Is this capacitor leaking

[CEAuke]

Hi guys

I'm trying to repair an electric device and I *think* the issue is capacitor leaking. However, I'm not sure if they pain goo over the capacitors for some reason or if it is in fact leaking. Can someone please have a look at the two pictures and let me know if I should replace the caps?

The first picture has the little cap between the white interface ports. The second picture clearly has some goo between the caps.

As a side note, do I just order the same voltage and uF rating (and maybe max temp?) or are all caps not created equal.

For a bonus point Is there a way to test a cap without an osciloscope

Thanks
CE

 

[MrAl]

Hi,

Post bigger pictures.

You can test the capacitance if you have a voltmeter and frequency generator or some other sine source.

 

[CEAuke]

I now took a few more pictures of the smaller one. The two large ones, I think have some intended goo between them. but the smaller one has a rough patch on the side (maybe overheated) and there's an uneven surface on the small 'transistor' next to it. So I'm wonder if THAT is maybe leakage.

 

[MrAl]

Hi,

Still kinda hard to tell as the pics are a little blurry in places we need to see clearly.

Maybe a few more pics, from all 4 directions (North, West, South, East).

What is that brown on the right side there?

That small part is a transistor.

 

[CEAuke]

And here a larger version of both of them with the suspicious goo. The brown stuff.
The problem is, as you can see on the interface connectors, these guys clearly paint some goo to SOME pieces but the interface ones are more reddish.

Apart from that: No bloating on top of the caps, no deformities from the top or the side.

 

[picbits]

The "goo" looks like glue to stabilize the rather large capacitors - without it, they may put excess stress on the solder joints during transit.

The small transistor in the circle appears to have already been replaced before.

 

[CEAuke]

Here's a view of the accused 'damage' on the side of the little one.

...and then a headscratcher for me not beging familiar with dual layer PCBs.
In the image below you can see that the cap connections only really has one connection. (on the negative pin)
So I guess the positive pin connects on the top side but the cap sits flush on the board. How on earth do you solder that? I guess if I need to replace it I can keep the cap on the long legs and just ensure that it's soldered on top of the board as well as the backside clearly does nothing. Am I right?

 

[CEAuke]

Here's the final photo that I have. Maybe someone replaced the transistor and damaged the
side of the cap without affecting the electronics.

There's a grey speck at the bottom of the cap where the insulation is melted through. Unfortunately I guess it can still relate to either of the two scenarios (cap overheat or bad transistor replacement)

 

[cowboybob]

... In the image below you can see that the cap connections only really has one connection. (on the negative pin)
So I guess the positive pin connects on the top side but the cap sits flush on the board. How on earth do you solder that? ...

CEAuke,

To facilitate a trace that must pass from one layer to another on a 2 layer PCB (even if only for a mechanical, component holding reason), a "via" is used. It's a simple, solder plated thru-hole top to bottom. On PCBs with more than 2 layers, some vias are invisible, but do not, generally, involve external components, just connections.

 

[picbits]

To me that "damaged" cap looks like someone caught it with a soldering iron and melted the outer plastic which replacing the transistor next to it.

I see two connectors for that cap on the underside of the PCB - I suspect if you heated both pads simultaneously the cap would lift out.

 

[tvtech]

It's friggin glue that changed color...caps look fine.

Another pet hate...why do some Ching factories out there insist Glue solves all connector/cap problems
Favourite irritation of mine....in order to get the plug out...you have to break the leads....so stupid.

Why put a connector there anyway???

Not in this case but worth thinking about....do not let Glue fool you..or confuse you..Cluck.

It is everywhere.

Cross Caps vent from the top and bulge there too....

Regards,
tvtech

 

[picbits]

Another pet hate...why do some Ching factories out there insist Glue solves all connector ........

Especially bad on the 30 pin LVDS connectors to flat panels - you get the glue off but end up knackering the connector and all the pins fall out !

 

[tvtech]

Heya Buddy..

And double sided or more layers too

Started out as straightforward...now becomes a Mission
Because of glue and all the stuff not understood..And the plug was not even the problem..

Makes me mad

 

[picbits]

I've just started using a Metcal PS-800 soldering iron and it makes life a lot easier for the double sided and larger copper surfaces.

 

[tvtech]

^^^Hi pic

Building boards with more than one layer is always nice and quick with a good iron...
The problem is fixing double sided or more layers...they are meant to be thrown away...not repaired..

What say you Buddy?

Regards,
tvtech

 

[CEAuke]

Hi TV

Cross Caps vent from the top and bulge there too....

What do you mean by that?
Edit: Aaah: maybe whatever a cross cap is, if it's broken I will see a bulge at the top and should see it venting there if it's broken?
I guess the conclusion is that the caps in the pictures are most likely working fine. Thanks for all the help! I'll continue my search. But I'm keepin an eye on these guys

 

[picbits]

They can still be bad without and outward signs but without an ESR meter you can't really tell.

 

[Mickster]

I think we need to start from scratch here and have the OP describe what the device is, what it is supposed to do, what it is not doing and then take it from there.
Globs of glue, dribbled, such as what appears to be Evo-stik or something similar, are sometimes used for 'mechanical stability' as indicated by picbits and tvtech.
Additionally, as indicated by picbits, the transistor appears to have already been replaced using a hot-air-pencil and during replacement, the tech scorched the side of C44.
Why did the transistor need replacing and has the transistor failed again, due to something else being at fault?

Let's start from the beginning.

 

[KeepItSimpleStupid]

My take:

The transistor was replaced and they damaged the cap with the iron and did not clean off the flux. Bad things CAN happen. Ask Murphy.

It's normal to use some type of glue or goo to hold large capacitors.

The concern about the capacitor in post #7 is that it's a thermal relief. Note the tiny radial connections to the large plane. This improves the ability for the wave soldering process to get the pad hot enough, Otherwise it would take forever to heat the large "copper plane". So, the thermal relief increases the soldering yield. Nothing fancy.

To check, just take an ohmmeter and measure from the seemingly "unconnected" pin to the large copper plane and yes, the left-handed connection is either on the top side or in the middle of the board. Replacing these without damage to the board is very tricky.

 

[CEAuke]

Hi guys

Remarkably after I reinserted this circuit board everything started working again. It must be all the good karma I got from this forum. I did find a small button battery that seems totally flat which must be used somewhere with internal memeory. I ordered new ones.

The device is an electric piano which has been giving me loads of problems lately. It started out to freeze randomly. Later it seemed to freeze when more notes are pressed (as opposed to only playing one or two notes at the same time). I once had issues with it and removing the power plug from a 6 plug combo and plugging straight into the wall seemed to resolve it. Then I moved and I had trouble again. Lately it continuously restarts on startup and there's a distinct (and expected) clicking sound when the amp switches on but it immediately restarts after that. This new symptom started after I once forgot to switch it off overnight.

I started reading about voltage regulators (for my raspberry pi and microchip developments) and then it struck me that maybe I do have some power regulation issue with my piano. So I started searching for a leaky cap.

The board I showed is the amplifier part of it. After a thorough review (with my limited electronic skills) of the power supply area theres only one cap that looks remotely suspicious but that may be flux rather than leaking.

The annoying bit is that it's sometimes working and sometimes not and it seems that if I dont use it for a few weeks, it's more likely to work than if I continiously try to get it to work.

Once I have the new button batteries I'll know if it's the cause of the problem or not. I guess disconnecting the amp and the battery at least reset the machine to the factory defaults.

TLDR: It's working now. May be a flat internal battery issue.