Workflow suggestions for refurbishing MC2125

Left meter behaves just fine when not in "Watts" mode which makes me suspect of portion of meter PCB circuit responsible for that calculation.

Michael

Michael,

My memory just kicked in. Had a similar problem on a 2125 that I restored for a friend. Turned out that an IC's on the meter board had failed. These are 133-037 which = LM301. I replaced them all as they are cheap enough.

-Jeff
 
@jeffs79: Great suggestion! I swapped the IC sets left (IC103 and IC105) and right (IC104 and IC106) and...the pegging problem went away. I am guessing a problem with oxidation in the sockets and/or a marginal part. Will replace these (0.87 a pop from Mouser) as well as cleaning the sockets and reflowing the solder joints.

Thanks again!

Michael
 
So I decided to pull the amp from service tonight after it was clear it's no longer working correctly. Both channels were pretty weak after the last round of bias tweaking. I opened it up, made sure the leads to the emitters were soldered well, and made a new set of readings. The results are below.

@1moreamp: I'm not sure what more point there is in taking measurements beyond these. What more needs to be determined? The left channel for sure has become pretty wonky---maybe as a result of all the poking and prodding---and the right channel has the funky diode that either needs to be cleaned or replaced (I'm thinking replace---the parts are cheap enough). I'd just as soon replace everything on both channels, including the pre-drivers on the power output PCBs and then bring the amp back to snuff again.

Michael

MC2125_EmitterVoltageDrops_06_09_15.jpg
 
Have you bench tested it under load? Have you looked for Hi frequency oscillation? Is the secondary power supply collapsing , how about the b- and b+ under load?

I do not know how to measure or fix sounding weak.....
 
Have you bench tested it under load? Have you looked for Hi frequency oscillation? Is the secondary power supply collapsing , how about the b- and b+ under load?

I do not know how to measure or fix sounding weak.....

I understand. Thanks for replying.

Where I am right now is that I've invested quite a bit of time and effort investigating the performance of this amp using the tools available to me and am unclear if there's any point to further testing. I don't have a dummy load nor an oscilloscope---though I clearly understand the value of having these available. I have been following the method suggested by 1moreamp to determine the health of the output stage, which involves using a DMM. Unless I am doing something dreadfully wrong or my understanding of the meaning of the voltage drops I've observed for the emitter resistors is incorrect, it appears to me that the amp is not (or no longer) behaving in a linear fashion. I am unable to make an adjustment to the bias trimmers that results in a balanced load across the emitter pairs. Other measurements I've made (and reported in this thread) include key components on the power output PCBs, including all the rail voltages. These appear to be more or less correct as per the schematic in the SM. So at this juncture, my inclination is to take a "blueprinting" approach and replace relevant components in the output chain and then reset bias with modern parts.
 
@jeffs79: Great suggestion! I swapped the IC sets left (IC103 and IC105) and right (IC104 and IC106) and...the pegging problem went away. I am guessing a problem with oxidation in the sockets and/or a marginal part. Will replace these (0.87 a pop from Mouser) as well as cleaning the sockets and reflowing the solder joints.

Thanks again!

Michael


Ditch the freaking IC sockets, They are unreliable. The only good IC sockets I ever saw in my life was the ones with fine teeth inside of them that latch onto the IC pins and they are solid gold plated and they cost like the dickens if you can even find them.
SO unless your fond of re-occurring issues and problems ditch the IC sockets in the round file.. and directly solder in the IC's, And or search till you find the sockets with gold plated pin sockets that have fine teeth inside of the socket itself. These are the only sockets I would ever consider using since they won't corrode, and the teeth lock the pins in place... But its best you directly solder the IC's in place.
The only people and places you see sockets used in the last thirty years in certain products where that chip is known for re-occurring and constant failure and need of replacement.


So NO IC Sockets, they are a weak link of reliability at the contact points where the pins contact the sockets, and the ones I see used in Mac amps corrode badly after 35 years or so...If you must use them use gold plated ones only they don't corrode at least.....Just a FYI heads on that been there done that issue...
 
So I decided to pull the amp from service tonight after it was clear it's no longer working correctly. Both channels were pretty weak after the last round of bias tweaking. I opened it up, made sure the leads to the emitters were soldered well, and made a new set of readings. The results are below.

@1moreamp: I'm not sure what more point there is in taking measurements beyond these. What more needs to be determined? The left channel for sure has become pretty wonky---maybe as a result of all the poking and prodding---and the right channel has the funky diode that either needs to be cleaned or replaced (I'm thinking replace---the parts are cheap enough). I'd just as soon replace everything on both channels, including the pre-drivers on the power output PCBs and then bring the amp back to snuff again.

Michael

MC2125_EmitterVoltageDrops_06_09_15.jpg


Weak sounding amp is not a quantifiable description, but I will try my best to work with it.

Now you have bias'ed both channels correctly you can now see that the left channel is really messed up numbers wise. Your right channel looks OK by the numbers.
You choice to rebuild is a probably fine for the left channel since I see issues with output balance and drive levels.
Now this does not mean your outputs are bad because if the drive circuitry is messed up and it appears to be that way then passing judgement on the outputs at this time may be premature.

I think your predriver circuitry may be "wonky" and I would rebuild that first and retest the final stages to see if the driver boards issues solved your output stage issues. This is economical thinking though and if your mind is made to up to "gut and go" the whole channel then well I don't blame you feeling this way. The rebuild of the whole channel method does shotgun all the issues promptly in most cases and with a amp this old well its very easy to go this way.

So yeah I will agree no more testing required the numbers are plain as day telling you there are issues in the Left channel for sure, and it will require teching out and or rebuilding it new again. I suggest the driver board first and retest because you will have to beta/gain test upwards of 15 to 30 outputs get get a closely matched set of output pairs, and I get mine directly from ON semi with no middle man via direct ordering < cheapest way to get fresh reliable factory direct parts I have found... Hope you can work this avenue out also...

c-dk has suggested your amp may be oscillating and this is possible, and only way I know of seeing that is with a oscilloscope to test and look for tell tale signs of the noise it will generate internal to the amp channel that you might not be able to hear since it will likely be very high frequency. But his input is very plausible since these older amps weren't that stable to begin with. So if you have access to a scope maybe you could look around first to make sure your not oscillating like c-dk thinks may be possible... Just to be sure. c-dk has always had some very good input IMHO..


When replacing outputs stages please replace all emitter resistors for general principals they may be weak and or off spec, Only with new emitters will you be safe and sure that the voltage drop your reading is accurate on your new output stage.. just another FYI this is the way I do things moment...lol...Wish I could hands on your amp for a half hour just to make absolutely sure about a few other things before you do a complete rebuild.. But I will stick with you best as possible with your available avenues...
 
Last edited:
I understand. Thanks for replying.

Where I am right now is that I've invested quite a bit of time and effort investigating the performance of this amp using the tools available to me and am unclear if there's any point to further testing. I don't have a dummy load nor an oscilloscope---though I clearly understand the value of having these available. I have been following the method suggested by 1moreamp to determine the health of the output stage, which involves using a DMM. Unless I am doing something dreadfully wrong or my understanding of the meaning of the voltage drops I've observed for the emitter resistors is incorrect, it appears to me that the amp is not (or no longer) behaving in a linear fashion. I am unable to make an adjustment to the bias trimmers that results in a balanced load across the emitter pairs. Other measurements I've made (and reported in this thread) include key components on the power output PCBs, including all the rail voltages. These appear to be more or less correct as per the schematic in the SM. So at this juncture, my inclination is to take a "blueprinting" approach and replace relevant components in the output chain and then reset bias with modern parts.

Back when this amp was new a McIntosh dealer had a choice.....buy the test gear Mac insisted you have in your service department.....or....send their kid to Harvard for a year.

To hope to repair these units without the correct test equipment is......well....... short sighted.

The amp needs to be driven into a steady state load and then tested for various possible issues under various gain levels......you should find a local tech that can properly test this for you. Chasing bias and offset will not solve your problem IF THERE IS ONE. You might just not like the sound of the amp........
 
Well, building a dummy load and driving it isn't a problem. And I have software to inject whatever in the way of signal. The scope is a bigger question. The investment looks to be around $200 plus probes and the damn things are also pretty big. Then if I don't use it again for months (or ever), well I suppose I can sell it. Or maybe I can find a hacker space near me that has one but then I need to pack the amp up and haul there and back. Oh, and then I need to learn how to properly use it and interpret the results. I guess the net/net is that I'm still mulling it over.
 
Well, building a dummy load and driving it isn't a problem. And I have software to inject whatever in the way of signal. The scope is a bigger question. The investment looks to be around $200 plus probes and the damn things are also pretty big. Then if I don't use it again for months (or ever), well I suppose I can sell it. Or maybe I can find a hacker space near me that has one but then I need to pack the amp up and haul there and back. Oh, and then I need to learn how to properly use it and interpret the results. I guess the net/net is that I'm still mulling it over.

I got a pair of non inductive 1KW dummy loads on sinks, and three Tek scopes at my home along with a Wavetek signal generator. And you have my PM response as to my location and my friends in the Bay Area near you that are BBB accredited and safe. Call me if you like, I visit the Bay Area weekly almost.
I am busy with a another AK'ers MC2105 right now but I should be finished with his amp by the weekend. His main rail caps had shorted out and vented so barring any other issues he should be a happy camper shortly and then I am free to look into your issues if you like...C:thmbsp:
 
Chasing bias and offset will not solve your problem IF THERE IS ONE. You might just not like the sound of the amp........


I agree chasing bias and offset numbers will not solve a problem, but it sure as hell will tell you, you got one!
His left channel numbers don't even compared to his right sooooo....:yes: I guess his amp has issues that spread beyond what the service manual talks about, and or describes how to deal with...Using the P3 kill-a-watt guesstimate meter thingamajig test method.......

Guess its time for a good and qualified tech to look inside and see whats going on and see why a high quality home amp by such a famous name can't show mirror like details from one channel to another after all these years of service...:thmbsp:
 
Moving ahead, here's the parts list for the rebuild of both channels...though I will probably fix the left one first and then evaluate the situation:

Pre-Drivers (Power Output PCB)

#132-148 x 2, Mouser #610-CEN-U57
#132-149 x 2, Mouser #610-CEN-U07

Note: These are Centronics U57 and U07, as per other threads on the topic. Dunno if there's yet a better replacement.

Drivers:

PNP, #132-151: MJ15025G x 2, Mouser #863-MJ15025G
NPN, #132-152: MJ15024G x 2, Mouser #863-MJ15024G

Outputs:

NPN, #132-164: MJ15003G x 4, Mouser #863-MJ15003G
PNP, #132-165: MJ15004G x 4, Mouser #863-MJ15004G

Emitter Resistors:

.15 Ohm/5W, #139-105 x 8, Mouser #588-15FR150E

Note: This is an Ohmite 10 Series non-inductive part, 1% tolerance.

Diodes:

#070-031 x 4, Mouser 625-1N5394-E3/54

Note: This is a Vishay 1.5A/300V rectifier. If there's a better replacement, please let me know

I'll also replace the two 12 Ohm/.5W resistors with metal film 1% parts.

Anything else? Anything out of place?
 
I got a pair of non inductive 1KW dummy loads on sinks, and three Tek scopes at my home along with a Wavetek signal generator. And you have my PM response as to my location and my friends in the Bay Area near you that are BBB accredited and safe. Call me if you like, I visit the Bay Area weekly almost.
I am busy with a another AK'ers MC2105 right now but I should be finished with his amp by the weekend. His main rail caps had shorted out and vented so barring any other issues he should be a happy camper shortly and then I am free to look into your issues if you like...C:thmbsp:

My plan at this point is as follows:

Order parts and rebuild the left channel and maybe the right one too.

Measure voltage drops on new parts and adjust bias.

Bring amp to you for a final evaluation.


I derive enormous pleasure from working on my vintage audio gear. I got my 2105 back up to snuff earlier in the year and I am hopeful I can do the same for the 2125, even if I need some professional assistance finishing it off.
 
Moving ahead, here's the parts list for the rebuild of both channels...though I will probably fix the left one first and then evaluate the situation:

Pre-Drivers (Power Output PCB)

#132-148 x 2, Mouser #610-CEN-U57
#132-149 x 2, Mouser #610-CEN-U07

Note: These are Centronics U57 and U07, as per other threads on the topic. Dunno if there's yet a better replacement.

Drivers:

PNP, #132-151: MJ15025G x 2, Mouser #863-MJ15025G
NPN, #132-152: MJ15024G x 2, Mouser #863-MJ15024G

Outputs:

NPN, #132-164: MJ15003G x 4, Mouser #863-MJ15003G
PNP, #132-165: MJ15004G x 4, Mouser #863-MJ15004G

Emitter Resistors:

.15 Ohm/5W, #139-105 x 8, Mouser #588-15FR150E

Note: This is an Ohmite 10 Series non-inductive part, 1% tolerance.

Diodes:

#070-031 x 4, Mouser 625-1N5394-E3/54

Note: This is a Vishay 1.5A/300V rectifier. If there's a better replacement, please let me know

I'll also replace the two 12 Ohm/.5W resistors with metal film 1% parts.

Anything else? Anything out of place?

My last input......

Do you know these parts are bad?

Continually shotgunning and not tracing the problem thru the driver board and output circuits is not the correct way to insure a proper repair.
 
@c_dk: Prior to looking at the output transistor emitter voltage drops, I measured most everything coming out of the power output PCBs for both channels. I also measured voltages for a number of components on the PCBs themselves. Below are the numbers. Nothing looked out of spec, as per the values on the schematic.

What other investigation would you have me do?

MC2125PowerPCBMeasurements.jpg
 
You have to look at the AC voltages, that is waveforms, under stress specifically at the low frequencies you claim are missing.
 
You have to look at the AC voltages, that is waveforms, under stress specifically at the low frequencies you claim are missing.

Got it.

I guess that will have to wait until I send the amp to @1moreamp (or for me to buy/locate a scope I can use at home). I still plan on going ahead with the rebuild of the output chain---I cannot possible see how it can be harmful, only, perhaps, inefficient from some points of view. Certainly as you suggest it's possible that the problem I found through the emitter study might be only the result of another problem upstream. I understand that and it means there will be more work ahead. Any time that might be wasted working this way is only my own, which I am happy to provide at a very low cost to myself.

I'm not sure how this will sound but, you know, I am 100% confident that I am going to get this amp back up to snuff before I am finished. I might not do it in the most efficient manner. I'll need to ask a lot of questions that have, to you all, obvious answers and then probably hem and haw, second guessing my own instincts. But it will get done.
 
Back
Top Bottom