Re-Capping

No. The caps in the photo are polycarbonate or polyester capacitors, not tantalum:

I like to buy replacement recap kits on eBay (because I'm lazy). So far, they all worked as expected. Where I had problems was when I tried to save money with used parts or salvaged parts, or didn't bother testing the replacement caps before soldering.

Is there any evidence of failing capacitors?

Just generally poor audio which I can't pin down to a specific stage. Tantalums are not noted for their reliability and seem to be the prime suspect in cases like this.

Yeah, I'm aware they're different types, Jeff. But AI said it would be fine to swap tants for polys in an audio amp chain of stages. However, I'm not sure polys can match the tants for their low impedance at high audio frequencies and I could end up with too much base at the end of the day. I know ideally I should replace like for like, but I've been looking for a reason to use up all the thousands of polys I have and AI said 'go right ahead you'll be fine to do that' hence my getting a second opinion here. I've had a lot of bum steers from AI - you have to double check every suggestion it makes no matter how logical it may sound!

You should be fine replacing the coupling caps.

Whether it works on the decaps sepends on whether the circuit requires the higher ESR of the tants for stability. Some do, some don't.

Cheers

Phil Hobbs

Thanks, Phil. You're the word of God AFAIC. :-) I'm also grateful to our other contributors of course. We often disagree over fine points like this and sometimes it's impossible to know whose approach is most appropriate in any given circumstance.

If the DC voltages on the various transistors are correct, the volume control isn't noisy and there is no hum when working from the mains, I would leave the capacitors alone and look elsewhere for the problem.

On 2026-01-21 12:46, Cursitor Doom wrote:> On Wed, 21 Jan 2026 12:42:50

-0500, Phil Hobbs > snipped-for-privacy@electrooptical.net wrote: >

Just what I need, a head bigger than the universe. :(

In places like the tone controls, where the impedance levels are higher, it might be worth paying attention to the "outside foil" dot, if there is one. (Wound caps usually have the dot, but stacked ones don't.)

Cheers

Phil Hobbs

I'm going to have to start from scratch again, anyway. The reference tape I made which I had implicit faith in despite not checking it for fidelity first turned out to have every bit as much distortion as that which the fault itself was giving rise to. Assumptions. The mother of all f*ck-ups! So now it's back to square 1. :-(

"That which is obviously correct, beyond any need of checking, is usually the problem". I forgot where I stole that.

If you seem to have hit a wall, and can progress no further along the current line of reasoning, then stop and check your assumptions. One or more assumptions will likely be wrong. If that doesn't yield results, verify your procedures to make sure you're not dividing by zero. If everything sounds distorted, remove your ear plugs, clean the wax from your ears, and turn off anything that might be producing out of band noises, such as ultrasonics or ringing.

The number 1 squared is still equal to 1. It also stays at 1 no matter how many times you square it. No matter how many times you repeat the mistake hiding in square 1, your answer will remain unchanged and wrong.

Tantalum caps do cause problems. However, the problems appear as noises, not distortion. "Tantalum capacitor causing scratchy noise in phono input?"

Yes, I agree. I'm the King of False Assumptions.

A useful link - I'm only sorry I couldn't read much of it before Facebook booted me off (I don't have an account with them). The fact is, this fault is presenting as distortion which I have previously described as sounding like a worn out vinyl record: scratchy and noisy in general. If that doesn't sound like a tant gone rogue, I don't know what does. The link you posted backs this theory up, so many thanks for that.

Yeah, I've had some good results with that stuff in the past. However, I've only used it when a fault develops after the equipment's been switched on for a while. I then spray it to see if it 'cures' the problem. Not sure how much use it would be in this case where the fault is present right from switch on. :-/

Freeze spray works well with semiconductors and not so well with capacitors and resistors. The big problem is condensation. If you spray a PCB (printed circuit board) with freeze spray, moisture will condense on the PCB. For any PBC or components that are not water tolerant, it will cause the circuitry to fail until you dry off the PCB. You can do that with a hot air gun or oven to accelerate evaporation.

Pre-heating the PCB before testing with freeze spray, is a bad idea. Most common components are not rated for thermal shock. Cracking components, such as MLCC caps, are possible. If you've heated the PCB to accelerate evaporation, I would let the PCB temperature drop to room temperature before repeating the freeze spray test.

You're not looking for a "cure" for the distortion problem. You're looking for something that causes the distortion to CHANGE. It doesn't matter if the distortion becomes better or worse. If the component is sensitive to changes in temperature, it's a likely culprit worthy of replacement.

Hint: Make a shield. Punch a hole in a sheet of cardboard or plastic. Place the hole over the likely culprit and spray it. The idea is to only apply freeze spray to one component at a time.

I think I'd just wait for ambient temp to do its work, tbh.

Good point! Thanks, Jeffrey. You're right of course (as ever).

I'd go for the mask idea in the endgame as it were, but I'd just have a spray around various small areas containing several components first, otherwise the process would take too long. I like to zoom in..... :)

Using a heat gun or hot air blower is not that difficult. I use one of several Type K thermocouple wire probes:

Y're welcome. I'm basically lazy. I like to reduce the list of potential culprits to a reasonable number of components and then replace everything. That's why I like recap kits (replacement capacitor kits) which are sometimes found on eBay. It takes me less time to replace all the caps than to figure out which one is causing problems.

If speed of repair is your goal, then throw money at the problem and replace everything. "Good, fast, cheap. Pick any two".

You are aware that often when AI doesn't actually know the answer it simply hallucinates something. And it doesn't TELL you that it doesn't know so how do you trust anything produced by it.

I often see Google AI suggestions that are total crap - repeats of the top ten sites, which are mirrors of each other. Back to Duck Duck form my search engine!

Try using data sheets and learning about the differences between capacitors or you will always be just another hack.

John :-#(#

I have reached that conclusion, yes. Hence why I was running the suggestion past the Panel here before being so rash as to run with it.

Yeah, but it does provide all that crap much faster and more efficiently than I could find it for myself.

No point. I'll always be "just another hack" due to my lack of manual dexterity alone. What's your excuse?