MX-1000 Please help solve one channel issues

rottalpha

Yamaholic
Subscriber
I have picked up another non-working MX-1000 a few months back. The amp would not come out of protection. Today I decided to take a shot at fixing it.
It turned out to be a bad left channel.
After some investigation, I found:

one blown driver (Q315 - 2SA1306), replaced with 2SA1837 and replaced its pair which was not bad, but replaced it anyway, with 2SC3503

one blown final transistor (Q317 - 2SC3280), replaced with MJL3281

Base-Emiter resistor for Q317 (R367 - 68ohm)
Emiter resistor for Q317 (R317 - 0.22ohm)

On the power supply board I replaced two transistors
Q513 (2SC3416) replaced with KSC3503
Q515 (2SA1352) replaced with KSA1381
D511 (1SS133) replaced with 1N4448

The amp comes out of protection now and I get sound on both channels, BUT...

I adjusted the right idling current (good channel) to 10mV - no problem
I tried to adjust the left channel, the channel I thought i fixed, and I get 20mV at the minimum, thus the idling current remains high on the Left channel, which I thought I fixed. I do not have to mention that the heat sink of the left channel is much hotter than that of the working channel.

I know I missed something, but do not know what. Does anyone have any suggestions? Thanks for your help in advance.
 
I don't like to mix transistors that are not exactly the same. To settle any doubts you can try to adjust the idling current without this transistor (and its complementary), remaining only 2SA/2SC pairs.
 
Do the old and new devices share an identical Idss? If not, they may not bias properly. Also, perform a complete replacement of all electrolytics on the power amp board and the smaller one underneigth. It's precautionary and only costs about $25.
 
Do the old and new devices share an identical Idss? If not, they may not bias properly. Also, perform a complete replacement of all electrolytics on the power amp board and the smaller one underneigth. It's precautionary and only costs about $25.

Thanks, I found no apparent bulging on any caps but I will eventually replace all electrolytics, including those black gates from the signal path. At this stage I am just trying to pinpoint the issue. I know that there is something else that is fried, which I have missed so far. I will get to work on it again, probably in the weekend.
 
My apologies. My intent was more towards other areas where a decrease in value may have safety consequences. Small signal capacitors should be fine, provided their leakage is low. Their capacitance will not dwindle if they are not blocking high bias voltages. Do you have a schematic to work with?
 
My apologies. My intent was more towards other areas where a decrease in value may have safety consequences. Small signal capacitors should be fine, provided their leakage is low. Their capacitance will not dwindle if they are not blocking high bias voltages. Do you have a schematic to work with?

Yes, I have a schematic. There is no bias adjustment in these amps. only Idling current. I am not a pro, but from what I understand, the bias is continuously adjusting via the H.C.A (Hyperbolic Conversion Amp.)
I like how these amps sound and their flexibility. The sound is very clear and has virtual zero distortion...up to the conservative rating of 260 watts.
You can pretty much drive any speakers with these, and up to three sets at one time (I never used it with three sets yet). Once I get this one working, it would be my 3rd MX-1000:D:D:D I should change my avatar to " MX-1000 hoarder":D:D:D...except that there may be a couple of guys here that have more than three of these LOL.
 
Howdy again. The bias voltage I'm pointing to is any voltage present in a small-signal stage, if applicable, not the actual current amplifier stage. Electrolytics that are charged with a voltage are what we call 'biased'. At turn on, before they fully charge they will conduct a direct current. Once they are charged, the only DC passing through them will be the leakage inherent in that capacitor type. I can't recall if this amplifier is AC or DC coupled internally. Provided the leakage is low and ESR is insignificant, they should be fine. You may find they are slightly higher capacity than stated on the shrink wrap and that is a normal phenomenon directly related to low biasing the dielectric for long periods. They can climb in capacity by about 25%, and it's harmless. Moves the pass-through F3 downward slightly.

Yeah, the bias scheme is quite interesting. It's sometimes confused with sliding bias, but it's not. It's more involved, dynamic tracking bias; ie modulates with the music signal to keep the duty cycle low...from what a guru explained. Krell tried it for a while by infringing an old Yamaha patent but they couldn't get it to work right. They went back to sliding bias like threshold did. Not as sophisticated or efficient as Yamaha's invention. But, I don't know exactly and info is scarce.
 
Last edited:
Howdy again. The bias voltage I'm pointing to is any voltage present in a small-signal stage, if applicable, not the actual current amplifier stage. Electrolytics that are charged with a voltage are what we call 'biased'. At turn on, before they fully charge they will conduct a direct current. Once they are charged, the only DC passing through them will be the leakage inherent in that capacitor type. I can't recall if this amplifier is AC or DC coupled internally. Provided the leakage is low and ESR is insignificant, they should be fine. You may find they are slightly higher capacity than stated on the shrink wrap and that is a normal phenomenon directly related to low biasing the dielectric for long periods. They can climb in capacity by about 25%, and it's harmless. Moves the pass-through F3 downward slightly.

Yeah, the bias scheme is quite interesting. It's sometimes confused with sliding bias, but it's not. It's more involved, dynamic tracking bias; ie modulates with the music signal to keep the duty cycle low...from what a guru explained. Krell tried it for a while by infringing an old Yamaha patent but they couldn't get it to work right. They went back to sliding bias like threshold did. Not as sophisticated or efficient as Yamaha's invention. But, I don't know exactly and info is scarce.

Thank you for the education. Really appreciated.
So you are saying that even if the caps would be somewhat out of spec, they would not cause the issue I am having, correct?

I have not yet tested any of the caps of this amp. Based on my previous experience with the other two amps, the PSU caps tested as spec (capacitance and low ESR). I only found a few of the Black Gates in the signal path that were bad (way out of spec - lower capacitance) and some that were above spec capacitance, but within 5%, indeed. Overall, the quality of components, including caps, that Yamaha used in these amps is top notch. After so many years and many of these are still running fine.
I will be replacing all electrolytics on this amp eventually, except the main PSU caps, for which I am still looking to find a viable alternative replacement solution since replacement caps of same physical and capacitance size, are pretty much unobtanium. I am toying with the idea of using a bank of 10,000uf caps as replacement for the main PSU caps. I just lack the time to do it and there is no urgency yet, since the originals seem to be just fine.
 
My pleasure. Yamaha's quality was excellent (among their higher lines such as this) and their designs were innovative.

Electrolytic's lifespans are rated on temperature, ripple current, etc and they are consumables. I recall this amp ran warm and idled at 80-90W consumption. That doesn't mean they are all over the hill, it doesn't mean any are other than those you found below rated specs. Some of the good types, like in Yamaha amps have already proven their 30 years of worth on the bench. But, it comes down to risk assessment where the owner has to decide if pushing the clock is worth it. I don't take risks like this, and I'm relieved that you're not going to either. It was pure unadulterated luck that the boards and everything else survived whatever mishap ensued with the previous owner, and that's like being given a second chance at life. If you can get the same new equivalent power caps with the same specifications for a decent cost, jump on them, although they might be OK. The others are where I would focus attention. You should check the smaller pF range capacitors too, since you have the tools at hand. They don't usually go, but it has happened in other electronic machinery.

Good to be careful with smaller value capacitors as a direct replacement for the larger power supply caps, because the lower equivalent series resistance will load the rectifier and transformer pretty good at startup. I suspect the resistance of multiple small ones will be a fraction of the large. Please check the schematic to see if there is a slow start on the amp, like a thermal varistor to limit inrush current.

Back on the bias issue - it's possibly the new device's Idss value or another value is a mismatch to the remaining semiconductors. A capacitor in the amplifier may also be responsible for the inability to set the bias where you want it. Electrolytic capacitors have a leakage value, and this represents the direct current they will pass even when charged. As time passes, they don't all degrade at the same rate because they each have unique jobs.

Do you have a scan of the schematic you can share? I haven't seen one of these in a while. How does the offset check out, and how many seconds does it take for protection to come off after power-on?

Best regards.
 
Last edited:
Thank you for the response.
" ...I recall this amp ran warm and idled at 80-90W consumption...."
Actually, comparing to other Yamaha amps, particularly to the M-80 and M-85 (which have similar specs), the MX series run much cooler. As far as the idle consumption, I have tested one of the working beasties and it idles at 76-77 watts.

If you ever get the chance to get one of these amps in NZ, you wont regret it.

I have a schematic from the AK database. See the following links:

http://akdatabase.org/AKview/displayimage.php?album=3&pos=180

http://akdatabase.org/AKview/displayimage.php?album=3&pos=179

If you look at the schematic and the components I have already replaced (see first post), you may get a better idea of what went wrong. I would appreciate any additional feedback.
Thank you for taking the time to look in to this.
 
Thank you for the schematic, it's a truly innovative and brilliantly executed circuit. It's bound to confound many.

Looking at the schematic, there's no capacitors directly in the series signal path- it's direct coupled. On the input there are 220pF to ground to form a low pass filter. That combined with the input stage capacitance, well, some preamp might not like that.

Yup, 80W idle consumption is right on.

D511, Q513, and Q515 are part of the VAS stage, ie the preamp so to speak. Scenario A: Some alien event caused the final pair of the VAS stage to go critical, taking out D511, Q513 and Q515. Q309 and Q311 in the HCA section survived. A path in the current amp was formed across Q315, R367, and Q317, and R317. Q315 and Q317 are on opposite sides and Q315 doesn't appear to be a good return path, but it happened.

Theoretically, a short in the output device could cascade backwards giving the effect seen here, too. What causes the shorts in semiconductors is prime interest, but not right now. I would check everything, and even reflow any suspect solder joints with a little extra lead solder. Do you have a scope and spectrum analyzer by chance?
 
Last edited:
Thank you for the schematic, it's a truly innovative and brilliantly executed circuit. It's bound to confound many.

Looking at the schematic, there's no capacitors directly in the series signal path- it's direct coupled. On the input there are 220pF to ground to form a low pass filter. That combined with the input stage capacitance, well, some preamp might not like that.

Yup, 80W idle consumption is right on.

D511, Q513, and Q515 are part of the VAS stage, ie the preamp so to speak. Scenario A: Some alien event caused the final pair of the VAS stage to go critical, taking out D511, Q513 and Q515. Q309 and Q311 in the HCA section survived. A path in the current amp was formed across Q315, R367, and Q317, and R317. Q315 and Q317 are on opposite sides and Q315 doesn't appear to be a good return path, but it happened.

Theoretically, a short in the output device could cascade backwards giving the effect seen here, too. What causes the shorts in semiconductors is prime interest, but not right now. I would check everything, and even reflow any suspect solder joints with a little extra lead solder. Do you have a scope and spectrum analyzer by chance?

Yes, I already re-flowed most joints. These amps are prone to cold joints. Cold joints are in many cases the cause of problems like this.
No, I do not have a scope or a spectrum analyzer. Twenty some years ago, I did, but this is an old hobby that I only recently resurrected. I no longer have access to that kind of equipment. I was planing on getting a new digital scope. They are very affordable comparing to the old tube tektronics..but for now I will just have to make use of a multi-meter and LCR/ESR meter.

I will get to work on the amp in the weekend and I will report with any additional findings or issues.
 
....so the beastie is cranking again. I finally made some time for it. The problem was on the power supply board and not on the sub board amp.


OK NIkko, so you were right. THANK YOU! :banana:
Bad electrolytic between the base of the two trannies I changed on the PSU board and one of their emitter resistor (C517 and R531 if you are curious to look them up on the schematic)

Thanks again everyone!
:banana::banana::banana:
 
My pleasure. Yamaha's quality was excellent (among their higher lines such as this) and their designs were innovative.

....

Good to be careful with smaller value capacitors as a direct replacement for the larger power supply caps, because the lower equivalent series resistance will load the rectifier and transformer pretty good at startup. I suspect the resistance of multiple small ones will be a fraction of the large. Please check the schematic to see if there is a slow start on the amp, like a thermal varistor to limit inrush current.

...
.


Hi Nikko,

This amp does not have a thermal varistor, but it looks to have a slow start.
Multiple capacitors is no the ideal solution, but it is so far the next best alternative for this amp and for the M series amps, that I could think of.
Yamaha used a physical size of electrolytic which are unotanium today, so replacing them with the same size ones is not a solution. Some have tried other, less desirable ways to solve the issue.

http://forums.phxaudiotape.com/show...ten-off-butt-Yamaha-M80-Rebuild-Thread/page13

Your thoughts on this would be greatly appreciated.

Thanks
 
Hello rottalpha, my hard drive went, so I've been away getting this new computer set up. The dual transformer's secondaries are interrupted by a switch and fuse block, go straight to the quad of rectifiers, then the main smoothing capacitors. There isn't any current limiting, but that isn't a problem. Can you please tell me the physical dimensions of the capacitors? Thanks alot.
 
Hello rottalpha, my hard drive went, so I've been away getting this new computer set up. The dual transformer's secondaries are interrupted by a switch and fuse block, go straight to the quad of rectifiers, then the main smoothing capacitors. There isn't any current limiting, but that isn't a problem. Can you please tell me the physical dimensions of the capacitors? Thanks alot.

Hi Nikko,

Thank you for responding!

They are 63.6mm dia x 69.4mm tall
40,000uf @ 50v and 47,000uf @63v, both values have same can size

there is about 5mm gap between them, so the replacement can not be much bigger in diameter. There are 14mm of gap between the top cover and the top of the cans.

The M' series are using similar cans, but they are positioned horizontally. I do not have an M' series amp here to take the measurements. All measurements I posted are for the MX-1000 / U. The MX series cans are positioned vertically.

As far as I know, none of the manufacturers have these size/capacitance under current production. This issue has been discussed by other AK'ers here, but with no real end solution found, so if you come up with one, it would be like finding the holly grail.

Thank you!

-Peter
 
Back
Top Bottom