• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Another M-80 Restoration Project

Another little reminder to those who are doing restorations on this class of amp, where BJT's are used in output stages, and this becomes critical in Class A operation, and that is the importance of properly matching output transistors. Thermal runaway, especially in Class A, is one of the hazards you have to prevent, and one reason you see "Chernobyl" in so many that lose the output side. Carefully match your pairs as best you can, and if differing rank, find replacements that do match. New thermal compound of low thermal resistance will also go a long way in the thermal coupling of the BJT to the heatsink, and after 30 years, certainly needs to be replaced. Replace the emitter resistors with close tolerance replacements. You'll not only go to reduce heat and keep the outputs happy, but go a long way to ensure bias remains stable, especially when using it in Class A operation.
 
Last edited:
DeltaB said:
Another little reminder to those who are doing restorations on this class of amp, where BJT's are used in output stages, and this becomes critical in Class A operation, and that is the importance of properly matching output transistors. Thermal runaway, especially in Class A, is one of the hazards you have to prevent, and one reason you see "Chernobyl" in so many that lose the output side. Carefully match your pairs as best you can, and if differing rank, find replacements that do match. New thermal compound of low thermal resistance will also go a long way in the thermal coupling of the BJT to the heatsink, and after 30 years, certainly needs to be replaced. Replace the emitter resistors with close tolerance replacements. You'll not only go to reduce heat and keep the outputs happy, but go a long way to ensure bias remains stable, especially when using it in Class A operation.
Click to expand...
Never worked on the Ms, did a bunch of work on the newer MX 1ks
I would say the first reason these M and MX go Chernobyl is because of cold solder joints.
Second reason in case of the Ms is the Sony glue
As far as matching outputs, of course, do it whenever possible, but won't the emiter resistors mostly take care of that??
I certainly do not believe that the original Yamaha supplied outputs would be so mismatched to cause thermal runaway.
@avionic has been inside more M amps than anyone I know. I would be curious to hear his opinion on this subject.

Cheers
Pete
 
rottalpha said:
Never worked on the Ms, did a bunch of work on the newer MX 1ks
I would say the first reason these M and MX go Chernobyl is because of cold solder joints.
Second reason in case of the Ms is the Sony glue
As far as matching outputs, of course, do it whenever possible, but won't the emiter resistors mostly take care of that??
I certainly do not believe that the original Yamaha supplied outputs would be so mismatched to cause thermal runaway.
@avionic has been inside more M amps than anyone I know. I would be curious to hear his opinion on this subject.

Cheers
Pete
Click to expand...
On the outputs used while in Class A, at constant current, the voltage across the emitter to base junction of a bipolar transistor decreases 2 mV for each 1 °C rise in temperature (reference being 25 °C). If the base to emitter voltage is held constant and the temperature rises, the current through the base to emitter will increase, and thus the collector current will also increase. Depending on the bias point, the power dissipated in the transistor may also increase, which will further increase its temperature and exacerbate the problem.

Certainly in the driver sections, poor solder connections, as you have stated lead to it's demise. And there need nothing more be said about the DSBG, it does it's own malicious work.
 
Last edited:
Hmm matching outputs.. Good luck in the near future.. Sanken MT-200's have been discontinued.
 
DeltaB said:
Yup, TO-3's to the rescue....
Click to expand...
TO-3Ps?
DeltaB said:
On the outputs used while in Class A, at constant current, the voltage across the emitter to base junction of a bipolar transistor decreases 2 mV for each 1 °C rise in temperature (reference being 25 °C). If the base to emitter voltage is held constant and the temperature rises, the current through the base to emitter will increase, and thus the collector current will also increase. Depending on the bias point, the power dissipated in the transistor may also increase, which will further increase its temperature and exacerbate the problem.

Certainly in the driver sections, poor solder connections, as you have stated lead to it's demise. And there need nothing more be said about the DSBG, it does it's own malicious work.
Click to expand...

Understood. In case of replacement output for the M amps because of their class A, I see how matching outputs helps (I personally do match components as a rule of thumb, regardless).

What I doubt is that the factory originals were that far off to cause thermal runaway. If they were, thermal runaway would have happened even when the amps were new (newer).


I still believe cold joints are the main culprit. You will find them everywhere, including on the PSU and output section. I am not sure what process Yamaha used to solder these things, but the M and MX amps, even some of the C and CX pres, are all notorious for cold joints and that is the main reason these things stop working or go Chernobyl in the case of the amps.

I have inside more than a dozen units of the era, mainly MX-1000, and very single one had cold solder joints. The non-working (Chernobyl) MX amps, all had cold joints on the output section. Note MX amps do not have the DSBG issue or a pure class A section, yet they can put out a fireworks show. Luckily they use TO-3Ps and those are still plentiful.
 
rottalpha said:
TO-3Ps?


Understood. In case of replacement output for the M amps because of their class A, I see how matching outputs helps (I personally do match components as a rule of thumb, regardless).

What I doubt is that the factory originals were that far off to cause thermal runaway. If they were, thermal runaway would have happened even when the amps were new (newer).


I still believe cold joints are the main culprit. You will find them everywhere, including on the PSU and output section. I am not sure what process Yamaha used to solder these things, but the M and MX amps, even some of the C and CX pres, are all notorious for cold joints and that is the main reason these things stop working or go Chernobyl in the case of the amps.

I have inside more than a dozen units of the era, mainly MX-1000, and very single one had cold solder joints. The non-working (Chernobyl) MX amps, all had cold joints on the output section. Note MX amps do not have the DSBG issue or a pure class A section, yet they can put out a fireworks show. Luckily they use TO-3Ps and those are still plentiful.
Click to expand...
Whether from the factory, or by a tech's repair, my outputs had both P and Y ranks mixed. As I have heard from other folks who do repair, this is not uncommon on many Yamaha devices outside of the Japanese market.
 
DeltaB said:
Whether from the factory, or by a tech's repair, my outputs had both P and Y ranks mixed. As I have heard from other folks who do repair, this is not uncommon on many Yamaha devices outside of the Japanese market.
Click to expand...
ouch, different ranks all toghether, most likely repaired IMHO. The higher gain ranks usually become unavailable after a few years. MFGs would not pick a component about to go out of production, so I beleive this is a classy case of repair the bare minimum, by replaceing just the fried trannies. Unfortunaltely it happens. Could have been factory service center repair too.
I am glad you resurected that monster the right way.
 
Some final thoughts concerning the M-80 after having some time to give some critical listening and variations over the components replaced, one of the greatest impacts of listening SQ came when I utilized WIMA FKP and MKP's in the ZDR circuit. While Yamaha did utilize a PP for C134, the FKP was the one that gave the best results from the scope. For some reason, while Yamaha did specify PP for C151 & C152, they stuffed the board with Mylar. Replacing all of the 0.047uf Mylars with MKP's, (C149~C152) especially in the Negative Feedback Loop (C103 & C104) the upper end opened up very well, especially when listening in Class A. If you want your cymbals to come back in Class A, this is the one that will do it for you.

Also, when matching the input sensitivity from the older C class equipment, (1.55v 6dBu) to my current system output, (1.0v 3dBu) it requires reduction of the 220 ohm (R103-R104) to 150 ohm. Now dynamics from digital content is properly scaled, with the 0dB reference point matching for both AVR and the M-80. (0dB reference on the RX-A2030 delivers the specified 34.5V output reference for 0db (no load) meter scale calibration)

Happy restorations everyone, and most of all, don't crush `em, rebuild them!
 
Last edited:
Guest125 said:
As rework continued today, one of the experiences I found when I removed the power boards, was that each of the pairings of finals in both the channel outputs and the voltage regulation circuit, were different lots and most disturbing, the rank. The outputs on the left were a rank of P and the right was Y, and the voltage pairs were one side 2SC2773/2SA1169 and the other 2SC2922/2SA1216, both of greatly different rank. B&D was kind enough to match 2 pairs of 2SC2922/2SA1216 Y's to match the one side, which leaves me with a matched set of 8 2SC2773/2SA1169 in both lot and rank to match outputs on both R&L sides after rework. I can only conclude that either no one was paying attention at the factory, or, this unit went through repair and the tech saddled them like that.

As a sidebar, I get asked at times, how tight do you place the screws, (and yes there is a torque spec) and for the MT-200 it is 7 inch-pounds. Anything over 10~12 inch-pounds yields diminishing returns, and will distort the package. SanKen recommends Dow SC-102 compound for use with these devices.
Click to expand...

Old thread, but a good one. Just going through my first M-80, and maybe I'm lucky but the outputs were the same type and rank.
upload_2021-7-26_9-19-39.png
 
You must log in or register to reply here.
Back
Top Bottom