Quick advice on ESR meter: did I really need 100KHz?

[mcinque]

During circuit troubleshooting, can the ESR of the electrolytics be measured accurately with a low frequency ESR meter? Or is recomended to use always 100KHz as test frequency?

Which criteria is used to choose 120Hz/1KHz/10KHz/100KHz on a expensive LCR bridge during ESR measurement?

I know that the more that the test frequency increases, less reactance will have the capacitor, but are 1KHz high enough to achieve a PASS or FAIL reading value on most electrolytics?

I guess the test frequency it's mostly a matter of where the tested capacitor is being used (switching, audio decoupling etc), am I wrong?

[w2aew]

During circuit troubleshooting, can the ESR of the electrolytics be measured accurately with a low frequency ESR meter? Or is recomended to use always 100KHz as test frequency?

Which criteria is used to choose 120Hz/1KHz/10KHz/100KHz on a expensive LCR bridge during ESR measurement?

I know that the more that the test frequency increases, less reactance will have the capacitor, but are 1KHz high enough to achieve a PASS or FAIL reading value on most electrolytics?

I guess the test frequency it's mostly a matter of where the tested capacitor is being used (switching, audio decoupling etc), am I wrong?

There are a couple of ways to look at this.  From a purely practical / applications standpoint, it makes sense to test for ESR at a frequency that is close to the frequency that that the capacitor will see in the circuit.  In general, the higher test frequencies are used for lower value capacitors - to help lower the capacitive reactance so that the ESR is more easily measured.

[mcinque]

Thank you Alan.

Since I want to buy a LCR bridge and a ESR meter for use in basic in-circuit troubleshooting, I've thought that the Agilent U1731C (1KHz max ESR test frequency) can be the right choice, but I'm not sure about its 1KHz frequency limitation.

Do you suggest that it can be a good choice or it's better to grab the U1733C that has 100KHz max test frequency? Or maybe is better to buy a single LCR bridge and a separate ESR meter like the Anatek Blue ESR?

[plesa]

1kHz is enough for basic testing and troubleshooting. You cannot use it for example for diode capacitance measurement. There is hack available to enable 100kHz, but it will require calibration.

[robrenz]

The DE-5000 is a great meter that cost a lot less than the U1733C (IET or not), has 100kHz and is true 4 wire connections which the Agilent is not.  The extra digit of resolution of the U1733C accomplishes next to nothing when you consider the accuracy there.

The U1733C is a very nice meter but you can get the same functionality and basically the same accuracy for a lot less money.

[G0HZU]

For in circuit testing I use a hastily prepared system involving a series resistor, a function generator and a sensitive AC voltmeter. I usually use a spectrum analyser as the AC voltmeter but with a fused diode clamp and dc block to protect it from damage. This method just gives an indication of impedance rather than ESR but it is very good at finding dead caps and the test frequency can be continuously varied as required.

A decent DMM could be used instead if it can measure AC voltage up to 100kHz. So the cost is virtually nothing if you already have the FG and the voltmeter.

For out of circuit testing I use a scope plus the same function generator and a series sense resistor. This lets me measure capacitance and ESR vs frequency. I've never ever owned a commercial ESR meter because most of them are limited in capability and they lack the versatility of the sense resistor + scope method. i.e. they only measure at one or two test frequencies.

For sure, you can spend a lot of money on an expensive LCR meter but you are usually limited to <100kHz.

[nowlan]

I thought the datasheet only specified 1khz.

[BravoV]

During circuit troubleshooting, can the ESR of the electrolytics be measured accurately with a low frequency ESR meter? Or is recomended to use always 100KHz as test frequency?

For this question, it is very likely that you're talking about electrolytics on "SWITCHING" PSU right ?

Then the answer is yes, most electrolytics caps especially at the secondary part, their ESR mostly are rated at 100 KHz as in their datasheet. Just download any branded low ESR "ELECTROLYTIC/ORGANIC" capacitor datasheet to see the freq at their rated ESR.

[mcinque]

thank you all for your replies.

Unfortunately I can't find the DE-5000 in my country (and I really dislike the battery holder solution that that meter use) but thank you for the advice.

I've tried the method suggested by you G0HZU, and it works fine.

I've noticed the differences in readings by changing the frequency, expecially with low vaule caps, like Alan said.

Looking at the readings, imo 1Khz isn't enough to achieve a clear result with low values.

I'd like however to have a *numerical* value to compare with the datasheet specifics, and I guess that a ESR meter it's the only quick solution available without doing math.

I was wondering if the 1733C worth the money to use it as a LCR bridge and ALSO as ESR meter for in-circuit testing or if is better to buy a cheaper LCR and a separate ESR meter. What do you think about this?

[robrenz]

One of your countrymen bought a DE-5000 as stated on this post I suggest you scan the whole thread as it is full of links.  This should be less than half the cost of the 1733C and basically the same performance.

IMO buy a  single LCR/ESR meter.

[kripton2035]

you can get one of these to italy : http://kripton2035.free.fr/LCR%20meters/lcr-deree5000-cl.html
direct search to ebay sellers : http://www.ebay.com/sch/i.html?_nkw=der+ee+de+5000&LH_PrefLoc=2&_sop=15
dont forget to get one with at least crocodile or tweezers probes it's difficult (i.e. expensive) to get them after.

anyway, for in circuit esr measurments, a dedicated tester as you can build or buy on my repository (http://kripton2035.free.fr/esr-repository.html) is always better : it has 100khz, protections against high voltages capacitors can have, and small voltage measurements to be in circuit friendly.

[ted572]

You need (must have) 100kHz to measure ESR, but use lower frequencies (generally 100Hz) to measure the capacitance of electrolytic and tantalum capacitors.  And BTW I prefer the U1733C.

[tautech]

You need (must have) 100kHz to measure ESR, but use lower frequencies (generally 100Hz) to measure the capacitance of electrolytic and tantalum capacitors.  And BTW I prefer the U1733C.

Not really.
ESR test frequency is relevant to the frequency of intended operation (ripple).
VERY different for Linear vs SMPS.

[Shock]

Higher test frequencies are useful for precision measurements of smaller caps. Consider if you're measuring capacitance as well, broader supported frequencies are a good thing on a good meter.

If your measuring larger caps like the big ones found in SMPS of TV's and computers, ESR test frequency of 100/120Hz is fine and this same frequency will give you a ballpark measurement of many caps, just not milliohm resolution and will not be so great for ultra low ESR caps.