Leaking Capacitor, or No?

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: lisa, luketeaford, Kent, Joe.

Post Reply
User avatar
wadesey
Common Wiggler
Posts: 59
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Leaking Capacitor, or No?

Post by wadesey » Fri Jun 26, 2020 12:03 am

When I was cleaning my Mono/Poly I took a look at the electrolytics and noticed green corrosion on the negative leg only of almost every one on the control side boards, and a few on the bottom boards. Mainly on these identical 10uf ones. I don't think any are noticeably bulging at the top or bottom and the big power supply ones look OK...

Could this be just from exposure, or am I kidding myself and they're all leaking? (no other components have that oxidation/corrosion...)
You do not have the required permissions to view the files attached to this post.

User avatar
papz
Super Deluxe Wiggler
Posts: 3201
Joined: Sat May 18, 2013 6:52 pm
Location: in a suitcase
Contact:

Re: Leaking Capacitor, or No?

Post by papz » Fri Jun 26, 2020 5:50 am

Yes leaking caps.
Should be replaced.
They exceeded their 3000 hours lifespan a few times since 1981.
Finest EMS gear service and Music Easel/208 goodies

I'm not a synths dealer and I'm not aware of an inexpensive Synthi secret market, please don't ask me if I know one for sale.
I don't offer support of any kind to people attempting to build clones of EMS equipment.

User avatar
wadesey
Common Wiggler
Posts: 59
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Leaking Capacitor, or No?

Post by wadesey » Fri Jun 26, 2020 12:45 pm

Thanks for the confirmation slap. I'm not looking forward to the expense, but I knew it was long overdue...

transistorresistor
Common Wiggler
Posts: 109
Joined: Mon Apr 27, 2020 9:38 pm

Re: Leaking Capacitor, or No?

Post by transistorresistor » Fri Jun 26, 2020 5:41 pm

wadesey wrote:
Fri Jun 26, 2020 12:03 am
and they're all leaking?

everyone listen up and tell all your friends:

The term "leaking capacitor" has absolutely *nothing* to do with gooey stuffs "leaking" out of the physical container that your cap is in. Oxidation on a leg has absolutely no direct correlation to whether or not a capacitor is "leaking" or not. Now sometimes goo does come out but the technical jargon to reference this officially is "letting the goo out" (this conjucates as "that cap let its goo out" among common variances) but that is the topic for another day. Please read on.

A capacitor can be looked at very basically as a dual function device that allows AC signals (your sound) to pass and BLOCKS DC voltage (your cv, the power rails, etc) from going through. The optimum performance of an audio circuit that employs capacitors is dependent on those capacitors blocking DC current.

The phrase "Leaking Capacitor" is short for "Capacitor Leaking DC".

That means the cap is passing both the stuff you want it to and the stuff you absolutely dont want it to and therefore is broken, or at a minimum performing out of spec enough to warrant replacement where the circuit can continue to limp along in degraded performance slowly until it eventually fails. Not ironically, the sound that some people identify as "vintage" very often is just the sound of tired leaking caps and THAT is most assuredly a discussion for another day.

As capacitors age, the material inside dries up and as it does capacitors slowly become resistors. When this starts to occur, the ability of DC to be blocked diminishes and DC voltage will LEAK through the cap. This is when its time to replace. This is what technicians are referring to when they say "leaky caps". Caps which are passing DC. Not goo. Direct Current. DC is not supposed to flow through, or leak through a cap under normal common circumstances.

Most electrolytics should be replaced in any device after ten years and any electrolytic cap that has been around long enough to have oxidized legs is a good candidate for replacement, but there is no way to look at a rusted cap and make the assumption that it is Leaky. I mean, damn good assumption, but I have tested caps that looked so wretched and were functioning to spec. Changed em anyway cause I was in there, but until you put the appropriate meter on a cap to measure for DC there is truly no way to just eye a cap and know if its performing to spec or not.

The oxidation on the leg in the photo is just that, oxidation, and it can come from any variety root causes however the "capacitor leaking" is not one of them.

This phrase is one of the most misunderstood things out there by people throwing terminology around. Holds court right next to Class-A discrete...

KSS
Super Deluxe Wiggler
Posts: 2449
Joined: Mon Jan 25, 2016 7:28 am

Re: Leaking Capacitor, or No?

Post by KSS » Sat Jun 27, 2020 6:49 am

transistorresistor wrote:
Fri Jun 26, 2020 5:41 pm
everyone listen up and tell all your friends:

The term "leaking capacitor" has absolutely *nothing* to do with gooey stuffs "leaking" out of the physical container that your cap is in. Oxidation on a leg has absolutely no direct correlation to whether or not a capacitor is "leaking" or not.
Sorry transistorresistor. This is simply false. You've written some very good things here, but starting with this leading false statement is *not* a good look. Let's get one thing perfectly clear right away. The word "leaking", as applied to electrolytic capacitors, has more than one meaning. Unfortunately you've conflated and confused things so much we'll have to take it line by line. Because there is in fact a direct correlation between the environmental factors -both micro-climate and general environs- and whether a capacitor has physically leaked. And when someone says they have -or asks if they have- a "leaking capacitor" the very first thing we need to find out is which meaning of leaking they are asking about.
Now sometimes goo does come out but the technical jargon to reference this officially is "letting the goo out" (this conjucates as "that cap let its goo out" among common variances) but that is the topic for another day. Please read on.
No. The technical jargon for this is "venting", and its hardly "the topic for another day" in a thread asking specifically whether the discoloration and oxides-salts buildup of the leads in a few photos is a physical indication of such venting.
A capacitor can be looked at very basically as a dual function device that allows AC signals (your sound) to pass and BLOCKS DC voltage (your cv, the power rails, etc) from going through. The optimum performance of an audio circuit that employs capacitors is dependent on those capacitors blocking DC current.
Agreed.
The phrase "Leaking Capacitor" is short for "Capacitor Leaking DC".
No, that is only ONE of the meanings of the term "leaking capacitor". It's a true statement alone. It's good that you're bringing up tis subject. However, in the context of this thread, it's misleading. Another equally valid meaning is reference to whether the capacitor in question has vented to a small or great degree. goo? Let's keep it real. Venting does not need to have any 'goo' present to be an indicator such venting has happened.
As capacitors age, the material inside dries up and as it does capacitors slowly become resistors. When this starts to occur, the ability of DC to be blocked diminishes and DC voltage will LEAK through the cap. This is when its time to replace.
Agreed.
This is what technicians are referring to when they say "leaky caps". Caps which are passing DC. Not goo. Direct Current. DC is not supposed to flow through, or leak through a cap under normal common circumstances.
To be correct, you would have to say this is one of the things technicians may be referring to". Or even, "usually technicians mean"
But to claim it as an always true blanket is not factual. The good point you again bring up is that an electrically leaky capacitor is not a good thing.

As capacitors age, a few things happen. Most are dependent upon environment and some will leave traces. Others may not. Internally the electrolyte is being absorbed into both the foil and the dielectric. This is the primary cause of the increasing electrical leakage and is also a reason for the always present physical leakage. Outgassing. Venting. Maybe with goo, maybe not. These two processes work together to bring about both the drying up and the shift towards resistor instead of capacitor. As the electrolyte is absorbed by the dielectric over time, there is an increase in heat also. This heat causes expansion, which leads to venting. This venting *can* be a factor in the type of lead discoloration and salts build up as seen in the OP photos. It's not necessarily goo. Gases also escape from the capacitor interior.
Most electrolytics should be replaced in any device after ten years and any electrolytic cap that has been around long enough to have oxidized legs is a good candidate for replacement, but there is no way to look at a rusted cap and make the assumption that it is -electrically- Leaky.
With my added word your formerly false statement is now generally true. I take exception with your "ten years", but note that your well-placed addition of the lead condition, and use of the word "candidate", since we both know the truth of the next line below.
I mean, damn good assumption, but I have tested caps that looked so wretched and were functioning to spec. Changed em anyway cause I was in there, but until you put the appropriate meter on a cap to measure for DC there is truly no way to just eye a cap and know if its performing to spec or not.
Excellent point. Entirely agree.
The oxidation on the leg in the photo is just that, oxidation, and it can come from any variety root causes however the "capacitor leaking" is not one of them.
No. Again you're forgetting there is more than one meaning to leaking as applied to an electrolytic capacitor. It *is* true that the lead condition of the photos can come from several different sources. But it's ALSO true that "one of them" *is* potnetially from physical leakage of the capactior.
This phrase is one of the most misunderstood things out there by people throwing terminology around.
I believe you're correct that many do not realize there are two different meanings for the term leaking capacitor. Thank you for bringing the lattice upon which to build a better understanding of the whole issue.

User avatar
mskala
Super Deluxe Wiggler
Posts: 2596
Joined: Thu Mar 27, 2014 7:33 am
Location: Toronto
Contact:

Re: Leaking Capacitor, or No?

Post by mskala » Sat Jun 27, 2020 7:01 am

I basically agree with KSS's points, and would add that electrical leakage may be a more common issue with tantalum electrolytic as opposed to aluminum electrolytic caps. Aluminum electrolytics tend to fail open, without a whole lot of electrical leakage, but with tantalum electrolytics, the leakage current may be considerable, and the heat from that accelerates the physical degradation of the cap leading to even more leakage current and more heat, and that's why they explode.

transistorresistor
Common Wiggler
Posts: 109
Joined: Mon Apr 27, 2020 9:38 pm

Re: Leaking Capacitor, or No?

Post by transistorresistor » Sat Jun 27, 2020 2:44 pm

Hi KSS-

Happily agree with your perspective on the points you make. Perhaps the sarcastic jokes we have at our shop bench about goo did not bring the entertainment value I had hoped, electrolytics almost always vent dry in the old stuff I see, we restore alot of 50's hifi here (or used to at least) and those old wax and paper caps can make a gooey mess. A conversation for a different day for sure.

My experience either w people I have worked for, contractors that work for me or colleagues across a dinner table or at a trade show is that nearly universally, a discussion of "leaky caps" is in reference to their electrical integrity in the circuit. Venting has always been for us just that, venting. Prior to this convo I never would have used the term leaking to describe a vented cap and would have presumed these two words to not be interchangable, hence my post. I have, for instance, never once in my life been in discussion about leaking caps (which as a guy that recapped recording consoles for years is probably more conversations than I care to admit haha) and ever once been asked if I was referring to physical or electrical leakage. Its a big world though, Ive only worked in studios in the US, Europe, and Scandanavia, perhaps Im in the minority but Im happy to be informed otherwise. I did some quick google and for sure there is reference which just seems awfully confusing to me, there's always been a clear distinction made in terminology in my discussions on this topic for the last 25 years. Enough so that I had the stones to correct someone on the internet!

Sitting here w pause to consider colloquialism in our industry, and perhaps its because all my colleagues and I are a bunch of tube heads, but my relationship to the concept of "leakage" is born in grid leak bias (I know you didnt ask but Im more of a fixed bias kind of guy) so leaking is deeply burned in the brain as referring to the trickling of current through a device.

Now I guess where this leads me is that over the years we have seen caps that may have vented but were not leaking DC. Caps upon inspection, just like the photo posted, that *clearly* looked problematic but you pull them, put them on the sencore and they are leaking zero DC and turn out to not be the failure that you are trying to find. That cap would usually get replaced since you pulled it to measure it anyway, but we'd never (never ever) refer to a cap that tested good on the sencore as "leaky" even if it had obviously vented. We'd for sure refer to it as garbage though life is too short to try to resolder a tightly trimmed radial cap on a 40 year old pc w lifting traces ;)

Over the years we'd get studio owners who were having problems w a channel on a console or tape machine or some outboard gear, theyd have a completely dead channel and approach us w some version of "I think my caps are leaking, can you check" and then you sarcastically charge your partner w "check to see if it let the goo out" because you know, if the owner was able to diagnose these problems he'd have no reason to call us... In the next moment you'd also caution him on not "letting the smoke out" upon applying power after the repair. Too much Zappa for us in the 90s. I risk sharing this anecdote for it to not work on the internet TWICE, but maybe someone out there will find it amusing.

Ive also done some limited touring as a guitar tech looking after a colleague's backline and have again and again heard this phrase, often uttered after a bad gig "hey man my sound was sucking tonight can you check for leaky caps" where the only exhausted response can be no dude, we rebuilt ALL these amps during production rehearsals remember? amp is good, maybe dont get so shitfaced before the gig tomorrow night. If you are lucky you get to work for the guy that comes to the side of the stage during soundcheck and says "well if you checked the bias and its good why does my amp still suck, maybe my caps are leaking". Yeah man, maybe.

It is a phrase that is loosely thrown around by completely non-technical folks in my life so I have a knee jerk to it, perhaps that reaction swings the discussion to an equally unhelpful place in the opposite direction. I am in fact only semi pro haha, I dont really know what Im doing. Years of solder monkey making consoles and amps work accidentally led to design and manufacturing, Im here to learn not lead, the products we make today are an order better than what we made 10 years ago and that is because of an open mind. It is indeed a bad look to have the information I try to share be construed as destructive not instructive, that is for certain.

So my clear, Goo Free thoughts for clarity:

1. a physical and visual examination can and should lead to candidacy for detailed testing and/or replacement

2. physical deterioration of a component should not necessarily lead to conclusions about electrical performance upon visual inspection alone, it is an excellent place to start though, see #1.

3. there is no substitute for test equipment when determining if a capacitor is leaking DC, see #2

I for sure learned number 3 the hard way when I was starting out working on consoles, shotgunning heaps of caps that looked terrible, sure that that was why the channel was dead only to replace them all to discover the problem was somewhere else. And then you get an LC53 and your life is forever changed like dry clothes after a rainy day.

number 2 was also a hard lesson learned (which yields to number 3) because in a thing like a recording console (and a synth), any circuit where there are PC mounted pots, you get the 7 guys that opened the thing up before you spraying all orders of aerosols into the pots instead of just replacing them, and you know, *sometimes* deoxit guy doesnt have the best aim and other times he sprays enough for it to *LEAK* out of the pot and run down the board and onto the surrounding components. So sometimes, when you open the case you see oxidation on cap legs and the cause can just be the accumulation of stuff sprayed or spilled in there that migrated to the components legs and slowly oxidized them. Sometimes you get components that may have not been tinned well when they were made so that kind of exposure will for sure be accelerated and the parts surrounding them look clean . Ive been sure caps that looked like that were trash. And sometimes its just "cleaner" which you know, sometimes in idiot world is wd40.

Regarding tantalums, I typically see them fail as shorts but man wet slug tantalum? Im pretty sure those things are gonna survive the end of time, Ive never seen one of those fail in any circuit.

Anyhow, delighted to bring the lattice, if we arent learning we are just biding out time. Naming my next band Bringer of Lattice and nobody will ever understand ;) So, tell us about the name of your band. Well, this one time on the internet...

User avatar
wadesey
Common Wiggler
Posts: 59
Joined: Mon Oct 10, 2011 3:01 pm
Location: Long Beach, CA

Re: Leaking Capacitor, or No?

Post by wadesey » Sat Jun 27, 2020 3:33 pm

Thanks for the informative and entertaining reads!

I think it's SUPER important to use accurate terminology and would hate to perpetuate the wrong usage. I guess "leaked" would be more clear for past tense "leaked goo" vs "leaky" or "leaking" for active electrical leakage. Either way, I'm describing it differently now for clarity.

And I definitely don't want to go in blindly replacing parts or spraying de-oxit everywhere to solve problems I don't understand, especially on this synth. There has been a noticeably rising noise floor the past 10 years or so, plus MG1 has started to vary pitch, even when pitch isn't selected, so it's becoming clear that some parts are starting to fail, finally.

KSS
Super Deluxe Wiggler
Posts: 2449
Joined: Mon Jan 25, 2016 7:28 am

Re: Leaking Capacitor, or No?

Post by KSS » Sat Jun 27, 2020 8:45 pm

Member Pelsea recently wrote in another thread about the different perspecitves of 'musicians' and 'engineers'. He taught EMusic at UC Santa Cruz for 30 years plus. While I dislike the overly simplified version of one v the other, I can appreciate his point. That people think differently and so talk and approach things ldifferently.

Seems although unlike you, I've heard and used leaky in both meanings since forever, we share similar beliefs and experiences. I'm no fan of 'shotgunning' repairs, and work hard to present better alternatives. Also agree on 'wet slug' tants. Never seen one fail.

And Deoxit? Best we not go there at all. I've known it to cause green crusty buildup and deterioration on PCBs and parts in old synths.. ;)

If "Lattice" ever performs nearby, I'll be sure to attend. I'll bring the bacon and tomatoes. "Bringer of Tomatoes" may not be an especially good name for a performer, though.

Post Reply

Return to “Music Tech DIY”