The AM Forum

In an effort to finally fix up a 1939-era broadcast receiver for a friend, I finally put it on the bench after it had been collecting dust for over a decade, following the e-mail that she was coming up from Dallas in a couple of months and wanted to pick up her radio while she was here. I was able to repair the power transformer (rotted insulation on the wire leads where they exited the transformer case) and wooden knobs (replace the missing metal insert cores that fasten the knobs to the shaft). Next, I found that the electrolytic filter caps are totally shot - one open and the other a dead short, while others are still untested.

So I dug into my junk box and pulled out a handful of 20 mfd 450 volt electrolytics.  Some never used, others salvaged from equipment. Using my Heathkit capacitor checker (one of my more worthwhile dumpster pulls, made usable with the "illegal" Heathkit manual someone scanned and forwarded to me via internet). I found that every capacitor I checked needed "re-forming".  The Heath does that well, by applying a variable voltage up to 600v to the  capacitor through enough current limiting resistance to limit the leakage current to a maximum of a few milliamps. The voltage is slowly ramped up until the capacitor either refuses to re-form, or else eventually takes the full rated working voltage with low leakage current while displaying close to the rated capacitance. Then the capacitor is left overnight to "cook" at full rated working voltage, and is ready for service the next day.

Since every capacitor that I pulled from my box of electrolytics and checked needed re-forming, I have decided that perhaps electrolytics in storage periodically need voltage applied to them for several hours to maintain them in good condition.  HV electrolytics are like vacuum tubes, difficult to find brand new these days, and when you find them, very expensive, plus the new ones are of dubious quality (for example, the exact replacement can that I bought for one of my 75A4s for nearly $50, lasted less than a year before one section went open-circuit on me).

Once the capacitor is fully re-formed, you need to supply only enough current to maintain full rated working voltage, which can be achieved using a variable voltage power supply and current limiting resistor - or even use the capacitor checker itself.

Once this restoration project is out of the way, I think I'll go through my entire collection of electrolytics, throw out all the bad ones that refuse to re-form, run the good ones through the re-forming procedure, and periodically "re-charge" them in banks according to working voltage. I'll rig up a jig to hold all my 450 volt ones, 350v, 600v, etc., all wired in parallel, and after each capacitor is individually re-formed, apply voltage to the whole bank and let them cook periodically to maintain them in good condition. Since each capacitor pulls only a few milliamps maximum, a small power supply should be sufficient to cook a batch of a dozen or two at a time.

In order to keep a supply of working HV electrolytics on hand so that they don't self-destruct while sitting on the shelf, this would seem like good regular routine maintenance.

This has been discussed ad nauseum on serious restoration forums and the learned conclusion is that reforming a used electrolytic just leads one down the path of false security. It may self destruct in a day, month or year(s) when least expected and often with spectacular results.

First time forming an old but unused electrolytic is often successful if less than roughly 30 years old as long as its not showing physical leakage out the vent. However I never take that chance with my own or customer electronics that Ive been working on for close to 50 years. For certain items I restuff the electrolytics and paper caps with new modern parts.

If that person is truly a friend and you have a few months to fix the radio I suggest you dig under the mattress for enough moldy dollar bills and purchase some name brand caps from Mouser, etc. I would stay far away from the hamfest and basement sellers offering no name Chinese dreck at bargain prices.

Since Ive never spent $50 on any electrolytic for my 75A4 or any other item Ive no answer for your problem other than to tell the person you got it from.

Carl

I personally don't fool around with old caps period. I buy them by the hundred so I get a decent price and charge retail for them. That’s all I need is a return for some old trash. If that happens I make no money and waste my time. Even if it is my own stuff I go new. Now I don’t have any problem with some old oil filled caps. I’m talking about 5kv and up no TV stuff. I think the new caps are way superior to the old stuff period but you have to buy quality. You get what you pay for. When I do repairs I find 90% of the time it is a power supply problem, and this usually has something to do with a cap.

73 N8QPC

Don,
        Reforming "aincient" caps is a total waste of time and reliability, period! ! !
If they broke down once, they are going to do it again. Modern 450v caps are still pretty cheap. I have "modern" caps that are 20+ years old and have NEVER had a problem with them. How would you feel if your friend picked up their radio and a few months later it went up in smoke because of a failed reformed cap!

Some of the replacement canned multi section caps are a bit pricey, but single axial or radial lead 450v caps are still just a few bucks. Absolutely not worth the risk of using old crappy reformed caps.

Like Carl said, this has been discussed ad nausium here, on other forms, and I have personally doscussed it many times at the Mid Atlantic Antique Radio Club meetings.
I tend to look at the ones who preach the merits of using reformed caps as either lazy or cheap asses! ! ! !

If you have modern caps, just keep them and use them as needed. If you have old
(60s on back) caps, toss them in the trash before you end up catching a dose of
"disposaphobia"

The Mouser website lists single-section 20 mfd/450v electrolytics at anywhere from $25.15 each to $38.36 each. I don't think the whole radio is worth much more than that.  It has a beautiful cabinet (a nice  piece of furniture) but the radio itself IMO is a piece of crap. If I could find new replacement capacitors at a reasonable price ($10 ea. or so - I am open to suggestions) I would use them.

The replacement can I bought for the 75A-4 came from Antique Electronic Supply in AZ. I don't recall the name of the vendor, but they claimed they had bought out a well reputed old company (Cornell-Doublier, Sangamo or one of the other old capacitor manufacturers) and continued to use the original tools and manufacturing process to turn out identical capacitors "superior" to the original ones. The new can was rated at higher voltage than the original, but was otherwise physically identical and a drop-in replacement. But after less than a year, a severe hum developed and I found the dead section.  I simply bridged a replacement out of my junk box (without even re-forming it) across the bad section, and it has worked perfectly for several years now.

I do plan to add a fuse to the primary circuit of the transformer just in case a cap shorts out. If the radio is playing well when she comes to pick it up, I figure I have done my part, since this is a freebie and I'm not charging her anything for more hours of work than I wish to think about. I'll let her know about the capacitors, and maybe she can find someone in the Dallas area who could replace them if they crap out.

After this experience, even if one purchased a bunch of brand new capacitors just to have some spares on hand, I would still recommend applying the working voltage periodically to electrolytics in storage to keep them from going bad on the shelf.

I re caped  my r 390A with caps from here.

http://www.justradios.com/capacitors.html (http://www.justradios.com/capacitors.html)


I was amassed at how small modern electrolytic  caps are.  

This company screwed up my order by sending me wrong caps. He replace them without charge and didn't want the other caps back.

Hey Don,

   Seems like a lot of hassel for a few bucks worth of caps.  "Let them Go"
new ones are only a 2 to 4 dollar investment and are far superior to that
old crap. Since you have a couple mos. of lead time you should just call
Tom at Hayseed.  You can get regular value/voltage Jobs mailed right
away, If it's a can it may take a couple weeks but the reliability is what
your looking for.

   The new ones will fit inside the cardboard tubes of the old ones if you
need that.  A 47uf 450v axial job is $3.50 and it wont pull 2 or 3 ma. Maybe
20 or 30 uA but I have not measured it.  I do know if you charge one up and
toss it on the bench for a week or so it will still "get ya". (Don't ask)


Here is my quote from Tom:

Sales at Hayseed Hamfest Co. wrote:

Oh, okay, I see, Dan.  You just need some regular axial electrolytics. That, I can fix you up with.  With today's tolerances, the standard value would be a 47 uF and I have some of those rated at 450V that are about 1-1/2" long and 3/4" dia.  Four (4) of them would run $15.60 + $3.50 shipping (unless you wanted to wait until your other caps were ready).

73

/Dan

Terry, the outfit in Ontario seems more like what I was looking for.

Dan, can you post some contact info for Tom at Hayseed Hamfest Co. ?

If it were my own radio, I would just use the old caps and replace them if/when necessary, but since it will reside in Dallas, she does need something reliable.

I'll probably stick in some of the old caps long enough to get the thing running in the next few days, and then investigate permanent replacements with new stock. Since the replacements will all go under-chassis, and in any case she is not interested in the guts inside the radio, no need to re-stuff old components.

Here ya go..........

sales@hayseedhamfest.com

NØJMY

Don,
       Also check Mouser and Digi-Key. Screw the can type. look for the axial or radial lead types. They usually carry several brands, look for the ones that are cheapest. Usually you can find 10@450, 22@450 and/or 33@450 for just a few bux each. 47uF seems to be the point where the price rises sharply. Hayseed usually has pretty good pricing, but it doesn't hurt to look around.

If you are putting them under the chassis, it really doesn't matter what they look like. The canned multi section type seem to be somewhat a specialty item and are always a bit pricey. I also sometimes find good deals at hamfests. I usually keep
50 -100 "in stock." One of the first things I do when repairing old gear is to replace the filter caps, and then any other wax caps that are anyway tied to the B+ line.
Even the screen bypass caps. If it's a waxed paper cap, it's outa there! ! !

The one time I reformed electrolytics was on a Hallicrafters SX-11.  I shoulda known better.  They lasted about a year, then shorted and smoked the power tranny.

I agree 100% with the strategy of using only new caps.  NEVER use old caps, NOS or not.

Yep, the only thing old electrolytics are good for is restuffing if you want to make it look like year one.

73DG

  Or   www.hayseedhamfest.com  fredricksburg, Ia.

The caps I buy from Mouser are either the Xicon/Lelon (house brand) or Nichicon. Axials are more expensive than radials but I use both; a Lelon 47uF 450V radial is $1.85. DigiKey and others carry similar stock. If ordering from Mouser ask for USPS First Class Mail for light packages otherwise the default is UPS, their delivery is real fast and there is no minimum.

Ive heard a few horror stories about the no name Chinese lytics from that Canadian outfit. If you just have to go that route dont run them more than about 2/3 of the voltage listed.

Carl

I think IC (Illinois Capacitor) is a good brand in electrolytics.  Look for iC as a logo on the side of the cap (if you have an assortment to chose from at a fest for example and want to avoid the stuff from China). 

Where are these caps manufactured? 

I use a lot of non-polarized caps by Solen from AES.  They are bulletproof, and come in very useful uF & V rating for power supply rebuilding.

BTW, if anyone is restoring an old BA and the can types are missing, I have hundreds of NOS ones and may have the style you need.  Free. ;)

73DG

On an island

Oh yea, that's the one. I should have known!

This is what I find on the Mouser web site for 20 mfd/450v electrolytics.  I must be overlooking something.


http://www.mouser.com/Passive-Components/Capacitors/Aluminum-Electrolytic-Capacitors/_/N-75hqt?P=1z0wqrdZ1z0vl82&SAP=true

Try 22 mfd/450 volts. They start at $1.23

http://www.mouser.com/Passive-Components/Capacitors/Aluminum-Electrolytic-Capacitors/_/N-75hqt?P=1z0wrkiZ1z0vl82

Try 22 uF, Don.  20 uF is no longer a standard value part and will be much more expensive.  'Lytics are very low tolerance, sometimes -20%/+100% so you're very safe with a 22 uF.

Mouser has a bazillion of different choices for around a buck each.  Go for the 105 degree caps instead of the 85 degree.

I do miss the days of getting a hardcopy paper catalog from Mouser and Digi-Key.
Sometimes the new is not always as good as the old. It made it a lot easier to thumb through the catalog and find what you wanted (and compare pricing at a glance) their on-site search utilities are not always your friend.

The electrolytics I got from Just Radios are "Miec" brand made by Ling In Electronics of Taiwan.  Being concerned about the quality I tested a bunch of all the types, tubular, dipped mylar, electrolytic and silver mica.  All of them have tested spot on as marked.  I've got no complaints so far.

Thanks for the tip.  I hadn't even thought about any distinction between 20 mfd and 22 mfd capacitors. I just assumed anything in the ball-park of 20 mfd would be about the same. Actually, one of the originals was 8 and the other 16mfd. I had a bunch of 20 mfd ones in the junkbox and those are the ones I re-formed with the capacitor checker, so 20 mfd popped in my mind with no further thought, for new replacements. When repairing an old radio, I often use twice or more filter capacitance than the original.  Before WW2, capacitance was hard to come by, and high value capacitors were expensive when they were available at all, so lower capacitances were almost universally used for economic, not technical  reasons.

Many pre-WWII transmitters used something like 2 mfd oil caps as power supply filters for the same reason. The dynamic regulation of those power supplies when used for CW transmitters and class-B modulators was atrocious, and often a cause of audio distortion and key clicks, but most hams of that day didn't have an oscilloscope to look at their waveforms, so ignorance was bliss.

Capacitor - Solen Fast Metalized Polyprop, 8.2 µF @ 630 V  $5.25 per @ AES.

Not a bad deal for a cap that isn't an electrolytic or oil, seem to last forever in stuff I fix.  Just the thing for restuffing old can types.

Values go up to 47uF at that voltage.  They also have them rated at 1kV.

73DG

 

Be careful about doing that for the input cap. It will raise the B+ causing higher current in the tube circuits as well as possibly be too high for the rectifier tube causing an arc. The max C is usually in the tube spec sheet. I try and limit the input to close to original and use a lot more at the output to kill ripple.

I never use the Mouser or any other large distributor on line catalog, they are a PITA and you miss way too much.

As to reforming etc: I'll stick to new electrolytic caps, not old stock. Still can't see how reforming will put the liquid, which has evaporated, back into the cap.

Mouser is the first place I look, still have a few of their paper catalogs. But looking online other caps are listed that might not be in the catalog. Why limit your choices??? And I keep an online account there, makes it easy to reorder if restoring same model RX or TX's

Craig,

<<On an island>>

There was an earthquake. 


Slab you must have good eyes. 

Oy-vey! ! ! ! !  What, you never heard of a magnifying glass  ???  ;D  ;D  ;D  ;D

The max C as specified in the tube spec sheet is usually well over twice what was actually used in an original pre-WWII power supply, even as referenced in the RCA manual of that same era. C was simply too expensive and bulky back then, so they used as little as they could get away with or fit into the space available. Many of the older BC sets used 4 to 8 mfd for the input cap with the type 80 rectifier (sometimes as little as 2 mfd), while post-War sets typically used 16 to 20 mfd for the first cap with the type 5Y3.  Exact same tube, just different base socket. Where you might get into trouble would be to try using 50 or 100 mfd for the input cap, although some of the tube type B&W TVs that I remember salvaging for parts back in the 50s did just that.

Excessive C at the input cap shouldn't boost the voltage significantly.  A capacitor input filter delivers DC voltage at near the peak voltage of the a.c. cycle, 1.414 X Vrms.  Once you have enough capacitance at the input to make the supply filter act like capacitor input, additional capacitance won't affect the voltage once the capacitor is fully charged. What strains the rectifier tube with too much input capacitance is the excessive current the capacitor pulls through the tube to initially charge it to full operating voltage. A large capacitor may act like a dead short for as long as a second or more, while the proper size cap acts like a dead short only for milliseconds. In every case, a capacitor input filter presents a dead short across the rectifier the instant a.c. voltage is first applied.

Tubes are intolerant of excessive peak current, but solid state rectifiers can handle it OK to some degree.  OTOH, tubes are much more forgiving than SS devices when the maximum p.i.v. is even slightly exceeded.

The other one hasnt been invaded yet so quality is still good ;D



Not really and the world didnt revolve around RCA.



The 80 started off with a 4uF rating at full voltage; over the years and also by manufacturers the spec was raised to 10uF.

The 5Y3 series varies all over the place from 10 to 40uF starting with the 5Y3G which is the only true 80 replacement to the 5Y3GB. The GT and GA versions also have varying ratings dependent upon manufacturer.

It really pays to read all the specs.

 
Different tube, different spec and also possibly a lower RMS. The minimum plate supply impedance also played an important role. Most B&W TV's Im familiar with used a 5U4 of some kind.




Supply impedance and the radios current draw both come into play when a widely varying DC load is involved. While a 4uF cap has little energy storage and the regulation is poor a 40uF will have much closer to the 1.414 of theory. This can be substantial if the radio is already using components at the edge of their specs. A higher voltage equates to higher current and hotter running power and audio transformers, especially in Hallidrifters with PP audio and failure prone iron, many consumer radios have similar problems.



Unless it has a controlled warmup characteristic.



Ask anyone who has used 866's. SS is rated at peaks per cycle and modern diodes such as the 1N5408 and 6A10 can handle substantial peaks.


 

Actually tubes such as the 80/5Y3 are not very tolerant of overvoltage and indirectly heated ones are even less so.
Ive measured 1N5408's and 6A10's as having a PIV of 300-500V above the spec without failure.

Carl

I don't believe there is any significant difference between the ratings of the consumer-grade 5Y3G and 5Y3GT, nor between different reputable manufacturers of the same tubes. I  have several books besides the RCA manual, including one by G-E and a couple of generic ones, and they all give approximately the same data.

The RCA book says that with high vacuum rectifiers, if the 1st capacitor is above 40 mfd, care should be taken to assure that the source impedance exceeds a certain minimum number of ohms (don't have the book right here with me for the exact figure, which I believe varies with tube type), but that shouldn't be a problem with those old broadcast receiver power transformers. Most of them were pretty poor in terms of voltage regulation, due to high leakage inductance resulting from their cheap design.

That's one of the reasons modern day HV power supplies using solid state diodes have as good regulation as they do (capacitive input supplies have had a reputation for notoriously poor voltage regulation); modern transformers are designed for low leakage inductance and pose a very low source impedance.  Most "vintage" transformers, designed for choke input filters, indeed have poor regulation when a capacitive input filter is used, because of the high leakage inductance due to physical construction and core material which results in a higher source impedance than a modern  transformer designed for the best possible regulation..

Maybe newer diodes that don't require the external equalising resistors and surge suppressing capacitors across each diode can take a certain amount of voltage beyond the PIV rating, but I never was able to keep a string of diodes (the older types using the caps and resistors) from popping like a string of fire-crackers even with the recommended resistors and capacitors, running at substantially less than their maximum rated voltage. This was particularly true when I tried to substitute SS diodes for tube type rectifiers in an existing choke input filter. One of the reasons is that the input capacitor presents a low impedance to the spikes as they exit the rectifier and takes them directly to ground; with choke input, the inductance of the filter choke presents a high impedance that allows a spike to build up to high enough PIV with the inductive kick to break down the diodes. Since I have plenty of tube type rectifiers on hand and filament transformers to run them, I gave up on the diode strings and just stick with hollow-state  rectifiers.  However, my BC1-T has used a couple of sets of plug-in SS replacements for the 8008 and 866A rectifiers working into choke input filters for almost a decade now without incident, so the designers of those direct-replacement modules must know something that I don't.

Choke input tends to work best with tube type rectifiers including MV and Xenon gas types, while capacitor input tends to work best with solid state. Capacitor input is not recommended for MV tubes because the peak current rating is easily exceeded, which according to the manufacturers significantly shortens the life of the tubes. It works OK with high vacuum rectifiers like the 5Y3/80 or even the transmitting type 836, because the higher internal resistance of the vacuum diode serves adequately as a current limiting resistor, so long as the input capacitance is not too high.

There was a similar thread on here 2 months ago started by Stu AB2EZ on re-forming electrolytic capacitors in which I made a post on the availability of high-capacitance film capacitors available today called DC link capacitors:

http://amfone.net/Amforum/index.php?topic=27929.0

The best way to avoid burning out a transformer or rectifier due to a shorted electrolytic (or other component) is to wire a fuse into the a.c. power circuit; there should be one there anyway. I have re-formed old caps simply because I  had them on hand and needed one right away, not after waiting several days for an order to arrive. Sometimes, I have re-formed old ones with the firm intention of ordering new ones as a permanent replacement once I got the equipment running or the prototype developed, then procrastinating or forgetting about it until the cap goes, or as in some cases, the re-formed cap never crapped out.

I have had electrolytics to open up about as often as I have had them to short out. I have also had them explode and shoot stinky fibrous crud all over the place or totally pollute the under side of the chassis with gunk.

Not a bad reason to replace them with something better but close in Farad value.

73DG

Ah Yes................. Capacitor cheese  ;D  ;D  ;D

Believe whatever you want, I was reading the original full spec sheets published over a few decades.


I touched on the source impedance yesterday, maybe you missed it. All full spec sheets include that and an engineer understands what it means. However techs or complete newbies working on grandpas heirloom often/never dont pay attention to details and just think bigger is better....and bang.
Many if not most ham receivers and transmitters used cheap transformers also. Some consumer sets had outstanding quality in their top of the line products.


Not a bad reason to replace them before even turning a set on. More transformers are destroyed by fools who find some old set and just plug it in without a clue in the world.  Many others claim since its been working fine for them for XX years it dont need no stinkin capacitors. Then sometime later they are begging for a replacement transformer or crying about the cost of a rewind. They get no sympathy from me.

Carl

Hence the need for the a.c. primary fuse.  Most "consumer" electronics stuff in the day omitted the primary fuse, just as they omit RFI filtering to-day, to save a minuscule percentage on the total per-unit manufacturing cost, and with no fuse, for the hope that the set would burn out after a few years and the customer would buy a new one.

Maybe that's the origin of the term "consumer" electronics.  They are supposed to self-destruct after a certain period of time to keep the industry flourishing. I don't "consume" electronic and radio equipment.  About the only retail products I "consume" are food, drink, clothing and fuel. Everything else I try to keep using as long as I possibly can, therefore admittedly, I am a piss-poor "consumer". If the majority of the public had my buying habits, the economy would have tanked decades before it did, since the whole system largely runs on wastefulness.

Where can I find copies of those data sheets that show significant electrical differences between the 80, 5Y3G and 5Y3GT? (Add to the list the 5Y4G, which is nothing but a 5Y3G with different pin arrangements on the octal socket). By "original spec sheets" for the 80 are you talking about the ones that came with the original round-top versions like the UX-280, CX-380, etc? They did upgrade many tube specs in the mid-30s when they changed the bulb style from the globe-shaped originals, but with the upgrade came full-speed mass production, so 99% of the tubes found to-day are the more modern types with the ST envelope or the smaller GT bulb.

Here's an example of a place with what appear to be decent rx filter caps:


http://cgi.ebay.com/Illinios-22uf-22-uf-350v-Axial-Capacitor-10-pcs-/190561144262

Says "new old stock" but doesn't say how old (5, 10, 20+ years)

You can go to Mouser and, most likely, get "fresh" stock of the same value/voltage/axial for $1.40 or $1.89 and even cheaper by the bucket.