Auto class A switch point (A-700)

muovimies

Super Member
I recently got my hands on an A-700 and giving it a look over I adjusted the auto class A switching point, but I used 6.8ohm load as opposed 8ohm by accident, which I think means I'm now trying to squeeze about 18W of Class A out of the thing as opposed to 15W - I adjusted the switch point for 11V ( @ 6.8ohm load meaning 18W) vs 11V (15W @ 8ohm load). That difference doesn't seem that significant to me, though taking losses into consideration things do add up... But not knowing how the circuit works, what triggers it, maybe it still gets triggered at 11V output no matter the load? Gah... I suppose I'll make it right the next time I open it, but I suppose in my normal listening I won't get past 10W really anyway.
 
Just set it correctly.
8 ohm loads aren't hard to find, and having the switching point incorrect can cause clipping / distortion when there are peaks in the waveform.
Are you playing the 1kHz test tone and setting the VR where the test point crosses between +/-16V?

If you are intending to use it regularly in Class A mode I'd strongly advise re-touching all solder joints (ideally with fresh solder, at least on transistors - reflowing the old stuff alone tends to re-crack faster in my experience) and ensuring very good ventilation around the unit. In fact if you're running it in Class A for extended periods of time (over ~45mins) I'd strongly recommend leaving the top cover off. In my honest opinion, these are incredible sounding amps, but they are under-specced from a thermal standpoint for their large heat generation - in particular the DE/FR model which has a plastic grille insert on the top cover - which in my opinion gives less airflow than the "regular" perforated steel version on the US/CA/AU models. The main outputs themselves will handle the heat just fine, but the problem is the cover traps in the heat and the other components on the board (along with the solder joints) do not fare well as a result. Along these same lines - if running regularly in Class A - I'd recommend setting the bias on the lower side of the 15.5 +/- 1mV recommendation. This bias setting (in AB) affects the Class A bias. Mine's on 14.5mV L/R stable in AB at the moment. I also wouldn't worry about the bias transistor seeing increased airflow - the bias transistors seem to have less influence on bias current on the A-700/A-1000 compared to the M-XX amps - keeping the main board / components cool is more important.

It's probably also wise to do the regular Yamaha tasks - removing the DSBG and cleaning up any corrosion damage. I think a recap is a wise idea on these units - in my experience the caps that will generally be out of spec are the 4x 330uF (Nichicon MUSE) and 2x 680uF (Nichicon MUSE KS - dark green) if the amp still contains all original caps. There is glue under all of those that you should clean (if not already sorted). There are also 4x 1000uF Nichicon Muse that are commonly out of spec - 2 on the main board and 2 on the EQ amp board (next to the volume control). These caps are all super easy to get to (except the 2 on EQ board - but even those are quite easy). All the smaller ones are usually in-spec so replacing those is non-critical.

Some of this may seem excessive (feel free to take or leave what you like) but I thought it would be good to share my experience since I run one as my daily amp almost purely in Class A mode, which I don't think many owners of these do. I think they are very capable of doing it for extended periods - and if you have revealing speakers there is an appreciable sonic improvement (noticeably deeper and wider soundstage) - but you have to improve the thermal situation if you wish for the amp to have a long and healthy life.

Enjoy!
 
Last edited:
Thanks a lot for the detailed reply.

Yes I set the switching point according to the instructions on the service manual using a 1kHz test tone for input, but I used 6.8ohm resistors as a load. I didn't realize my mistake until later when I already had the cover back on and everything set up and had a 'doh' moment. The thing is it was probably on spec originally because it switched at 9.3V volt output - as 9.3V/6.8ohm roughly equates to same current as 11V/8ohm - but I don't know how the circuit works so it might be a coincidence :) Didn't notice anything funky in class A but I figured I'll just use it in A/B until I set it properly, luckily it was not a difficult job. Unfortunately I don't seem to have any 8 ohm resistors laying about. It might be this amp was checked at some point fairly recently, because the bias was within spec for both channels and I've rarely found that to be the case with amps this old. Or maybe they just used high quality trimmers.

Mine is an European model with the plastic grille. A bit surprised there's no side vents on the top cover, having some on the right hand side I think should improve air flow greatly, it seems it would create a very natural input path for cooler replacement air. Replacing the caps you mentioned would seem like a good idea. So far first impressions are quite positive, not awestruck like I was with the A-520, because I had very low expectations for that one and in turn it created very high expectations for the A-700 :) But you were right, the A-700 definitely cured the issue with bass I was having with the A-520 - the bass goes deeper and is more articulate. An otherwise more controlled sound too, after the A-520 the A-700 sounds almost too controlled sometimes if that makes any sense. Highs are better, especially in class A mode, in comparison the A-520 is bit coarse. Phono stage is noticeably more quiet in MC mode, the A-520 could get a bit hissy at higher volumes. Other than that not a big difference between them. I'd imagine pretty much any 80s op-amp based design could be improved significantly by finding modern replacements for the ICs.

Based on these two models I feel pretty confident in saying Yamaha did an excellent job on both the A00 & A20 series. With the earlier series having the better build quality unsurprisingly.

Btw. have you replaced the speaker relay on any Yamahas of the period and if so, do you think replacing a seemingly fine relay would be worth it? No issues but they are getting old at this point and it could be one of those things you don't realize what you're missing until you do it.
 
No worries!

I think generally all Yamaha products (especially around '78-'86 or so) sound great and strive for the same "natural" sound, so any differences are going to be small. I think the differences do add up though, and on top of the things you mentioned, I found the biggest difference (particularly in A mode) was the sound staging / spatial representation in the treble. I personally find these amps pair very nicely with any B&W speakers (both very old and very new), but particularly the newer models with the S2 Nautilus tweeter will exploit the awesome spatial capabilities of this amp. It certainly is a very clean and accurate amp and I certainly know what you mean (I own both too). Once you get used to the sound it's hard to go back in my experience.

Certainly I think the top cover is the death of these amps. I think the heatsink itself is sufficient and the power supply area is strong and generous, but the cover just traps the heat in. It's obviously not a problem (at all) in AB mode, but I think in A it's a much happier amp with some airflow to stop the heat pooling on the main board. I use a ~silent/slow fan for mine to help. My first one blew up from being left on overnight in Class A and overheating (EU model also), so I've been much more careful with the 2nd one and it's been running great. It has definitely been a learning experience, but has been worth it.

I don't think there's any need to replace the relay unless there appear to be problems with it, the stock Panasonic/Matsushita item is of high quality. You can clean the contacts by gently leaving to soak with de-oxit applied (only to contacts) and using a very fine emery cloth on a chopped up credit card or something to slide/buff along the surfaces. The plastic cover pops off. Matching replacements are available though.

I found the same as you on both of my 700s - the bias setting was correct and accurate, neither had been touched for a while. They seem to hold their adjustment well once set. The trimpots are the same as any other Yamaha amp of the era, not the greatest quality, but personally I've had no problems with them (I de-ox and clean them before setting). I have now replaced mine on all Yamaha amps though (from recommendation of other AKers) with sealed Vishay or Bourns sealed trimpots. Multiturns offer more precision but the single turns are good too.

I can collate and share my parts list later if you like, I meant to do it at some point anyway. I managed to cut down the number of liquid electrolytics substantially by replacing them with more modern caps which should keep filtering in-spec much longer (and with lower ESR).

I don't have any experience replacing the ICs (I don't use the phono stage - they only affect that right?) but I believe you are correct that it can be improved - AK user "Karl vd Berg" has talked about this or done it (maybe on this amp) some time ago. I vaguely recall he posted his suggested replacements in a thread also.
 
Well the heatsink itself looked pretty substantial like you said, but the U-shape it forms seems like the perfect way to heat everything on the PCB in the middle of that shape. I see in the A-720 they changed to a more conventional straight heat sink (a lot skimpier one though, it seems from photos). If I end up running this a lot in class A, I'm thinking at least drilling some holes to the right side of the cover and maybe doing something about the top grille. An upward blowing fan in the top grille would be one option, but not sure I want to go that far. The good thing of this example being not at all minty externally is I won't feel bad modifying it.

Replacing trim pots I think is always a good idea and multiturn makes for a lot less pain in dialing in adjustments, but unlike some other amps I've had, the ones Yamaha used don't seem too prone to misbehaving. Some of the worse ones get into such condition with age that all you have to do is to give the amp a little bump to make the values go all over the place.

I think a parts list would be awesome if you already have one put together - at this point I think I'm just going to replace the trouble prone caps you mentioned - but I'm sure it would be a great reference for many.

Regarding the phono stage, yes, there's an op amp that doesn't affect any other inputs, it's taking care of the basic gain for MM cartridges, the additional 20dB or so for MCs is provided by a pair of FETs. I took a little peek at the schematics and it seems there's another opamp in the tonecontrol section. In the later A-520 (& it seems A-720 too) the tone controls are in the feedback loop for the main amp, instead of having a separate "tone amp" section. I think that's the better design and my ears seem to agree with that, I'd say having the tone controls in circuit makes next to no difference in the A-520 when set flat, while with the A-700 the tonal balance stays the same, but there's a noticeable difference in clarity for the worse. Too bad I looked at the schematic before listening critically to the tone controls, so I had some bias going in...

Out of curiosity I checked the service manual for A-720. There seems to be no separate adjustment for anything related to Auto Class A - in the service manual Class A is just covered in the biasing adjustment - 5mV at test point for A/B and 60mV (+/- 30mV!) with Class A engaged. I don't understand what the 'switching point' in the A-700 is for, looking at the schematic there's somekind of active circuit with an op-amp, I'm not sure but maybe the 'switching point' adjusts gain for the op-amp which sets bias when Auto Class A is on? I'm curious but over my head here... and now I wanna hear an A-720 to see how it compares with the A-700. I'm kind of under the impression most people prefer the A-700 though.
 
Rest assured the A-700 is the superior amp. I wouldn't bother with the A-720 if you have the A-700. The lack of bass control from the A-520 will return, and the spatial depth/width will be reduced. You'll notice the big MT-200 devices are gone too, replaced by a larger amount of smaller devices. This was the beginning of the cost-cutting era for Yamaha.

I don't use the tone controls. If you're zero-ing them you can use the "direct" switch which bypasses the entire TC board (I'm sure you already know this - just unsure why you wouldn't just bypass the lot if you are concerned by their influence at zero position). I certainly think the amp sounds better with the direct switch engaged, and in my experience these amps are very neutral and accurate so if there is a need to use tone controls I'd be looking sternly at your speakers and/or room acoustics.

The switching point adjustment was removed in the later A-720/1020 models (the M-XX amps don't have this either). What it adjusts (to my knowledge) is the crossover point from Class A to Class AB. Obviously the PNP and NPN output devices reproduce half of the output waveform each, and in Class A since the bias current is increased substantially, the headroom before clipping would occur is reduced substantially (but obviously is still sufficient for most real world listening). The "Auto Class-A" system results in any "overrun" of the waveform being reproduced by the other device. Having this adjustment is cool because it means that the Class A operating range can be maximised accurately while avoiding running into the clipping range. Aside from this, of course you can still check the Class A bias at the same bias test points on the A-700/1000.

I'll put together a parts list a bit later. I recommend Nichicon (like stock) for all those caps mentioned, KZ works well. For the 680uF you'll need to use Nichicon PW or KG "superthrough". I think it's wise to use a higher voltage rating provided fit is OK.
 
My A-700 will be getting a recap and new ICs for the phono section soon. I also discovered a very random scratchy noise issue that's probably a bad solder joint somewhere, or the magic glue has rotted something away - occasionally the amp gives out a faint scrathy sound if bumped, it's not very high in volume and the volume control has no effect on it. Usually the problem goes away giving it a couple light taps, after which further tapping will not generate any noise. I'll try to find a precise location for the issue once I've got the amp opened up on the table, hopefully I can track it down.

If you can remember, what's the easiest way to get to the EQ amp board? Maybe it's self evident once I dive in there, but in case it's not some tips might save me some headache.
 
If I remember correctly, you can remove the input selector board to access the top of the EQ board (some screws at the front after removing faceplate), and the bottom panel to access the solder joints. The volume control can be removed by unscrewing the nut at the front which gives better access to the EQ board also.

Regarding the scratching noises, it may be a bad solder joint like you said, but the culprit is often the balance control (still in-circuit with tone bypass engaged); de-oxiting it should help. I'd do the same treatment on all other switches / pots at the same time.
 
You remember correct :) That's how I did it but I have to say it wasn't fun working in there with all the cabling etc. getting on the way in the tight space. I did remove the side supports as well so that I could get the whole 'front assembly' a bit further out, as much as all the cabling allowed. To gain unlimited access would've meant a lot more disassembly, but I managed to squeeze the necessary parts in. I changed the phono stage ICs and electrolytics. I also did a full recap on the rest of the amp, except for the 2.2uF caps because I had accidentally left them out of my order. Everything else is new now, Nichicon Muses (KZ) mostly and KT (high temp) for all the small caps near the heatsink. Also for the 330uF caps, not sure why I had ordered KTs for those, but that's what I did so in they went. The higher temp rating is surely a plus, but it seems a position where it would make sense to go for SQ premium :)

The recap did improve sound noticeably though not in a dramatic way - ie. it's not like a whole different amp or anything, it sounds the same but a bit clearer all over the spectrum. The phono stage upgrade was huge on the other hand, I don't know if it's more down to the new caps (KZ for the 1000uF and ES for the bipolars) or the new op-amps, but it's one or two steps above what it was. As it was, I prefered the PhonoBox SEII over it, but after the upgrade it went totally the other way. It retained the nice & coherent sound it had, but got rid of the 'meh' dynamics and lack of life and is more detailed and revealing as well now. I went with NJM2068 for the opamp because it was the best I could find in SIP8 casing. It no longer masks the benefits from Class A mode.

Didn't really find a reason for the scratchy sound, I checked parts for glue damage but apart from a little surface oxidation on one jumper and leg of a resistor there wasn't any - the damage that was there was so minor I don't think it would affect anything, it was really just thin surface layer on the metal that cleaned up nicely. I also examined the solder joints and did find a couple large suspect ones that I redid. One of them was on a ground strip and I suspect that might have been the culprit. The problem hasn't occurred since the recap, but it was very random before as well, so it's possible it's still lurking there somewhere, but nothing I'd loose sleep over. I've done the pots and the balance control isn't & hasn't been scratchy at all, so I think it's fine. The poor ground connection could fit, the noise was kind of faint and volume control had no effect on it.
 
Good to hear about the improvements and that you're happy with it.
Where'd you get the NJM2068, and is it just 1x needed (on the EQ amp board)?
I have a dead phono stage on one of my amps, maybe that's an excuse to fix it even though I don't use it.

Do you make use of the Direct / Tone Bypass switch? I find it helps overall SQ.

I've been using Panasonic OSCON solid-electrolyte capacitors for a few builds recently and have found them to be excellent. I've managed to reduce the number of liquid electrolytic capacitors from 48 down to just 8 on this amp (or 10 including the 2x power filtering caps). I also added two large heatsinks for daily Class-A use (amongst some other small changes).

A_700_oscon_2.jpg
 
Last edited:
Yep, I got the plain L-version from Mouser. The stock amp was SIP9, but pins 2-9 are identical to SIP8 1-8 - so just leave pin 1 empty and install the SIP8 IC on pins 2-9. In the SIP9 versions pin 1 is a double for Vcc and 1&9 are shorted on the PCB. I hope that made sense. That dual op-amp handles the audio duty for both channels, there's another SIP8 one on the board near the big caps under the volume control, but that's a dc servo for the other one and shouldn't affect sound quality. The difference in sound is pretty night and day in terms of detail and dynamics, stock it sounded very lackluster in comparison, but drifted caps might have something to do with that as well. If there's any downside to the new op-amps, it's there's a slightest hint of metallic quality to some upper frequencies, but it's subtle and it's like a grain of sand compared against a sand castle of improvement :D

And yeah, I run the amp in direct mode basically all the time, I only touch the tone controls for some very old recordings to make them more listeneable, but I don't have that many of those.

Those OSCON caps look interesting, are they more durable than normal electrolytes? Do you think they sound subjectively different? I think I need to look them up. In hindsight maybe it would've been wise to replace at least the 0.47uF caps with film. I also looked at the 4.7uF ones but getting them in film would've been kind of expensive and they tend to get large. But 'expensive' is kind of relative and with the amount of money we are talking, I do think the biggest expense is the time invested in changing them, so... in any case in the end I just decided to go like for like, I figured that realistically, if the originals gave 30+ years of service, these new ones should pretty much last until the end of life for the amp. I could still go film for the 2.2uF ones, but I don't think those positions have much of an effect on sound. I don't have a cap meter, but getting a cheap one would have it's uses at times and it would be interesting to measure all the original caps just out of curiosity.
 
Thanks for the details and link, I'll grab a few on my next mouser order.
SIP8 pin thing makes sense, cheers for the explanation.

Those OSCON caps look interesting, are they more durable than normal electrolytes? Do you think they sound subjectively different? I think I need to look them up. In hindsight maybe it would've been wise to replace at least the 0.47uF caps with film. I also looked at the 4.7uF ones but getting them in film would've been kind of expensive and they tend to get large. But 'expensive' is kind of relative and with the amount of money we are talking, I do think the biggest expense is the time invested in changing them, so...
Yeah they have a longer lifespan (5000h) and higher temperature rating (105*C), but the other reason I used them is that they have significantly lower ESR than liquid electrolytics. I think subjectively in terms of SQ they sound slightly cleaner but the difference is small.

I used Wima MKS2 wherever possible (0.47, 2.2, 4.7, 10uF from memory) and Panasonic ECQ-P for the 2.2uF 100V (C227-C230).

I think using Nichicon electrolytics everywhere is absolutely fine though, and just a general recap (pre-OSCON) did make things sound cleaner across the board just like you said. In my case the amp is almost always in Class-A and sees heavy daily use so I just wanted it as a durable as possible.

In my opinion (re: temperature ratings), it's far better to give the amp good ventilation / leave it open in Class-A mode than just using 105*C caps. I feel 85*C caps are fine, and that good ventilation is more important (for solder joints and other components).

One thing I'd like to try in the future (something Oilmaster suggested may offer a minor audible benefit) is replacing the fluoro green caps often across fusible resistors (and maybe the resistors themselves as he also suggested, they do seem to drift in spec a little). I believe the stock caps are polyester/mylar, whereas ECQ-P are polypropylene. I'm sure the improvements are marginal/nonexistent at this point but it's part of the fun to try. :)

I don't have a cap meter, but getting a cheap one would have it's uses at times and it would be interesting to measure all the original caps just out of curiosity.
You could get a cheap one off ebay, it's called the Mega328 (~$12). Something I found interesting (and annoying) is that virtually all new Nichicon caps measured below their labeled value, eg ~920uF for a new 1000uF KZ cap. I know that's within spec, but on a few occasions even the ESR was higher than the original 30y/o part (I found this on those Nichicon ES caps for the EQ board). I'm sure it's no big deal, I used them anyway.

For whatever it's worth, I found on Yamaha amps with (large) factory-fitted Blackgates, they still had capacitance higher than spec, and ESR lower than any replacements I could buy. Perhaps the BG cult following was justified!
 
Last edited:
Well, I think I might want to go over some of those caps again and try different brands or going film... I first replaced all the larger electrolytics (330-1000uF, maybe the 100uF ones as well) and listened to the amp some. I recall some improvement but quite subtle. Replacing all the small caps near the heatsink area (0.47, 4.7 & maybe some 10uF or so if I recall correctly) and I noticed a bigger change - I wasn't expecting much difference from these caps, but considering for the others I replaced "audiograde" Nichicon with audiograde Nichicon, and here I replaced some unknown caps with audiograde Nichicon, maybe it makes sense. The originals had no branding on them but they looked like maybe they could be Nippon Chemi-cons. The change was most obvious in bass & treble as an increased clarity and sense of "control" - however what I also noticed was a slight metallic quality in the upper midrange somewhere, most easily heard on male voices gaining a bit of a robotic quality to them. I first thought it was the phono stage, but actually that doesn't seem to be the case. Like I said it's subtle and part of it seems also just about being revealing of source material - it really depends on the recording and analogue era stuff is less likely to suffer from it, though some do. It did improve after a day or two (at first it was apparent on pretty much all vocals, now only some) after that I haven't really noticed further change. I'm pretty sure it's not all in my head :)

Those caps I replaced with Nichicon KTs and now I'm thinking maybe they are the culprit but on the other hand switching new caps for different new caps would seem a bit silly. I'm only even considering it because I'm trying to get the absolute best out of this thing and because I'm curious. It's slightly more noticeable in class AB than class A, so to go on a limb and guess, it might be harmonic distortion I'm hearing (3rd order perhaps - I've found that get's audible at very low levels as a sort of metallic 'zing'). It's possible it was always there but it was just revealed from the increased clarity from the recap. Not sure yet if this is something I'm going to try to do anything about - in any case I think I'll give it more running time first. Overall I think it's definitely a better sounding amp now than it was before the recap.

I agree with you about ventilation and it's not like the caps are going to heat anywhere near 85*C even, I mainly went for the 105*C ones because they should have longer life. I haven't yet done anything about ventilation, I'm thinking making a completely new top-cover from some aluminum mesh screen could be the way to go. I might get in touch with a local metalwork school and ask them if they take commissions like that. I could make a rudimentary one myself I suppose, but they'd have all the tools and could probably make a 1:1 copy of the original, just out of mesh screen. Having a screen with around ~50% open area I think would basically equate to running the amp completely without cover. In the mean time just removing the plastic screen cover completely I believe provides good ventilation, but it's not a very good looking solution. It's getting a bit involved, but on the other hand why not and it's a good amp.
 
I think the mesh screen is a good idea, post some pics if you go through with that. I run mine open with a low rpm inaudible fan pointed at it, but it's in a cabinet so it's not visibly obvious that it's open. I always thought it'd be nice with a neat mesh screen raised above the amp itself like the Yamaha M-4 but it'd take some work to make neatly.

Like you I found most caps do have an audible change over the first few days before they settle into how they'll sound permanently, and I found different caps take different amounts of time to settle in as well.

I just documented the capacitors I've used in past A-700 builds, notably replacing most liquid electrolytic caps with solid polymer (Panasonic OS-CON) and polyester film (Wima MKS2) caps. I've also replaced the stock mylar/polyester film caps with polypropylene items (Panasonic ECQ-P / ECW, and Wima MKP10 / MKP2 / FKS2). They are in a google doc here, I'll continue to update in the future as I rebuild more of these amps.

Some minor notes:
  • This documents all capacitors on A-700 main amp board (except ceramic and polypropylene, which can stay), and electrolytics on EQ amp board. TC board not documented since it's disabled with "direct" switch.
  • I've used higher voltage ratings for most caps where possible
  • C233 is an unmarked cap on the main board diagram just above C203 (both 0.47uF 50V)
  • For C179-C182 I used 120uF instead of 100uF as they tend to drop in capacitance over time (120uF is still within 20% spec). Optional.
 
Last edited:
Just a little update, the amp has seen a few dozen more hours and the slightly metallic tinge to some voices I was hearing seems to be gone. Sounds really good. Either I got used to the sound or the caps broke in, I'm thinking might have been a bit of both, but I'm sure the sound has also changed, not just my head :) Just as I was getting happy to stay put with the A-700, an Onkyo Integra A-8800 (same as A-8690 I think) fell into my lap for a price I couldn't refuse. Haven't hooked it up on the main system yet, but seems a promising piece. A top model from the late 80s (apart from the "Grand Integra") w/ a built in DAC, weighs about 32lbs. Similar power rating to the A-700. Built quality is definitely a tier up from the Yamaha as far as look and feel go. One of the nicest feeling volume pots I've ever encountered with a weighty knob and just the right amount of smooth resistance :) Quick impression is the Yamaha is certainly more light on it's feet, which seems like a Yamaha characteristics... not sure how to describe it, it's not that they sound bass light or anything, just 'agile' and open. The Onkyo seems more thick and meaty. But like I said no listening with the main speakers yet, gonna open it up first, check bias etc.
 
Back
Top Bottom