• 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

Capacitor matching in a Power AMP

Bogdan Baba

Bogdan Baba

Member
I have a Mitsubishi M-A01 that I'm recapping. I am in no way an expert at this but so far I recapped about 20 devices so I do know a little bit.

This particular AMP has one requirement the space is super tight I have no clearance what so ever.
I need to replace 2x 20.000 UF at 55v with something.
I checked digikey and mouser and none of the 20.000 at 63 or 100vdc would fit in the case.

I purchased two Nichicon 22000 at 65v but I'm not sure if i can use them they are 10% more UF would that work?

The alternative would be to buy 4 10000 UF caps at 63v and then design and 3dprint a adapter for all 4.
Would it be an issue if I have two caps in parallel and then on the circuit board they are in series.

I would solder then together and then run leads to the board.
 
Register to hide this ad
What is the tolerance code of the originals ? Of the new Nichicons ?
 
I've seen a number of new Nichicons be closer to the low end of the spec rather than nominal, so they're probably more like 20000uf anyway. :dunno:
 
I do not know what the original tolerance is these are ~1981 caps. They are also Nichicon.

Yesterday I finished recapping the power amp with the 22000 UF I tested the amp for around 30 minutes and did all the adjustments based on the manual
After I did all the adjustments I close it up and flipped it up and qued a song (was listening on headphones) I immediately noticed one channel was out (no led vu meters on) and smelled something.
I pulled the power immediately then open up the case.
The new caps barely fit in the case and they don't fit at all in the original space. I 3d printed some adapter but I'm afraid I might have pushed something together underneath the caps when I closed the case. as it was such a tight fit the top was pushing on the caps.

After opening the top and bottom i powered it on with a dimbulb and nothing it was all working fine again so i decided to stop ****ing with it and ask smarter people for advice.
Today i removed the caps and found the source of the smell it was with a 100 ohm resistor not blown but def cooked. (in the photos location + resistor)

I don't know for sure if it was the case closing or the bigger caps that almost fried that resistor. This is why I'm asking the question.



WhatsApp Image 2022-05-05 at 3.20.47 PM (1).jpeg WhatsApp Image 2022-05-05 at 3.20.58 PM (1).jpeg WhatsApp Image 2022-05-05 at 3.20.58 PM.jpeg WhatsApp Image 2022-05-05 at 3.21.12 PM.jpeg WhatsApp Image 2022-05-05 at 3.21.13 PM.jpeg WhatsApp Image 2022-05-05 at 3.26.35 PM.jpeg
 
I'd say not from the caps themselves unless, perhaps, one of them was installed backward or pushed something to short out. Just touching the cap shouldn't as they appear to be sleeved with plastic.

You mentioned adjustments...
 
I did all 4.
1) => done and verified
2) => done and verified
3) => Current was 2.09v at C402 and 1.968 at C302 i wanted to change the voltage but moving VR302 did nothing. Going in the schematic VR301 and VR 401 control C302 and C402 also it's odd that Vr 402 would control that cap voltage and the voltage for VU .
4) => was able to set the voltage across both resistors.

WhatsApp Image 2022-05-05 at 4.37.39 PM.jpeg WhatsApp Image 2022-05-05 at 4.40.58 PM.jpeg
 
Last edited:
I ordered a Capacitor tester and I'm going to compare the capacity between the Original Caps and the New ones. Ideally should I just go with the 22000 or do 2x 10K in parallel ?
 
Lavane said:
What's the diameter of the original cap?
Click to expand...
Old Caps 40 mm x 101
New caps 50.6mm x 100

There is enough clearance to mount a 40 x 112 as long as it fits exactly in the mounting holes. If the cap is larger at the base and an adapter is needed 40 => 50 like the one i made you lose 10 mm. You need 10 mm from the board to the tallest component.
 
IMO, chasing the capacitance is spinning wheels. 22000 vs 20000 is not what caused that resistor to burn. Focus elsewhere.

Turning a trim pot and observing no change is a bit scary. My first inclination is something in the circuit/adjustments caused it.

What is the number of the resistor that burned?
 
Last edited:
The 22 vs 20 isn't worth wasting a brain cell thinking about. No big deal. I can't see how the cap can could have touched the resistor, but that's the only logical explanation, if the problem is cap related. I'd just replace the resistor, but reverse it so the body is away from the cap and flatten the lead on the other side so it can't hit anything. Most likely problem is the lack of change when you adjusted the trimmer. That suggests there was a problem in the circuit. A good rule is "explain all anomalies before continuing! I can't figure out the connections on the German schematic that's online and I don't see how the trimmer problem suggests something else, other than maybe a bad trimmer. I'd do an in-circuit test with an ohmmeter. If you replace the resistor and everything adjusts OK, it might well be OK.
 
Last edited:
ConradH said:
The 22 vs 20 isn't worth wasting a brain cell thinking about. No big deal. I can't see how the cap can could have touched the resistor, but that's the only logical explanation, if the problem is cap related. I'd just replace the resistor, but reverse it so the body is away from the cap and flatten the lead on the other side so it can't hit anything. Most likely problem is the lack of change when you adjusted the trimmer. That suggests there was a problem in the circuit. A good rule is "explain all anomalies before continuing!
Click to expand...

Understood thank you! I assumed that the Instructions were incorrect.
There are two variable resistors vr301 and vr401 that are not mentioned anywhere in the service manual. I didn't really touch them.

Instruction for Step 3 and Step 4 mention the exact same Variable Resistors VR302, VR402 Adjusting VR302 / VR402 results in no voltage change across C302/C402 that is why i assumed yesterday that the instructions were wrong but i was too freaked out to go back and validate my ideea.
 
Bogdan Baba said:
I do not know what the original tolerance is these are ~1981 caps. They are also Nichicon.
Click to expand...
Looks like they are code M = +/- 20%.. Its printed right under the operating temperature of 85°C..
The "Gold Tunes" are KG(M) as well +/- 20 % as well..:thumbsup:
 
What's the lead spacing for those big filter caps? I found some 22,000uf 63v, 35mm x 70mm with 10mm lead spacing over at Partsconnexion. They are M-Lytic brand. Good caps, I've used them in a few tube amp builds.
 
Lavane said:
What's the lead spacing for those big filter caps? I found some 22,000uf 63v, 35mm x 70mm with 10mm lead spacing over at Partsconnexion. They are M-Lytic brand. Good caps, I've used them in a few tube amp builds.
Click to expand...
Original caps 18
New Caps 19
I found the caps you were talking about they would be ok but since i already purchased the 22000 Nichicon's and i 3d printed an adapter i'll stick to that. thank you so much for the Partsconnexion didn't know about it
 
I want to thank everyone for their replies I finally put the amp back and found out what exactly shorted the resistor and did all the adjustments again.

What shorted the resistor? I'm happy you asked
I wanted to lengthen the wires from the power transformer because it was super difficult to open this power amp. The new wires I purchased were a bit thicker then the existing ones and had silicon sleeves.
The bottom cover had a rubber square that was pressing the motherboard from below to balance the heavy caps on top. Unfortunately I didn't pay attention to this and the extended wire was being pressed into the motherboard by the rubber square and the pins of resistors/diodes and what ever was under it where pressing into the silicone sleeve and made it though it to contact the wire. I inspected the wire and found small puncture holes.

So yea that happened I'm just happy I didn't **** the whole device up... not a very expensive lesson to learn.

The instructions for the Adjustments at step 3 were wrong in the manual it's actually the vr 301/401 that adjust that step.
 
You must log in or register to reply here.
Back
Top Bottom