rig. in actual use the heatsink needs to be cooled from
ideally outside air . Recalling the design of FM transmitters they often have the heatsink underneath
and a fan outside the case - the question is to be blowing hot air out or drawing cool air in - not sure about which
will need to do some research.

Bksabath

09-11-2018, 10:44

Plan is drawing cool air in from the bottom blowing out on the top

Yes it a test rig

On Martin amplifiers I have 4 mosfets in parallel, one may gogle up "Current Hoggin" and the effect of temperature on those dudas

http://www.firstwatt.com/pdf/art_matching.pdf

I can measure the VGS of the mosfets at a point where they start conducting at an arbitrary current whit say 5 volts across the resistor and a any other point say 10 volts across the resistor

The red meter may not be that good and the resistors may not be exactly 10 Homes so there is going to be an error there but as long as that error is constant...

The fan /heater arrangement guarantee that the temperature will be constant and the working conditions are similar to when in use
I can measure different mosfets and select the ones that are the best match to each other for VGS and maybe trans-conductance

Alternatively if you have children's you can sell them and buy a curve tracer :D

Said this The DCA pro is a nice piece of kit but Limited current wise

Light Dependant Resistor

09-11-2018, 14:03

What I would be doing is first measuring the resistance variation of not the mosfet being tested
but the resistor itself. You are relying on and so is Nelson, that the resistance is somehow fixed.
when involved a heated environment. The truth is resistance changes with temperature, that is, it is rising upward.
You are relying on the higher wattage of the resistor to compensate - which it does, but measuring what
resistance variation you are getting then is in terms of matching mosfets a error term, so needs to be allowed for.

Instead of R1 in the Pass schematic, insert a proper current source which if being varied has the
setting resistor close to the coolest position ie near the fan. A simple LM338 TO3 Vref/R current source
ie 1.25/R is going to be more accurate than a resistor I think. failing that, locate the R1 equivalent resistors
in a spot not involving temperature change.

Bksabath

09-11-2018, 14:18

Well the resistor is bolted to the same plate and so at the same temperature all the time I am running the test at 50 C so that ambient temperature will not make much of a difference and the plate is a 1/2 in thick piece off copper
If I time the various steps and do the same for each DUT the error that, I agree whit you is there, will be the same.
I may even repeat the test on each alternate Tuesday and Friday :D

Is it not the Mosfet my current source?

Can you post pictures or drawings please?

One is alwayse willing to learn

Light Dependant Resistor

09-11-2018, 15:19

You could measure resistance change at ambient to 50 degrees C - noting the heatsink is also transferring heat to the resistor
I have attached diagram for the N channel and P channel current source using a LM338 replacing resistor R1, however requiring
also a set resistance to attain the current. In this case 1.25/R see Figure 36 http://www.ti.com/lit/ds/symlink/lm338.pdf
The resistor to provide the regulated current needs to be positioned as close to the delivery point as possible - suggesting a
hole with said resistor underneath ( not heating itself ) and short wiring to come up to the positive measuring point.

If that looks too hard, just discern ie measure if the resistor is adding a error, and locate the resistor or resistors appropriately
so they remain at the same temperature and therefore no longer add error.

pc141263

10-11-2018, 12:35

got the amps and power supply home today. I purchased a record 40 years ago. quadrophenia. I have owned this album for 40 years and I have never heard it this well. the sound is amazing.

Bksabath

10-11-2018, 13:00

Nice one Paul
Really happy U like it
What other kit U use ? and pictures please, we like hi fi porn here and pages do no get sticky :D

One must remember that "I have never heard it this well" in this case is only possible to Nelson Pass
Not only those must be one of the simplest amplifier circuits around they sound pretty good and one is encouraged to build them and learn in the process.

Bksabath

10-11-2018, 13:48

Bksabath

10-11-2018, 14:22

Just to make one thing clear just in case

From day one when it was said "I would like a pair of amplifiers like yours" "and me 2" the deal was that I would help Paul and Martin to build the amplifiers ,following on what was said by Nelson Pass, this is purely a DIY venture

I said at the time that I will help, whit my only retribution being the pleasure on building them and maybe a bath room retile job :D, being DIY both Paul and Martin would have to do any part of the work that may be possible if they are capable to do or willing to learn, Martin has been learning how to solder for example, that they will be given home work and that they will be making the decisions on what components to use and understand how the components work and what they do.

Based on that I am posting here. It is a handy place to keep notes and maybe help others a little bit
I understand that it may be to simple for many members
rest assured that I know very well that this would not been possible on other forums where there would have been red biro marks all over the place and that I am grateful to Admin moderators and such for the opportunity to use this space

Bksabath

10-11-2018, 14:58

How I am testing Martin 2 quads

open the packets and mark them a to H
Put them on the hot plate to warm up a bit and starting whit A plug the in set VGS to get 10 V across the resistor and wait 5 minutes write down VGS reduce VGS to get 5 V and write it down reduce VGS to get 50mV write down and repeat for B
do the dishes while I wait 5 minutes for B to setle down

open a bottle of vino while I wait and then repeat for C
start celebration of my 41 anniversary of being 16 repeat for D
maybe put a post in while I wait for C and repeat

...

Bksabath

10-11-2018, 16:27

24635246362463724638
Martin F5

This is plan A

4 nice feet and the perforated plate so that fan can suck air from bottom and blow it up

1 flat piece of acrylic to hold the capacitors in place this will be whit top flush to bottom of sinks and have a hole for the FAN / FAN's and keep the soft start board ,under the trafo in the front, isolated from accidental contact it also stop air from top going back in to fan from top
Also at front the rectifiers and earth lift

side view whit Power boards

back view whit input board mounted on sinks and the speaker protection on the side ( the amplifiers will be a mirror image of the other so that back panel also acrylic will have this configuration : Speakers , Input , Power and viceversa

at the top another flat piece of acrylic whit a hole so that sinks go trough and mounted so the top is flush whit the sinks

another nice piece of acrylic in the front

We reserve the right not to tell you what the side panels are made off … no not Granite

How do U like the granite front on Pauls?

Inside also as many flashing coloured LED as possible

so we need 4 pieces of acrylic and 2 or sumthink else

Overall size for flat acrylic same as perforated plate size of front and back same as plate in width and as tall as needed

Light Dependant Resistor

13-11-2018, 12:09

Thanks Chris
LM338 quite impressive

Could you please complete the SCH whit the meter to measure the VGS and Current and what U use to change the VGS please

I am sure many, me included, would really appreciate this

Yes the resistor and it's temperature will add an error but as the temperature is constant and fairly hi compared to the ambient the error can be kept fairly constant

What I am trying to do is have and idea of where to use Mossfet A and B and see if it would be better to use mossfet C or D instead or if any combination (out of the possible 6 in this case, how many if one has 8 dudas instead?)would be beterer

1)Would you agree that what I am doing not can be done whit parts one already have (if you don't have a CPU cooler you can borrow one from your kid play station or such :D)?
2) that it would help complete novices in to understanding how a Mossfet works ?
3) that maybe it could be a bit more accurate when comparing 2 devices than what was posted on the " How to match devices " Pdf if one accept that there is a error but that as long as the test are carried out in the same way it will give an rough idea on how similar the working conditions of 2 mossfet could be and therefore help in choosing the best pair or quads out of a batch?

Hi Sandro

For a ideal power supply the current supplied should be independant of voltage.
We can see parallel 1 ohm resistors making half an ohm so with the later schematics
32 volt supply, VxR possibility of 16 amps current supplied in each.

However this is assuming Voltage and for that matter current via the resistors are stable
so very much relying on brute strength of the power supply to deliver that level of current.

With Mosfets we find their ability to have transfer of signal arriving at the gate in both the
source and the drain legs to be a major feature of their construction. It is one of the reasons
they are so good, vs bipolar transistors where there is phase change in the collector – but
equally signal is there too phased different in emitter same as base and collector 180 degrees
different. Factoring into this too is degeneration

What this tells us is that there is guaranteed to be modulation of audio
though- out the power supply – but only really needed in terms of the schematic in the output.

My efforts at showing use of LM338 represents a significant departure to assist the power supply
to deliver current independent of voltage. However the 338 is resricted to 5 amps output unless
arranged with a pass transistor, however 5 amps thankfully is its minimum and it will provide
average of 8 amps meaning one TO3 device per mosfet is eminently sensible. To see this measure
static resistance from input to output or input to adjustment of a 338 you should see many mega
ohms of separation. Now compare that to half an ohm separation from
the power supply in the F5 – see the difference, I add or digress this as soggy capacitance vs
semiconductor separation brilliance

So I just do not get Nelsons resistor approach to separate supply from audio load, using particularly
mosfets. It looks to me brave, but not at all real world practical.

The 338 is arranged then as a current regulatorworked out by dividing the reference voltage which is 1.25 volts
by a set resistor Its advantage is the device being tested see’s a fixed current independent of voltage. Voltage
can go up and down but current remains precisely fixed – just what is needed.

To change its current delivery just increase or reduce the setting resistance value.

So for say just over 5 amps, which is to me a sensible figure the resistor would be 0.22 ohms

But you need to ask yourself why you are testing or matching mosfets
it appears there is slightly better reliability with them matched like this

The better answer is to supply each mosfet with a LM338 on a heatsink arranged as a current regulator.
The advantage each mosfet receives precise current vs vague changes with power supply going up and down with
half ohm resistors then wavering around as well. Most of your need for precise matching disappears,
as each device is precisely controlled with its maximum current delivery prearranged. but by all means do the matching as well

I think you will find the TO3 LM338’s arranged as current regulators on appropriate heatsinks, I think exactly what is needed for the F5.

Bksabath

13-11-2018, 18:23

Must say I am a bit lost will take a while to digest this … but thanks Chris

Back to Martin F5
Brute force PSU indeed each mono-block 3 X 47000 uF in C R CC or probably C R C L C on each rail whit a 600 VA 24 V transformer give us 35 V rails possible voltage swing of around 64 V before clipping

Martin configuration is whit 4 IRF240 and 4 IRF9240 in parallel
the transistors have been purchased from another forum shop at 2 different times, they supply kits of transistors good for the 50 W as in Paul amps so matched pairs, and that is why I am going to try and make matching the best possible

here is what I have for the IRF 9240
test at 50C 1 A IDS (current flowing from drain to source)

VGS (voltage between gate and source that cause that current to flow)
A= 4360 B=4356 C=4280 D=4286 E=4210 F=4307 G=4275 H=4284

the same patern repeat for 500mA and 50mA
I can get really nice pairs out of this but not quads
worst case would be using A and E on the same channel
say I bias A at 700 mA or 350mV across the resistors then the bias for E would be 862 mA or 430 mV and this is far to close to the point where the diodes start conducting bypassing the resistors so it would be quite possible that all the current will flow on just one mossfet (current hoghin ) and shortly after this will blow up

On Paul's whit pairs as matched it was not a problem he said he likes them to be reliable so we kept the BIAS low (but I was running them over 250W when I had that small fire)

Whit Martins is going to be a different story, I know that as soon as he got them he is going to start to fiddle whit them :D

Thing is now we know that a potential problem is there Martin will have to decide 1) get same more transistors and match them
2)get rid of the Diodes

1 is easily solved IRF9240 and IRF240 are readily available I just need to place the hi current on the PCB whit the test point so that the bias will be lower for the others and keep it on the 700 mA mark =350mV

All said I am using fqp19N20c (TO 220) and fqa12p20 on my F5 and bias is set at 1.3A they have been working like that for 8 / 9 years

the original design was calling for FQA19N20C FQA12P20 those are obsolete, and unobtanium … ARE THEY ? ;)

Bksabath

13-11-2018, 18:44

2466224663

those 2 pictures of one of my F5 mono blocks
2 FQP19N20C and 2 FQA12P20
inside there 50mF on each rail and home brew speaker protection the common power supply in a separate box has a 1000 VA 24 traffo and a further 100mF rail in CLC where the L is a 1kg of 2mm copper wire X2

the only reason I can pull so much bias 1.3 A on each transistor is easily explained by the size of the heat sink and that copper heat spreader

heat spreader is a story by itself at 1/2 in thick It toke me a week to surface lap them whit the sinks
when that was done the spreader and the sink stuck due to the vacuum formed between them for good measure I have used a special graphite paste to improve on heat transfer
Well it must have been very special at £35 gram

The transistors are clamped whit captive bolts so as not to strip the treads

Bksabath

13-11-2018, 18:45

PS one may notice the white lighter on top of the sinks

Light Dependant Resistor

13-11-2018, 21:03

Hi Sandro
LM338 is of course a positive device, my suggestions at post above to replace the fixed 1 ohm /2 resistors is fine for the positive rail
with LM338 doing precise current regulation, The negative rail though needs some additional thought. It is possible to also use a 338
in a negative rail for voltage regulation, I will need to test to see if it can be configured for current regulation, but suggests instead a 337
with a pass transistor or mosfet on each to get the target 5 amps of current.

Bksabath

16-11-2018, 13:53

Martin I got your "transistors" matched
at 1A 50C

Left 4340 / 4341 /4342/4344 and 4339/4345/4348/4348

Right 4380/4374/4370/4365 and 4356 /4355/4354/4359

of the group buy I am using 1 out of the 8 plenty of good pairs there but no quads or poor match between the 9240 and the 240

I am going to call it a day I need same beverage after that

Hifidog

17-11-2018, 14:31

Nice, I know how hard it has been getting the matching quads, patience of a saint!

It is my turn to be patient now. 4kg of enameled copper to wind for the inductors and a few meters of copper bar to polish, clean and nickel plate.

The parts for the case are ordered so its all go from now on.

Bksabath

28-11-2018, 12:37

Simon sez :D I love winding coils on another tread
24767
4 Kg off 2.62 cooper to be turned in to chokes for the L leg on the CRCLC on the power supply
haw fun was that ?

On another front just got a delivery of 100 FQA19N20C

I already have the other ones but had to use the TO220 version of those on my F5
That is going to get a major overahaul after 8/9 years of service

24768

No need to say I will have same spare if you are interested on same quads send me a pm they will be matched whit the rig on the previous post first came first served limited stock

I tell you now they are not cheap but sound beterer than the IRF counterparts

Bksabath

02-12-2018, 11:29

Time to clean up

24809

and put things togheter

2480724808

Bksabath

11-12-2018, 19:47

2485624857

Got the inductors today

I Kg each of 2.6mm copper

sq225917

11-12-2018, 20:09

I could have done you nicer ones....

Bksabath

12-12-2018, 19:33

I could have done you nicer ones....
Tanks for the kind offer

I let you know when to expect the 4 Kg Cooper delivery

It toke 4 to 5 hours for Martin and me to wind them up

The aim was to get a similar size to the caps which we did and we learned 99 wrong ways to do a thing

The potting from what I heard was worth of an episode of Mothers do have them :D

sq225917

12-12-2018, 19:43

A big drill and a wooden bobbin.

Hifidog

12-12-2018, 21:44

I thought using a drill would work until the courier delivered the copper. Not an easy job by an means, we had a nice rig but still was very difficult to wind. They are tight though and at around 25 meters each and 1kg each I think we done well. The potting was a nightmare and I learnt how not to do it. It took 1kg of epoxy but final product will function well even though some polish wouldn't go amiss.

Hifidog

12-12-2018, 22:28

My hands are still aching;)

Bksabath

12-01-2019, 12:45

25056

Teaser :D

Tread has been dormant for a while most of the case is done and soon they will be up and running
Martin and I decided not to post till later

sq225917

12-01-2019, 16:13

I said big drill, probably a hand crank shaft of correct size would be better though. What value do the coils need to be?

Bksabath

12-01-2019, 17:31

25059

Wire was 2.62 od mm that gave us a resistance of 0.047 for a 22 meter length or a 1 Kg I let you do the math and get CSA

It was wound on a 15 mm od former 100mm tall and end diameter was about 55 od

we chose the cable size so to give us the right resistance for the 1 Kg weight and one never find inductance meter when needed

In the picture the gigamagic (device that let you do that) the Dudas (nuts) are m12 so 19mm spanner, the all thing was held on the vice one hand turning the spanner the other guiding the copper third hand to keep tension and fourth hand to unwind the coil

Toke about 6 hours to get the coil wound up

As mentioned before potting was worth an episode of Some mothers do ave them

Bksabath

12-01-2019, 17:35

Yes drill was suggested but I decided to go sveth and thears way rather than risk loosing a limb

Hifidog

12-01-2019, 18:58

Nice! Any excuse for selfie for aye��

Bksabath

12-01-2019, 19:57

Nice! Any excuse for selfie for aye��
Nah that would cause image to distort and text to look blured and Crappy
In reality it is / will be completley beterer ;)

sq225917

12-01-2019, 21:38

A slow lathe is your friend, thick leather gloves, would have been all over in a minute or two.

Bksabath

16-01-2019, 15:25

Bud idea to wear glowes on a late
Only way to do it is by hand IMO

especialy whit a 5.3 mm SQ ,Murphy Law apply (Shit happens and the Universe is out to get you) it is much faster by hand than a visit to A&E whit bodily part missing

Still I wish I had a Late

sq225917

16-01-2019, 18:19

Foot on kill switch, 30rpm, what could possibly go wrong...

walpurgis

16-01-2019, 18:42

Foot on kill switch, 30rpm, what could possibly go wrong...

Yeah. Wear a tie too, why not? :)

Bksabath

19-01-2019, 20:51

Only thing that one should wear when working on a late is underpants and or Brasserie all rest is optional extra

Note Underpants or Brasserie will exclude one or the other unless you are like one Fitter I use to know that went on holiday to Brasill and came back whit a pair of bubbies
Do we need to be politically correct one the Brexit is over ?

Any way Martin and I decided not to post more about the BOX
I have been working on the finishing touches

This is the temperature controller

250952509625097
Temperature controller
Prerequisite:
The amplifier must reach working temperature quickly
When the temperature is reached 2 cooling fans turn on andkeep the temperature at a set value
Component used are when possible are the same as used troughout on other part of the amplifier
Bling whenever possible 😊
How it works
From the positive supply rail, a Mossfet switches on and supplya variable voltage to the fans
The speed of the fans is controlled by 2 thermistors placedon the heat sink
As the temperature increase the thermistor resistance dropthose are in series whit a variable resistor VR 1 that set a voltage that throughthe gate resistor eventually turn on the Mossfet as the voltage increase
Around 3 volts are required by the Mossfet to startconducting
As the Mossfet output voltage is low the 2 fans’1s ,2 X 12volts units will stall also as the amplifier must reach a working temperature quicklythe supply to the Mossfet is switched on and off
The operation of the relay will also switch on 2 Led oneindicating low temperature the other cooking temperature
The relay part of the circuit is the same as the one used onthe speaker protection board
The point at which the relay turns on is set by Vr 2
A third Led switches on if the temperature ever goes to hiand is set by Vr3
Led 4 is the Speaker protection Led and is on this sameboard
The board control circuit is supply by a 7812-voltageregulator as 12 volts are required by the Vu meter boards that I will notdiscuss at present
Q1 Q3 KSC945 Q2 MPSA06
R1 1K R2 100K
Vr1 100 K thermistors 4.7K
RL 1 12 V relay
Vr2 Vr3 and all other to be assessed

Bksabath

19-01-2019, 20:55

Sorry for crappy photos + copy and paste make same words stick togheterer :)

Comments especialy as resistor values are most velcome as I am building a couple of protos (see circuit boards) to start whit

Bksabath

10-02-2019, 15:46

2528125282

Fan controller now completed
I had to change the previous design as the relay was chattering a bit

Bksabath

14-02-2019, 16:05

25308

Another view of the fan controller board (Silver plated of course)

25309

One off the 4 VU meter Bling

25310

All drilling is done now is time for assembly

2531125312

Bksabath

25-03-2019, 12:15

25640

Sorrrrrrrrrrrrrrrrrrrry could not resist :D

Bksabath

17-04-2019, 11:30

25814

Work in progress

Bksabath

14-05-2019, 04:44

First listen happen last week
26052

Left channel finished there is same minor work to do left

26053

right channel is going back on bench today to get boxed in
26054

Hifidog

31-05-2019, 19:52

Here some pictures of the amps in the final position/system.261952619626197261952619626197

Alex_UK

31-05-2019, 19:53

Wow, they look amazing!

Bksabath

01-06-2019, 13:04

They sound OK as well Yes?

where one go from here ? :scratch:
I bought 32 small capcitors yesterday:D and already have 100 or so each of the FQA and FQP

Hifidog

10-06-2019, 17:32

I have been enjoying the amplifiers for a couple of weeks now and to say I am pleased is a huge understatement. Usually when I compare components I would listen for the improvements and differences in presentation but with these it is not even comparable to what Im used to or even anything I have heard before. It is a case of 'what amplifiers'? they do not sound like Hifi in a traditional sense. The sound is unlike recorded playback but more like studio monitoring or a live performance in my listening room. The textures, dynamics, emotion and so much more is akin to being there with the performer, so in comparison to anything I have heard its more real and enjoyable than ever. In the realm of hifi I can understand what improvements there are in terms of control, dynamics, pace, texture and detail on all points I am completely satisfied with the abilities of these amplifiers. I really can not find any room for improvement other than from my accompanying components will now need to be fine tuned and a new cartridge has been installed with a new phono stage to complete the system at a level I feel I could never of achieved with conventional options available on the market and certainly within my budget. Some friends have also heard them and the response has been nothing short of WOW, they have shared my feelings, in the fact they just dont sound like hifi.

My system consists of DIY F5 V3 TURBO monoblocks, Wilson Benesch Discovery speakers, Pioneer PLC 590 turntable with Pioneer PA5000 tonearm and Supex 900 MC super cartridge, Musical Fidelity M6 Encore connect preamp and streamer, Trichord Dino mk3 with NCPSU, Puritan mains conditioner for source, James Audio mains conditioner for power amps, Audioquest NRG10 mains leads throughout, QED Signiture interconnects and QED Supremus speaker cables.

Bksabath

10-06-2019, 20:41

Well now one need to listen to that Supex
Not tried the Solid silver cable ?:D

thingfish

10-06-2019, 20:48

Sorry ..... But being a bit of a newbie.
And without wishing to be rude..........

WTF is this thread all about

Ta.

Bksabath

11-06-2019, 02:27

Pictures may explain
It is a record of things I am building keeping busy and out of troubles :D and making a few peoples happy

Bigman80

11-06-2019, 03:55

Sorry ..... But being a bit of a newbie.
And without wishing to be rude..........

WTF is this thread all about

Ta.Sandro and his friend have built DIY F5 V3 Turbo amplifiers.

And done quite a fantastic Job of it by all accounts.

Well done guys. Love the finished look.

Bksabath

11-06-2019, 13:30

Thanks Oliver
I hope one day you get a chance to listen to them

Martin one's exceed any exspetation we both had
I will have them back as we decided to do a bit more work on the Fan controllers
Must say when same xxxx start a post whit "no wishing to be rude" he only make me wish he stick to other treads maybe even other forums.
The 150 years old slate for the side's deserves same better pictures it is quite a nice material to work whit and it does take quite well thread inserts (V coils)
The acrylic on the other hand is a real pain to work whit

Next here there is going to be a secret project that I have on the back burner for a few years , say no more

A D.I.Y. cable ...
And I am hitching to build a pair of F6
I already have 704,000 uF capacitors for it

Bigman80

11-06-2019, 14:13

Bksabath

12-06-2019, 13:06

Next here there is going to be a secret project that I have on the back burner for a few years , say no more

A D.I.Y. cable ...

Sorry for quoting myself same progress on that front
The dude that sells Gekko cables had same surplus cable on sale on Ebay
Martin and I got 4 meters each at £12 meter

I put it on my system a few days ago whit same phono Plugs and have just got a text phone / phone call from Martin he finished his
You can scrap the QED Signature cables from Martin list of toy's and the 2 meters spotfire cable I have may came up for sale
I just have to wait to get same super duper KLE phono plugs to finish off mine

:D
Do not bother to look at Ebay the cable is not longer listed I got another 10 meters of it before Martin beat me to it
We have a cable guru on AOS and PM have been sent
One practically can scrap the Diy cables from my list of things to do

I was wondering the other day about this tread and the title
"The Bug is back" blipping right spot on and now I can confirm it is contagious...;)

One is going to loose a friend if he call to talk about is FiFI at 12 AM (PM for same)
I just finished a Night shift of a 72 hour week F.F.S. :D

Bksabath

27-11-2019, 11:16

27026

It has been a long time since my last post here
Things are going to pick up again

Teaser enclosed

tapid

27-11-2019, 11:40

27026

It has been a long time since my last post here
Things are going to pick up again

Teaser enclosed Good

Lawrence001

27-11-2019, 12:02

27026

It has been a long time since my last post here
Things are going to pick up again

Teaser enclosedOoh that's the big, rare variant of the Howard isn't it? Always wondered how they sounded.

Sent from my POT-LX1 using Tapatalk

Bksabath

27-11-2019, 13:19

It,s a bit of a long story …
Hey that is why I am posting here :doh:

Got them yesterday well actually Martin did

Drop them of at my place and 5 seconds later I wanted them
We put on a bit of Charlie Byrd (CCS 8002 ) either cek the number or … the 45 RPM DDR
up and down up and down up and down .... "any reference to the clock work orrange is purely intentional"
and i fell in love

I got a pair of maggies for F. S. (and the paradise) but I have never heard that feeling so good

Things fall apart when listening to Dead can dance The wind that shok the barley (MOFI 2-001)

There is a broken woofer and this is the perfect tune to test for that

At present Martin is in the back bedroom (Toni bedroom) working on it
27027
Enclosed a picture of Toni

Bksabath

27-11-2019, 13:30

PS as posted on same where helse

If you have another pair of Castle speakers laiing around I want them

Bksabath

01-12-2019, 13:53

was re reading the previous post
It was meant to say
While plaiyng the Bird Drums track "one can ear the stick hitting the drum and then the drum skin moving up and down bla bla bla
I have posted same videos on youtube
Quality is prety low but I am sure I get the idea
Winchester doing Byrd
https://www.youtube.com/watch?v=i2p-L2R9jnE

Broken woofer on the RH hand speaker
https://youtu.be/1fbjg8967G0

Fixed woofer
https://www.youtube.com/watch?v=RSGEncbbq8k

Bksabath

01-12-2019, 14:08

May take a while for the movies to show You tube need to decide if you little children may get exited

Any way the plan is to do a Martin on the remaining woofer's

While listening the only criticism was from both of us that the tweeters are a bit splashy, metal domes one of which is crushed, so we are waiting for same soft dome to arrive

Bksabath

13-01-2020, 12:20

Time to start a new project :lol:

It is going to be a big one

27201

3 box F5 (may not be a turbo)

27202

at present 32 X 22mF 40 V PEH200
27204
and 4 heat sinks

On those, I managed to get 100 FQA19N20 and already had the FQA12P20, I will fit 4 output devices, as said before it may not be a turbo, as the diodes on the turbo version limit the Bias to around 700 mV and I may be able to push more than 240W dissipation on those heat sinks

Bksabath

22-01-2020, 15:30

First lay out
Buss bar for the ground 38 X 380 X 5
and the 4 buss bar's for the rail voltage's

27259

The chokes will be 1Kg of 2.62 mm Od cooper about 5.4 CSA

27260

Rough size for now 300 mm wide 400 long 220 hi

27261

It is a bit tight but it all depends on the size of the Chokes I may try doing common mode chokes
positive and negative rail's ,2 wires, wound together around the same core
From past experience it is going to be a 2 man job :rolleyes:

27262

Hifidog

22-01-2020, 16:01

Nice! Looks like it's going to be a big boy.

Made in 1968

22-01-2020, 16:20

Will the tin of Tomatoes add a certain flavour to the sound?

:D

Bksabath

22-01-2020, 17:12

Will the tin of Tomatoes add a certain flavour to the sound?

:D

The cleverest thing I can say, to stay on par, is that I run out of Can's of IPA

JohnJo

22-01-2020, 17:50

Looks very interesting Sandro, I’ll be watching this thread with interest :)

Bksabath

23-01-2020, 12:51

27267

First draft of the Power boards
This is the N channel
Only difference whit the P channel is the diode connection at the bottom

I may not use the turbo depending on how much Bias/power dissipation the heat sink can take
Bias whit the diodes is limited at around 350 mA across the 1 homes resistors. If the turbos (diodes) are not used it could go a fair bit up
On my old F5 is set at about 1.3 A

The difference between the turbo and the "normal" is More power out at hi volume or more class A power at low volumes from what I do understand

Bksabath

03-02-2020, 11:12

There is going to be same delay on getting the boards made in China

Bksabath

29-02-2020, 08:11

27424test rig

27426 fan controller circuit

I have 100 each of the Fairchild FQA19N20 and FQA12P20as in the 2012 F5 Turbo article those are both obsolete components and duringthe year’s alternatives have been IrF
240 and 9240
http://www.firstwatt.com/pdf/art_f5_turbo.pdf (http://www.firstwatt.com/pdf/art_f5_turbo.pdf)
on my now 12 years old F5 I have used the TO220 versionFQP19N20 as I could not get hold of the TO3P version
To make things simple and keep the engine analogy thePositive and negative Mossfet work as the pistons on a boxer engine one pullwhile the other push around a central point, 0 volts, the class A operation ofthe amplifier, which gives no cross over distortion, is limited by the Biascurrent.
How much current depends on how good the interfacebetween Mossfet and heat sink is and how much heat the heat sink can sink.
One can get rid of 120 W of heat on a 300X170X80 heatsink whit same help by fan cooling
Or about 770 mA bias on 4 Mossfet
I do not know at present how much I can get on the new400x220x83 sinks I got, I hope not to use fans on this project, but the morebias the better.
I am using the best interface I can get kerafoll red as Iam finding quite difficult to get same CVD diamond pads ��
The bias circuit is quite simple more about it on thearticle

www.firstwatt.com/pdf/art_mos_test.pdf ·
The originalF5 was rated at 25 W whit 2 devices I am going to use 8, 4 FQA12P20 and 4FQA19N20 I may not use the Turbo diodes but space for those is on the boardsthat will not be made for same time now

The use of 4devices on each supply rail calls for matching, lacking the founds for a curvetracer, I am going to use my Atlas DCA Pro for a first sorting and then my hotplate.

Each Mossfetwill be tested for Voltage between source and gate at various points, as the parameterschange whit the temperature of the devices the testing will be done as much aspractical at a constant temperature of around 50C
3 repetitionof the same circuit are in use
A) Plate heater is a Mossfet used as heat sourcefor the copper plate, 1 A current at 35 Volts = 35 Watts
B) Fan controller again a Mossfet used to drive aCPU cooling fan the speed of which is adjusted whit a thermistor, the CPUcoolers are rated at 150W, so that the temperature is as constant as possible and does not change whit the total current dissipated
C) DUT circuit The VGS of the Mossfet under testwill be measured at different currents allowing time for the plate temperatureto remain at 50 C
If one is remembering Homes Lawit is quite easy to measure current whit pretty good precision by measuring thevolt drop across a known resistor 1 V / 1 Homes = 1 A

I will record for each of the 200or so Mossfet the VGS at the point at which it starts conducting and then 100300 350 400 600 800 1000 mA, any small error caused by EG the tolerance of theresistor will be common to all devices the resultant component matching shouldbe a few orders of magnitude better than what is in the article
I am popping out for a while Imay be same time

Bksabath

11-05-2020, 11:14

I have been working on the speaker protection
the boards on Martin amplifier's use standard relays , First year birthday BTW

There is much better on the market. The Amplimo LR-24

they have 2 set of contacts one designed for hi power and one for low level signal that open and close whit a tiny delay I have been using those on MYF5 for the past 12 years or so the design has changed over the years but Amplimo told me that the specks are exactly the same

27721 First draft of the boards

27722

The diagram is the classic for the UPC 1237

I have added a Led Blinking circuit, as in my experience a blinking Led draw much more attention than a steady one

May not be very elegant, my knowledge of electronics is not that god
It is nothing more than an astable multy vibrator giga magic whit 2 PNP BC 556 and a BC 337 to break the charge / discharge cycle of the 10uF capacitors
the control voltage for this is taken from pin 6 of the UPC 1237

When the relays are open there is full voltage on this pin the BC 337 is closed Led is on off on off, as pin 6 sinks the relays the voltage go down and the BC 337 stay open charge discharge of the capacitors stop and the LED is steady on

27723

So far is all working on the bread board

Bksabath

11-05-2020, 13:16

27724

Space for the Speakers terminal's and board size

this work well whit a stereo build The UPC 1237 can sink 80mA at pin 6.
a 10 v drop across R1 (10/330) =sink 30 mA whit supply rail at 32 voltage across the coils 22v

the speaker terminal's are spaced at 60mm

Bksabath

11-05-2020, 13:55

27730

spades or bare wire 45

the boards fitted directly to the speakers terminals
M6 holes

Bksabath

11-05-2020, 15:41

Just had most interesting conversation about connecting 2 pairs of speakers on the same channel

Dude : hi I got bla bla bla 16 channels AV amp bla bla bla

Ok grab red speaker 1 terminal and put on the red amp then spk1 black to spk2 red and spk2 black goes to amplifier black

Dude: ok simple enough

yes 5 minute job really but then again … you got manweb phone number handy

Dude : :scratch:

after all you can do like last time Just tell them that the light went off it would not be the first time you blow up a 400A fuse at the substation by mistake

Bksabath

12-05-2020, 14:46

27743
Home brew boards

Yes Ebay ones look pretty and cost less
But can you do this ?

277402774127742

Bksabath

13-05-2020, 08:46

27745

Ready for gerbers

Bksabath

14-06-2020, 13:57

27892 Just started whit the box construction it is 3 mm Aluminium plate 4 pieces 300 X 400 for the monoblock V 3

It does not look like much and has a few scratches straight out of the bag once drilled it is going to be electrical etched and anodised

And the chokes between the capacitors banks nothing more than a 0.076 Homes resistor after all
27893

The water clear resin is curing pretty well may even put a few LED in those :D

Bksabath

18-06-2020, 11:09

279502795127952

Case construction
Sinks 220 X 400 X 86
Plate 300X400 X 3
Angle 1/2 in 25 mm may be better

Bksabath

20-06-2020, 12:35

2796327964

The back panels

Just put the plugs the other way around for left and right channels

Bksabath

07-07-2020, 10:13

28071
Building the enclosures
this is a 3 box build
Left and right amplifiers and separate enclosure for the transformers
28072
The speaker protection is mounted directly on the speaker terminal
28073
Back panel whit same pretty nice silver plated CHK post
28074
They are going to be bigish as the sinks are 400 X 220 X 86
28075
The boards are designed for a stereo set and designed around the Amplimo speakers relay

Bksabath

24-07-2020, 12:34

Just had the aluminium panel anodised
28189
28190

2819128192

And the front granite tiles cut and polished
28193

Those just need the holes for fixing and for the speaker protection LED

Close up I am really pleased whit how it is turning out and must mention Aleo Plating in Liverpool for the fantastic anodising job

Hifidog

24-07-2020, 19:31

a thing of beauty! They look even better in the flesh.

Bksabath

25-07-2020, 14:31

I still think that yours are really dobby Horroshow (what not seen clockwork orange ?)

2819828199

I have been making up the capacitors banks
I am going for only 180,000 uF on each rail
2820028201
Ground buss bar cut polished and silver plated

viddy that turn up pretty horroshow as well

28202

Hifidog

27-07-2020, 18:33

the plating looks perfect. Is it silver?

Bksabath

28-07-2020, 21:50

yes spezial thing

Can you lend me your silver bars maybe we could try to cooper plate them as copper plated silver should not suffer from skin effect

Need to ask our mate that used to design particle accelerators and partner to pop around for a listen

Bksabath

09-08-2020, 13:15

2830028301

track to listen to : Time is the revelator

Do not waste time whit the silver plating

I have read the instructions 12 time'S thanks for asking
Tiffo Silver star is a waste of time and money

Bksabath

09-08-2020, 13:25

28302
So I built a Little power supply for one saying WTF ? yes Brunel be proud of it is only a 1000 VA traffo
28303
Taking things slow at the moment 8V 150 mA

28304

To make my own Nickel acetate

That is a 6 litre batch been running for last 20 minutes

I am going to let it run for a couple of hours and then see if it turns green

For ones thinking WTF gogle up DIY Nickel Plating

Bksabath

09-08-2020, 13:29

283052830628307

Missing pictures ?

Bksabath

10-08-2020, 10:25

28313

Its working pretty well

Nickel Acetate is a pretty green
Plated an old piece of scrap and is really good polish really well the plating is quite thick

I am going to live it for a few days and see

Tiffo Silver was turning yellow after a week or so and polishing would return it to the bare cooper

Bksabath

18-08-2020, 17:55

Nickel Plating sorted

28374

This the part of the power supply inside the mono block's
283752837628377

The transformers DC filter and a few more capacitors will be on Box 3

I do not need to build that at present as one for my old F5 is working

28378

You may have noticed the heat sink at the back

The input / Bias board will be mounted on it

I have 2 spare from last build and sink is way oversized but may make my own in future whit a few mods

Bksabath

03-09-2020, 05:35

Wow over 17K views on this thread

28459

Last pieces I need for this project ,apart from PSU that I have already

the power boards have been ordered and are on the way
I kept the "Turbo" diodes so that one can make 50W on a 300 mm heat sink or the 100W on the 400mm
I may not fit the diodes on this build as I am not sure what bias will give me 50 C on the sinks, the more bias the more class A, and the diodes limit this as they bypass the resistors at around 330 mA
BTW. I am running 1.2 A on the old ones

Bksabath

03-09-2020, 17:49

28464

I made same progress on re drawing the Input board but is to early to show the new design
Fitted to the sink is a spare board from a different forum

I am not quite so happy about it because of track length and spacing but is good enough for yours :lol:

Time will tell

2846528466

And this is the part of the power supply that is fitted inside each of the mono blocks
the only thing missing are the by pass capacitors

Bksabath

09-09-2020, 12:38

28511I have same caps that have been sitting in the loft for 10 years or such
28512
28513
this is a 10 year old solder joint
don't skimp on this stuff 3.6 % silver is well worth the Dosh

28514

I am using Rega and Qued speaker cable for the wiring

WTF People ask yourself "Is my wiring looks pretty"

if your answer is no start again and ditch the red blue and what not single cores

and silicon rings around the caps

Bksabath

16-09-2020, 12:34

2858628587
2858828589

The Power boards have arrived few touch ups before I order same more

They look Bumpy :)

That is 10 U (ounces of copper) or to put it on other words 5 times the normal decent quality boards

The spacing for the mossfets is 100 mm

Right thermal load for 50 C at the right bias for a quad on 400 mm sinks or a pair on a 300 mm sink

Bksabath

16-09-2020, 12:39

PS if any one is interested send me a PM

Bksabath

16-09-2020, 12:43

28590

Pictures are not that good
I am lacking the skill of many in this forum …

Bksabath

16-09-2020, 13:40

285912859228593

Time to finish the front panels

Bksabath

19-09-2020, 15:50

28617

I left it for last the last bit of drilling for the transistor mounting

28618

I am using 2 M5 bolts the kerafol Keraterm 86/82 I am using is good for 60 Nm and has a thermal conductivity of 6.5 W/mK

Same time ago I was looking for CVD Diamond and at the time it was not made for this application now is available under the trade name Diafilm TM220 whit a conductivity of 2200 W mK (is it real)

Either I win the lottery or they decide that I had a good idea after all

But any way compare a M5 bolt to the M3 screw and think about it

28619
28620

That is where the thermistor go right against the squarish tab camming out of the moss fet body there will be same sort of heat shrink around both to maintain contact

That is the hottest spot nearest to where all the switching happens

I seen them all over the place from open air to near the penny washers normally used to fit the dudas

Bksabath

24-09-2020, 11:42

Nearly 18 K view's on this tread
Constructive comments are welcome

+ If any one is interested I can arrange for a group buy for the PCB's and any bits and pieces

2863028631

One off the 3 box set up :D

They are big

Bksabath

12-10-2020, 12:18

well over 18K views on this tread
I feel that I am letting you down :)
Thing is I got a new toy today after waiting for over a month for it to be delivered

I am Waiting for HiFI Dog to pop around and lend same moral support as I am so scared to mess it up when fitting it to the V
Will let you know how it sounds later 1.1 mV output on the Paradise ???
And if not the paradise will it overload the XP10?
But then the F5 need same serious output to be driven 2870728708

Later …

Bksabath

10-11-2020, 13:54

28825
Just the last hurdle
I am matching up 300 Mosfets
I may be same time :)

But I am going to take the rest of the day of
I got a bottle of amarone and a new toy for my birthday

28826

maximax

17-11-2020, 23:21

Hi, if you don’t mind me asking - those heat sinks with the wave profile:
https://uploads.tapatalk-cdn.com/20201117/06668f685d126f574b895e0a6f5b8bc8.jpg
Where did you get them from? Price? Thanks!

Bksabath

20-11-2020, 07:28

maximax welcome to AOF
It all depends on what and how you using them for
send me a PM whit more info
In the mean time

http://www.abl-heatsinks.co.uk/index.php?page=extrudedproduct&product=204

https://www.gdrectifiers.co.uk/uploads/PS285_PS385_PS485_PS585_PS685_PS835.pdf

Here is a link or 2
I got the ones from link 2 on this build
In regard to price all I can say is quite a lot
I have not tried them so far but I can say based on experience that the 170 X 300 is good for twin (2 + 2) mono blocks at 770 mA Bias or quads (4+4) Monos at 770 mA but whit cooling (link 1) £35.26 each

One thing I can say for sure is that I can not see how the specified parts (on another forum) can possibly work but apparently they do maybe it is all due to the TO3 being fitted whit a single M3 screw and washer :lol: :lol: :lol:

My old F5 whit over the top sinks and cooper diffusers (400 x 300 single sink 4 mosfets biased at 1200 mA each) run pretty cool till one try a mate Polk cobra speaker cable they get pretty hot on a 3 / 4 hour session but sound so good
If you get a chance the Cobra is the best speaker cable I have ever tried but it may kill your amplifiers

maximax

29-11-2020, 22:19

Belatedly many thanks for the detailed response. I was aiming to build the sit-3x and a passively cooled Hqplayer PC. However, I have a feeling that work will sidetrack me next year. Anyhow, the development of the sit-3x appears to be dragging on, so no rush.

Those folks even sell a 10cm deep heat sink:
https://www.bal-group.com/heatsink_detail/240
No idea about prices.

Bksabath

30-11-2020, 07:25

28951

Just matched 200 FQA1920C and got lucky whit this batch any of the rows is matched at 500 mA at 50 C
The I run a test whit the DCA Pro DCA75 and pick the 4 s' I need

I still have to do a couple more runs on the FQA12P20 batch

I have same spares if any one is interested :D

I decided to get the trafos as well before brexit just in case I had to pay import duty on those

What no one told you that you will have to pay taxes when you buy stuff from the EU during the referendum campaign?

I got a couple of 800 VA 28+28 V toroidy supreme on the way the DC voltage on the rails should be around 38 V say 28 V RMS = 200W on 4 Homs
The 800 VA will not flinch at running 14 A

Sit X3 love to ear it when is done
Yes was looking at that sink as well If i remember is not wavy surface area is what matter are you going to force cool it whit fans ?

majex45

30-11-2020, 18:34

maximax

30-11-2020, 20:59

No forced cooling for me. So are you saying that wavy profile is preferable for completely passive cooling?

Re eu taxes, yes you will probably have to pay import duties ie vat unless there is a deal to that effect, but you won’t have to pay vat at source. The standard vat rate is 23% in Poland.

Bksabath

01-12-2020, 19:20

I was somewhat involved with CVD heat spreaders back in the day. I once held a diamond about 3 inches in diameter by about an eigth inch thick. It felt cold as it sucked the heat from my fingers but was slightly grey and not completely transparent. We used them to spread the heat from high power solid state lasers. Ah, happy days.

I am really interested to try them I have been on to Element 6 way before they started making them for the purpose ( not on TO3 size that would be a special order) , they stole my idea :D
They do make a few different types and performance is fantastic not only in regard to the temperature but also mechanical the Kerafol 82/83 which as far as I know is the best interface only takes 60 Nm of pressure

As I have no alphabet after my name they were not really helpful so I have to vin the lottery so that I can make the machine to make my own

Bksabath

01-12-2020, 19:30

No forced cooling for me. So are you saying that wavy profile is preferable for completely passive cooling?

Cool.... seen a few pictures of the sit 3 both whit fans at bottom and fans on top, fan sucking is bad idea IMO as it is going to get hot blowing is much cooler

If you where to use the fans I have a simple circuit to share as the one on Hifidog amps

to exchange heat the surface area of the sink is probably the most important if not a really thick base plate

I am not sure which ( the 83 or the 100 ) is betterer

Bksabath

01-12-2020, 19:39

Re eu taxes, yes you will probably have to pay import duties ie vat unless there is a deal to that effect, but you won’t have to pay vat at source. The standard vat rate is 23% in Poland.

+ the charge that the shipping company charge to make the payment not counting the delays while your thing is held at custom (4 weeks ) + (2 weeks ) for the letter whit the 17 digits number to arrive
+ you must be polite at all time because due to the current situation they can send same one over to beat the crap out of you

But not one told you this when you voted at the referendum

maximax

01-12-2020, 21:52

We should stay off politics here, but since that door is currently open, may I casually remark that it sounds very much like you voted for the wrong cause in 2016, no excuses. The money behind leave just wanted to get rid of eu style consumer protection, so more £££ for them.

Anyhow, amplifiers... the sit-3x project is not looking good, the guy behind it appears to be shifting towards a non-sit implementation because his two channels measure and sound different and he can’t make sense of it I believe. No criticism, he did rather well so far, and the 2sk182 is maybe not the best known quality. Maybe it’ll be the singing bush amp for me then. But no rush.

Bksabath

02-12-2020, 19:05

Not being politic at all They are all the same just by wanting to be voted for place them in the wrong place and wrong job

Well maybe Papa will publish his F7

It all boils down to politics even me telling you build an F5 in the mean time :lol::lol::lol:

maximax

02-12-2020, 23:56

Pass has already said that he will issue F7 kits one the Sony vfet stuff is out of the way. Output transistors are not publicly available I think but he has a large supply, unlike the vfets.

Bksabath

25-12-2020, 17:50

29060

Just got the transformers and plan has changed a bit
I got 28 V 800 VA units
Rail voltage is going to be around 39 V
There is going to be a fair bit of work to do

I had a couple of hours to spare

29061

so I made 2 RC filters in series 10Kr 300 nF and 10KR 500 nF
and whit bits from the loft and a roll of Duck tape :)

29062

I made myself a sub woofer
Quite pleased whit it so far. is it going to be the next project?

I have read that it is difficult to pair the maggies to a sub woofer

Advice is most welcome in this regards

Bksabath

29-12-2020, 09:23

29071

I am working on the soft start board

whit 2 800 Va transformers the surge current at power up is far to hi to safely rate the fuses

I am going to use 2 relays for each transformer the first one power the transformer trough a thermistor the second relay to bypass the thermistor

Four delay on timers will take care of the sequence

It has it's own supply from the mains whit a small block transformer

M6NTL

30-12-2020, 12:30

Bksabath

31-12-2020, 10:31

Thanks Mark

I do not like the “blue” PCB version
The Zenner supply is quite …all that said I have used the blue boards on 2 previous build and they are adequate
Besides any PSU whit 300 VA or more need a soft start
I have 2 X 800 VA traffos
29073
I am going to start whit same simple circuit for the on off signals
The blue boards use 12 V relays that do need 2 times as much current than the 24 V
The ones I am using are "solid" and have spare terminals for the Thermistors

I am looking at the ULN 2003 supply can go up to 50 V and is an array of 7 Darlington’s Each output is capable of 500 mA
Its das look quite interesting the 3 spare outputs can be used as on off and standby function
I will start a simpler version of the timing circuit that closes the relays in sequence
I need 150mS delay between RL1 and RL2 probably a longer time for RL3 to kick in and trigger RL4 after another 150Ms
The function of RL1 and RL3 is to power on the transformers
While RL 2 and 4 do the Thermistor by Pass
There is also the possibility to Include an ARDUINO Daughter board
This is going to be to far above my head but what about using the Arduino as speaker protection?

Bksabath

09-02-2021, 11:21

2919229192Nelson Pass on his favorite amps, playing with distortion, and his 21 inch subwoofer - YouTube (https://www.youtube.com/watch?v=-Prz6IpHlSg)

Well while I was standing by the door waiting for the postman to drop of bit and pieces for the F6 I had a one of those strokes of genius :)

The Magies do same things really well but they could do whit a tweeny litlle bit of bottom end

So I got 2 Beyma 15 in woofers and started searching for same 15 in Pipe

Unless you are over the pond or willing to pay for export and such I could not find any

Is it going to stop me?

I am making one out of glue and paper

Pipe Od 400 mm and 2440 mm long, which turn out to be the space between the Magies

The woofers are going to be face to face so Isobaric like and the pipe is going to be a transmission line whit 34Hz resonant frequency
PMC do transmission lines but I tend to walk straight out of the door when I ear them we shall see

Gogle them up !

Bksabath

09-02-2021, 11:23

PS F5 is still work in progress It will be done do not you worry about that

maximax

09-02-2021, 11:31

Will your glue paper pipe be strong enough and/or have usable resonant properties? If it doesn’t have to be perfectly round inside, mitred battens (via railway or table saw) could be glued together.
I couldn’t even begin to contemplate such a project, waf and space issues. The Ripol approach used by voxativ for their woofer assists is much more space friendly.

Bksabath

09-02-2021, 11:47

I hope so it is an open ended tube after all and being made whit scrap paper most of the money is spent on the PVA glue besides it is going to be at least 10mm thick

The woofers are 15ML100 not cheap but not to expensive either and one must have a couple of spare woofers yes?

I solved the waf problem 3 year ago , or did she solved the FiFi problem ?:scratch:

I am going to ang it of the ceiling behind the telly as I do not have the head room

Bksabath

14-03-2021, 02:41

Today I had beans

29293
29294

Sorry wrong forum
Today I made a circle

Bksabath

14-03-2021, 03:19

I nearly finished the Pipe'O tube

Ingredients

As before
A tube former to work on

2.4 meters by 390 internal diameter

29295

Decent PVA glue this can be diluted a loot and is ideal

29296

Due to the current situation no free news paper so I have used this for the first layers cut in to strips between 4 and 5 inches wide
follow the grain and when soaking it whit PVA water solution it will stretch more in one direction cut the strips so that as it dry and shrink making the all thing tight
brush over whit wet brush to get rid of air pockets overlap each strip about 50% and do not fuss much about lining them up
cover the tube whit first run of strips and then turn and cover the tube whit more strips and then turn and cover the tube whit more strips and then turn and cover the tube whit more strips and then turn and cover the tube whit more strips
till you run out of it letting the layers dry in between

29297

I tried a roll of this as well both papers work ok if I am going to do this ever again I stick whit the brown heavy paper
one may need 2 rolls of this

29298

to finish off cover the tube whit a couple of layers of inexpensive decorating cloth
When buying this from the hardware store grab a few rolls and then find a scale in the shop and pick the heavy one's as they do vary a lot and short ones will not cover the all of the tube
again lay one layer on stretch it as much as you can then "paint" over it whit the PVA solution

It does take a loads of time surface area is 2.4 m X 1.4 m and considering a 50 % overlap on both directions a loads of elbow grease

Bksabath

17-03-2021, 08:14

I had a first listen to the El Pipe O yesterday !!!!!

Few things got wrong
One of the woofer's is faulty out of the box and I glued my hair while working on the Circle
Here is a cross section of the Pipe

29310

Circle in place and glued to the tube
Rivets not working to well will need same better way
29311

First woofer mounted then the is a 9mm spacer and the second woofer will face the first one the magnet on the outside and the 2 connected out of phase in Parallel

29312

And here it is suspended from brackets and hidden by the telly :)

29313

I am using this as cross over for now whit the Sub input taken from one of the Magies

29314

The cross over need same work :)
I will post more when I get the second woofer and done same testing and calibration
It sound pretty good BTW

Speed is brilliant and the integration whit the Magnepan is spot on, was playing at silly levels yesterday and the single woofer was barely moving but it's contribution to life and impact on the reproduction made a night to day difference and all this whit an uncanny ability for it to be disappear Incidentally ben tuned at 33Hz it is 2.4 meters long the output at both ends is in phase and the ends happen to be right behind the Magies

(I moved them a few inches out of the way to take the picture)

Already I don't know how I was living whiteout it

Try this

MEUTE - Hey Hey (Dennis Ferrer Rework) - YouTube (https://www.youtube.com/watch?v=NYtjttnp1Rs)

Hifidog

17-03-2021, 21:09

I heard this yesterday and its come together nicely, it is very well integrated and has a real snap to it that helps with integration. No sluggish boxy bass here at all, im looking forward to hearing in Isobaric and with the new crossover, also I think some additional power would really get the best from it. Not like other subs but has its benefits for sure, directionality and integration being huge bonuses, if I ran my BK monolith at the same crossover frequency directionality would be an issue.

Bksabath

18-03-2021, 08:16

Tanks Martin
I think that we can bot reply to maximax of the 09/02 (did it really take me that long to glue it on) yes the tube is strong enough no vibration or such from there can't say how much surprised I am

When the vibration and noise was camming from yesterday was from the glass cabinet and the dog next door when I put the signal generator and set the output at the woofer terminals at 1 watt
No spelling mistakes one watt 2.86 volts ! most probably rooms modes set in my room being 5 meters in long this is whit the POA4400 at 3 o' clock not as while we were listening whit it at 11

Frequency response is right down to 10 Hz no spelling mistake

Disregard the numbers on the picture they are totally meaningless just look at the frequency response ( it will not measure below 10 Hz)29326

the peak at 33 Hz is at the resonant frequency of the pipe it all will be sorted whit a better low pas filter tuned at present at 47 Hz me think

I sent the faulty woofer away yesterday after a shorth call to the shop they are going to be black listed !!!!

Main thing is already. I think I have finally found a way to put same punch to the , can stress how much those are the best speakers I have heard now, MG 1.7i.
No merit of mine here the praise should go to Nelson Pass , you must read the article he posted about the EL Pipe O el-pipe-o.indd (firstwatt.com) (https://www.firstwatt.com/pdf/art_elpipeo.pdf)

Al I did was building it, Who said being stubborn is not a virtue ?
still as I write this I know where I may get a pair of 3.6. :D

Yes a 400 W D class be nice and a pair be even betterer another tube behind the telly would make them blend in and disappear like

WHERE DID I PUT MY HAMMER !!!!!

maximax

18-03-2021, 08:34

I only skimmed the article but the two constructions discussed there differ from yours. Also, did you stuff your pipe? Maybe you can integrate the pipe with your AV furniture, but a much larger TV would also do the trick.
Why is this pipe better than eg a decent 15” in a closed box? Would an Earl Geddes approach using more conventional subs work with Maggies?

Bksabath

18-03-2021, 10:46

My front wall view

29327

Hi Max bear whit me...
At the beginning there was no sound... sorry different forum

I named this thread the bug is back as it refers to the itch I have to make things This disease went away for a while till Martin and Paul told me that they would like an amplifier like mine
Big relapse I am afraid

Me and Martin can't resist the chance of experimenting whit what are already hair rising duck feathers systems plus considering "the current situations" ,this must be the top excuse for time wasting at present, I had plenty of beer money and time to spare

So Max to answer your question that you are most welcome to ask
Near the end of the article Nelson write that after blowing 2 of the 4 woofers they just placed them in the pipe whit the magnet of the woofer on the floor supporting the pipe

Quote:>
Reconstruction Rather than spend another $800 on woofers, we decided to try single woofers on each side, so we took apart the tubes and reconfigured them as 10 foot tubes with a woofer at the bottom. We made a nice cylindrical coupler out of MDF to mate the woofer to the tube (Figure 19) and sat them on the woofer’s magnet on blocks of granite (Figures 20 and 21). On top of the Fig. 2

So no stuffing at present because
1 does it need it ? tube as it is seems to work pretty well
would it slow down ,and it does as a side effect, the air flow and muddy the sound ?

I have asked those myself

So far I have the former material that can be reused again and again I probably have wasted £50 on glue and paper it would have been half of that If one collect old news papers
Considering this
Nelson uses 21 in drivers

Plan was to get 2 15in in Isobaric mode to achieve the same frequency response but reduce the diameter, apart from that length is about the same, awing the telly 16 inches from the wall is ok

I discovered that one woofer is working pretty well frequency response wise
directionality is not a problem at such low frequency it would probably be noticed going above 120Hz current filter is at 47 Hz

Being an hollow round tube there are no resonances from the cabinet, the snag whit it being a transmission line the length of is a function of the frequency response

I have heard same 5 figures PMC that made me puke by the way

I can't answer to your question about the Earl Geddes , I am sure Martin could I will do same research on it any way so many tanks for posting
I have found this Multiple Subwoofer Placement | What's Best Audio and Video Forum. The Best High End Audio Forum on the planet! (whatsbestforum.com) (https://www.whatsbestforum.com/threads/multiple-subwoofer-placement.1005/)

Most probably it would cost more than £50 I am budgeting for the glue
The tubes being suspended seems not to cause any problem whit introducing unwanted vibration
what would be the effect of one large cabinet? or more ?
and what about reflections ?
Next step

I need to invest same money on the glue I need about 20 litters of it. at this quantity it became a bulk purchase and turns out to be cheaper still apart from that and same old news papers I have all I need including the brackets
2 tubes will look much better the round shape helps whit diffraction from the wall and 2 of them will blend in much more I was going to do same treatment to it any way and those 2 painting need to be moved away they do deserve a better place
I am using a Denon POA 4400 as amplifier at present, i got it about 15 years ago for £80 :eyebrows:

But I am already building 2 new F5's so the 2 old one's can be used to the subs or to finance the purchase of 2 400W class D that would pay for themselves considering the energy bills

Bksabath

18-03-2021, 12:27

Just re-reading my post

Sorry is quite long and I repeat myself

must be a side effect of the "current situation"

;)

But hey ! there is method in my madness

Bksabath

20-03-2021, 18:19

Back to work on the F5T

I have already made A first prototype of the soft start circuit
The first interaction was using an Arduino Nano it does work quite well but
1 There is so much one can do whit the Nano and it would have been wasted
2 I was wondering about the reliability and noise issues

So V2 use discrete components

Here the schematics of the control circuit

29335

Nothing complicated about it
Part one of the circuit is a toggle switch circuit

29336

At start up the Mossfet is off no power flow Led 1 is off
The only power lowing is a few micro-A due to leakage current
When pushed the momentary push button toggle the circuit the Mossfet conduct
Led 1 turn on 5 volts supply is sourced to the 4 cascaded timing circuits

Timing circuits

There are 4 cascaded circuits as I am using 4 relays for the 2 800 VA transformers

As a rule of thumb any Transformer of 300VA or more need a thermistor
1 to safely size the fuses 2 to stop blowing your mains

29337

In the picture R27 any from 100K and up charges the capacitor
Once the voltage is hi enough the NPN transistor causes the PNP transistor to conduct and the Led lights up
The voltage at point Ref3 make the relay turn on
Mains power for Transformer 1 is now on
Ref3 also supply voltage to the resistor that charge the capacitor on the next stage
Ref4 make the second relay turn on and the thermistor is by-passed
And so on for Ref5 and Ref 6
On the drawing:
the changeover switch is instead a momentary push button no power is present once is pressed and realised

Led 1 to 5 could be replaced by a single led or omitted I do not really like to have a Christmas tree on the PSU enclosure

Bksabath

20-03-2021, 18:36

Soft start board

This is the prototype
29338
I can make single layer boards easy enough at home

29339

But a 2-layer job done professionally is much beterer
Size can be shrunk a fair bit
From 134X198 I am down to 116X178 and a bit more need sawing off

On the third picture the first take on the new board
I will need to place the 18 transistors and 27 resistors on it
Piece of cake that I will have to swallow tomorrow

29340

I am going to get a few made as the boards are cheaper in larger quantities
Please PM if you are interested
It will work whit any amplifier that has one transformer over 300VA or two

Bksabath

21-03-2021, 16:25

Board done
Gerber on its way to manufacturer

29344

Bksabath

02-04-2021, 11:02

The version one of the soft start PCB has landed

29362

More about this later

In the mean time the revised matching rig
29363

I am using a o.1 % 30W 1 Homes resistor
And a new clamping method

The old bolt thingy was a pain whit threads stripping after a while and having to use a spanner to tighten up the Mosfet
The new spring clamp is not only much quicker it also make the clamping force constant so one less chance of errors between one DUT and the next

29364

It is based on Nelson Pass article "Practical Mosfet Testing for Audio

The only difference is that the DUT's are measured at a constant temperature

I am running a first selection of P's and N's channels at 30 V 500 mA and recording the voltage between the gate and the source pin (VGS)

I am using an arbitrary figure for the current. I recon that 500mA will be the right bias for my heat sinks and it will give me whit 4 + 4 dudas about 4 Amps Class A at the speakers

This is what I have so far whit another batch on the bench and one on its way

2936529366

Before I pick the 8 P's + 8 N's I need
There is going to be quite a bit more of matching to do as I am not only looking for matched VGS but also for matched GM

The P and N channels at the end will not be 100% but I am doing my best besides I have been told, by an Authoritative source quote "You will have to live whit it"
Please note Authoritative has capital A and is bold underlined as source could not be any higher :D

After all this is the 4th F5 I am building. it is taking a long time but I aiming for End game

To the 23000 views on this thread my tanks for your patience! not long to go now the weather is turning nice and "current situation" be over soon

Bksabath

06-04-2021, 19:27

Soft start board finished and tested

29384

Minor mess on the voltage regulator's

I made a mistake and pins are crossed over the regulator's

29385

I added same diodes on the Led outputs so that I can use one LED whit 5 large resistors
One single Blue Led on the front panel that get brighter by stages do me

Led must be blue in best Pass Lab tradition ,or was it two? :D

Any way one can clearly ear the Relays Kicking in one by one

29386

I have 5 spare boards up for grabs at cost price first came first served
Send a PM ...

Bksabath

29-04-2021, 19:36

29425294262942729428

Same more pictures of the PSU

maximax

30-04-2021, 22:15

Are these motor run caps?
Obtaining mains transformers is a pain now, toroidy doesn’t appear interested in quoting for small orders and Canterbury Windings is booked out for months… maybe it is indeed time to go smps for class a amps, as some on diyaudio are doing.

Bksabath

01-05-2021, 09:25

0900766b81717062.pdf (rs-online.com) (https://docs.rs-online.com/7661/0900766b81717062.pdf)

Rifa PEH200 on this build only 36 X 22000 uF

Never had any problem ordering Toroidy via the web site, do not ring the grumpy old git :) and ask for discounts he is a grumpy old git and try to charge you more.
Once you place the order Pay pall take care of the rest
Tips
When you order them get the bases and ask for no badge normally they are in the bag and one can fit them after so that they are straight

I have made the mistake to get traffos from the like of Canterbury windings and it was a big mistake as they buzz a lot. If you do you must specify no noise for audio use vacuum resin impregnated 12 times and good luck whit that, which is a pity because I rather buy the gear from here even if it was a bit more expensive

You are only going to buy the trafos once in a life time so get the best you can

There are other decent brands out there but either they cost 2 times as much or are not half as good as the Toroidy + one have to pay custom duty and that pisses me off, even better order them now or you may end up paying custom and such in a few months

maximax

01-05-2021, 10:00

Thanks v useful. Maybe you ordered from toroidy before the start of this year? You cannot web-order to the uk anymore, you have to email for a custom quote. And there should also be import duties unless the order is beneath… £150?? Not sure what the threshold is.

Bksabath

01-05-2021, 12:54

Ops my last order was in November 2020

I got trafos for paradise as well may change depending on "current situation" I hope

Maybe dude was selling to many to the UK :)

Bksabath

16-05-2021, 04:34

2944829449

PSU fully wired and tested
It is on the bench powered up
I am going to live it on for a while on no load to form the capacitors
Rails are all at 40.62 Volts, no load, those tarafos are so good

Bksabath

09-06-2021, 09:22

I am still working on this do no despair
Second pass of matching and the last 200 on the bench as we "speak"

I have stopped any purchase of parts from China due to "the current situation" AKA the Fauci virus

29507
29508
2950929510

I have made same small modifications to the test rig
as you may notice I now have 3 pots to dial in the VGS the central one is a 1K 25 turns that let me dial in the 500 mA current precisely
at the risk of repeating myself I am measuring the VGS at 500 mA 30 V 50 C

The bottom most white board has 200 Fqa19N20C whit VGS from 3600 to 3700 mV

On the bench a further 200 FQA12P20 Vgs around 3800 mV

VGS can and is I have found up to 4700mV on other batches I have

The lower the VGS the higher the GM or gain of the mossfets

It looks like I struck gold whit those low batches
yes there is 200 mV difference between the N and P sides and I have asked same help to a reputable source :D

I copied in the email

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Very nice build. file:///C:/Users/Sandr/AppData/Local/Temp/msohtmlclip1/01/clip_image002.gif

You are not realistically going to get the same Vgs and Gm figures from N and P types,
so we have learned to settle for matching within polarities.

If you want to try to address this, you can use different quantities of P vs N, or alternatively
adjust Source resistor values. In the Pass Labs forum at diyAudio Lynn Quam has shown yet
another approach with his SIT3X circuits, where the drive to complementary types is altered
a bit with local “Schade” feedback, and he gets some good results.

It’s a relatively minor problem, but interesting to pursue.

As to interviews, I just respond to invitations, although lately I need to be careful of being
over-exposed....

best,

np

>>>>>>>>>>>>>>>>>>>
so as soon as I have tested again the P channels
I should be able to pick the best matches and solder them on

Bksabath

10-06-2021, 13:14

29514295152951629517

I have finished matching the FQA12P20

I have groups of matched massfets at 500 mA in 10 mV rows

I am going to use the DCA Pro to pick the quads I need
On the graph the 4 that are going on the board's for one channel
of the 17 I have that measure up at 3.840 V I got 3 very good quads
Victory !!! well almost there
I am whacked for today so I am to call it a day

Besides I got an invite to "world fist listening session"
If I get permission, you may be in for same HiFi porn later

PS

From picture 1 quite a few rows there for quite a few super duper quads

I got spares :D

Bksabath

01-09-2021, 10:33

Hi there :D

It has been pretty quiet here lately
Life happens outside this thread
Any way there is no point posting whit little to report, a catch-up post is now well due.
I had a call Martin a few weeks ago both amplifier dead or dying WTF?
Back ground
Those have been running for nearly 2 years now driving a pair of, most excellent, W.B. Discoveries whit the best sounding “Polk Cobra” Cables
I am not sure of what was happening and what caused the problem, I was not there, but as it turns out one of the mono blocks was blowing fuses and the other one had smoke and discoloration on the Fukushima Futaba plate resistors on the CRCLC supply circuit.
Not sure if it was a passive pre whit a DAC supplied by a car battery a fault whit the mains or the Cobras causing oscillation, what strange is that both amplifiers went at the same time.
At first look the boards Mossfet and Jfet look ok
Martin is drooling over the Fairchild I have and may new 10 U PCB’s that wont fit, why?
Any way we decided as I had the amps to bits to build a new Plinth and in the process to get rid of over 25 Kg of 100-year-old slate
I am a point when we will start to build them back so here are a couple of teasers

2970729708

It should be ready pretty soon I hope
As Martin is not listening to his FiFI at present as the spare £6K Musical Fidelity he has "sound like shit":lol:

Hifidog

01-09-2021, 20:31

I blame the dog for it spontaneously deciding it didn't want to work, either that or the massive amount of refurbishments going on in our street and our recent power outages.

I do want the Fairchild's, but for good reason. The matching is incredibly close and I would be surprised if I could get some as closely matched as your set BKsabath.

They are going to be better than ever in any case so every cloud has a silver lining I suppose.

Thanks for the help and the Fairchild's mate :smoking:

Bksabath

02-09-2021, 06:31

Cool by me, at a price:eyebrows: pity the 10 U boards do not fit I have made them for 400 mm heath sinks

As soon as the mods are done I am going to get them going whit the IRF. I could not find any faulty part so far, your Jfets and such seems ok lets see what up
2 things came to mind
1) I am looking forward to compare the IRF 240 9240 whit the Fairchild's on my new ones then maybe swap them after a one to one comparison.
So my plan would be rebuild yours whiteout changing to much see what happens when you got them back as you may kill them again:lol:
I would look for the compensation for the Cobras. I believe Nelson wrote about it as he likes those cables a lot
Then when mine are done and you really like them I can lend you the my old F5's while we swap the power stage if you can't bear to have to listen to the MF
I am planning to re do the Input boards as well whit the 10 U copper weight, those would fit and I want try the new linear system Jfet so it may be worth doing it this way?

hey what up whit the Musical Fidelity amp any way what is wrong whit it ? :D

Bksabath

22-09-2021, 11:01

29769

new fan controller / soft start wired up and tested :)

Bksabath

13-02-2022, 18:09

I has been a long time
Life happens...
30090

I have finished Martin amplifiers rebuild they have been back home for a couple of months now
I let him post pictures of is system ...
And I am at the last stages of finishing my F5' Turbo V3
After building them up and making sure all nuts and bolts fits
I left the capacitors reform at full voltage for 36 hours
And today I started dry testing the Power assembly to make sure all is horoshow

300913009230093
The little power supply is same what limited still I am getting about 3 A when 330 mV flow trough the 1 homes resistors whit a 4 homes dummy load
More to follows eventually

Bksabath

23-07-2022, 18:23

To avoid being boring I'm not repeating the firs 2 lines of my last post

30474

One of the monobloc's is finished and has been running for a few months quite pleased whit it no sacred smoke so far and the soft start had quite a few cycles so is not going to break any time soon

I have a stasis PA7 on the bench and restoring it priority number uno

Any way from time to time one got to chill out so Carnegie Hall SHOW !!!

As I posted one another thread those did best sound in show IMO but then my cartridge is only second best :lol:

Curvi-BMR MK2

30475

What I heard at the show despite the room and too much too loud tunes (one listen to PMC like stuff just in passing and it still FXXXk your ears up

Not those

30476

Staging wise the Magies are pretty good
Those are also as good if not better in many respects
Staging Yes but also each instrument is nailed down in place and this let you know how many oboes are playing on the Bolero (DECA SXL6065)
first correct answer get prize

And then consider Just 1 driver 1 inductor and 1 cap and 1 enclosure (If that does not give WIFE and hard on or whatever work for them ... I don't know)

Any way I had them here for a few hours whit Chris (big tanks for letting me listen to them at home)

My mate Rob was also here PS The Cubardi is so much better after one let it breath a few hours
His comment : Best I heard your system so far

They are back home now in Manchester and I
I am back home and the Magies sound DULL

Goggle them up and if you get a chance Listen to those

Best in show IMO
yes ok you all have at least 30 K or more and can buy same fancy things whit loads of chips and stuff

One have to have a chat whit the bank manager...