Testing Capacitors

[SuperChickenPlucker]

I built a VJr head about a year ago - completely gutted it and used a turret board kit with a Hammond 125ESE. It sounded great with the Weber 8" alnico speaker I bought to go with it.

I also have a Fender DRII that I've been playing recently, and when I went back to the VJr, it sounded thin, flabby, and harsh in comparison. I thought perhaps it was the 8" speaker that made such a difference in tone, but when I hooked the 8" up to the DRII, it sounded fantastic. So I hooked up the VJr to the Celestion G12H in the DRII cab and it sounded just as thin, flabby and harsh as thru the 8".

So I broke the VJr open and started testing components and voltages and came up with a few odd readings. I used a Fluke 179 DMM to test the resistors and capacitors. All the resistors tested close to stated values with the exception of R1/1M which actually tests 9.82k (yes, it's marked Brn-Blk-Grn, odd it would test 10k), and R16 and 17 which are marked as 100R but actually measure 50R.

Measuring the caps yielded wild variations from the marked values, so I'm wondering if a DMM can be used to reliably measure capacitance. C1,C2(coupling caps)/.0022uf gave no reading at all, but did measure 350k ohms - so I think they must be blown. C3, C4(cathode caps)/22uf both read 115uf. C5/1000uf tested at 7000uf(!?). Filter cap C6/47uf tested at 80uf, and C11/0.1uf tested at 25uf.

Is it possible the cap values are that far off, or is the meter unable to accurately test the caps?

[Alex/Tubewonder]

Measuring passive components in circuit will almost always give erroneous readings. The components are connected to other components and what you are actually measuring is combined effect of "everything". Some components have one side at very high impedance like plate or cathode resistors, grid leak resistors can be connected to pots etc. Lift one component lead from the circuit if the reading is off and measure again.

[Enzo]

I agree, disconnect one end of each part you want to test. Your meter has no way to ignore the parallel circuit paths the parts find themselves in.

ANd be aware that caps can hold onto small charges a along time. Nothing dangerous, but even a tenth of a volt ina cap will totally distort both reasistance readings and capacitance readings. In fact, the resistance measurement will actually PLACE a small charge on the cap.

COnsider that the difference in tone can be related to the circuit design, the transformers in use, the power suply design, etc, etc. I doubt you will find a "bad" part.

[SuperChickenPlucker]

Thanks for your replies.

Have you ever had components damaged from too much heat when soldering them in place? And how would one know without disassembling and testing the part. Then it has to be soldered back in place, subjecting it to more heat. I guess the components must be pretty sturdy since folks here do it all the time.

Being relatively new to amp construction, that is my concern - that I damaged a part during construction by overheating it.

What are your thoughts?

[jrfrond]

Any cap that tests a steady 350K ohms is leaky, and will sport zero capacitance.

Garden-variety electrolytic caps are usually +80/-20% tolerance, but some of yours sound WAY out of spec. However, to be accurate, you must test them OUT-of-circuit, discharging them first. Be forewarned that this too is NOT a guarantee of performance. Many times (more often than I care to admit), electronic components that pass static testing will fail in-circuit when energized. These are times when troubleshooting skills, as well as a good stock of available replacement components, become necessary.

[steeveejay]

I built an ESR & low ohms meter to measure electro's in circuit . It measures the
"equivalent series resistance" of the cap . faulty caps dont lose their capacitance significantly , but their esr does go way hi . a capacitance meter will give you a low reading when the cap is near open circuit . electrolyte has electrical resistance which is very low , when it deteriorates , the esr creeps up . the cap acts like it has a series resistor . when the esr spiral up so does the heat due to ripple current , which leads to the ultimate failure ,, the dark side of the force !!

[Mars Amp Repair]

Well folks,
I like many here have been in the biz forever. I've worked on radios, vcr's, one tube color cameras, you name it. I personally have never trusted a capacitance checker for anything. There are so many weird parameters of a capacitor that a tester really only tells you what you would easily find by either bridging or substituting the cap.

I find for most circuits, esp power supplies, you can either merely bridge the new cap across the old one to see if the result improves or just disconnect one end & substitute a new cap in. Even if you dont' have the exact value, you will undoubtedly see an improvement if the old cap is bad. Substituting the cap is the only way to test caps in say an oscillator circuit (unless the cap is open).

Also it depends on your symptoms as to what caps you would obviously suspect. If you have an really old amp, you might as well replace all the filters as well as the coupling caps, esp if the coupling caps are of the oil variety as in some of the older Japanese amps.

Just IMHO, glen

[Jaybird]

This may sound wierd, but I think that the sound your not liking when comparing to your Fender is the 125ESE OT. I personally like the 125CSE. It is more suited to the EVJ in my opinion. I used a 125ESE in one of the Tweed champs I built and the sound is weak and thin compared to the Tweed I built using a 125CSE. Night and day. The 125DSE might be good but I have not tried one to know. The CSE is an 8 Watt,the DSE a 10, and the ESE 15 watts. That can make a difference in your tone. It may not be your caps at all. With the OT more is not always better.

[SuperChickenPlucker]

This may sound wierd, but I think that the sound your not liking when comparing to your Fender is the 125ESE OT. I personally like the 125CSE. It is more suited to the EVJ in my opinion. I used a 125ESE in one of the Tweed champs I built and the sound is weak and thin compared to the Tweed I built using a 125CSE. Night and day. The 125DSE might be good but I have not tried one to know. The CSE is an 8 Watt,the DSE a 10, and the ESE 15 watts. That can make a difference in your tone. It may not be your caps at all. With the OT more is not always better.

That's an excellent thought. I think I may try that out.

[Enzo]

As a matter of practicality, I am with Glen here. If I suspect a cap, it is usually far faster to tack in a sub than to set up the equipment to characterize the old one. PArts are cheaper than labor.

On the other hand if you are trying to figure why two identical amps sound different, then there is a case for component characterization. But measuring parts bewteen two different brand, different model amps seems pointless.

[jrfrond]

Well folks,
I like many here have been in the biz forever. I've worked on radios, vcr's, one tube color cameras, you name it. I personally have never trusted a capacitance checker for anything. There are so many weird parameters of a capacitor that a tester really only tells you what you would easily find by either bridging or substituting the cap.

I find for most circuits, esp power supplies, you can either merely bridge the new cap across the old one to see if the result improves or just disconnect one end & substitute a new cap in. Even if you dont' have the exact value, you will undoubtedly see an improvement if the old cap is bad. Substituting the cap is the only way to test caps in say an oscillator circuit (unless the cap is open).

Also it depends on your symptoms as to what caps you would obviously suspect. If you have an really old amp, you might as well replace all the filters as well as the coupling caps, esp if the coupling caps are of the oil variety as in some of the older Japanese amps.

Just IMHO, glen

Pretty much what I meant when I said "as well as a good stock of available replacement components". Subbing is a valid form of troubleshooting. I also don't accept the results of cap checkers as gospel. This is where experience becomes paramount. For the record, I've tried lots of test gadgets over the years, and most of them just plain suck!

[Enzo]

SOme years back, we went out and paid the megabucks for the fancy Sencore cap tester. What is it called, Z-meter? I think it was model LC something. It does everything. Leakage tests up to something like 500v, ESR tests, measure capacitance, power factor, dielectric absorbtion, probably more. Also checks inductors.

After the novelty wore off, pretty much all we ever used it for was to measure inductors. And about the only time we needed to measure inductors was to replace one on a speaker crossover. Replacements came in values of 2, 3, 4 etc. So when we needed a 3.6mHy inductor, we'd unwind a 4 while monitoring on the Sencore. Not sure that is the best money we ever spent. I suppose it might have come in useful now and then on a small inductor in some EQ circuit. But that never came up.

[jrfrond]

The handiest thing I've purchased for the shop is a Wavetek LCR55 DVM. It does the usual DVM functions plus (duh!), LCR functions. The cap and inductor checker works pretty well, but what is REALLY handy is the milliohm function, which steers us towards shorts pretty easily once you get the feel of it. Veryuseful for finding those shorted .1uF coupling caps used all over the place in IC-based circuits. You just need to find the path of least resistance. We also have an old Leader Transistor Tester, used mostly to match transistors from stock for power amps. The last gadget is a StudioMaster MIDI Analyzer, no longer made. It's a little down-and-dirty, self-contained, battery-operated box that uses idiot lights to track MIDI functions. Very cool. That's about it.

[Gtr_tech]

I have a couple Sencore LC77s. I use them quite a bit. They make a handy reference when a customer is doubting advise to replace filter caps. I have them look at the vent seals bulging out, then pull one and show them the leakage reading on the sencore. Shortens the time going in circles with someone saying "well it always worked before"....yea, but how much longer *will* it work in its current condition?

Woulda come in handy for ringing flybacks if I was still doing arcade game repair, but last time I did that I got burned so bad I doubt I'll *ever* get caught up in that again.

[Enzo]

Oh lord, I STILL have video monitor customers. Pain in the butt too, since they are all 27" these days.