Hi All

This topic has been brought up many times, but I wanted to get some ideas on this.

I have some 47uF/450V caps that are now a 1 1/2 old, and when I use my Heathkit C-3 to test, the eye gradually (and I mean gradually) opens...but in the course of a minute or two doesn;t get as fully opened as at let's say 150V.

This is expected...

I have no problem throwing them out...now that I am entering retirement, I live next to Antique Electronic Supply, so I can get new caps the next day.

Some folks do say that "freshening up" these caps is OK, others say "sure, go ahead. but failure is around the corner".

I would do the experiment....but, I don't want to use my C-3/950B for concern of burning out the pots, etc.

Is there a safe way to use my DC power supply with an ammeter and resistance load? If so, can someone share how they have done that?

Thanks!

Steve

There is a government document online (somewhere, sorry) that describes the process, what to do and what to look for if the cap is beyond "re-forming".

Chas

You are opening up a big can of worms here............you will get flamed bad by even mentioning reforming electrolytics by most folks........A friend of mine made a small device and I have one that uses a HV power supply and I have had great luck reforming.


I will copy from the PDF of the reformer.........

The plates inside an electrolytic capacitor slowly lose their insulating oxide layer as the cap sits unused
over a number of years. This occurs if the cap has been sitting on a shelf or inside an electronic device.
As the insulating oxide layer depletes, the caps internal resistance lowers, eventually reaching near zero
ohms over time.
Permanent damage occurs when a sufficiently high current is presented to an electrolytic capacitor with
depleted oxide layers. The heat generated by the low internal plate resistance causes the plates to
deform and/or the electrolytic solution to become superheated, and thus turn to a gas. The electrolytic
gas escapes from the capacitor, or it explodes. Once this has happened nothing can be done, except to
replace the capacitor.

This is what we want to avoid, thereby successfully resurrecting the cap, with no damage done, during
the reforming process. If a capacitor is properly reformed it can be brought back to its original
specifications and perform normally for a lifetime.
There are many techniques and much literature on the subject. In common is the requirement to try to
limit the current during the reforming period. Some suggest the use of variacs, light bulbs, or resistors in
series with the unit or high voltage supply. What is in common with these techniques is the attempt to
limit the current, hence the heating, to allow the plates to rebuild their insulating oxide layers without
overheating.
How the above techniques fail, is that they can still damage the capacitors. By not knowing how quickly,
or slowly, the oxide layers are reforming they can inadvertently overcurrent and overheat the cap. A
variac on the line side of an electronic unit that needs reforming cannot control the current actually
consumed by the capacitor. The voltage and current that the capacitor is subjected to depends upon the
type of rectifier in the unit, solid state .vs. tube rectifier. This method often touted by restorers of
vintage gear is clearly hazardous and often fails to properly reform the caps which will hasten their
demise.

The solution:
By using a series constant current source connected to a power supply, set to the working voltage of the
electrolytic to be reformed, a capacitor can be reformed safely and effectively with nearly 100% success
rates. The constant current source insures a safe maximum current and allows the capacitor to reform
with no risk of damage to the capacitor, or to you, (you will often see pictures of people reforming a
capacitor in protective enclosures in case the capacitor explodes or outgasses).
With a low constant current there is no risk of overheating or internal arcing damaging the plates.
Further, the voltage the capacitor has across it is dependent upon its internal resistance, which
increases as the oxide layer builds up, the more resistance the more voltage.
The idea here is to deliver a constant 5ma current maximum to the capacitor. When the capacitor has
fully reformed the current will drop off sharply to a leakage current value, usually on the order of
microamps.

Such reformed caps "should" be checked some time later, the reforming may not "hold" any reasonable time, effort wasted.

If they are used every once in a while they will be fine. Its not like over night the oxide will release. Brand new caps on the shelf at a electronics store can do the same thing. The military has a plan in place where they take caps on the shelf and run them thru a reformer every so often.



Todd

Caps that are 1.5 years old should not be bad. Just let them charge to full rated voltage with a 10K or so current limiting resistor.

Hello guys,
I sure the shelf life is better then 1 1/2 years if not then would be tossing almost every one my caps in the garbage .

Sincerely Rich

Thanks!

this is exactly what I was hoping for in my question!

I can buy new caps; what I can't "buy" is the knowledge that comes from experimentation and learning the principles through the process.

I just measured the current through one of the caps (not from a well known manufacturer) that the eye did not fully open on...it went from about 8 milliamp DC to about 500 microamps DC after maybe 5 seconds..I didn't hold it longer to see how far it fell.

Fresher caps (from a well known manufacturer!!) went to less than 200 microAmps DC in 5 seconds....certainly within tolerance.

I am learning new things already!

Steve

I have 50 year old Sprague can electrolytics to go to less than 40 uA in ten seconds. I have Aerovax cans that days hours to reach 10 mA leakage, and they are bad again in a week. I would not trust any electrolytic that stays near a mA at rated voltage after being charged for a minute or two.

Todd, could you send me any other info on putting together that reformer?

Thanks!

Steve

PM me your email and I will send you the PDF file!


Todd

If you look at the spec sheets for just about any modern electrolytic, there is a specification for leakage current. There is almost always a disclaimer that if the capacitors have been in storage for more than six months, they will fail that leakage spec unless voltage is applied until the leakage current stabilizes. Reforming electrolytics is an inherent part of their operation which usually goes on all the time without any intervention. The process only gets a bad rap when people attempt to use it on junk caps that have already reached the ends of their useful lives and can’t be saved. Reforming only works on caps that are good to begin with but just need to be brought out of hibernation, as I would expect your 1-1/2 year old caps to be.

So what you do is take the rated leakage spec, double it (which is normally considered end of life in engineering terms), then choose a resistor which won’t pass more than that at rated working voltage. Use a bench supply to deliver that voltage to the cap and capacitor in series. For example if spec sheet leakage for a particular 40-uF, 450-V cap is 440 uA, double that would be 880 uA. Then 450/.00088 = 511-k ohms. If you have a variable voltage supply you can use a lower value resistor and turn the voltage up as the cap reforms, but under no circumstances should the current be allowed to exceed 2-mA.

After 15 minutes using Chris's suggestion, the 47uF/450 V cap was reading a leakage of less than 150µA after 30 seconds.

I will put it away in a bag marked with the date and check it again in 7 days.

I like experiments...makes me think I am back in undergraduate labs....

Steve

Pop Electronics Oct. 70, page 46 Reforming your electrolytics. https://www.americanradiohistory.com/Ar ... 970-10.pdf

I think I will build this! Using diodes to multiple the voltages ala Cockcroft seems to me to be the way..rather than using my older Heath P-3.

Now to find a 1P12T switch rated for 600V!! Maybe someone has suggestions...

Steve

I might build one as well. It seems better than burning up a vintage capacitor tester to do it

You don't need to build anything the Eico will do a fine job. Let the cap sit for a couple minutes before you get a final reading. Check to make sure the voltage on the Eico is within about 10% with the cap on it. I assume you are using the electrolytic range only. The Eico limits the current so no need to add any resistors.

I would use the Eico 950B for reforming the caps as well. As long as the unit has been fully restored internally, it will do one of the functions it was designed to do.

I no longer use my 950B but it's a good instrument for this.

I use the GR 1644 megohm bridge, which limits current and even measures leakage. Any voltage up to 1000V.