M-45 Recap and Service
- Thread starter WannaBeHiFi
- Start date
WannaBeHiFi
Active Member
I have. Of the 8 removed so far, all have measured Capacitance at specified rating. I can measure F, ESR, and Vloss, though I'm not sure why I should.
WannaBeHiFi
Active Member
Hello Darkness, my old friend...
WannaBeHiFi
Active Member
C103, C104:
All of the signal passes through these 47uF 16V Rubicon Black Gate F capacitors.
These lil' darlins' are near 40 years old! How they holdin' up?
1: 45.5uF, ESR = 0.31Ohm, Vloss = 0.6%
2: 48.97uF, ESR = 0.29Ohm, Vloss = 0.8%
To be replaced by Elna Silmic II
1: 47.53uF, ESR = 0.52Ohm, Vloss = 0.8%
2: 47.27uF, ESR = 0.56Ohm, Vloss = 0.9%
What, if anything, should a neophyte read from this data?
'Cause, Ain't Broke Don't Fix It is pitchin' a fit!
And Mr. Fixit says you can't argue with physics and these "lil' darlins" should have been put out to pasture a long time ago.
All of the signal passes through these 47uF 16V Rubicon Black Gate F capacitors.
These lil' darlins' are near 40 years old! How they holdin' up?
1: 45.5uF, ESR = 0.31Ohm, Vloss = 0.6%
2: 48.97uF, ESR = 0.29Ohm, Vloss = 0.8%
To be replaced by Elna Silmic II
1: 47.53uF, ESR = 0.52Ohm, Vloss = 0.8%
2: 47.27uF, ESR = 0.56Ohm, Vloss = 0.9%
What, if anything, should a neophyte read from this data?
'Cause, Ain't Broke Don't Fix It is pitchin' a fit!
And Mr. Fixit says you can't argue with physics and these "lil' darlins" should have been put out to pasture a long time ago.
Sparkplug
Super Member
Based on the data available (applied physics), the OEM caps are good and doing their job. Anything beyond that becomes a personal choice. The reason I asked for this data was to re-confirm what I’ve been seeing working on equipment here myself. I don’t recap by default anymore except for the power supply circuit board. I have wasted enough money changing zeros with zeros. I have never taken a bad cap out of any equipment, ever, not one, we’re talking hundreds of caps if not more.
Last edited:
Imho you got lucky with the Black Gates. So far almost every one that I have seen has been moderately to severely degraded.
If you want to get comfortable with what you are doing just decide on your goals, and if you are going to want to learn something, and if you are going to want to do this sort of thing again in the future. And if you want to insure good operation for many years in the future.
There is only one way that I know of to accurately check the condition of a cap, and most other components, and that is to un-solder at least one leg.
Personally, since I have to go that far, and since I will want to re-solder each and every connection in the unit, I just remove the part.
I do test each and every old and new part, and document them. I want to build a picture of the condition of the unit, and I want to sort and match the new parts. I want to make the time spent worthwhile, so I have found that it is best to be methodical and to try to stay with "best practices" as much as possible.
Documenting what you see and are doing gives you something to work with in x number of years when you see another one of these amps and you have forgotten the details, and it gives you something to work with in case something goes wrong with this one.
I am never, ever, going to re-install a 30+ year elco, except possibly the main PSU filter caps if I can't find a suitable replacement. I do not re-install old carbon film resistors, diodes or zener diodes unless I have no other choice, and if I do I document where they are.
Those parts have been removed, in most cases I have new items on hand, and they are not expensive. I have already subjected the traces to heat while removing the part and only want to heat them once again, so for me, it makes sense to put in new, tested parts.
Re-installing old elcos is just pre-programming the next point of failure in my opinion.
In the end the equipment is your property, to do with as you want. Make your choices and don't let anyone else dictate your choices. But if your goal here is to learn something, then by all means learn the right things.
I miss frosty mug root beer in the summer from the A&W restaurant. Served by cuties on roller skates.
Just reminded me...Back to business...
I've seen the old black gates be the first part to fail in two chernobyl'd M series here. One BG cap was an unrecognizable board stain.
I rarely test caps that old. Right to the trash can & in with the new.
Good signal path choice IMO. I like those and the Nichicon UFG fine gold jones there. I thought the Silmic II's sounded better, but were discontinued a few years ago and I thought they were all gone by now.
. Caps are so damned subjective nowadays it's an endless toil of choice, especially in the signal path. I haven't tried Torin's but am tempted to. The old black gates bad a graphite layer on their plates, the elna silmics used silk if I understand it right. Some guys say F it and threw in panasonics and broke them in for awhile for good sound. Anything new is better than the antiques in there now waiting to fail IMO. Poof
Last edited:
WannaBeHiFi
Active Member
A&W WTFI miss frosty mug root beer in the summer from the A&W restaurant. Served by cuties on roller skates.
That, my friend, was a rather nice Imperial Stout.
WannaBeHiFi
Active Member
Hey Y'all! I've made some progress.
Main PCB: All the electrolytic caps have been replaced except for C159, the 0.1uF 50V and C171, C172 the 22000uF 69V. The unit came out of protection on DBT.
Adjustments. Warmed up on main power.
Lch: R155 came out of the gate idling at 5mV.
Rch: R156 was idling at 3.1mV. NOW, I understand why one might want to replace VR103, 104. Twitchy lil' bastards they are.
I finally got R156 to settle in about 5mv.
Then I engaged Auto Class A. Left channel was idling about 44mV. Right a little less.
I had to fiddle with VR 104 several times because it kept drifting down. I'd run it in Class A for a while, then let it settle to idle. After several cycles I had R155, R156 idling about 5mV and low 40's in Class A. There is a remark that check point voltage applies to all of the large emitter resistors. Thats where things get interesting.
Idle voltage, back to front.
Left Channel
R159-4.7mV
R161-4.4
R155-5.0 Check Point
R157-4.2 SM recommends 5 +- 1mV.
Right Channel
R160-7.1mV
R162-0
R156-5.7 Check Point
R158-3.6 Hmmmmm
Topside R162 looks like all the others.
Down under there is a bit of darkness at the joint.
Do I have a bad connection, bad resistor, or a problem somewhere else? Please and Thank You.
Main PCB: All the electrolytic caps have been replaced except for C159, the 0.1uF 50V and C171, C172 the 22000uF 69V. The unit came out of protection on DBT.
Adjustments. Warmed up on main power.
Lch: R155 came out of the gate idling at 5mV.
Rch: R156 was idling at 3.1mV. NOW, I understand why one might want to replace VR103, 104. Twitchy lil' bastards they are.
I finally got R156 to settle in about 5mv.
Then I engaged Auto Class A. Left channel was idling about 44mV. Right a little less.
I had to fiddle with VR 104 several times because it kept drifting down. I'd run it in Class A for a while, then let it settle to idle. After several cycles I had R155, R156 idling about 5mV and low 40's in Class A. There is a remark that check point voltage applies to all of the large emitter resistors. Thats where things get interesting.
Idle voltage, back to front.
Left Channel
R159-4.7mV
R161-4.4
R155-5.0 Check Point
R157-4.2 SM recommends 5 +- 1mV.
Right Channel
R160-7.1mV
R162-0
R156-5.7 Check Point
R158-3.6 Hmmmmm
Topside R162 looks like all the others.
Down under there is a bit of darkness at the joint.
Do I have a bad connection, bad resistor, or a problem somewhere else? Please and Thank You.
Attachments
Nice. No magic smoke.
I spray cleaned my trim pots at first and let them dry a day or two. Eventually I replaced them though a year or two later cuz everyone else is doing it, lol. Reminder, they're fragile, if they go open/high resistance stuff blows up.
It looks fine to me. It's just residual flux from the out put transistor pads. If it bugs ya, a toothbrush dipped in IPA should clean it up. If it's solidified I'd poke at it with wood or plastic.
Looks like ya touched up the light solders there , good.
Setting the bias on my M series was quirky. I set the bias after 45 minutes and let it stabilize in AB. I won't touch the class A button during that time.
WannaBeHiFi
Active Member
Yea, but 0V across R162 can't be right.
It worked before. We're just re-capping with kid gloves.
WannaBeHiFi
Active Member
Yes, That's the connection below.
Since that pic I've removed R162. It tested per spec. And in the process resoldered. With resistor and connection eliminated I'm leaning toward bad Q130. My reasoning is Q128 and Q130 both get the same base current off R154. Since Q128 is passing idle current across R160 and there is no voltage across R162, then Q130 must be at fault? Yes? No?
Attachments
WannaBeHiFi
Active Member
C171, C172
***Brief intermission for unrelated ramblings
You've got to put on your sailin' shoes
Put on your sailing shoes
Everyone will start to cheer
When you put on your sailin' shoes.
One of Feat's better albums. Almost every song written by Lowell George. What a talent! Bloody shame he was taken too early.
I raise my Milk Stout in honor of Little Feat and Lowell. Got No Shadow really ought to be cranked up, but.....
My mood is dark as this beer. It appears I've F'dup this amp. Prior to this exercise in madness the amp was working. R needed a bit more gain to balance with L but...
So, if Q130 is Chairman Mao, I just don't want to talk to him now. And if you don't get that, well, you ought to listen to more Little Feat.
*****
Returning to the conversation at hand...
From what I can tell, there are no direct replacements to be found from reputable vendors. I've been thankin' about "Modifications". And I came across another discussion where Mr. Avionic revealed the tip of using 0.25in. solder wick as a connector. Well, that opens a host of rather limited options.
The limit is the mounting hardware. The existing caps are 50mm W x 80mm H. There is a lot of vertical room. But, near as I can tell, we're stuck at 50mm W. Unless there is an option I've not considered.
This is what I've found.
Which is the "better" option. Both are general purpose. DigiKey list the same two caps. The only audio grade Nichicon listed on Parts Connextion. But it's way too large to fit in the chassis.
Whis is better? Lower ripple current, or lower ESR?
Is higher heat rating a factor?
Should I just leave it the F alone?
Medicated Minds want to know?
***Brief intermission for unrelated ramblings
You've got to put on your sailin' shoes
Put on your sailing shoes
Everyone will start to cheer
When you put on your sailin' shoes.
One of Feat's better albums. Almost every song written by Lowell George. What a talent! Bloody shame he was taken too early.
I raise my Milk Stout in honor of Little Feat and Lowell. Got No Shadow really ought to be cranked up, but.....
My mood is dark as this beer. It appears I've F'dup this amp. Prior to this exercise in madness the amp was working. R needed a bit more gain to balance with L but...
So, if Q130 is Chairman Mao, I just don't want to talk to him now. And if you don't get that, well, you ought to listen to more Little Feat.
*****
Returning to the conversation at hand...
From what I can tell, there are no direct replacements to be found from reputable vendors. I've been thankin' about "Modifications". And I came across another discussion where Mr. Avionic revealed the tip of using 0.25in. solder wick as a connector. Well, that opens a host of rather limited options.
The limit is the mounting hardware. The existing caps are 50mm W x 80mm H. There is a lot of vertical room. But, near as I can tell, we're stuck at 50mm W. Unless there is an option I've not considered.
This is what I've found.
Which is the "better" option. Both are general purpose. DigiKey list the same two caps. The only audio grade Nichicon listed on Parts Connextion. But it's way too large to fit in the chassis.
Whis is better? Lower ripple current, or lower ESR?
Is higher heat rating a factor?
Should I just leave it the F alone?
Medicated Minds want to know?
