Mackie CR4 amplifier no sound

[momaka]

Originally Posted by UserXP

I unsoldered one leg of capacitor C39, labelled as 50V1uF. The ESR meter displays:

965nF
ESR=1.5ohm
Vloss= 0.4%
...
To me, the ESR data looks normal for a capacitor with those ratings.

Yes, both ESR and capacitance look normal.

However, the problem with many of the cheap crap brands is that their leakage current can go high once voltage is applied. Well, that can happen to any electrolytic capacitor, really. But it just happens more to the cheap crap brands.

Originally Posted by UserXP

Does this capacitor need to be 1uF? I have Panasonic 50V2.2uF I think FR series at my disposal. Will it be fine to use it, or should I go for the original 1uF?

No, the cap doesn't have to be 1 uF. The higher the capacitance, the slower the amplifier will switch between STB, Mute, and Play operation. So with around 2x higher capacity, you'll probably get 2x slower transition between these states... which you should not notice at all, because even with the 1 uF cap, the switch between the states is almost instant. So with 2.2 uF, it will be "slightly less instant" (that's a highly technical term, OK. ) I'll give an explanation of what C39 does a little further bellow.

Originally Posted by UserXP

If yes, what brand/series would you recommend for a cap of this function and this close to the heatsink?

Any brand and series will do.

However, because these caps are close to the heatsink and are exposed to heat, I'd suggest going with 105C -rated caps. GP 105C caps will probably be fine, but the Panny FR you have should be a nice treat, since those are quite reliable and long life.

Originally Posted by budm

At this point I would get a couple 1uF for this unit and the working unit since more likely the one in working unit will fail later on as well

+1000

In fact, I suggest getting as many of the caps replaced as possible, especially the ones in the hot areas. All the caps look like no-name cheap brands, and sooner or later they will fail. So probably a good idea to recap the working unit, too.

Originally Posted by UserXP

What I don't understand is that one pad of C39 measures +20V and the other (the one on pin 5 lead) measures +13.90V. What kind of a capacitor is this, how can it have two positive supply rails?

So, the function of this cap is kind of similar to C74, except instead of providing a delay (by taking time to charge through a resistor) for the power-on mute, it provides a delay when the amp is switched between STB/Mute and play when the headphones are plugged or unplugged. Otherwise, going abruptly between these states could cause crackle to be heard through the speaker, as you noticed already.

Just remember this: capacitors act sort of similar to very tiny tiny batteries in the circuit and as such, they don't like to allow sudden voltage changes across their terminals. So whenever something in the circuit tries to change voltage abruptly, the capacitor will smooth that voltage change into a more gradual one. When you add a resistor in series (in the case of C39, it's resistor R91 and C-E junction of Q2), the change of voltage across the terminals of C39 becomes even slower/"smoother". So this allows the amp to go more slowly between STB, Mute, and Play operation.

Originally Posted by UserXP

I guess the temperature might be influencing the capacitors. They are some "Cap-Top" brand, all of them. The small ones are especially prone to drying due to heat.

Yes, no-name / crap brands and heat don't mix well. I find that in a lot of 90's and 2000's budget/mid-tier audio gear, if there are any no-name or crap cap brands, the ones located closest to any heat sources are usually the first ones to fail. And even the ones that aren't can sometimes fail too. Like you noted, it kind of varies with the batch. So you can have the same cheap brand of caps last a long time in one unit and fail rather quickly in the next. For this reason, I always recommend to use only Japanese capacitors from the big 5 (Panasonic, Rubycon, Nichicon, United Chemicon, and Sanyo/Suncon) - at least if you don't want to have to do this repair anytime soon in the future again.

Originally Posted by UserXP

I'd like to put one that is similar to the other caps: black sleeve, white lettering and a cros vent. Of course, this is just for rare visual purposes.

How about replace the rest of the crap caps with good ones then? That's one way to get a more "uniform" look.
In all honesty, though, I don't know why you should care about that. Unless the unit has a clear window with lights inside pretty much begging people to stare... you're just not going to see the insides again. So IMO, you should put whatever cap brand actually does the job reliably rather than focus on the looks.

Otherwise, if you want some really cool-looking caps, I have some bright-yellow CapXon KM caps from an Apevia PSU. They look kind of catchy with their yellow sleeves... but no telling when they will fail (well, some of them have failed already, which is why I removed them all and replaced with good Japanese brands.)

[UserXP]

Momaka, thank you for this informative post and all the help, especially your patience, really from all of you. I must have asked all sorts of noob questions and thank you for making concrete, clear and wise input on them.

You are absolutely right, changing all the caps would probably be the best step. I was thinking about that when I had issues with my original working unit, but then somebody posted that they replaced all the caps and got weird voltages in some areas, so I was reluctant, especially since I managed to solve the problem. I guess in each build there is a bad cap and it's a lottery which area it's going to appear in. I will need to examine all of the caps on this PCB. Some are quite obstructed by other components and many are glued, all this hiding their ratings. And of course the looks is not an imperative, I'm just so happy we got this thing working again.
I will order some other ratings of FR caps if I can find them on Farnell. Hopefully, I will find most of the caps values on those segments from the schematics.
I will look tomorrow for some Panasonic 50V1uF caps and order them.

[UserXP]

Guys, just an update on this. The problem was definitively the C39 capacitor. I had Panasonic 50V2.2uF FR version, but I wanted to stay within the amplifier's specs. So since you mentioned it can be a general purpose capacitor, I found some Rubycon 50V1uF 105C 5000 hours YXF series capacitors on the local Farnell website and ordered them. I am delighted to say they have arrived and I have installed one today. The amplifier works great now and hopefully this cap will cope well with the heat the amplifier produces. The voltage on the mute pin now hovers around +13.72V in play mode, which is great! Thank you all so much, for identifying the part, and especially you momaka for your thorough and patient step-by-step instructions.

[momaka]

No problems!

Rubycon YXF is actually one of the most entry-level low-ESR series, so it is better than regular GP caps. Better yet, it's still rated for long life (5000 hours for 16-100V rated caps in 6.3 mm dia. or smaller... and more for bigger caps), so that's actually a pretty good choice.

Should replace the caps now in the good unit too, while you're at it and still remember what the issue is, rather than try to figure it all out again some years down the road. And you'll probably have more peace on your mind knowing it's been fixed properly too.