LCR meter opinion

[nctnico]

Bottom line: saying LCR meter can't measure capacitors at higher frequencies doesn't make sense. It screams measurement error to me.

I don´t tell that. 

It´s important to do measures according to the specs for getting more or less constant reproduceable results you or others can work with.
Specs what the selected cap concerns and what the measure device concerns.
Measuring values with a frequency higher/lower than the frequency on which the values based according to the datasheet means simply, that what you measure then is not specified.

No. You are grossly mis-interpreting capacitor specifications! According to your logic capacitors should only be used at the frequency stated in the datasheet which is completely bonkers. The real problem with one or two datapoints in the datasheet is that you need an LCR meter to get yourself more datapoints if you are interested in those. Keep in mind that manufacturers want their parts to look good so they are more likely have left high dielectric loss figures out of the tables you posted for cosmetic reasons rather than not being able to measure them. OR that the losses at the maximum voltage rating exceed the thermal rating of the part; but that doesn't rule out using a higher voltage (physically larger) part at a lower voltage but higher frequency. The latter would require you to use an LCR meter to measure the losses at the maximum frequency you want to use the capacitor at.

And again, your wildly varying test results are measurement errors. I'm pretty sure that an LCR meter designed to go up to 100kHz can measure a capacitor with a value in single digits uf within a few percent of error (assuming the LCR meter has a reasonable base accuracy) IF the LCR meter is used properly.

[ogden]

Sorry but you are not convincing me by showing capacitor specifications. It is not like you can't use these capacitors above the frequencies used to measure / specify them.

We talk about measurements, not use. Specified at some frequency does not mean it can't be used or won't work at higher frequencies. Thou if you want *assured* performance at higher frequency than your capacitor specified at, you are using wrong component.

Bottom line: saying LCR meter can't measure capacitors at higher frequencies doesn't make sense. Your results scream measurement / equipment error to me. Are the Kelvin clips actually Kelvin clips? How about shielding?

Look into DER EE5000 datasheet - at 100KHz frequency max 10uF, 3% accuracy. Keysight U1733C has 5% accuracy for up-to 20uF at 100KHz. It's because of LCR meter, not capacitor limitations. Isn't that obvious for you already?

For example: this 10uf 0402 capacitor is specified up to several GHz: https://www.murata.com/en-eu/products/productdetail?partno=GRM155D70G106ME18%23

Capacitor indeed can be measured at any frequency, even up-to several GHz, but not using common LCR meter! For that impedance analyzers are used, like: https://www.hioki.com/global/products/lcr-meters/3-ghz/id_6618

[nctnico]

Sorry but you are not convincing me by showing capacitor specifications. It is not like you can't use these capacitors above the frequencies used to measure / specify them.

We talk about measurements, not use. Specified at some frequency does not mean it can't be used or won't work at higher frequencies. Thou if you want *assured* performance at higher frequency than your capacitor specified at, you are using wrong component.

That is what I would call a contradiction. You will want assured performance in any application!

I gave a perfect example which you appearantly have missed. Again: If my application doesn't need a low loss-factor at a given frequency, I'm not going to care that the loss factor isn't specified for a certain capacitor at that frequency. Just check the material specifications (graphs and links have been posted before) and measure a few samples to make sure what kind of ball-park figure the value is at to prevent potential problems.

Think about how audio equipment uses electrolytics in the audio path. In consumer equipment none of those capacitors are specified at 20kHz.

Bottom line: saying LCR meter can't measure capacitors at higher frequencies doesn't make sense. Your results scream measurement / equipment error to me. Are the Kelvin clips actually Kelvin clips? How about shielding?

Look into DER EE5000 datasheet - at 100KHz frequency max 10uF, 3% accuracy. Keysight U1733C has 5% accuracy for up-to 20uF at 100KHz. It's because of LCR meter, not capacitor limitations. Isn't that obvious for you already?

You are repeating exactly what I'm stating! And the conclusion is the same: measurement error in the case where an 8.2uf capacitor shows results which are way off (far outside what you would expect based on the LCR meter specification). In the end an LCR meter is less easy to use than it appears.

[ogden]

Sorry but you are not convincing me by showing capacitor specifications. It is not like you can't use these capacitors above the frequencies used to measure / specify them.

We talk about measurements, not use. Specified at some frequency does not mean it can't be used or won't work at higher frequencies. Thou if you want *assured* performance at higher frequency than your capacitor specified at, you are using wrong component.

That is what I would call a contradiction. You will want assured performance in any application!

Application?      It does not matter - opponent can't read on purpose or not, I am out of such kind of discussion. GLHF

[nctnico]

Application?      It does not matter - opponent can't read on purpose or not, I am out of such kind of discussion. GLHF

Now you are being silly.

Back to the core of the discussion: Are you seriously going to suggest that a reading which is several 100's of percent off from what is expected (based on what the test equipment is capable of) indicates that the piece of test equipment is unsuitable (despite it's specifications saying it is) and not a case of user error?

[ogden]

Application?      It does not matter - opponent can't read on purpose or not, I am out of such kind of discussion. GLHF

Now you are being silly.

Back to the core of the discussion: Are you seriously going to suggest that a reading which is several 100's of percent off from what is expected (based on what the test equipment is capable of) indicates that the piece of test equipment is unsuitable (despite it's specifications saying it is) and not a case of user error?

You are wandering all over the place. When we are talking about assured performance of *capacitor* at higher frequencies - we were talking about *capacitors*, not about equipment capabilities.

[nctnico]

I'm following your lead. You came up with 'assured performance' which I countered with the hint that you'll need to use a component that is fit for purpose anyway. But maybe you missed that hint.

Anyway, if your beef is with capacitor specs: in general the specifications for capacitors leave a lot to be desired and in many cases you have to look further than just the datasheet. Capacitor datasheets just provide a few data points from which you can determine the performance over a wider range based on construction and dielectric. For example: the datasheet for MLCC capacitors typically doesn't specify the leakage current. But this isn't zero and for some circuits this leakage is not insignificant. You'll need to dive into the properties of the dielectric to get a ballpark figure (and perhaps measure a bunch of samples). So yes, an LCR meter is an important tool to measure components to supplement information that is not in the datasheet.

[Martin72]

and in many cases you have to look further than just the datasheet

But not with those instruments we´re talking about and not in "many cases".
It´s always the same when you´re "on stage".

Nect week I´ll continue testing between the three LCRs and several caps with values between 47nf and 4.7µF.
In the table, I´ll put the dissipation factor sheet vs measuring, when it´s possible because of the limited resolution of the keysight and DE5000.
Btw., is a schmematic of the DE5000 known/ avaible ?

[nctnico]

Before taking any more measurements: check the Kelvin clips and connectors for damage and do an open / short offset calibration after changing the measurement frequency. You should not get any readings which exceed the LCR meter's accuracy specification with those capacitors.

[ogden]

I'm following your lead. You came up with 'assured performance' which I countered with the hint that you'll need to use a component that is fit for purpose anyway. But maybe you missed that hint.

What would "if you want *assured* performance at higher frequency than your capacitor specified at, you are using wrong component" mean, huh? Whatever you are smoking - you shall consider to quit.

[nctnico]

I'm following your lead. You came up with 'assured performance' which I countered with the hint that you'll need to use a component that is fit for purpose anyway. But maybe you missed that hint.

What would "if you want *assured* performance at higher frequency than your capacitor specified at, you are using wrong component" mean, huh? Whatever you are smoking - you shall consider to quit.

Did you ever see a capacitor specification / datasheet stating a maximum operating frequency? IMHO you are mis-interpreting capacitor datasheets by mistaken datapoints for operational limits.

[ogden]

Did you ever see a capacitor specification / datasheet stating a maximum operating frequency? IMHO you are mis-interpreting capacitor datasheets by mistaken datapoints for operational limits.

Specified at particular frequency does not mean it is maximum operating frequency. Are you OK?

[nctnico]

Did you ever see a capacitor specification / datasheet stating a maximum operating frequency? IMHO you are mis-interpreting capacitor datasheets by mistaken datapoints for operational limits.

Specified at particular frequency does not mean it is maximum operating frequency. Are you OK?

You are contradicting yourself again. At one hand you say assured performance based on data points in a datasheet are not to be exceeded (no guarantees beyond that) and on the hand point your are claiming that it is fine to use a capacitor at any frequency. What is it? Make up your mind!

[ogden]

Did you ever see a capacitor specification / datasheet stating a maximum operating frequency? IMHO you are mis-interpreting capacitor datasheets by mistaken datapoints for operational limits.

Specified at particular frequency does not mean it is maximum operating frequency. Are you OK?

You are contradicting yourself again. At one hand you say assured performance based on data points in a datasheet are not to be exceeded (no guarantees beyond that) and on the hand point your are claiming that it is fine to use a capacitor at any frequency. What is it? Make up your mind!

I was saying: you *can* use component at ANY frequency, yet if you want assured performance at specified frequency - you better have specification for it. I am explaining this like third time here 

[edit] I also said: "Specified at some frequency does not mean it can't be used or won't work at higher frequencies. ".  Explaining in your terms: No guarantees beyond data points in datasheet does not mean one can't use part beyond that, just performance won't be assured. Comprehend?

[Martin72]

Before taking any more measurements: check the Kelvin clips and connectors for damage and do an open / short offset calibration after changing the measurement frequency.

I think I know what I´m doing since 22yrs, but thank you for the advices.

You should not get any readings which exceed the LCR meter's accuracy specification with those capacitors.

This was already proved, at 1 and 10khz you get results which are very familar in comparison with the two other LCRs.
At 100khz it´s drifting a little bit, but it´s still in the specs of every LCR.
What the displaying capacitance value concerns.
But that wasn´t the point in this thread.
Mainpoint starts with the statement of a user, after several recommendations to go for the DE5000, that this LCR is a good one, but have problems on 100khz.
The "problem" turned out to be a probably wrong display of the D-factor in the 100Khz range.
This is irrelevant in more ways than one, but it does give the impression that the LCR has a flaw.
Yes, in theory it is -BTW, I found out, that the U1733C got this too.
But in practice...
Point one:
You won´t measure µF caps with 100khz - why should you.
Point two and most important:
The resolution of the D-factor display itself.
"0.001" - If you exclude the basic digit errors, this would be the highest resolution.
0.001 or 1x10-³...
This resolution is useless for foilcaps.
So if the LCR shows "nonsense" at 100khz or "0.000", it´s the same:
You can´t see the real value.
Same when you use lower testfrequencies, the usually ones for foil caps in µf-range, 1 or 10khz.
You´ll see "0.000" or "0.001"
Therefore it is a problem in theory, not a real one which can stop you buying the DE5000.
Displaying the capacitive value was never the issue here.

[cdev]

itd great but people should realize that most of the info you usually seek to get from L and C components could also be derived with a small and inexpensive nanovna2 too, using the right fixture.. (for series or shunt components). If I was going to choose between the two I would choose the nanovna. I spend a lot more time playing around with it than any piece of test equipment I have had previously. But the nanovna2 - although its not as rapid to test a passive component for something like esr, you will also need to make a "test fixture".

But it will show you the numbers you need to know and do it in a useful way.

The DE-5000 will be able to test components at (much) higher frequencies than most LCR meters but its still very limited to the in depth data you'll get with a nanovna2. If you are into RF stuff, that decides it. If you are into more generic electronics, maybe the DE-5000 will be better. I suspect it would be they are two specialized tools each does well at what its intended to do.

I dont know. Until I got a nanovna2 not much later DE-5000 was the only LCR meter Ive owned besides the AVR Transistor tester.. (which is very useful, especially for its cost, too, its really cheap. It will give you a very rapid interesting fast evaluation of what any component with two or three connections is. By probing them with a bunch of different signals. Simply having the ability to measure low L was a big thing for me. And it turns out DE-5000 does very well at that. But so does the nanovna2 it turned out.

If you are really on a tight budget also look at the AVRTT. For $20 or less, It will act as an LCR meter within limits. Also give you ESR. Accuracy is surprisingly good.  But, it does not support 4 wire measurement, as far as I know. Maybe a new version does. They are constantly improving it.

[Martin72]

I don´t know this meter, but what I know in general:
You won´t cover all your needs with only one instrument.
For example inductor measure...
We (Company I work for) use at last three methods to measure them.
General purpose LCR Meter, a (selfbuild) voltage-time-area tool and the good old resonant-frequency solution, with a little bit power...

[ogden]

VNA has it's own impedance limitations. One can't replace another. For component measurements I still suggest LCR meter, especially for >1uF capacitors, >1mH inductors and literally any resistor, unless you are looking for hi-frequency parasitic impedances of such. VNA is RF territory. Thou there is one capacitor-related measurement I suggest VNA any day: power plane impedance analysis. If you try - don't forget to use DC blocking capacitor! Without such you may damage instrument.

[probe]

Hi all,

For me the DER-EE DE-5000 LCR kit ordered at eleshop.eu and shipped to Italy would be €240, a set of kelvin clips around €20, so €260 total. If I were to order a Tonghui TH2810+ from Batterfly.com it would be €667 shipped, including Kelvin clips, fixture, short and test report.

So, €400 more for the Tonghui. I'm leaning towards a benchtop, but the Tonghui is 10kHz and the next next step up to 100kHz would hit €1000.

What would be your advice? general hobbyist use, mostly HAM radio stuff. (building a QDX now, when available will order a (tr)u-SDX).

[Shock]

My advice would be to avoid the DE5000 kit and just look at getting a DE5000 with TL21 adapter. The kit adds almost $200 to the price for no additional measurement functionality.

For most people to start off with they are fine. Roughly their base cost is $90 plus accessories on the ebay.com site. A fancy display and pc connectivity is mostly a gimmick on LCR meters for home use. Moneys better spent on a decent diy test lead or fixture setup in my opinion (it's the most painful aspect of a measurement).

[Martin72]

My tests will be expanded by another relative cheap LCR measure device.
Just bought an ET4410 today, will arrive end of next week.

[trobbins]

If the OP is in to DIY and short on cash then I'd recommend setting up their PC and soundcard as an impedance measurement rig.  I have a decade old USB soundcard with 192kHz sampling and it accurately measures L, C and R from circa 2Hz to 90kHz as an impedance plot with free software like REW.  The accuracy performance relates to the reference resistor tolerance used during calibration, so a 0.1% resistor allows better than 1% capacitance and inductance measurements.  The measurement plot also provides enhanced information on parts, especially inductors where the first self-resonance will typically show up, as well as the shunt capacitance value.

There are quite a few ESR meters suitable for DIY - I use a circuit by JAY_DIDDY_B in the projects area of this forum that properly extracts the resistive component of capacitors (no matter what type), as well as measuring resistance levels down to 100 microohm resolution to about 1milliohm accuracy. 

Combined with a cheapo 4-digit DMM like the Aneng AN8009, that covers rapid and accurate measurement of R and C, and a more informative measurement of L than the common practise of purchasing what can be an expensive commercial LCR meter.

[Martin72]

My tests will be expanded by another relative cheap LCR measure device.
Just bought an ET4410 today, will arrive end of next week.

It´s here (on postal office, will get it tomorrow)!! After only 48h, although I´ve ordered it from Banggood.
While ordering, I had the choice between "CN" and "CZ" Location - "CZ" was a little bit cheaper, so I´ve choosed this.
Now I know what it means....

Meanwhile I´ve expand the cap test, adding some electrolyt and ceramic capacitors.
So table will start with 470p, 1n ceramic, then 47n, 100n FKP, 220n MKP, 470n Snubber MKP, 1µ MKP250V, 1µ MKP700V, 2.2µ MKP, 3.3µ MKP, 4.7µ MKP, 8.2µMKP, 10µ elektrolyt, 100µ elektrolyt, 1000µ elektrolyt.
Cap-tests finished with the DE5000, U1733C and ST2830 will follow, at last the ET4410.

[ogden]

Did you get it? Please share your impressions

[Martin72]

Hi,

Sorry forget it here....

New thread:

https://www.eevblog.com/forum/testgear/east-tester-et4410-desktop-lcr-meter/