View Full Version : Yamaha NS1000M - Tweaks
Thanks for the update Ken! Looking forward to reading about your progress in the "active domain"! :)
I tried the Scanspeak and Yamaha 12” Bass units in my semi active project, running them with Quad Bookshelf units, so have experienced their different traits under similar conditions.
http://www.jkwynn.co.uk///Project_Images/SAP/sap_12A.JPG
http://www.jkwynn.co.uk///Project_Images/SAP/sap_14C.JPG
My concerns about the Scanspeaks ability to perform in the upper bass and crossover region were unfounded and it out performed the Yamaha in every area. Obviously from the specification, it would have deeper bass, but its noticeably a lot deeper, which gives a strong platform to work with. Its faster and tighter, not that the Yamaha is poor in these areas, but the Scanspeak is better and can only improve in a more rigid cabinet. The big surprise was in its upper region, where it is subtle and tuneful, blending seamlessly with the Quad mid range. Comparatively the Yamaha bass is muddy and slow, not that I would ever have said that prior to the comparison, as I like the Yamaha's a lot. The weakest link with the Yamaha is the tweeter, ragged low end and falls off rapid at 15kHz. The Yam mid range is crossed quite high at 6kHz to deal with this. With a shallow filter slope the Mid range driver dominates well up the spectrum, the tweeter contributes little to the program as can be demonstrated by switching it in and out during play.
I'm building a Hypex DLCP/UcD Pre/Crossover/Power unit in another thread. http://theartofsound.net/forum/showthread.php?51046-Hypex-DLCP-(DSP)-with-6-Channel-UcD-Amp-Project
The plan for my next experiment, is to use this DSP set up to try the Yamaha Mid range with the Scanspeak Bass and Fountek Ribbon Tweeter. It's a modern take on what Yamaha produced and is the NS-1000M revisited. It has great potential, but only trying it will confirm this and satisfy my curiosity. The Fountek NeoX 2.0 Ribbon can be crossed from as low as 2300 Hz 2nd order. It might be nice to move the mid range crossover point down and away from 6kHz, where sibilance lies. 4kHz with a steeper slope might be a good starting point.
Looking at the Fountek Spec sheets, for the Neo X Series, the NeoX 2.0 has the flattest, smoothest response in the range. I was impressed with the NeoX 1.0 used by Quad and the NeoX 2.0 looks a bit better. So my plan is:
1). Obtain a pair of NeoX 2.0 Ribbon Tweeters.
2). Make Router templates for the Tweeter and Yamaha Mid Range.
3). Construct a temporary Baffle to mount the Mid/Tweeter.
4). Construct a lash up of the active Hypex DLCP to run this set up, using my Nakamichi Power amp.
So I have quite a bit to get on with. :)
This project has changed a lot during its time, so much so, that it is no longer about the Yamaha NS-1000m or any of its components, so I've started a new project for an active 3-way and thought I would wind this project up.
I decided to put the NS-1000M back to its original (Passive) specification, I still had the original x-overs so this was not very difficult to achieve.
http://www.jkwynn.co.uk/Project_Images/Yam/yam_37.JPG
I've learnt a lot during my experimenting, the most important thing, is that the NS-1000m has it's own character and this shines through whether you are passive/active/digital. Like all speakers, they are not perfect, they have little in the way of deep bass, but what they do have, is plenty of punch in the mid bass due to a rise in response at this point. With a bit of room reinforcement they are good for 50Hz, maybe just under. Where they shine is in the mid range and top end, with clarity and imaging that is rare to find. My original thought was to put the speakers back to original spec and sell them on, trouble is, I'm enjoying them too much.
My advice to any owner of the NS-1000m who is thinking of tweaking, is something I learnt from Paul Coupe (Reference Fidelity Components) and applies to any classic speaker, keep it simple and swap like for like, or you risk loosing the magic you enjoyed in the first place.
To this end, let me guide you through the changes I made and destroy a few myths along the way.
The simple re-cap I completed at the beginning of this project freshens things up and brings the speakers to a level, which is probably not much better than the day they were made, but more importantly, does not change the speakers character, so lets go through what I did to achieve the crossovers pictured below.
http://www.jkwynn.co.uk/Project_Images/Yam/yam_37A.JPG
My advice on how to freshen up the NS-1000m
1. Change the spring loaded speaker terminals, nothing fancy needed, just some Gold plated Brass posts, as fitted earlier in the project - Avoid cheap white metal posts, if you buy off the Bay, check them by scratching off the Gold at the solder tip with a file. I fell foul of this and purchased some which turned out not to be Brass and had to be changed.
2. Replace the flimsy wires that connect the posts to the PCB, again nothing fancy, I used some Van Damme 2.5mm square speaker cable.
3. Replace the caps for the Mid Range, these are the large silver cans, all identical 3.5uf. There are six (wired in parallel to achieve 21uf) and used in series with the mid driver in the circuit, there is also a single unit used parallel across the circuit. Contrary to common belief THESE ARE NOT ELECTROLYTIC CAPACITORS. If they were they would be around 30,000uf total and not 21uf. They are metallised paper in oil and are a form of film cap, so should be replaced with similar. I chose to use Ansar Supersound Poplyprop Caps. After some experimenting, I found using 3uf of ClarityCap ESA as part of the 21uf total, gave a slightly fuller, richer mid range and removed a bit of grain, but the difference is small, don't expect a huge difference. Do not use electrolytics to replace these caps, it will seriously degrade the sound quality.
4. Replace the two electrolytic caps in the Bass circuit (94uf total) with similar, I chose to use a pair of 47uf Mundorf E-Cap bypolar electrolytics. Do not be tempted to use PP film caps here, they have a much lower ESR value and will change the characteristics of the circuit.
5. Replace the film cap (2.7uf) used in series with the Tweeter, use a high quality film cap here, I used a Mundorf Supreme.
Notes:
A). The level controls need to be in the circuit, if you bypass or remove these, you will have to alter the design accordingly to compensate. Don't ask me how, as I didn't measure the L-Pads, just leave them in circuit and you will be fine.
B). I spent time measuring and matching the bundles of caps to get the desired values, you can probably see the true value scribbled on the side of some in the picture.
.
Edit: If you want the last drop out of the speakers, you can bypass the 2.7uF series cap for the Tweeter and the 21uF bunch in series with the Mid Range with a 0.01uF Vishay MKP1837. I'm not a big fan of bypassing and not had much success with it in the past, but it does seem to work in this configuration, taking the last bit of grain out of the Mundorf Supreme on the Tweeter and opening the sound stage up in the Mid Range. Wouldn't have bet money that such a tiny cap would have made any difference, but my ears are telling me otherwise.
That's it, nothing more complicated is required, the results are superb and without altering the character of the speakers. I'm going to spend some time enjoying mine again. If only they went 15Hz deeper in the bass! ah well, that would make them just about perfect in my eyes and nothing in this game is quite that. :)
Good overview and summary. Worth reading by anyone with an interest on these speakers
Thanks Grant, I thought putting it all together in a simple, straightforward form would be useful to folk.
I had to stamp on the miss-information that the big silver caps are electrolytics, as they are not.
I keep seeing this repeated, even Troels Grevesen got this wrong when he did a crossover upgrade for the NS-1000m on his site.
Excellent summary and nice to hear from you Ken:thumbsup: I remember you starting this project ages ago. I am sure you suggestions will be a useful resource for anyone maintaining the NS-1000m.
Funnily enough I have been in correspondence with Spendor today regarding something very similar as my speakers are 30 years old and I thought I would investigate wether they needed new caps or tweeters as mine contain ferro fluid. Spend or said don't touch anything unless it's broke and the speakers are finely calibrated so changing anything will make them sound different. The caps are on a passive crossover so they felt they should still be ok. I think I will leave well alone as they sound fine and doing anything may bugger up the sound. I will only start surgery if they actually fail in some area!
its very easy to get sucked into replacing caps after reading countless forum posts from people who have stated they heard an amazing difference when done,the night and day comment may also appear.
the truth is,well for me anyway,ive never heard a great difference,i didnt with ditton 44's,i didn't with ditton 66's either,both had a full re cap and internal wire change.
with this in mind im not touching my ns1000's especially as the caps are glued in.
sq225917
05-01-2018, 20:02
Stock lytics in the xo for the ns1000m has esr of just over 1ohm.
If it ain't broke don't fix it!
Leave well alone if they are sounding good James.
I've been beavering away, doing all kinds of stuff, but mainly getting to grips with my measuring gear and DSP software. I've always found the difference between a tweeter being wired normal/reversed to be quite subtle and not always obvious. Stick a mic on the job and whoooa! where did that canyon in the response come from? All good stuff and adding to my knowledge as I go.
I really am enjoying the NS-1000m in stock passive form, especially with vinyl, no multi-channel active set up and digital conversion to get in the way, just a simple two channel set up.
Do I miss the extra bass depth and dynamics with my active set up, yes. Do I miss the extra zing and shimmer the ribbon tweeter gives, yes. Do I enjoy the music any more with it, probably not. Hey ho. :D
"with this in mind im not touching my ns1000's especially as the caps are glued in."
Yes, I had to rip mine apart removing them, that's how I first discovered they were not electrolytics.
Stock lytics in the xo for the ns1000m has esr of just over 1ohm.
The only lytics in the stock crossover are the two in parallel across the bass, all others are film caps.
What frequency did you measure these at?
"with this in mind im not touching my ns1000's especially as the caps are glued in."
Yes, I had to rip mine apart removing them, that's how I first discovered they were not electrolytics.
if its the same glue they used on the binding post plates i thought about hot water to soften the glue up,a hair dryer worked wonders on my binding posts also.
The glue almost looks like evostick contact cement, but may be a hot melt of some kind.
A couple of mine just cracked out of the glue without a problem, the rest ripped apart under the pressure.
Just as a guide for folks, ESR is talked about a lot with re-capping projects, the picture below is of a 1uf electrolytic measured using DATS v2 software.
Notice the massive difference in ESR dependent upon the frequency it sees.
PP caps do not have any where near the same level of ESR and strapping a fixed resistor in series with it will only work for a fixed frequency and is not going to give the variable level due to frequency change. This is why we change like for like with caps.
http://www.jkwynn.co.uk/Pics/DATS_ESR.JPG
Again, I'll put this up for reference.
This first plot is of the NS-1000m bass, measured near field to avoid room effects.
http://www.jkwynn.co.uk//Pics/NS-1000m_bass_7cm.JPG
I've not shown the trace, but there are no phase changes. You can see the elevated bass around 65-115Hz which gives them the punchy feel, then it falls off rapid from around 60Hz (f3 just over 50Hz), pretty much identical to the Yamaha results and also those of Troels Grevesen, when he said “the NS-1000m doesn't do much below 50Hz” he wasn't wrong, however.........
http://www.jkwynn.co.uk//Pics/NS-1000m_bass_myspot.JPG
This plot from my listening spot shows the room taking effect and bass re-enforcement down to 40Hz. This explains why at a NEBO meet, in a very large room, with the speakers well away from the corners and rear wall, there seemed little bass, they would be working more like the near field response. Drop them back into my smallish listening room and they sound totally different. You can also see a suck out centered around 65Hz, there is a phase change centered here which will account for this and is probably down to room reflections/cancellations, so you win some, you loose some.
So although on paper, the bass output is pretty weak, the reality in your average size listening room is somewhat better. :)
To my ear, I always thought this was the case, it's nice to be able to confirm it.
Note: Measured using REW, my mic is calibrated down to 5Hz.
StanleyB
06-01-2018, 15:30
I've not shown the trace, but there are no phase changes. You can see the elevated bass around 65-115Hz which gives them the punchy feel, then it falls off rapid from around 60Hz (f3 just over 50Hz), pretty much identical to the Yamaha results and also those of Troels Grevesen, when he said “the NS-1000m doesn't do much below 50Hz” he wasn't wrong, however.........
The NS1000M was produced over several decades, and I have always questioned the frequency response between the earlier and later versions.
The level controls need to be in the circuit, if you bypass or remove these, you will have to alter the design accordingly to compensate. Don't ask me how, as I didn't measure the L-Pads, just leave them in circuit and you will be fine.
It’s not difficult to measure the L-pads and they can be replaced by two resistors (you might need more than two to get exactly the right values). If one does measure them the pots can be seen to be less than ideally matched and do not present an even impedance. Of course, once they are bypassed it doesn’t allow you to adjust them so easily.
walpurgis
06-01-2018, 16:11
That's correct. 'L' pads have an insertion loss. Their bypass or removal adds a couple of db or so to the output of each affected driver. Without correction, this would be output over and above what would be achieved at max 'L' pad setting. So probably resulting in a very bright, forward sound (nasty).
That's correct. 'L' pads have an insertion loss. Their bypass or removal adds a couple of db or so to the output of each affected driver. Without correction, this would be output over and above what would be achieved at max 'L' pad setting. So probably resulting in a very bright, forward sound (nasty).
Yes, this is what I was getting at.
The NS1000M was produced over several decades, and I have always questioned the frequency response between the earlier and later versions.
As far as I am aware, the design/components never changed, so the response should have been the same throughout its production life.
Never seen anything to suggest otherwise.
It’s not difficult to measure the L-pads and they can be replaced by two resistors (you might need more than two to get exactly the right values). If one does measure them the pots can be seen to be less than ideally matched and do not present an even impedance. Of course, once they are bypassed it doesn’t allow you to adjust them so easily.
I tried using fixed resistors right at the beginning of the project, to be honest, I couldn't hear much, if any difference, one way or the other, so as my pots are smooth and not scratchy I put them back in, when reverting back to passive. The ability to adjust does come in handy now and again.
.
The caps I used in the example above replicate the sound of the original silver cans exactly, if this is what you want then use the Ansar Supersound as indicated.
The stock NS-1000m have always had the reputation of being a bit Amp fussy and needing something rich sounding, with a highish damping factor, like a Luxman, to sound their best. When using my Nakamichi Power Amp (fairly neutral) with the stock speakers, I had to turn the mid range down by 2dB on the level controls and I have to do the same with the Ansar caps fitted. The metal dome mid range just has too much top end bias otherwise.
I have a lot of respect for the work done on HumbleHomeMadeHiFi in the cap tests and I nearly always agree with the results on the caps I have experience with. The comments they make about the Ansar Supersound is that they concentrate on the upper frequencies and are therefor not particularly neutral. I decided to try a more neutral cap, just for the location in series with the mid range driver (21uF) and see how that faired.
I settled on the fairly modest entry level Mundorf MCap MKP. The largest size (diameter) that will fit into the moulded pockets on the rear of the Yamaha x-over plate is the 4.7uF. So I used 4x 4.7uF and a single 2.2uF to make the 21uF needed. When I measured the individual caps for both x-overs, two were just over the nominal value and the rest were all just under. They were all closer than 1% which is a lot better than the stated 3% and something also picked up in the "Humble" tests. I added a 0.1uF to one bunch and a 0.22uF to the other, this gave me totals of 20.996uF and 21.018uF which is more than close enough, note I didn't use a bypass on these. I left all the other caps as previous, including the bypass used on the Mundorf Supreme.
http://www.jkwynn.co.uk/Project_Images/Yam/yam_37B.JPG
The caps fitted in nice and neat, a bit of silicone sealant was used to hold them in place and reduce vibrations. I ran them in for a couple of days before giving them a listen.
I can not over state how much of an improvement these caps make over either the stock Silver cans or the Ansar Supersounds. They work so much better with the mid range driver. The level controls can be set to normal, they are much more Amp friendly in this respect. The more neutral balance removes the top end focus and gives a slightly richer mid range with better (more realistic) piano in the lower part of the middle of the key board. This has been one of my pet hates on every pair of NS-1000m that I have heard, these caps cure it. They make for a very easy listen, without being over lush, sound stage has improved further and sibilance has all but disappeared (unless on the recording).
I was originally putting my NS-1000m back to stock spec to sell on, that decision may have now changed, as I am enjoying listening to them like never before.
This final combination is a winner, I can definitely recommend it. If you can use a soldering Iron, this simple capacitor replacement is actually an upgrade and will give your speakers a new, more neutral and natural sound. If you have a really bright sounding Amp using a percentage of ClarityCap ESA for the mid range bunch will help, but if your system is neutral this last combination is the dogs whatsits.
http://www.jkwynn.co.uk/Project_Images/Yam/yam_09.jpg
The caps have been fitted, Ansar Supersound (PP) for the midrange bandpass, a Mundorf Supreme (PP) for the series tweeter and Mundorf Ecap (Lytics) for the parallel pair in the low frequency filter. The Supremes are a bit pricey, but I only needed a 2.7uf so not too painful, its the big one, with the two Mundorf Lytics behind it. The three small Ansars on the right are 1.5 + 1 + 1 = 3.5 (3.48 actual). The Ansars on the lower left should make 21uf and are 8 + 6.8 + 6.8 = 21.6 (21.29 actual). The Lytics 47 + 47 = 94 measured 95.4uf. The 2.7uf Supreme measured 2.71, so all values are very close to ideal.
I replaced the bell wire between the binding posts and the x-over with van damme 2.5mm square OFC, I soldered it to the back of the posts.
I left the internal wire as standard, as best as I could measure, the bass driver is wired with 15 AWG which should be up to the job.
hi ken,
did you use plain or raw for the ecaps?
I used Plain Mundorf e-Caps, which are low loss bi-polar electrolytics, for the large parallel values.
Note that I have had even better results using Janzen Superior Z-Caps instead of Ansar Supersound/Mundorf Supreme, for the Polyprop positions and this is what I have in my latest passive crossover. The top end is just a bit cleaner and slightly more detailed from the mids upward. :)
I used Plain Mundorf e-Caps, which are low loss bi-polar electrolytics, for the large parallel values.
Note that I have had even better results using Janzen Superior Z-Caps instead of Ansar Supersound/Mundorf Supreme, for the Polyprop positions and this is what I have in my latest passive crossover. The top end is just a bit cleaner and slightly more detailed from the mids upward. :)I used Jantzen Aluminium Z-Foil caps in my Apogee crossovers. They are good, as Humble asserted. TBH I may swap them out for Jupiter copper foil but they are huge and won't fit in the current box.
Jantzen coils too. I really rate their coils they look a lot nicer than the Mundorf ones I am also using. Great quality for the money. Better than Danish bacon, by a good margin:)
Respect for messing with your crossovers. Have you tried modelling the crossover in a simulation package?
I was very impressed by the Jantzen coils I used in my NS-1000s. If you look at the crossovers of commercial high-end speakers you will often see Jantzen coils being used. For mid I used a combo of ClarityCap ESA and Mundorf Supreme paralleled together. The tweeter is all Mundorf. This combo sounds suitably airy with no enhancement of sibilance and a detailed presentation, lacking in unnatural edge.
I used predictive software on the crossovers and also measurements to do the final tweaking, apart from the slopes being steeper for the mid/bass its more or less stock passive design. I used active for the mid/bass hand over on my final version.
I too have used Jantzen air coils, on this and several other projects, I rate them highly.
After reading several cap comparison articles, including Humble, I changed my Mundorf Supreme on the Tweeter for a Jantzen Superior Z-Cap, it is slightly smoother/less sibilant with a smidgen more detail, top end is its strong point, we are talking very small differences, but easily noticeable on torture tracks.
tpaxadpom
18-12-2018, 01:14
https://i.ibb.co/RQLP18c/20181215-110552.jpg (https://ibb.co/RQLP18c)
Picked up a week ago. Transmission line acoustic enclosure. I would like to build a bracket/box for tweeter and midrange drivers to figure out if it is worth investing money into. Really like how it sounds as is in my garage. Haven't had a chance to test it in my dedicated 2 channel setup (needs some serious cleaning before I will take them inside).
My experimentation with the NS-1000M has concluded.
My final and best results were with a semi active set up, which can be seen here:
https://theartofsound.net/forum/showthread.php?66160-NS-1000M-Upgraded
Please Lock thread.
Powered by vBulletin® Version 4.2.3 Copyright © 2025 vBulletin Solutions, Inc. All rights reserved.