Translation needed or mini puzle

[Kiriakos-GR]

I found this text, that it supposed that this "dude" describes by using half worlds a method about testing capacitors with out ESR meter.
Now read it, and tell me what he does ... 

Probably he uses AC voltage , but he does not say how much, and then I am loosing him.
I think that he speaks about one know capacitor by putting a second ( for testing)  in line ... and then again I am loosing him.. 


You can detect errors with a capacitance meter, and a ESR meter (effective series resistance)
I do not have an ESR meter so I use a 400v 470nf cap in series with my multimeter (on the AC volts range)
reading lots of good working sets will give you a feel of what a good cap may read for your meter/cap
and then when you meet a bad one you will see the difference
I also have an oscilloscope to look at the waveform (ideally a flat line)

[amspire]

I think this person is looking at the AC voltage across the capacitor under actual circuit load. The 470nF capacitor is just to eliminate the DC voltage across the capacitor under test.

If you were familiar with one particular device, then you could get to know what to expect in a good device, and you would see the difference in a device with a bad capacitor, if it was, say, a rectifier filter capacitor.

So if you were repairing a lot of similar devices, it may be a workable technique.

I cannot see the method being a lot of use as a general technique. No use at all in detecting a problem in say a preamplifier audio coupling electrolytic.

Now there are ways to test ESR without dedicated ESR meters.  You could apply a 5v 100KHz squarewave via a 10nF capacitor to a capacitor under test. This puts about 30mA through the capacitor. Measure the voltage across the capacitor under test. Any capacitor over 20uF should have less then 10 mV voltage with 30mA current.  Large filter caps will have less then 1mV across it.

Lots of other ways you can do it - like using an oscilloscope to look at the step voltage across the cap when you apply a 10uS pulse via a resistor.

The good ESR meters will keep the maximum test voltage very low, so they can test capacitors in-circuit with no chance of blowing up other components. It is very easy with a rough test setup to be mislead by spurious voltages, and again the ESR meters are designed to be as trouble free as possible. And they are very quick and simple to use.

Richard

[alm]

Looks to me like he's just measuring ripple in circuit. Hence The comment about a scope showing a flat line. Of course ripple also depends on the circuit, eg. current draw. It's not very helpful if there are lots of them in series. 

[Kiriakos-GR]

I think this person is looking at the AC voltage across the capacitor under actual circuit load. The 470nF capacitor is just to eliminate the DC voltage across the capacitor under test.

Richard

Hi Richard.
Well this method it supposed to be used on analogue TV sets,
usually the low voltage set like  150V  20V 12V 9V comes from an simple transformer and the AC to DC conversion,
it is made just by a simple diode ( not bridge).

Theoretically speaking, what it would be the AC voltage that a single diode will not convert in to DC,
if the source is 150V  ?     
Its known that the capacitor acts as filter so to eliminate the called ripple.

Your estimation about test at full load must be correct.
Alm is correct , that he is trying to measure ripple.

The first conclusion could be that if the capacitor falls sort under load so to eliminate an major part of ripple,
it should be damaged.

And now the new question would be how much ripple per volt is acceptable ?
Can we make a rule of the thump about it ?

[alm]

Old service manuals often mentioned ripple measurement and would provide reference values. Without understanding the circuit, it's hard to make any useful estimate IMO. If there is a linear regulator, for example, the ripple only needs to be low enough so the regulator doesn't drop out of regulation. If it's powering something like a fan, you may barely need any smoothing. If it's an unregulated supply powering an amplifier with almost zero CMRR, it may need to be very low.

[Kiriakos-GR]

unfortunately even the  B&O service manual from my TV set that is a model of 1989,
it does not have such info.

The only positive is that by owning the U1272A I can use the mixed mode "AC/DC volts",
and get both readings with a single move.