Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: The Souvenir Shop :: Contributors :: Advertise


It is currently Apr Fri 25, 2014 3:24 am


All times are UTC [ DST ]



Post New Topic Post Reply  [ 14 posts ] 
Author Message
 Post subject: Mica Capacitor Failures that aren't found by Test Equipment
PostPosted: May Thu 22, 2008 9:36 pm 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 20669
Location: Detroit, MI USA
Since many Forum members don't regularly visit the TV Forum, I thought it would be a good idea to mention this here, since it will also apply to radios.

Several of us have recently reported finding failures of mica capacitors which still check perfect when tested with good quality equipment at their rated voltage. This has produced some difficult to diagnose symptoms in the TV's the caps were used in, and I am certain the same thing can and does happen in radios.

Obviously it is not a good thing when parts test perfect, but won't work in a circuit. This can cause a lot of frustration as well, particularly for those with less experience. When using a tester such as one of the later Sprague TO- series, and the caps read full scale on the leakage resistance test indicating they have no leakage, and capacitance is right on the money, then one would like to assume they will work in a circuit.

I'm at the point where I'm ready to make the statement that we should probably start automatically replacing all the mica caps in sets which have unexplained symptoms after everything else has been tested, and the paper and electrolytic caps replaced. Frequency determining caps in the local oscillator circuit of a radio may be the exception if they have no B+ voltage applied, but it looks to me like any mica cap used as a bypass or coupling cap is suspect.

I know that some are already routinely replacing all the mica caps in their sets at the same time they do a recap of paper and electrolytics. That's not a bad idea either, particularly on sets which are from the early 1950's and before, since that seems to be the vintage where they are most likely to be failing.

_________________
Dennis

Experience is what you gain when the results aren't what you were expecting.


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 22, 2008 10:00 pm 
Silent Key

Joined: Jan Thu 01, 1970 1:00 am
Posts: 34328
Location: Sandpoint, IDAHO 83864
Dennis- I am in agreement with you, but I have a question. Just what is the failure mode and possible cause of these strange failures? Is it an increase in ESR, though minute at such low values of capacity like a mica capacitor would be? Or is it some sort of internal breakdown under AC signals?

As a tidbit, I have found many small micas that were bad when the seals between the body of the capacitor and the leads broke from handling the capacitor. Some of the dot coded capacitors always seems to be mounted upside down, so you have to rotate the body of the capacitor to make out its value. Doing this will often break the seal between the wires and the body, so if you are going to check them, you might as well replace them at the same time, as they will be ruined after you check them.
Curt

_________________
Curt, N7AH
(Connoisseur of the cold 807) CW forever!


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 22, 2008 10:03 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1908
Location: Hinsdale, IL, USA
Dennis,
As you know finding caps that fail only under the load of the circuit is difficult.

In all my year as a bench tech (from 1970 to 1990), I never once used a cap tester or tube tester for diagnosing a set. A good DVM and scope were my primary diagnosing tools.
I always used part substitution once I suspected a particular part was bad. To be efficient you must understand how the circuit works and be able narrow down the possible parts that could cause the problem. When time is money it's faster to sub a known good part than to try and test it. For hobby work and for the fun of it it's nice to test components, but if you trust the testers implicitly you will run into these cases where the testers lead you astray.

In general never believe a tester when testing high current or high voltage components, just sub the suspected part. You will save yourself hours of grief. JMO

_________________
Chuck D. KB9UMF


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 22, 2008 10:05 pm 
Member

Joined: Dec Sat 08, 2007 9:51 pm
Posts: 1783
Location: Nashville TN
Hi great post.Ionce had a radio drive me nuts for weeks (had a dead spot on the toip band between 14 to 20 mhz.After replaceing all caps still same.Ended up finding a bad mica cap,replaced it radio worked fine.Ireplace all caps after that.Angelo 8)

_________________
Be happy ,enjoy life as its so short!


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 22, 2008 10:25 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 15450
Location: Somers, CT
I am finding a few that are "noisy." Moving the part body causes electrical noise. I assume the lead stress is being transferred to the internal mechanical compression-type connection to the metal plates, and that flexing is causing the electrical noise.

Pete


Top
 Profile  
 
 Post subject:
PostPosted: May Mon 26, 2008 5:32 pm 
Member

Joined: Mar Tue 18, 2008 12:15 pm
Posts: 100
And then there was the time that I replaced a cap with a brand new one, which didn't fix the problem. After another hour or two I ended up coming full circle, back to the original cap I had replaced. You guessed it, the NEW part was BAD :lol: :lol: :lol:
Some days you're the statue,
other days you're the pidgeon!
-Dave


Top
 Profile  
 
 Post subject:
PostPosted: May Mon 26, 2008 8:12 pm 
Member
User avatar

Joined: Jul Fri 13, 2007 1:48 am
Posts: 2175
Location: Washington DC
Peter Bertini wrote:
I am finding a few that are "noisy." Moving the part body causes electrical noise. I assume the lead stress is being transferred to the internal mechanical compression-type connection to the metal plates, and that flexing is causing the electrical noise.

Pete


Right on. I had one that went either "open" or "ok" with the slightest touch of small screwdriver. Now I always punch, wiggle, and vibrate all parts under a non-working radio before I start down the path of trouble shooting.

_________________
John O.

"Apolitical and staying that way"


Top
 Profile  
 
 Post subject:
PostPosted: May Tue 27, 2008 1:33 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 26016
Location: Pocasset, Cape Cod, MA
>>When using a tester such as one of the later Sprague TO- series, and the caps read full scale on the leakage resistance test indicating they have no leakage, and capacitance is right on the money,

I wonder if the problem is series resistance from a poor connection. That might not show up on a capacitance measurement, not at 60Hz anyhow, since the resistance would be small in comparison to the capacitor's own reactance.

A modern digital meter might show it.


Top
 Profile  
 
 Post subject:
PostPosted: May Tue 27, 2008 1:51 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 20669
Location: Detroit, MI USA
I'm thinking that Alan is probably close to the correct answer to this situation. When tested on vintage equipment at 60Hz, they work, but at higher frequencies (15750 in TV and beyond) there are issues which cause the capacitor to look like something else to the circuit. That's the only theory I can come up with.

I have broken open some mica caps which wouldn't work in a circuit but checked good, and saw nothing obviously wrong. I wonder if it is even possible to have ESR at 15KHz in a 100 pf mica? And exactly what device would we need to measure that with? My ESR tester runs at 100KHz IIRC, but is only good for electrolytics, not low values of pf capacitors. I did check some that wouldn't work for value on two relatively modern solid state devices, including a DMM and a dedicated cap tester, but didn't find anything out of the ordinary.

The one troublesome mica cap in an early RCA TV chassis was coupling between the sync amp and sync separator. It introduced just enough phase shift in the signal to make the horizontal blanking bar creep into the left side of the picture, but didn't affect sync stability or vertical sync. So at least in that application, something was happening at H frequency but not at 60 Hz.

_________________
Dennis

Experience is what you gain when the results aren't what you were expecting.


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 28, 2008 5:15 pm 
Member

Joined: Apr Sat 22, 2006 10:46 pm
Posts: 998
Location: Waterloo, Iowa
What I never understood, years ago the lore in the radio repair manuals trended towards "you'll ruin the alignment if you mess with these, and they hardly ever fail".

Alignment should be in the ballpark if the correct value is used - and alignment isn't that big a deal anyway, not for consumer radios
anyway.

I found definitely audible improvements, if not measurable with anything I own, by replacing micas in frequency determining circuits. Better selectivity and sensitivity, and improved warm-up drift. In short, mica replacement can't hurt anything, except maybe your pocketbook.

It may be that 60 years worth of crud on the _exterior_ is causing problems, or a combination of other factors. (nearby lightning strikes can't be good for the front end) I've found micas that show no leakage on old-school cap testers, but replacement of those same caps really perks the set up. Seems to me they were low pF Micamolds but I'd have to look.


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 29, 2008 12:45 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 6263
Location: Beaver Falls, PA. USA
Maybe something is happening to the mica dielectric after all these years. I'm finding an increasing number of micas are testing leaky now; I have even found NOS units which wouldn't pass a leakage test.

It might be an interesting experiment to set up a signal generator and oscilloscope, and see what happens to that capacitor when you feed a 15khz signal through it.

Tim KA3JRT


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 29, 2008 1:27 am 
Silent Key

Joined: Jan Thu 01, 1970 1:00 am
Posts: 34328
Location: Sandpoint, IDAHO 83864
I may be all wet, but let me say what I think is going on.

Most of us are very familiar with the silver migration problems in the mica capacitors in some IF transformers. I think that the same thing goes on with regular mica capacitors that have silver as the conductor in them. The migration may not be complete to where it constitutes any real leakage or a short, so it does not show up on tests that apply a DC voltage to them, even though it may be several hundred volts.

But what happens when they have AC signal voltage on them? Under an influence of AC, a capacitor will show different characteristics than it will on DC. And as the frequency increases, this will change. So at high frequencies, you have a capacitor with a built in shunting resistance and this resistance varies with frequency. I am not talking about the impedance or reactance of the capacitor, I am talking about the resistance caused by leakage, if you want to call it that.
Curt

_________________
Curt, N7AH
(Connoisseur of the cold 807) CW forever!


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 29, 2008 1:51 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 26016
Location: Pocasset, Cape Cod, MA
100kHz won't be high enough for an ESR check; the capacitor's own reactance needs to be an ohm or less at the working frequency, if the additional series resistance is to show up.

15kHz is within the range of a GR 1650 bridge (at reduced accuracy) or there are probably RF bridges that would do. A little higher frequency, and a Q meter would work. There are modern bridges that will work too; I know there are fancy HP bridges though you don't always get a choice of operating frequency.


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 29, 2008 1:51 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2848
Location: Ga
When I started restoring tube radios that comments were not to change the micas but after working on an RCA with the black losenge type mica caps, I don't hesitate replacing them.


Top
 Profile  
 
Post New Topic Post Reply  [ 14 posts ]  Moderators: Marcc, Norm Leal

All times are UTC [ DST ]


Who is online

Users browsing this forum: Bill njb, Google Adsense [Bot], NumberMaj 1 and 10 guests



Search for:
Jump to:  














Privacy Policy :: Powered by phpBB