View Full Version : SL-1200 DC Power Supply DIY
Sideband at 1.050Khz is ex the Hi Fi News record.
Cardas 1Khz is OK.
That's probably why it's only at one side...
@ Dimmie
Test record was High Fi Sound stereo Test Record HFS69, Side 2 Test 1. All I have for info is the cover rear.
@ Coen,
Yes Im rusty using FFT I should have noticed that !
Tried ARTA but no better.
I dont think Im making much progress in obtaining meaningfull measurements. There are too many variables associated with using a record as a reference tone source and low end sampling scopes
The answer I want is probably available via one of the servo error signals inside or on pin on the AN6680.
I still havent been able to get hold of a Platter at a reasonable price. I could machine out the centre to give me the motor only and allow free scoping of the electronics with the platter running.
I will probably just have to content myself with my ears for now.
The big consolation is that my Deck is now very very good.
It hasnt cost me a lot of money to get there.
...... Im still using the std arm with my own silicon arm damper...... Do I now look at some mods to the techie arm ?
dimkasta
01-02-2017, 19:54
Gogo rewire it. Single wire run from cart to rca is a nice improvement. Especially with some nice thin teflon/silver wire. I have even removed the pins from the headshell so that I ran the wire in a single piece. I just cannot change headshells easily, but who cares :)
BTW, I had done some measurements some time ago through the power entry hole. The probes can easily get in through there.
I just had to tape them down to keep them from obstructing the platter rotation.
Tried ARTA but no better.Try Spectrum Lab http://www.qsl.net/dl4yhf/spectra1.html (free). I did make some measurements and looks like it can be useful. I will upload several spectra catched from FG in nearest time.
Thanks will down load a copy ASAP
dimkasta
02-02-2017, 13:29
Where are you guys measuring? At pin 24 of 6680?
So, promized pictures captured at the pin 23 AN6680 - i.e. FG signal amplified by Q203 and internal 6680 amplifier-filter*. I used Spectrum Lab, two small windows "Components" and "Configuration" will show how to obtain spectrum.
First one - platter 'as is' with my resin mat, second and third screenshots captured with added mass on platter (400 and 850 g, I used 2 and 4 old bearings from my mini-lathe). All measured data are with power upgrades described above.
Main interested peaks are:
50.5 Hz (note: it's NOT mains 50 Hz). Main peak width is wery important thing, but it depends on FG coil and disk magnet and I'm afraid it can't give us useful information.
55 Hz (i.e. 50.5+4.5) - sideband 4.5 Hz, motor torque signal (3 phase, 12 stator coils and 16 rotor magnet poles @33 rpm)
64 Hz (50.5+3*4.5) - 3rd harmonic of 4.5 Hz.
Using this signal we can see changes in sidebands amplitude even when resin mat removed or attached.
* Note: signal after internal 6680 amplifier has some distortion (components at 101 Hz etc.), but we can watch it in real time. Will be better to take signal after Q203, amplify it and use.
dimkasta
02-02-2017, 19:05
I found some time and tried to measure the FG signal.
Unfortunately, the parts are not in a position that allows the probe to fit without obstructing the platter.
I eventually managed to get a measurement (using some jumpers) between R119 and R220.
I got a steady 50Hz (20ms) signal which is good in general, but the signal was dirty and I do not know how much of the dirt was because of the jumpers. The noise appeared pretty random.
I tried applying some digital high pass filters to locate any harmonics or peaks, but I could not see any.
Tomorrow I will try to solder some permanent test points and retry.
I forgot one more peak between 4.5 and 13.5 Hz: approx. 6.7 Hz.
Tried ARTA but no better.
I will probably just have to content myself with my ears for now.
I gave up on measurements, too little space and without a reference it is hard to know what's right even with a proper measurement. Then how do these electrical signals translate to sound? Kind of mystery to me. Btw I focussed on time domain measurements.
I trust my ears but I know the limitations.
Ultimately measurements and listening are the only way to make any further progress.
The big consolation is that my Deck is now very very good.
It hasnt cost me a lot of money to get there.
I and of wondered if you a planning to offer regulator pcbs or something. The LT devices have specs that far better match the load regulation requirements of the components on the deck than a 317.
My own experience is that not only the regulator and its implementation matters but also the grounding of input, ref and output. This does not always follow plain logic and it seems to matter a lot.
...... Im still using the std arm with my own silicon arm damper...... Do I now look at some mods to the techie arm ?
Tonearm and cartridge lust are always short around the corner. Forget cheap "upgrades" here. I didn't think the Jelco nor the Rega added any value here save for a different sound.
The stock tonearm is actually a very solid design and execution. Bearings are chatter free afaik and the weight suits may carts. The only thing I swapped so far is the cable part from the tonearm pcb to the phono stage. Just cheap and simple "shielded low noise microphone cable (OFC)" with nice plugs. The sure helps though I think the internal wiring is a candidate for replacement.
Yes measurements are not easy, but if I can get an old platter I will still machine it to allow access to the electronics.
I did think of PCB's but there does not seem to be much interest apart from 3 or 4 of us. and we all have regs built.
I have built these very easily using vero board with repeatable results using the minimum number of components.
I could post detailed photos and sketches of the layout if that was helpful.
I have come to respect the LT regulators, and their lower dropout voltage meant I could use a quality off the shelf mains transformer @18-0-18v (allowing for regulation and mains input tolerances)
They are more expensive than the older regs.
I have been looking at modifying a rega arm or building an audiomods version , but Im moving into territory where the is spend more than the deck cost.(all my mods are DIY except a S/H SDS platter plate)
The silicon damper gave me a worthwhile improvement, as did fitting a soft copper washer in place of the rubber gasket between headshell and arm. Is this an indicator that a weak link is the break in mechanical connection to provide a removeable head assembly. If one looks at the dissipation of energy in the form of sound vibration across differing materials and how best to absorb that energy. I might get a clue.
Im debating the fitting of a fixed head support onto the Techie arm, this could be welded on using a low temperature weld rod like lumiweld. Drilling multi holes in the arm and filling with Foam.
A few hours drawing and machining,it wont cost me a lot if it doesnt work
I think Qwin did fit a straight arm to a technics arm bearing using an alum tube with carbon fibre outer sleeve and a Linn head shell mount. Maybe I should drop him a PM .
dimkasta
03-02-2017, 15:00
There could be more people lurking silently around these places. :)
It's just that the 317 already works nicely so there might not be much interest to get things even further, unless an equally easy solution is provided.
Smds are unfortunately a brick wall for most diy people.
Perhaps we can work on something with lt1963 or lm1085?
In any case, as we discussed I can help with pcbs, and even with soldering of smds if this is wjat we decide to work with
I have built these very easily using vero board with repeatable results using the minimum number of components.
I could post detailed photos and sketches of the layout if that was helpful.
That would be very nice!
I have been looking at modifying a rega arm or building an audiomods version , but Im moving into territory where the is spend more than the deck cost.(all my mods are DIY except a S/H SDS platter plate)
The silicon damper gave me a worthwhile improvement, as did fitting a soft copper washer in place of the rubber gasket between headshell and arm. Is this an indicator that a weak link is the break in mechanical connection to provide a removeable head assembly. If one looks at the dissipation of energy in the form of sound vibration across differing materials and how best to absorb that energy. I might get a clue.
I removed the rubber gasket once but thought is was much better left in. Is the copper washer a diy effort?
I noted that the headshell leaves considerable axial movement even when clamped.
MOD update:
I proceeded with the "final frontier mod" but encountered some misfortune. Unlike other mods where getting the new reg up and running was sufficient for an improvement it did cost extra effort to arrive at a satisfactory sound. It sounded quite all right without the reg in place but with the new reg every little detail seems to matter. I arrived at a 317 reg that is set up very similar to the main reg for the 21V board voltage. The sound is unfortunately very unstable (sounds different each time I listen) and I suspect the NOS black gate buffer cap to be responsible for this. This is the cap I had lingering around and put in the circuit, they are notorious for their long "break in". Therefore I'm reluctant to call this mod a success yet.
The copper washer is a DIY effort. You can buy a washer of the correct dia ( 12.5mmm OD 8.4mmm ID, 1mm thick or 1/2 " * 5/16" * 1/32") and file a grove in it to just slip over the locating pin.
Before installing it anneal the copper to make it soft. This allows it to form to the shape required to match the techie components.
To anneal copper, take it to red heat on the gas then let it cool, it will remains " soft " for 12 -24 hrs then slowly harden again to its normal state.
Given the wide tolerance of the techie detachable head, the pin is a bit smaller than the locating grove.
,I find it difficult to get a repeatable azimuth setting when changing or removing /replacing the head. I keep mine set hard to the right and hold the head in that position when tightening the clamp ring.
@Ianmac I found a proper copper washer at the nearby carparts store. I filed a slit and heated it red hot on the gas furnace. At first the surface turned silvery but that came off as a powder later. With the washer and new tonearm balance (the copper washer is much heavier) I listened to some tunes. My first impression is that the sound is cleaned up and more dynamic but also a little rough. It reminded me of my first cd player. A second listen this morning learned that the sound is also lacking proper highs. So I won some and lost some. Likely this mod works better in conjunctuon with the viscous damper. For now I prefer the stock rubber washer.
I think Qwin did fit a straight arm to a technics arm bearing using an alum tube with carbon fibre outer sleeve and a Linn head shell mount. Maybe I should drop him a PM .
Do give that mod some serious thought. I did it to my deck (along with better tonearm wire) and it was a very nice improvement. When completed, the arm is lighter and better balanced because the headshell end is lighter than the stock arm. Where this really shines is it's lack of resonance. The aluminum/carbon fiber tube is so much less resonant than the stock tube it isn't even funny. The cost? Less than the price of a good armboard. Scroll down to the eighth pic The Techie tonearm mod (http://theartofsound.net/forum/showthread.php?29887-My-Second-“Techie”-–-well-all-of-it!)
dimkasta
13-02-2017, 08:44
Keep an eye on your cart's compliance and the resonance frequency with your tonearm's mass
Keep an eye on your cart's compliance and the resonance frequency with your tonearm's mass
Well, I'd like to say I did that :whistle:
The aluminum/carbon fiber tonearm tube is both lighter and less resonant. Is that a good thing? My cart is an AT120e and I had to add a 2g weight because the headshell end of the tonearm, now lighter, couldn't get the stylus to set down on the record.
dimkasta
20-02-2017, 18:26
If you did not check the resonance frequency with a test record then you didn't :)
Don't get me wrong. This has nothing to do with your DIY efforts. And nothing to do with actual ringing of the arm.
It just has to do with how your new arm compliments your cartridge.
With commercial arms, you can put your cart's compliance and the effective mass (not weight) of the arm on a calculator and get the resonant frequency. Your goal is to put this frequency at ~8-12Hz. This will make it high enough to not let them resonate with rubble, and low enough so that they do not resonate with bass from your music.
With a DIY arm you have to check it with a record.
If you do not have a test record, an indication that you have an issue with amplified rubble is if you see the cones of your speakers wobble.
But since you decreased the mass, chances are that you are closer to bass frequencies. Check it if you have more issues with jumping/skipping on bassy passages
I don't have any issues with jumping/skipping on bassy passages but I would like to check it to know it's current resonant frequency. Where would I get such a test record?
Guys, please pardon us for going off on a tangent here.
dimkasta
21-02-2017, 16:50
Yes it would be nice to deduct the effective mass, so that you know which carts it is going to work best with.
I am using the Hi-Fi news test record.
I got it from amazon a few years back
I want to find another one with a better channel separation test, but that is different discussion
dimkasta
01-03-2017, 21:30
I found some time today to replace my old 3-capacitor pcb with my new 2-cap one.
I did some quick listening on low volume, but can't say I noticed anything significantly different. Perhaps some thinness, but it could be from the fresh soldering and parts. I will leave it running tonight and do some proper listening tomorrow. Later this week I will also install the snubber to see if it's worth the trouble.
dimkasta
05-03-2017, 20:21
This weekend was a relaxed one so I had much time to listen to music.
Going to a 2-cap filter on the PSU was definitely a setback. The sound was kinda mushier. But I will compensate for it with an extra RC stage inside the turntable when my regs are finalized. This will also help with transients by reducing the effect that the umbilical cable has.
For now, I finally installed the Quasimodo snubber.
And I am just basking in the awesomeness that it brought to the sound. Such a big and nice improvement on the clarity of the sound.
If you do not know what I am talking about, here is the link to the post with my initial measurements and the schematic/parts.
http://theartofsound.net/forum/showthread.php?26620-SL-1200-DC-Power-Supply-DIY&p=811715#post811715
Thanks again to Mark Johnson (http://www.diyaudio.com/forums/power-supplies/243100-simple-no-math-transformer-snubber-using-quasimodo-test-jig.html) for a very nice testing jig.
This is my go at an external power supply for the SL-1200.
My most sincere thanks for this thread, it was extremely useful in this process - http://theartofsound.net/forum/showthread.php?22954-A-quick-visual-guide-to-externalising-PSU-from-your-Technics-SL-1200
After a conversation with Wayne Colburn of Pass Labs, learning that he has made a SP-10 PSU for in-house use, a few things became apparent - Get the bridge diodes out of the TT and just have DC on the umbilical. Use 1N4007 diodes, for whatever reason they sound the best, even better than exotic high-speed diodes. Snub the bridge on the AC side with a .22uF film cap. Use multiple stages of RC filtering.
The biggest improvement is of course, no transformer vibration on the platter, easily done merely by locating the transformer outboard of the TT chassis, and this is what I was originally planning to do - but converting umbilical to DC wasn't much more work.
Does it sound good? Of course! And the best part is that I'm not even done yet, the stock regulator hasn't been replaced... (mainly because I ran out of time...) :) :) :)
UPDATE - The first regulator modification can be found here --http://theartofsound.net/forum/showthread.php?26620-SL-1200-DC-Power-Supply-DIY&p=467311#post467311
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1827_zpsb905cfd8.jpg
The first thing to be done is remove the bridge diode and snubber caps.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1826_zpsdf1211cd.jpg
These were located in the holes now visible.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1828_2_zps7c058cea.jpg
An umbilical is added. The white 3rd wire is capped off, but still there if I need it for something in the future. The bare metal wire is the shield drain, to be attached to ground.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1831_zpse61eadf0.jpg
Fused IEC plug, switch, microphone connectors used for the umbilical.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1832_zps1e41498a.jpg
All to be placed into this ABS enclosure.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1835_zps77f2b3ee.jpg
There is no reason not to use the stock transformer, it's the proper voltage, soft-mounted, has a beautiful mu-metal shield, and is free with the purchase of a turntable.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1833_zpsa5955da8.jpg
(3) 4700uF 63V volt capacitors, (2) 10ohm 3W resistors, .1uF polypropylene bypass cap, all wired up in a CRCRC filter. Why those values? I had them on hand. If I needed to buy new, I would probably use similar values.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1836_zps8c42643c.jpg
And here it is with the bridge to the side. (The yellow is the transformer secondary.) The eagle-eyed among you will notice the 1K bleeder resistor across the cap bank.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1838_zpsbf5cbe75.jpg
The final configuration of the umbilical attachment to the PCB. Of note the shield drain wire (the bare one) is not connected to anything at the PSU box. The shield is just at ground potential.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1837_zps120ec36f.jpg
The insides after the modification. I utilized the now unused AC power switch to switch the strobe.
http://i198.photobucket.com/albums/aa276/aeroplane_album/IMG_1840_zpsf2550805.jpg
A few zip ties to attach the wire to the studs for strain relief, and of course the obligatory ferrite clamp.
http://i198.photobucket.com/albums/aa276/aeroplane_album/SLPSU_zps2831b828.jpg
Schematic of the Power Supply.
This is stage one of my PSU modifications, a different regulator of some sort will be next.
Questions or comments warmly solicited.
Thank you for looking.
Dear 6L6,
I know your power supply project has been done a long time ago, I am so sorry.
I have a SL1200 MK3D, I searched for power supply modification and I found your thread. Actually, I can mod my turntable followed your instruction here.
However, I would like to control the speed so I want to keep the original speed led (in the main power switch of the turntable), but I do not know how I can do. Please give some advices!
Thank you so much!
tungh
The PSU is only that, it does not control the functions of the deck. The switch on the schematic only connects the PSU to the mains supply, it does not make the deck spin.
The Start/Stop and 33/45 rpm controls are momentary push button switches that trigger these functions, they connect to the function Chip mounted on the main PCB. These controls are connected to the PCB by the harness and multi pin connector, shown bottom left of the picture above. You can not run the deck without them, or replace them with latching switches, they have to be momentary.
If you want to remove other items that might generate noise in the system, remove the speed fader and strobe LED and in particular the strobe transistor that causes the 50Hz flicker. Even if you remove the LED's the transistor is still in circuit and firing, so it needs removing, as in theory, this could add noise. The fader and strobe are not required for normal HiFi operation, the deck is self monitoring and maintains very accurate speed, these controls are for DJ use when track merging. EG. When the fader is set to "0" the deck should run at optimum speed. Adjusting the Fader cons the function chip into thinking the deck is not running at the right speed and it compensates, thereby creating a speed change. If you take the fader out, the deck just runs at the correct speed, be aware that if doing this, you have to bridge two of the pins on the fader harness header on the main PCB, or the speed control function will be seen as open circuit by the function chip.
Note: My own PSU/Regulator project is based on Jim's circuits.
As Ken says, there is no change in functionality to the table controls other than the on/off switch. The pitch slider works, the pitch LED works, the the strobe works, etc... The table merely sounds a LOT better because of the transformer being removed from the chassis. :) :) :)
Dear Ken and 6L6 (I do not know about your real name, but I guest you are JIM)
Everything is very clear and I could start my project soon. Thank you so much!
However, as Ken says, I should take off the both fader board and led. I will explorer how I can do that (I am not at home now). If I am not wrong, it links to the mainboard by another multi pin connector, so I will disconnect them.
For the strobe light, I think when I turn on the outside power, the strobe light is always on, is it right? Because, the DC power will lead from the mainboard to the light, it is not connected to the switch directly. I will try to keep it just for my experience, after that, I will find its wires and cut it off.
For the switch, I would like to keep it enable. But, could I use the switch to turn the mainboard ON/OFF? It means I will turn on or off the 30v DC from the PSU outside before the DC could reach the mainboard. I do not want AC and DC power together in a wire shell...
Sorry about my silly questions, because I do not know exactly what is included inside the small board under the switch and I am not a professional in DIY. I am just curious and want to challenge myself in this small project.
Thanks and best regards,
tungh
Dear Jim,
I started my project with the Talema transformer (In 230v - Out 30v) and your schematic. It worked without burning (kidding), everything seems to be good. When I check the output of transformer, it is about 32-34v AC. But the DC output is just 24v DC. Is it normal? Or I have something wrong? I use the 2.2K for BLEEDER.
Please help me!
Thanks
30v AC rectified will give you 40v DC before the filter bank. The bleeder won't make any difference. What value did you use for the filter resistors? And please post some well-lit, in-focus photos of your PSU please, can't help without them.
30v AC rectified will give you 40v DC before the filter bank. The bleeder won't make any difference. What value did you use for the filter resistors? And please post some well-lit, in-focus photos of your PSU please, can't help without them.
Dear Jim,
I attached the pictures in this post... As you can see, I changed the bleeder back to 1K. I made the serial connected of 2 secondary transformer to get 30v and I have 37v AC before diode bridge.
27173 27174 27175 27176
All those connections look good. How is the bridge connected to the caps?
All those connections look good. How is the bridge connected to the caps?
Thanks Jim, I tried to copy from your schematic and pictures.
As you can see on the left hand of the picture I attached, the red wire was connected to the output positive pole (+) of the diode bridge and the white wire is connected to the negative pole (-) of the diode bridge. I am at the office now, so I can not take a picture. Sorry about that!
27177
After so many years I'm finally planning to do power supply mod. It will be one with own voltage regulation (with tracking pre-regulator configuration). I'm reviewing multiple threads on the topic. And I have one doubt. Everyone I saw was mentioning 21V voltage on regulator output while Technics service manual states 20V. So why to use 1V greater value. 5% it's not much so probably should not hurt but why not to do just 20V?
27851
Never mind, I found it. Newer manual for mk2-a version states 21V.
dimkasta
13-09-2024, 17:22
Almost 10 years later I managed to finish the project to my liking...
Here is my latest PSU and reg
The PSU is a plain full wave rectifier, followed by the quasimodo snubber, and then a CRC filter. I reduced the last R in Jim's CRCRC to 1R since the 10R was starving the electronics while breaking.
And I moved the last RC close to the regulator and inside the TT
The regulator sits on top of the input hole of the TT, offering an easy place for a DC socket (currently using a plain RCA socket)
I also added a PPTC and currently experimenting with a TVC for extra protection
The rest of the regulator follows Jim's design, apart from an extra 100nF ceramic at the input (as specified in the datasheet) and some quality of life parts like leds for easy monitoring, test points for the voltages around the regulator and a choice for a screw terminal for easier experimentation
31435
31433
31434
Powered by vBulletin® Version 4.2.3 Copyright © 2026 vBulletin Solutions, Inc. All rights reserved.