|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
207.173.74.124
Ok, so I've been holding off on this one for a while now... I have for years been extremely skeptical of people who, with their trusty Variacs, proceed to 'reform' old capacitors into usability. Aside from the faulty logic surrounding 'forming up capacitors to voltage' and other such things that simply don't jibe with the physics of it all, not one of these people has bother to explained any mechanisms by which a leaky/defective/weak capacitor can somehow be rescued by slowly raising the working voltages.
Attached to this post is a picture of one such capacitor. It was claimed by the previous owner that this cap had been 'reformed' on a Variac, and was as good as new. Indeed, when I purchased this device (a Stroboconn Tuner) this cap did in fact work (sort of). In fact, it had already failed long ago and the amplifier was suffering from poor frequency response and motorboating. As I racked up several hours at a time powered on for this unit, the cap became worse and worse. And I should point out - had no visible signs of failure. No bulging, nothing escaping from the vent hole at the bottom (more about that later).
Meanwhile, I had wasted plenty of time replacing and testing components that were just fine, overlooking this obvious culprit because somebody had claimed it was fine due to their Variac treatment. On a hunch I ordered a new can cap from CE since I had ruled out everything else. Once the new cap was in place, the symptoms vanished. Holding the old cap in my hands, and looking at it closely, I saw no signs of failure. So, I decided to cut it open...
This one as a 1976 date code. A lot newer than some of the caps people claim to have rescued..
So, I am just going to go ahead and say it. There is no such thing as reforming a capacitor. Once the electrolyte inside has either dried out or oozed out of the cap, it's done for. Capacitance will slowly degrade until either the amp malfunctions or the cap shorts out. None of those things are good for your tubes or transformers.
If you depend on 'reformed' caps for your device, you are simply fooling yourself and wasting time by thinking that ramping up the voltage does anything at all beneficial to the capacitor. Replace old electrolytics is the moral of the story.
This is very likely what the inside of your 'reformed' caps looks like, even with no signs of bulging or venting.
Follow Ups:
Hello all,
Thanks for the thoughtful replies. A couple points. First, I should have titled this differently. I am referring to reforming of USED capacitors, not NOS. On that subject, I discovered by reading some of the articles posted here that proper procedure to reform a cap is by applying full working voltage through a current limiting resistor. Bringing amps up on a Variac doesn't really mimic this at all; you are simply ramping up voltage rather than forming it to the full rated voltage. Additionally, you have no current limiting resistor in place so any leakage current is going to be unchecked as you do this.Also, regarding the 6db or so of headroom I got from the EL84M from Jim McShane, I should note that part of the reason for such an impressive gain in, well, gain is that the old tubes were no longer drawing all that much current. They were drawing about 17.5 - 19ma a piece, which is pretty cold for this amp. After putting in the new tubes, the draw went to 23-24ma per tube and that is no doubt where the headroom is coming from. They are great tubes to be sure, I just didn't want to leave the impression that there was some magic here. No magic, just higher current pull than my tired old Dutch Amperex 7189s. EDIT -> 6db is a seat-of-the-pants measurement based on my ears alone. I particularly noticed a dramatic increase in the ability for my S-5000 to pass the bottom octaves, and was frankly astonished by how much less large bass notes robbed power from mid-frequencies than before. This thing can handle 90's hip-hop records now with little complaint and startling authority.
Here are my overall points:
1) Reforming caps is a gamble at best, and a long-odds gamble when you're talking about used, heat and ripple stressed caps.
2) The potential damage to the amp if the cap fails hard is not worth the savings.
3) The time you have to invest in reforming makes replacement even more economical
4) Even a cap with no leakage is likely to have less capacitance due to drying of the electrolyte
5) Bringing up amps on a Variac does not simulate the reforming process, it merely ramps up voltage rather than apply full voltage through a current limiting resistor.
6) A brand-new can cap costs less than a fried 5AR4It's all of those reasons that I find reforming old used caps in amps to be a losing proposition.
Edits: 12/23/15
I'm just passing on the words of a well respected designer. This refers to low voltage caps, like up to 50v or so:First rule of electrolytics is: buy only fresh. They have liquid inside, which dries out. The problem starts even with storing them too long. After 2 years the leakage current gets more difficult to clear, and that is certainly a problem in my designs. Never buy caps on ebay. They may be old, they may be fakes.
If you do want to know, test them on a bench power supply that shows voltage and current. Apply the voltage you intend, and hook up the cap through a 1K resistor. Leave it connected, and measure the voltage across the 1K and wait till current is <1uA. should only take an hour or so. Stubborn ones can be chucked. The 1K is to sense current, so measure the volts across that: 1mV is 1ľA in this case. Or use 10K to get better resolution.
Edits: 12/23/15
You'll likely never find an electrolytic cap with under 1 ua. leakage, I think he meant ma.One reason that you need to apply the high voltage is that these caps are self-healing. If leakage develops due to contact (like a small short) between internal elements it can clear itself by burning open. Once it does the leakage is reduced. you can see it happen on occasion if you watch caps being reformed - the current will jump up for a short period then fall back below where it was before it jumped up. That was due to the current passing through a "short" then stopping after the short was burned open.
It should also be said that the rate of liquid electrolyte loss varies with temperature and other storage factors. Caps kept in a cool dry environment will last longer on the shelf than caps stored in hot moist conditions. Yes, fresh is better; but good quality caps stored under proper conditions can be stored longer and be used successfully.
I am still using caps I bought and stored 20+ years ago. Most caps from the lot reformed successfully with no hot spots or bulging of the can, and had leakage current under .25 ma. after reforming at their voltage rating plus 10% (that's their surge voltage rating) at room temperature. They are Marcon photoflash electrolytics, 750 uf @ 330 volts if anyone is interested.
Edits: 12/24/15
I worked for a company that sold equipment to the military that would sit in warehouses for years until needed. It was shipped with variac procedure to reform the caps.
I have seen capacitors like that in old amps that had lots of DC leakage, drawing enough leakage current to nearly blow the fuse. It is common when the amp has not been turned on in a long time. Sometimes they will even get noticeably warm. You can ramp them up slowly with a variac and sometimes reverse that effect.
That being said, when caps behave that way, I don't trust them. I replace them even if I was able to get them to stop leaking.
It can be a viable method to get an old amp running just long enough to test the transformers, though.
Whether or not "reforming" is legitimate or not, can (and has been) debated.
What is NOT debatable, is the damage that can be done to the amp if the "reforming" doesn't work and the caps have catastrophic failure.
I have never heard of anyone having to raid their 401k in order to recap their amplifier - It ain't a deal ($$$$) breaker. You can attempt to reform the caps in your amp and it may or may not work, or you can just buy new parts and be assured it WILL work.......
I vote for doing it right the first time.
Meat; It's the right thing to do. Romans 14:2
"Whether or not "reforming" is legitimate or not, can (and has been) debated."
But once again let me say there is NO DOUBT that re-forming of an electrolytic cap is a legitimate process, and is recognized as such by the major capacitor makers.
But reforming is not magic! It does not restore a cap to proper performance in every case - some are just too far gone. And it can be done improperly and that can be trouble. There is a way to determine if the reforming process has been successful, and there is also a need to physically examine the cap, checking for warm/hot spots during reformation.
I've personally applied my reforming process to 100s of caps, and I'd say about 80% were successfully reformed. The 20% that were rejected were tossed either due to high DC current flow through the cap even after the proper reforming process was done; or I felt hot spots on the cap during the reforming process. A few vented on their own or started to distort the case.
That DC current thing is important. It's heat build up inside that causes those "exploding cans". The heat is a result of current flow through the cap. For instance, a cap exposed to 400 volts that has 1 ma of leakage has to dissipate .001 x 400, or 0.4 watts. At 10 ma leakage it becomes 4 watts and that is a significant amount of heat - more current flow = more heat. It could cause a cap to vent to relieve the internal pressure. And yes, I am ignoring the impact of the ripple the cap sees to keep this simple.
The caps I've successfully reformed show DC current leakage at a fraction of 1 ma. - usually under .25 ma. at 10% over maximum rated voltage on the cap.
!
The Mind has No Firewall~ U.S. Army War College.
I have continually seen advice on this site that if you get a vintage piece of equipment that is more than 15 or 20 years old you can bet the caps should be replaced if you want to use it on a regular basis. I have also seen the advice (and followed it for years) that if you want to listen to it for awhile first to see if it is worth the bother of recapping, then bring it up slowly on a variac so you give the caps a chance to wake up. This has worked very well for me for over 20 years.
I don't think any serious restorer of vintage gear really expects the caps to magically come back to full useful life; it's just meant to be a temporary step in the process.
You can tell if the cap or can is too light, due to the electrolytic fluid evaporation. When in doubt, I'll remove one or two big ones in a unit. Feel the weight versus NOS good ones of that era. Too light? I toss them & replace all with all new 'lytics. As they prolly will not re-form or filter very well.Paper caps usually show signs of "bleeding" or "pimples" on the edges. I just toss these.
Edits: 12/21/15
I agree that old capacitors are not going to be magically restored by "reforming". However brand new capacitors that have set on the shelf for 2 to 3 years require it. This is standard knowledge on products such as UPS's and variable speed drives.
The following probably has more info than you care to know. Note info on storage performance....and that typical life is pretty much never expected to exceed 15 years.
http://www.nichicon.co.jp/english/products/pdf/aluminum.pdf
Not all caps can be successfully reformed - but don't assume that means caps can't be reformed. It is a legitimate process when applied properly and can improve capacitor performance a good deal. But it cannot save a capacitor that is too far gone, which appears to be the case with the one pictured in the original post.
Below is a link to a brief but accurate "how to" on reforming electrolytic caps. There are MANY more, including posts on OEM websites such as Panasonic, etc.
It has to be done right, and its limitations have to be understood - but reforming is a perfectly legitimate process.
Can success be determined with an ESR meter? What about a square-wave generator and an oscilloscope? (Assuming you have also checked for a short with an ohmmeter.)Dave
Edits: 12/22/15
Hi Jim,
Thanks for your reply. Also, thanks for those excellent Sovtek EL84Ms you sold me recently, I had no idea they would add like 6db to my headroom on the S-5000...
At any rate, my question is this - what is supposedly going on, physically, inside the capacitor during 'reforming'? I cannot envision a physical process taking place other than deterioration. The electrolyte certainly isn't becoming less dried out, which is irreversible. What physical properties are changing by this action? Furthermore, how could any such physical change not be a form of stress and fatigue on the cap?
These things are basically paper, foil, and an electrolyte goo. I cannot envision how any changes to their physical properties could be anything other than detrimental, nor can I see how applying voltage gradually would correct that. The idea behind reforming caps that you see from manufacturers refers to old caps that are unused old stock. Caps that have never seen their appropriate working voltage since the factory, and more importantly, caps that don't have years and years of heat and ripple current under their belts. In those cases, I think the best case is that you manage to get them to work without internal leakage (shorts) but I don't see how you'll ever reverse the deterioration of the electrolyte, the resulting loss of capacitance, etc. Reforming them to create the magic oxide layer on the foil doesn't ring true in cases of old, used caps IMO.
And let me ask this - all the supposed cases of success at reforming caps in old amps... How does one even know that there was a cause/effect relationship? If you bring some caps up on a variac instead of just applying full power right off the bat, how can you know that the result would have been any different had you just applied full power? You can't A/B the same capacitor, so there is no way to establish causality whatsoever.
+ 6db headroom !! Convinced me.
I just ordered a quad of EL84-M from McShane Design for my "big transformer" S-5000.
Can't wait.
Meat; It's the right thing to do. Romans 14:2
Electrolytic caps. employ a thin layer of aluminum oxide as the dielectric. That layer has to be thick enough to withstand applied DC voltage. Forming, by slowly ramping voltage up, thickens the layer.
As you showed, if the caustic, conductive, paste has dried out, NOTHING will help.
Eli D.
Another case of audio snake oil debunked.
The universe is made of electrons, protons, neutrons, and morons.
Well then what do you think of cryo treated circuit breakers in your main electrical panel being sold as "audio grade" ?
Man I gotsta sell me some of those! A testament to the fact that you CAN sell anything if you properly target your sales.
...and had the entire house and its contents cryo'd. Problem solved.
The universe is made of electrons, protons, neutrons, and morons.
:)
Hey Sherwood,
I own a variovac and kind of believed that you could reform electrolytic caps. Not any more!
That was a great lesson with your picture as concrete evidence. Will it do anything for all the smaller caps used in a amp?
This is another, I think (fallacy), break in period for certain electronics areas. How about cables. How is a cable in Hifi supposed to sound better after a break in period? If no one touches the physical connection how does it get any better.The first thing I read in my basic electricity book was conductance. The only thing that I can think of to make electrons move faster in an untouched component cable is heat. With a cable just sitting there connected in your system, how is it able to perform better with time lest the application of heat. Does the copper get better with age like Scotch?
If someone can answer that I'm all ears. Thanks...Mark Korda
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: