Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Jul Wed 24, 2019 9:32 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 9 posts ] 
Author Message
 Post subject: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 2:02 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 44
Location: Ithaca, NY USA
Got a Sprague Tel-Ohmike TO-3 recently & I've been putting it through its paces, checking all kinds of paper & mica caps. It's unrestored at the moment (except for the power supply caps), so I'm assuming that the reference or "standard" caps are all leaky & off value to some extent. I found something very interesting in the .001 to .5uF range, which is the least accurate. On that range, good caps (newer epoxy & Orange Drop types) around .05uF won't open the "eye" sufficiently...with the best ones it doesn't open at all, just gets less tightly closed. Leaky wax paper caps open the "eye" just fine...the leakier the better... & their readings are more accurate! Here's what I'm thinking...it's basically a bridge circuit. If I understand this right, when it's balanced, some combination of resistances & capacitances on both arms of the bridge hits a ratio that causes the voltage to the eye tube to go to zero, so that it opens. Now, if the reference capacitor for that range is leaky, it's like putting a parallel resistance across it. In order for a test cap near that value to balance the bridge, it probably needs a similar parallel resistance (i.e. it needs to be leaky too!). I tried testing a good cap & putting a 470K resistor across it...sure enough, the eye opens part way.

Somewhere I remember reading that it's supposed to be the other way: when a cap checker tests a very leaky cap, you get that sort of indefinite "eye" opening, but a good cap should open the eye right up. If my theory is right, that would make sense...in a good tester, the "standard" cap is not leaky, so it would be leaky test caps that give an indefinite reading.

I'll probably end up replacing the "standard" caps, but I wanted to know if this problem could be caused by anything else. Any thoughts?


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 2:19 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18544
Location: Albuquerque, NM 87123
I suggest you fully restore the TO-3, then test it.


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 2:30 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 44
Location: Ithaca, NY USA
Johnnysan wrote:
I suggest you fully restore the TO-3, then test it.


But I want to know what causes the problem. Shotgunning a piece of vintage equipment may fix it (or may introduce new problems), but it isn't troubleshooting. I want to understand how the circuitry works.


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 2:58 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 3930
Location: Littleton, MA
dme3 wrote:
I'll probably end up replacing the "standard" caps, but I wanted to know if this problem could be caused by anything else. Any thoughts?

You've correctly diagnosed the likely cause. The only way to be sure is to replace the standard caps (and any other paper or electrolytic caps). Trying to troubleshoot a circuit full of bad components is just a waste of time.

_________________
Steve Byan http://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 5:02 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 44
Location: Ithaca, NY USA
OK, for the record I replaced the two .01uF caps that made up the .02uF standard, and now everything is just about spot on in that range, with "eye" action looking great.


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 6:39 pm 
Member

Joined: Jan Tue 10, 2012 8:39 am
Posts: 240
The bridge, to be balanced, must have the unknown a precise ratio of the standard. If the standard is leaky, the unknown should also be. Replacing the standard with a good capacitor should restore proper operation.

I have worked on several of these service grade bridges and they all seem to need new standard capacitors. Not so for many of the standard resistors, particularly the wire wound types. I worked on one recently where the wire standards were far more accurate than the bridge would warrant.

These units proliferated in the 1940s and served a purpose, although accurate measurement wasn't their biggest characteristic. You can see that by the wide and crowded dial scale. A change of perhaps 1% wouldn't be very readable. But that's not the purpose for which they were made.

Accurate capacitor measurement was perhaps reserved for laboratory quality equipment. The problems of strays, parasitic impedances, frequency dependence, and accuracy of standards prevented economical manufacture of accurate devices. Add to that the inevitable deterioration of components and the result isn't so good.

I have a very old Cornell-Dubilier decade capacitance box. On the back is a large label with hand written values, apparently measured when the unit was new. I measured the values, perhaps 60 years later, and they are still spot on. Congratulations to the company for a job well done. Not so for the standards in the EICO and Solar low cost units. I haven't checked Heath or Sprague units lately but am not optimistic.

I have a few ways of measuring capacitors; perhaps the most accurate is the GR 1658 bridge. The 1650-A is good but not as accurate. For measurement at radio frequencies, the Boonton Q meter does well. But for electrolytic capacitors, the service grade bridges are entirely adequate, although they won't measure ESR. The cheap Chinese do-everything gadgets based on Mega328 are pretty good for casual testing although they don't do leakage. Good leakage testing is the province of my GR 1644 Megohm bridge, which gives results for stresses up to 1000 Volts.

So capacitor testing isn't a cut and dried process, but requires judgment and care. Michael Faraday gave us a can of worms for sure.


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 9:17 pm 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 8544
Location: Long Island
Quote:
But I want to know what causes the problem. Shotgunning a piece of vintage equipment may fix it (or may introduce new problems), but it isn't troubleshooting. I want to understand how the circuitry works.


Your initial assessment is pretty much spot-on. We know basically how Wheatstone bridge operates on DC, right? When the currents are balanced in the arms on each side of the bridge, there is no difference in potential between the terminals that the galvanometer or null detector is connected to, so it indicates "null." Well when a bridge is operated on AC, two conditions have to be met to get a null, not just one. First, the currents have to balance as before. But then in addition, the phases of the currents have to be in balance too.

In an AC circuit that contains only pure resistance, the current and voltage rise and fall together at exactly the same time as the cycles progress. The phase of the current to the voltage would therefore be zero. But if the resistance is replaced with pure capacitance, the current will no longer be simultaneous. Instead, the current would be ahead of the voltage by a phase angle of 90 degrees. Now if there is a combination of resistance and capacitance together, which is the case if a capacitor has significant leakage, that phase angle will be more than zero but less than 90 degrees. There are many good textbooks that go into the deep theory if one is interested.

The way capacitor testers such as the TO-3 work is, the "standard" or reference cap is in an arm on one side of the bridge, and the unknown capacitor is in an arm on the other side of the bridge. The other arms are the variable pot with the pointer knob, and some ratio or range resistors. If both capacitors are perfect, the phases of the currents on both sides of the bridge will balance and a good null results. But if the standard capacitor is leaky, the phase on its side of the bridge will be off, so you will get a better null when you test a bad cap than when you test a good one--just as you saw.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 9:44 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 18544
Location: Albuquerque, NM 87123
dme3 wrote:
OK, for the record I replaced the two .01uF caps that made up the .02uF standard, and now everything is just about spot on in that range, with "eye" action looking great.


When I make up standard capacitors for capacitance bridges I usually buy 100 or so capacitors in .01uf and 1uf. I use an accurate capacitance meter and mark all the capacitors, then match them up by twos (one low, one high reading) to get precise .02uf and 2uf. Much more precise than the original specs of 2 or 5%.

If you just grabbed two .01uf and came up with .0200uf, my congratulations to you for your luck.


Top
 Profile  
 
 Post subject: Re: Tel-Ohmike likes leaky caps!
PostPosted: Jun Tue 18, 2019 10:05 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 44
Location: Ithaca, NY USA
Johnnysan wrote:
dme3 wrote:
OK, for the record I replaced the two .01uF caps that made up the .02uF standard, and now everything is just about spot on in that range, with "eye" action looking great.


When I make up standard capacitors for capacitance bridges I usually buy 100 or so capacitors in .01uf and 1uf. I use an accurate capacitance meter and mark all the capacitors, then match them up by twos (one low, one high reading) to get precise .02uf and 2uf. Much more precise than the original specs of 2 or 5%.

If you just grabbed two .01uf and came up with .0200uf, my congratulations to you for your luck.


I'm almost positive that's what Sprague did in 1950 when they built this thing. The .01 caps look like regular Sprague wax paper types right off the production line. I'm sure they matched up a low-reading sample with a high-reading sample and it was much cheaper than making precision caps in those days.

I had two identical-looking Sprague caps in my parts stash...they tested about the same and both had very low leakage, so I put 'em together and hoped for the best. Guess I lucked out.

Thanks to bob91343 and Chris108 for your explanations...I never studied electronics formally and I'm sorta proud of myself for stumbling upon the basic idea after reading the Cliff's Notes version of bridge theory. Just wanted to be sure I was on the right track.


Top
 Profile  
 
Post New Topic Post Reply  [ 9 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: Google [Bot], Jthorusen and 6 guests



Search for:
Jump to:  


























Privacy Policy :: Powered by phpBB