View Full Version : Accuphase M-60 Love
How can you not love that?
What's more, when I put these monos up against a Luxman M-800A driving Apogee Duettas, they come out very well indeed. Better channel separation, higher power (half a Kilowatt into 4 Ohms).
But it's poorly.
Can I fix it? We will see.https://uploads.tapatalk-cdn.com/20260206/fdfb5ce29531aec31905b12af2541cbb.jpg
First off some Gemini AI derived info:
Based on your serial number K8X548, we can break down the vintage of your Accuphase M-60 quite accurately.
Using the classic Accuphase dating convention for that era:
Serial Decoding: K8X548
* K: This is the model series prefix used for the M-60 and other units from that specific mid-to-late 70s production run.
* 8: This represents the year of manufacture. For this series, it corresponds to 1978.
* X: This is the month code. In the Japanese system of that era, Accuphase often used X, Y, and Z to represent the final months of the year. X typically stands for October.
* 548: This is your specific unit's production sequence number.
Your amplifier was likely manufactured in October 1978.
Why this matters for your repair:
By 1978, the M-60 was at the peak of its production cycle (it was released in July 1975). This means your unit benefited from any "running changes" or minor circuit refinements Accuphase made during the first three years of the model's life.
https://uploads.tapatalk-cdn.com/20260206/a3bc511f6443740afbfd16b79a4073a3.jpg
Soft start circuit. This supposedly 2.2 Ohm resistor is date coded 1993. Somebody replaced the original ceramic resistor with this. Granted it has lasted for some years, but this sucker is the wrong type. That's a metal chassis mount resistor designed to use the chassis for heat dissipation, I believe.
As it is no longer to spec, the switch on relay, which you can hear when it works, no longer works. So l am going to try putting a new ceramic one in.https://uploads.tapatalk-cdn.com/20260206/eac2c29c1278292b94a16617b0cda509.jpghttps://uploads.tapatalk-cdn.com/20260206/8e1d18844951bccacdda0357410628f3.jpghttps://uploads.tapatalk-cdn.com/20260206/47a3aa65875cd5e1da74f238301359de.jpg
https://uploads.tapatalk-cdn.com/20260206/8ad8a80aeb262e24826a8afba185b09e.jpg
That LCR meter is ace. You can check an awful lot out with that with the amp off. £100 Ali Express.
Everything I have checked using it seems fine, apart from that resistor. Basically I am surprised at just how 'on spec' these old components seem to be, including the caps. They obviously used good shit.
That LCR meter is ace. You can check an awful lot out with that with the amp off. £100 Ali Express.
Everything I have checked using it seems fine, apart from that resistor. Basically I am surprised at just how 'on spec' these old components seem to be, including the caps. They obviously used good shit.
Funny that I was just admiring your meter :)
You can't beat the old amps, so much better than a biscuit tin with a power button and a blue LED.
Resistor in. Did it work? Nope.
Last time I switched it on the circuit breaker tripped. Need to check that is still working. The switch on relay could also be roasted I expect.
But not today.
https://uploads.tapatalk-cdn.com/20260206/7e6921b7e59a1d566e8094ecd12f6f45.jpghttps://uploads.tapatalk-cdn.com/20260206/d133729871219d146c9c26e3674ddf37.jpg
Funny that I was just admiring your meter :)
You can't beat the old amps, so much better than a biscuit tin with a power button and a blue LED.TBH I don't like the colour of modern Accuphase kit. At all. The silver facia of the M-60 is nice, though, I think, as is the simple styling.
Meters are lovely to watch in the dark and when cranked up it is quite disturbing to see how much the Duettas will drink.
Quick continuity check on that circuit breaker says it's OK. Hm.
Got it. For some reason that resistor broke. I bought 4 so will try another.
Edit: no I won't. I found the part in the manual it should be 20W not 10. Gemini had told me to order 10W but has admitted it was wrong.
Suspecting a relay fault or cap failure that is used as a timer to change its state, or the resistor itself. Maybe. Just guessing so far.
Resistor is part 15.
https://uploads.tapatalk-cdn.com/20260206/6260266dcfa13299aabac650680c1a2e.jpg
This monoblock had a bit of a mash up in power transistors. No way on this planet were these matched. TBH I think one of the types in there wasn't right anyway - the AIs didn't like it. Can't remember which one it was.
So about a week ago I slammed a new set in. Photo of one side only for these. The AIs had better be right about these. Both Gemini and Grok think it is right.
I have also located the soft start delay cap plus two related transistors in the service manual to check later. Be good if something was obviously wrong with these items cos that probably means I have it sussed.https://uploads.tapatalk-cdn.com/20260207/24575e7ead8ca01841f83caed68fd277.jpghttps://uploads.tapatalk-cdn.com/20260207/74d1049174adbc3d4ca0db4aaf9a840b.jpghttps://uploads.tapatalk-cdn.com/20260207/ce33f5efe17931e1989674321c01e49f.jpg
Sussed it. The yellow ITT cap is fine I was pretty sure I had tested all the caps. It's the 4.7uF soft start cap. The two transistors below it - one is definitely knackered the collector / emitter is not open circuit, which it should be. The other one is fine.
These two transistors control the relay state depending on the cap charge state.
Degree of certainty is very high I would say. Probably a good idea to do all the transistors on the board in one go.
I spoke to two electronics repair guys and they would not touch this amp despite trying to talk them into it.
Fault finding this really didn't prove difficult, so their loss I guess.
https://uploads.tapatalk-cdn.com/20260208/b363adf3c10d6cf53c7fcc9a435901ca.jpghttps://uploads.tapatalk-cdn.com/20260208/2abdcfb7ceb5760a4b4e4e0c80dc6515.jpg
I wonder what sort of repairs they will do?
I mean it's an old amp, you have the schematic, there's no surface mount, it's all neat and tidy. Is there some reason I don't know?
They were both Weston-super-mare based.
One guy wouldn't touch it because it was high end and it weighed too much. He did sound pretty old, though.
I'm struggling to remember the other phone call, but I know it was the second time I had called him (the other was for the 211 amp fault) only to be turned away again.
I could have tried Alan (Firebottle) but I just decided to get into DIY figuring it'd be cool to sort stuff out myself.
There's a high likelihood more failures will occur with the M-60s and I just don't want to sell them. So it's probably a good decision long term.
Makes sense. I wouldn't have the courage.
I wonder what sort of repairs they will do?
I mean it's an old amp, you have the schematic, there's no surface mount, it's all neat and tidy. Is there some reason I don't know?
Point well made
Firebottle
09-02-2026, 08:41
You're doing brilliantly Justin, well done.
:cool:
Thanks, Alan:)
To make it clear to others the resistors that were failing are only designed to take strain until the relay clicks after 3 seconds. Cos it never does they get more than they can handle, so 10W or 20W ones will fail. A cheap fail-safe in a way, I guess.
Ordered some bits. These suckers are definitely not dying on my watch.
Been looking at an Accuphase DF-75 active crossover. One on eBay for 6K. I reckon I can build one with an 8 inch OLED to control various parameters with a Raspberry Pi 5, plus some A/D, DSP and D/A boards.
Most of the time will probably be spent writing the GUI.
Cost probably £1K and absolutely bespoke to my needs. Performance probably close to the Accuphase.
Finding the time is another matter.
OK whilst I think that protection board is knackered as described and I now have the parts to fix it, I thought let's put the protection board in from the working amp with a new 20W resistor.
Result? The resistor blew again.
The relay the protection board controls is this one, and it clicks almost insanely loudly 3 seconds after power up in a working amp.
https://uploads.tapatalk-cdn.com/20260225/4f7689b3bac1acad1d3dd995f75f926c.jpg
I basically misunderstood things a bit. There's another relay right by the power switch that is connected to the resistor that keeps blowing. I removed it and it is unquestionably broken i.e. the coil is open circuit.
This relay is immediate on power button depression. If it doesn't do so it fries the resistor.
Just ordered another from eBay for £10.
FWIW the Alps picture is the on/off switch which has a really nice feel to it.
https://uploads.tapatalk-cdn.com/20260225/31d35c26c1a456d7d5d018d14adc1df6.jpghttps://uploads.tapatalk-cdn.com/20260225/6e0e18f44548354a140651f7a87cb68c.jpghttps://uploads.tapatalk-cdn.com/20260225/d4e75cd95d0ad84880d9ab22ed06a27e.jpghttps://uploads.tapatalk-cdn.com/20260225/ea98e883c18b5b226084c3c9be839559.jpg
Looking at the two protection boards from both amps it's quite obvious they have had quite a bit of work done on one of them over the years as the components are not consistent.
While one of the transistors is definitely broken as I identified earlier in the thread, I find it very hard to believe both that and this o/c relay failed at the same time. Need to double check if that failed transistor really is doing what I thought it was.
I'm enjoying watching you progress with this.
Lovely to see the engineering involved in the build of the amp and makes a great case for these big old behemoths of amps that parts are relatively standard, easily replaceable, service manuals available and it's not a case of bin the SMD boards and buy whole new PCB.
Once up and running do you intend to replace all the lytic caps on the PCBs? I'd have guessed that at least some have dried out by now due to heat?
All the caps measure fine. If it ain't broke don't fix it.
The huge ELNA reservoir caps all measure almost 33,000 uF. Way over the rated spec of 22,000uF. To replace them would probably mean even higher capacitance for the same size caps these days, but that's a dangerous game as it may alter the sound more than you might think.
The best ploy, I think, is just to fix stuff when it breaks or measures out of spec.
All high power transistors will be new OnSemi devices, though, with better specs than the originals. It's only £100 for a set for each mono, just need to do the other one. This has to be done as the mismatching and wrongness is unacceptable to my mind. If they sound like shite, I'll just track down a full set of NOS Toshibas.
What does amaze me, as you point out, is you do appear to be able to acquire all the bits even in 2026.
The fact that the relay replacement was only £10 on eBay is mind blowing. I was expecting around £50 cos it looks nicely made.
Notice the Hitachi cable in the relay. That had me looking up Hitachi hi-fi history. Love the copper, too.
I am assuming the £10 is just over stock of old components nobody really wants anymore. Though that is a cheap listing even for eBay.
The mofo relay is definitely that larger OMRON, though. That's a beast. It's the load or loudspeaker relay, if you like.
Wakefield Turntables
27-02-2026, 19:21
Great thread, keep it up!
If you insist:)
The Cure.
No cool Hitachi branded cable:(https://uploads.tapatalk-cdn.com/20260228/b602d3a4224f09c6e2a243bc48a0fa0f.jpghttps://uploads.tapatalk-cdn.com/20260228/9053ddfb909922625bf52ef0cbc8431c.jpg
Had a bit of a panic seeing the Made In Indonesia logo. Was this a clone? No, apparently it is a legit place of manufacture for OMRON, though it is a Japanese company.
Showing an AI the side pic above it told me it was a fake and NOT to use it. Showing the picture below it told me it was genuine.
Never fully trust AI.
I am betting it is real.
Will start trying to put it in this afternoon.https://uploads.tapatalk-cdn.com/20260301/72647f0b872af2dc03f4970cdd88eaff.jpg
Doesn't really seem worth the bother of faking that.
Doesn't really seem worth the bother of faking that.Co-pilot says:
There are counterfeit LY‑series relays out there, but the important nuance is this:
Counterfeits exist in the general industrial market,
but there is no evidence of a widespread, organised counterfeit problem specifically targeting the LY2‑DC24.
What does happen is more subtle:
---
[emoji3456] 1. The LY2 is one of the most widely cloned relays in the world
Because it’s a 40‑year‑old, high‑volume industrial part, dozens of factories (especially in China) make:
- “LY2‑compatible” relays
- “LY2‑style” relays
- “LY2 pattern” relays
These are not marketed as Omron, but they look similar and are often sold on eBay/AliExpress as “Omron‑type”.
This is why your first photo looked generic:
it had the shape of an LY2, but not the branding.
This is not a “fake Omron” so much as a generic clone.
---
[emoji3456] 2. Actual counterfeit Omron‑branded LY2 relays are rare
Omron is not a high‑margin target like:
- Siemens PLC modules
- Schneider contactors
- Allen‑Bradley industrial controls
So there is very little incentive to counterfeit Omron LY2 relays with fake logos.
When counterfeits do appear, they are usually:
- Poorly printed
- Wrong font
- Wrong schematic
- Wrong pin numbering
- Wrong case moulding
- Wrong coil resistance
Your second photo shows correct Omron branding, correct moulding, correct schematic, correct font, and correct internal layout.
That is what a genuine Omron looks like.
---
[emoji3456] 3. Omron does manufacture LY2 relays in Indonesia
This is official and documented.
Omron has factories in:
- Japan
- Indonesia
- China
- Malaysia
So “Made in Indonesia” is normal for LY2‑0 and LY2‑D series.
---
[emoji3456] 4. The relay you have now is genuine
Your bottom‑view photo shows:
- Correct Omron logo
- Correct LY2‑0 model
- Correct 24 VDC coil marking
- Correct Omron schematic
- Correct moulding
- Correct internal armature and spring layout
This matches known‑good Omron LY2 units.
The earlier side photo simply didn’t show the logo clearly.
---
[emoji3456] 5. Summary
- Yes, LY2 clones exist.
- No, there is no major counterfeit epidemic for LY2.
- Yes, your relay is genuine Omron.
- Yes, it is safe to install in the Accuphase M‑60 soft‑start position.
---
Before you fit it, do you want to confirm the coil resistance and contact continuity so we know it’s electrically healthy before powering the amp?
Inflation adjusted Gemini AI estimates the current retail price would be £15-22K.
Food for thought.
Dodgy flash pic of the other side. I've posted these for reference in case some future dude is interested in doing the same.
The thermal compound I used is insanely expensive I think and you can see the excess which I will wipe away. This stuff doesn't dry out.
The new relay is inside the housing behind the resistor. However, it needs screwing back in which is very fiddly. Will do it tomorrow. If it stays powered up, then the DC offset needs trimming followed by transistor bias adjustment.
That ceramic resistor I think must be wider than the original sand cast item. I will probably add an insulator either side to be safe.
Also, I just used some existing mains cable wire I had hanging around to do the job (yellow and green stuff). This is not following convention, but that won't hurt.
https://uploads.tapatalk-cdn.com/20260307/b5d91c0e85c7c262f3142ee02e132e1b.jpghttps://uploads.tapatalk-cdn.com/20260307/401efd6cff968401a017c237eedb600a.jpg
Did it work?
Nope.
Ego bruised.
97% sure it was all plumbed in correctly.
Something else must be busted.
That's a shame but you'll track it down.
You have spare amps to use while all this is going on?
Yup Luxman M-800A and Austria Analogue 211 monoblocks.
It's a challenge. Will get there eventually.
Right. as long as your not slumming it then ;)
Nope. Really like swapping different amps in. Cures yearning for upgrades all the time cos you get fed up with an option.
This wisdom has been learnt through many years. Works for me, anyway;)
I have two sets of spare monoblock amplifiers I can use should my current amps develop a fault. I don't chop and change amplifiers just for the sake of it.
Barry don't you swap your alternatives in every now and then?
If not, why not?
FWIW this M-60 fix may not be fast. It may span over months. However, if I get some time off, expect rapid attention and a quick sort out. I haven't been giving it much time, really. An hour or two max per week.
Barry don't you swap your alternatives in every now and then?
If not, why not?
FWIW this M-60 fix may not be fast. It may span over months. However, if I get some time off, expect rapid attention and a quick sort out. I haven't been giving it much time, really. An hour or two max per week.
No, not now. I used to be a bit of an amplifier 'swapper', even to the extent of trying out classic amplifiers such as the Radford STA-15 III, or the Electrocompaniet "The Two Channel Audio Amplifier", both of which, in their own way, sounded excellent with my Quad 57 ESLs.
These days however, I'm quite happy to leave my Quad 510s in situ with a set of 'monoblocked' and upgraded (to Keith Snook Level 3) Quad 405s, and Mark Levinson ML-2 amps in reserve should the 510s develop a problem.
Currently just the 3 sets of M-60s available on eBay.
All the same pair. All different prices. All from the same 99.5% feedback seller.
Curious.
https://uploads.tapatalk-cdn.com/20260316/eb0a652c796e44c62da69cb5e852718c.jpg
Back on the job. Just trying some cross comparison unpowered checks between working and busted units. Now using Claude to try and suss it as Grok said it was best at electronics.
https://uploads.tapatalk-cdn.com/20260722/132c01224a814a9e897cca99922f6d05.jpg
Spent 3-4 hours cross checking these driver boards. There are some discrepancies in values between about 4 components I have duly noted.
I don't reckon they are massive enough to be the issue.
Logic is pointing to the transformer primary and indeed Claude thinks that's it. But my in-circuit super basic cross comparison tests say no.
The working mono still has the 2.2 Ohm 20W resistor from 1976 near that power switch, the one that keeps blowing in the non-working mono that had already been dogdily replaced before I started looking at the issue. That means this problem has happened before in all likelihood. Without replacing the transformer, or so it seems.
Gonna get the tranny covers off and have a look to see if one looks different to the other. I have another idea. Will the working mono work with the busted mono's protection and driver boards?
This is proving to be a really likeable problem:) I just love it when I haven't got a fucking clue what is going on;)https://uploads.tapatalk-cdn.com/20260723/d33517c91bc2d1bdef5f982cc89e39c2.jpghttps://uploads.tapatalk-cdn.com/20260723/ed26787e7ac147d54ac7c4170d3501d7.jpg
Wakefield Turntables
26-07-2026, 09:23
Great work Justin, been a while since I checked into this thread. I really hope that you managed to get these sorted. :thumbsup:
Not great work yet.
Claude is effectively very good at admitting it isn't sure about much. It's not the AI for this problem, despite what Grok said. And its analysis of circuit board pics is below average and it knows it.
Copilot has just come up with a great transformer primary test. I think it is onto something. It agrees with me that indeed it is not a primary winding short issue.
I'd explain it but it's a ball ache post that most would not understand or read but there is a notable difference between the two monos that is significant I believe.
Replacing a transformer is not a job I want to do. But I probably am mad enough to get some bespoke ones made if needs be. The issue isn't the money, more the hassle factor.
ATM it is not looking like it is needed thank god.
The red wires in the pic are secondary transformer outputs that should have 64.5V on them to feed a couple of diodes in the rectifier circuit. These should sum to give 129V across the diode terminals. In the working amp they do, in the busted one they don't - it's 0V.
Which implies the secondary is open circuit in the busted amp.
Fingers keep pointing at the tranny. https://uploads.tapatalk-cdn.com/20260814/ff3dadd4ac2e517f802f1dc21d3b8365.jpghttps://uploads.tapatalk-cdn.com/20260814/39bb47196b53a3eb4a8b89ad802540b4.jpg
Firebottle
15-08-2026, 15:49
That's a bit of bad luck with a faulty tranny, but at least you have found a fault.
How many other secondary windings does the transformer have, and what are their voltages? You might be able to find a replacement, or at least have one made.
That's a bit of bad luck with a faulty tranny, but at least you have found a fault.Yeah. I think it went from short to open circuit as it was blowing that resistor I was on about. But it isn't anymore.
The blue and brown wires are two more secondary outputs that seem fine, Barry.
I will take the tranny cover off tomorrow and prey it's an interface fault between the tranny and the red wires but I highly doubt that even if AI thinks it is likely.
Whilst the M-60s can dump a shed load of power into the Apogees, it really does get extremely hot doing so. It's probably been killed by repetitive heat/cool cycles. Or a transistor took it out, maybe. I will probably never know.
There was an 'event' yesterday when measuring the other working amp's diodes. Massive flash and burning plus soot left to prove it. Thought I had fried the amp, but the speaker protection switch fired open and it the amp seems OK.
Gotta be careful with those probes, LOL. I'll post a pic and you will see it must have been fairly extreme. It was;)
And yes, Barry, I'll get it rewired.
Diode fire mess. Check out the leads connected to the left hand diode and the burn on the base. Wish I had it on video it was impressive!
Also transformer cover picture. It is a UK voltage unit but it can also be set to run at 120V or 100V.
https://uploads.tapatalk-cdn.com/20260816/a0155e923d9f5223998c6a35d93a6373.jpghttps://uploads.tapatalk-cdn.com/20260816/3b6d773f5d2e4d01bd04d4f59375dfc7.jpg
One extracted tranny. Gonna try and get a quote from trans-amp tomoz.
It's full of potting compound. Makes it a bit more challenging I would guess.
I really didn't wanna do this, but I am confident I can get it back in once fixed.https://uploads.tapatalk-cdn.com/20260817/5958255892ae2dac3d5fd83614cdf662.jpghttps://uploads.tapatalk-cdn.com/20260817/af01c1f31e35e0b3539bac382f5d5fbb.jpg
I'm gonna say it I think Accuphase missed something here.
A 50th Anniversary re-issue of these son of a bitch amplifiers would have been an amazing act.
IMHO still outstanding in 2026 and I am so looking forward to getting them working again.
I'd be curious to hear a current P-7500 but it's not a monoblock. Still, it has an appreciably better spec. But massively so? Honestly? Not really.
Isolated, this transformer denies absolutely any existence of a fault as far as I can determine. Correct continuity and sensible resistance measurements on a really extensive set of tests.
In circuit and power on, I had the right AC voltage selection range on my multimeter telling me over and over again there was 0V on the busted mono versus 129-130V on the good one at the diode junction.
I did that so many times because I absolutely did not want to remove the transformer. And I just didn't think it was a likely fault.
It's definitely like living in a world where you just cannot rely on your senses.
I am starting to realise why no one is keen on working on these things.
Plus point is there is no need to send this thing away to get it fixed. But it is a bit of work to plumb it back in.
I should probably ask Alan (Firebottle) if he wants to get involved to be quite frank because I am rank amateur at this sort of stuff.
That said, I am learning stuff here. If it's not fixed in a month or three Alan, I will be in touch to see if you are up for it. If I can't suss it rest assured the pay will be good if not better than that.
I admire your willingness to have a go at it yourself. At least they're old enough to be all discrete so must be fixable but that's where having decades of experience like Alan has comes in to play. I think it's a lot easier to build them rather than fix them.
I have visions of old Japanese ladies building these things for some reason.
It was probably very easy to make mistakes. Some of the wiring junctions are very crowded.
It is like debugging a very difficult software problem in a way. In that sense I can relate to it. It's very easy to come up with false conclusions using a debugger in a large, complex program, for instance.
Whatever. Ordered some heat shrink to plumb it back in.
100 inch VU meters, an upside down busted Accuphase with no heart and a glass of Drambuie...
Smiles.https://uploads.tapatalk-cdn.com/20260821/909d23f83ef4484b5c18d6cb3b7551b3.jpghttps://uploads.tapatalk-cdn.com/20260821/f92b903ab274dd7e05ebe765fa6b7ba5.jpg
Firebottle
22-08-2026, 07:52
No probs Justin, I am free to travel now once the eye is back to normal and I can drive again.
I will come to you initially.
Sounds cool Alan it's not as if you don't know where you are going.
Let me plumb that tranny back in. Just this minute tested continuity from the two red 64.5V wires in the amp and they are fine to the diode. As expected.
I still can't believe I kept measuring 0V on the diode.
Sounds cool Alan it's not as if you don't know where you are going.
Let me plumb that tranny back in. Just this minute tested continuity from the two red 64.5V wires in the amp and they are fine to the diode. As expected.
I still can't believe I kept measuring 0V on the diode.
If you are measuring the full voltage when open circuit, and zero voltage when connected to the diodes, then the diodes must have shorted out and will need replacing.
I said to Claude: The leads to the diode at the cut point are fine. So we are saying what I measured again and again can't have been true. I am seriously struggling with this. I literally went from one amp to the other in the same session. 0V on the busted amp, 130V on the working one. Clear as day.
Claude said: Let's step back, because I think there's a way to reconcile everything you've measured without doubting yourself — and it's something we touched on very early today but never fully resolved.
A shorted diode would produce exactly this pattern. Here's why: if D001 or D002 has failed shorted (not open), it effectively places a dead short directly across that half of the transformer's secondary winding whenever the circuit is complete and powered. Under a genuine short, voltage at that point collapses toward 0V (V=IR, and R≈0 across a short) — while current shoots up massively, which is exactly what would blow R003 on the primary side. This unifies both of your major findings into one single fault: the 0V reading was real, and it explains the original symptom that started this whole investigation.
It also explains why the winding tested perfectly healthy once removed: a shorted diode isn't a transformer fault at all — the winding was never the problem. It only showed 0V because something downstream was short-circuiting it under load, not because the winding itself was open. Out of circuit, with nothing loading it, of course it measured fine.
Why this wasn't caught earlier: you did test these diodes for shorts, but the test relied on comparing against the working unit's continuity pattern — and the working unit's diodes showed the same "beeps across all three legs" behavior, which you and I concluded was normal packaging (case tied to one terminal internally). If that packaging conclusion was right, a genuinely shorted diode on the faulty unit might have looked identical to healthy-but-normally-tied-together packaging, and the continuity test simply couldn't tell them apart. That's a real limitation of that test, not a mistake either of us made carelessly.
How to test this properly now: with the diode fully isolated (not in-circuit, ideally desoldered or at minimum both leads lifted), measure it in diode-test mode, not continuity/beep mode — you're looking for a proper forward voltage drop in one direction and a clean open in the other. A stable low-resistance reading in both directions on an isolated diode (not just a case-related beep) would be unambiguous confirmation of a short — no packaging ambiguity possible once it's truly isolated.
This feels like the missing piece — worth testing before anything else.
I'll check it out as per Claude's advice.
Essentially if this is the issue, and Claude has been telling me it isn't and it's my measurements until now, then it's pretty much the same failure I found in that 211 monoblock I fixed a while ago, Barry.
Any success with the diode measurements Justin?
I haven't had the time, Barry, or enough will to make the time. I have pencilled in Friday for investigation.
The diodes are fine, Barry.
I did go as far as isolating one but IMHO that was a complete waste of time. Each package appears to contain two diodes connected to a common pin. Each measure 500mV isolated one way and massive resistance the other.
Isolating is just a fools game despite AI insistence. Why? Because in-circuit you can see the diodes charging up caps one way round with the diode probes, but not the other.
We just need to see how one mono behaves differently to the other.
I could post a load of stuff showing Claude talking rubbish about this, but it is pointless. It came round to my way of seeing it eventually.
That said, I did learn a bit from doing this.
Still perplexed I haven't sussed this out yet. The poor old Accuphase is turning into an amateur's test bench.
I need to get that tranny back in and reinstate the diode network. Not looking forward to the tranny loads of wires to do.https://uploads.tapatalk-cdn.com/20260829/a529834dda4befca9f915ee674cdf395.jpg
Curiouser and Curiouser! :scratch:
So the transformer is OK, and (both?) the diode packages are OK. That leaves the large reservoir capacitors and the four disc ceramic capacitors to check.
If they all check out OK, then the only thing left is to detach the amplifier supply rails from the capacitors to see if the capacitors retain their voltage.
There isn't much I haven't checked, or at least think I have checked as being fine with an LCR and multimeter with no power applied.
I will put it back together and send Alan a PM when done.
Phil M-W
29-08-2026, 20:35
Justin, looking at the pictures I can see that the diode packs appear to be isolated from the chassis with mica washers and some thermal grease. It would be worth checking that the washers are not damaged (perhaps have pinholes in them or a piece of swarf punching a hole through them). Since the metal bodies will be connected to the DC rails, a short there will pull the transformer outputs down to a diode drop above the chassis potential, presumably ground. I don't know if this area has been disassembled previously but if the thermal grease has been replaced, somebody may have used silver loaded grease, typically silvery grey in colour, and popular for use with CPUs and heatsinks in computers, relatively cheap and very readily available. This grease is (highly) electrically conductive and if carelessly applied can bypass the insulating washers. The correct grease would be aluminium oxide loaded grease, which is white.
Just my 2 cents.
Phil
Phil, I just got continuity to chassis for literally a second or two from the base of one of those diodes using the continuity tester on my multimeter, which admittedly is a cheap Maplin job that is probably 30 years old.
But then it just goes/disappears. Even after discharging the orange caps you can see in the pics.
I thought software was hard but I really am realising hardware can be an absolute bitch.
Sometimes, you just need to hire someone who has been in the field for a few decades and pay for what they know.
At least, that's the conclusion I am coming to for this case.
These things are worth saving because they quite literally are outstanding examples of their ilk for the time period.
Like the best classic cars.
FWIW the short beep is apparently the meter charging the mica which acts as a capacitor. Only happens on the one diode, oddly. Will do some resistance checks, but now I think about it I found this attribute a few weeks ago. I think I concluded it was OK.
Phil M-W
30-08-2026, 12:36
I forgot to mention to check the insulators around the mounting screws when I posted last night. Hmm, I'm not convinced that the capacitance between the rectifier and the chassis because of the mica washer would take long enough to charge, that the meter would be able to register it. The orange ceramic caps across the diodes might be enough though and could cause that. It's also possible, that the fault doesn't show until some volts are present. Get that a lot with car ignition systems. You should really not be doing this sort of test with external connections to the diodes, the external components are likely to mislead you.
To rule it out it might be worth temporarily replacing those rectifiers with a bridge rectifier like this (ebay item no. 285437590123). It shouldn't break the bank and will prove whether or not the fault lies there.
As for keeping old things running, I share your sentiment, It just feels wrong to throw away something for a trivial reason. It's why I'm trying to create a sub-woofer using an EMI speaker the came in to my possession 40+ years ago, there's nothing wrong with it and I'm fairly convinced it will do the job, in spite of being pressured to throw it away and use a B139 instead. The people who think that are missing the point and are simply focused on the immediate cost rather than the value and the environmental impact of creating yet more landfill.
Same goes for classic cars, I've got one that probably is and another that may be getting there. I doubt I will be parting with them anytime soon.
As I said: you have to check one thing at a time and in isolation of its neighbours. Laborious but essential if you don't want the neighbouring connected components to mislead you.
Phil that bridge rectifier is very similar to the one I used to fix one of of my 211 monoblocks a few months ago and documented on this forum.
I kinda feel your point that what's causing this issue is something that is failing when power is applied but looks fine when tested with power off.
You might not know I replaced all the output transistors and used what seemed to be incredibly expensive thermal paste. I went for the best stuff I could find, though.
Phil M-W
30-08-2026, 19:38
Phil that bridge rectifier is very similar to the one I used to fix one of of my 211 monoblocks a few months ago and documented on this forum.
Justin, I've only been on this forum 2 or 3 weeks and so far have only looked at 2 or 3 threads that caught my interest. If you are aware that they're available that's great.
I kinda feel your point that what's causing this issue is something that is failing when power is applied but looks fine when tested with power off.
Since everything seems to test OK when tested in isolation, it's what I'm leaning towards. It's what I would be looking at.
You might not know I replaced all the output transistors and used what seemed to be incredibly expensive thermal paste. I went for the best stuff I could find, though.
Ah, didn't know that, or it hadn't filtered through to my brain. What colour was the thermal paste? If it's grey or silver in colour, then it's likely to be silver loaded. It will be more expensive than the white aluminium oxide paste, simply because it has silver in it. It's generally more readily available because it's used by computer builders and there are a lot of them still about. It is preferred by them because it has better thermal conductivity, which if you're trying to prevent a CPU going into, in some cases a literal meltdown, it matters. The down side is, it's also electrically conductive (aluminium oxide is an insulator), which is why the instructions typically tell you not to apply too much, and if you do, to make sure you clean the excess off the board.
I don't know what you used, but it may be something to think about.
Phil
CHIP-QUIK TC3 was the paste.
https://sansui-vintage.com/product/rectifier-diode-sg-5ts-for-sansui-au20000-ba3000-nos/ also has SG-5TR.
Not cheap, but hey, be nice to use the original parts.
I wish I had a variac. That way I could verify that the diode network works until a given voltage point. We know they are good for 500mV but that's sod all.
It's tempting to go for it and just order the diodes and plumb them in since they are both isolated now.
The new transistors went in post amp being busted. When doing my testing and comparing across monoblocks, neither amp has the driver or protection boards installed. So that scope rules out the transistors.
Phil M-W
31-08-2026, 12:21
I looked up the specs of the chip quik, it says it's insulating so I guess that theory can be ruled out. It was worth finding that out though, one less thing to worry about.
I presume you are aware that the SG-5TR and SG-5TS are different, one is common cathode and the other is common anode, so you can't replace one with the other. I don't know which is which because I can't find a data sheet for either of them and I can't work it out from the pictures because I don't know which wire is the positive rail and which is the negative.
A variac would be handy but not just for the reason you're thinking of, it will reduce the amount of stress the components are put under during fault conditions like you have here. The 500mV sounds like it might be the forward voltage drop of the diodes when they are conducting. It seems a bit low but if the diodes are passing considerable current, which I guess they are, they will be heating up quite a bit, so that could explain it.
If it were me, I would buy that bridge rectifier and fit it temporarily. If it makes no difference, then the problem doesn't lie with the original rectifiers, it's somewhere else further along. If it does make a difference, then there is an issue with the rectifiers (or their mounting) that needs further investigation. It's a process of dealing with the possibilities one at a time and working through step by step.
As far as originality goes, I'm a bit ambivalent about it. Personally speaking, I think there are places where it matters and places where it doesn't, or perhaps I should say, isn't detrimental. Other people will have different opinions on this, but there are certainly situations where components, that were originally fitted, were only just up to the job, but they were the best that was available at the time. In this case replacing them with something newer may be a good move for reliability reasons and availability may force your hand anyway.
Taking an example from the classic car world, I periodically do work on Aston Martins, and a surprising number of early, high value ones (DB4,5 and 6s) are fitted with electronic ignition and in some cases, also well disguised electronic fuel injection. These are prize winning concours cars, where originality matters, but these modifications mean the cars can stay on the road and be used and enjoyed, rather than being a 4 wheel doorstop. I suppose it all depends on your priorities.
I also know owners that never drive them, and are quite happy to sit and look at them and polish them periodically. I don't get it, but it's their car and their call, I think they're missing the point of ownership, but that's just my opinion. There is one owner who has this approach that I can relate too. He's registered blind and consequently doesn't have a driving license. He bought his as a barn find and restored it, and (with the help of one of his children) has restored it to show worthy condition and by all accounts has done a very good job.
The 500mV is just the test voltage applied by my multimeter I think.
Re NOS parts I'd rather use them. Swapping all the power transistors for modern equivalents as I have done may well sound odd as they won't perform the same as the original parts, even though on paper they are better. It's a calculated risk I have taken. If they sound shite I will be forced to get some NOS ones. They are available.
I might get a variac - looks like £140 does it. For the moment I just need to get that tranny back in - waiting for some wire cutters to arrive as the ones I have aren't up to it. The tranny is unbelievably heavy. It can never have been cheap to make.
Thanks for your advice it is carefully being considered.
Firebottle
01-09-2026, 14:50
Justin do you want to borrow my variac? It isn't getting any use atm.
I will be coming down the M5 in a week to 10 days, depending when this gas bubble in my eye eventually is absorbed.
I can swing by on my way down to south Wales.
Alan just let me know when you will be coming. I dunno if you are up for a debug session when you arrive but that would be cool. With both monos disassembled to a greater or lesser extent a perusal by yourself would probably be invaluable.
I plan to plumb the transformer back in but leave the two 64.5V windings open and check them first. Then wire them back in and check that dual diode snubbered rectifier setup yet again now I understand it better if the transformer really is OK.
It probably is best I buy a variac though.
I just want to get to a state where I can maintain what I own.
Thanks for the offer. Might be well worth bringing it.
I've taken about 70% of this circuit out of the equation and I still can't debug it.
I am doing the equivalent of software unit, integration and system tests on the 30% I have isolated the fault to and I am still clueless.
Amazed.
And totally intrigued.https://uploads.tapatalk-cdn.com/20260904/578f5cac5b0e005f9cd138e606649e80.jpg
Extracted transformer and diode network re-instated. Hence all the blue heat shrink on the right hand side.
Loads of flux mess and solder splutter to clean up but I am sure that's wired correctly.
See a flaw in that anyone and please tell me. It's an amateur job but I am sure all those connections are sound.
Bridge rectifier also re-instated but no heat shrink as per the original set up.
I wonder if anyone from Accuphase has picked up on this thread. If so, send your best engineer to my address by first class air ticket immediately:) LOL.https://uploads.tapatalk-cdn.com/20260904/7e9c2e6132cd6e313a8bf2648d700f8b.jpg
Patrick Dixon
05-09-2026, 13:50
This thread worries me a bit. Best case is likely that you avoid doing any more damage to the amp and worst is that you damage yourself. So I reckon you are better waiting for Alan to come and look.
As an electronic engineer turned software developer I think there are are a few important differences:
You can't electrocute yourself with software.
You can't damage the code.
If you have a problem at the start of the code execution it's not going to destroy the rest of the code (generally).
The subsequent code does not affect the code you are currently executing.
So with that in mind, you at least need to isolate parts of the circuit so that you can check if they are working or not. The last thing you want to do is to destroy that mains transformer because that will be very hard and very expensive to replace/fix.
I haven't read through the whole thread again, so I can't remember whether you'd confirmed that the issue was with the PSU or the load?
If it's with the PSU, isolate it from the rest of the circuit before checking, and at least current limit the supply. Ideally feed the bridge/caps with a test supply voltage rather than just connecting to mains through your transformer.
If it's with the active (load) circuitry, disconnect it from the PSU and feed it with a current limited PSU so you have some control over what's happening.
You can't just put a few printf or debug statements in there and hope the problem becomes obvious, particularly with a highish power analogue circuit.
I'm not really one for loads of fancy test kit, but I'd want at least a scope, a lamp limiter and a lab PSU plus your multimeter to start on this.
I wouldn't worry if I were you Patrick.
I do have a fair idea of what I am doing. I might be a software engineer, but I do have decades of messing around with hardware on and off for embedded systems.
Scopes, logic analysers, multimeters etc.
I fixed my 211 monoblocks a while back. Riskier than this.
However, I am NOT an electronics expert. I just know enough not to kill myself, LOL.
I have just ordered a variac, FWIW. I have a scope, a multimeter and an LCR meter.
Patrick Dixon
05-09-2026, 17:39
I'm not sure why you would need help from anyone from Accuphase then, but I'll leave you to it. Good luck!
That was a joke Pat!
I am talking to Alan privately, BTW.
I keep looking at the condition of my M-60s in awe.
Somebody very possibly long dead now kept this particular pair in a very nice environment.
I just haven't seen a pair in better condition, especially with regards to the faceplates.
The fact that one isn't working is making me quite obsessively determined to sort it out.
I cancelled a variac order, but have ordered it again.
Also considering a dim bulb unit. Such an apparently cool thing for very little money.
To be fair, though, I am probably averaging about an hour a week if that with occasional bursts of enthusiasm.
Powered by vBulletin® Version 4.2.3 Copyright © 2026 vBulletin Solutions, Inc. All rights reserved.