DER EE DE-5000 vs MESR-100 Accuracy

[hgg]

Hello,

I have just received a brand new DER EE DE-500 LCR meter and I was comparing measurements
with the MESR-100 and the UNI-T UT61E.

I got some crazy results.  For example when measuring the ESR of a 0.1uF capacitor,
the DE-5000 showed 5.08 Ohms and the MESR-100 24.4 Ohms!  Both measurements
were taken with 100Khz.  Which one is right ?


DE-5000


MESR-100

The DE-5000 could not measure the ESR of some capacitors at all at 100Khz, like a3300uF, 470uF,
330uF & 220uF.  When I lowered the frequency to 100Hz it could measure them.

The MESR-100 did not have any problems at all with any capacitor.

The other thing is that the Uni-T was almost always closer to the written values on capacitors. 
For example below is a comparison between the DE-5000 and the UNI-T 61E:

Capacitance               DER EE DE-5000         UNI-T61E

1500uF                       1583                          1670
2200uF                       1940                          2056
470uF                         390                            415
220uF                         192                            197
330uF                         301                            326
10uF                           10                              10.4
4.7uF                          4.65                           4.8
0.1uF                          0.91                           0.93
6pF                             6                                6

The mighty UT61E compares really well with the dedicated ESR meter...
The most striking result thought is the difference in the measurement of the ESR
5 Ohms and 24 Ohms !  Can you really trust this meter?  Do you think I might have
received a lemon?

Any other DE-5000 owner?  Have you compared this meter with some reference values
to check its accuracy?

Anyway, I bought this meter only to measure inductance and it seems that the inductance
measurements were close enough to the written values.

I would be very interested to hear any other comments from DE-5000 owners.
Thanks.

[Wytnucls]

Electrolytic capacitors should be measured at 100/120Hz. Compare ESR/D against the datasheet to confirm proper LCR meter function.

[hgg]

Hi,

Actually the one I have from Panasonic is measured at 100Khz and I've seen others as well in the
data-sheets.

Its a brand new Panasonic 150uF, 6.3V Electrolytic Capacitor.
Data Sheet: http://docs-europe.electrocomponents.com/webdocs/0bec/0900766b80becf93.pdf

The ESR in the datasheet is 0.5 Ohms @ 100Khz
At 100Khz I measure 0.34 Ohms.  40% off...
The funny thing is the measured capacitance at different frequencies.

For the 150uF we have:

100hz    ->  141.5 uF
    1Khz  ->  130 uF
  10Khz  ->  104 uF  and at
100Khz  ->  16 uF !

Is this normal?

(It looks like that capacitance should be measured at 100-120Hz and ESR at 100Khz.)

[Wytnucls]

Capacitance is usually measured at 120Hz, with a 20% tolerance.
Tan d (or D on your LCR meter) is a better indication of the health of that capacitor.

So you measured 141.5uF, which is well within the 20% tolerance of a 150uF cap.
An ESR of 0.34 Ohm is probably correct, as it is close to the published max value of 0.5 Ohm.
You should also get a D (dissipation factor) of less than 0.22 for that same cap, at 120Hz.

[hgg]

Ok,

The Tan D in the datasheet is 0.22 max and its an upper value for a healthy capacitor.
Its not an absolute value that I need in order to check the accuracy of the LCR meter.

Anyway, in the auto LCR mode it defaults to Capacitance & 1Khz frequency and I get
a TanD of 0.160, but in the datasheet TanD is measured at 120Hz. 

If I select Capacitance manually and set the frequency to 120Hz I get a Tan D of 0.047

Does this mean that the capacitor is healthy with a Tan D way below the max limit?
I still don't know how you can measure the capacitance accuracy of your LCR meter...

[alm]

As you state, tan(D) (and ESR) are maximum values, so you'd expect the typical cap to be better than that, especially when new. So it sounds like that cap is fine.

You can test the accuracy of your LCR meter by comparing it to another, more accurate, LCR meter or by measuring standard capacitors/inductors (expensive, I believe GR made same). Testing resistance accuracy is obviously trivial. You could also try adding low value series resistors to a cap, and observe the increase in ESR. For low capacitance values you could get some C0G ceramic caps, these have lower tolerances and less frequency dependence. Reasonably priced ones only go up to a few nF, however.

[wraper]

The ESR in the datasheet is 0.5 Ohms @ 100Khz
At 100Khz I measure 0.34 Ohms.  40% off...
(It looks like that capacitance should be measured at 100-120Hz and ESR at 100Khz.)

Datasheet specifies max ESR, so nothing is off.

[hgg]

I've tried adding a small wire of 0.08 Ohms in series with the capacitor and the ESR increased
by 0.08 Ohms indeed, so it works ok but that is only a relative check.

So in order to check the absolute accuracy you either need a very well calibrated meter
(which by the way was calibrated how?) or the actual value that calibrated the accurate
meter...  Chasing your tail... :-)

There should be another way.

[dfmischler]

There should be another way.

Try this: Measuring Capacitor ESR

FYI - The MESR-100 specifications say it is for capacitors >= 1uF, and that the error will be larger on .1 uF.

[mos6502]

It's very easy to check how accurate the ESR measurement is. Take a cap, measure the ESR, then put a resistor (preferably metal film), put it in series with the capacitor, measure again. The second measurement should be equal to the first + the resistance of the resistor.

I'm going out on a limb here, but seeing as the DE-5000 is actually a proper instrument, and the MESR-100 looks like it's been slapped together in a Chinese basement, I'd say the DE-5000 is the one that gives accurate readings here.

The DE-5000's accuracy is given in the manual for all the ranges. What's the accuracy spec for the MESR-100? Oh, that's right, there is none. It's essentially just a random number generator.

The DE-5000 has several frequency/capacitance ranges, like every other LCR meter. You can't measure large capacitances at high frequencies, because the impedance of the DUT would be too low and thus the resulting voltage would be too low to measure. And it would be pointless, as well. That's why capacitor datasheets specify the capacitance at 120 Hz for electrolytics or 1kHz for ceramics or film caps.

Yes, the capacitance changes with frequency depending on the type of cap. Electrolytics are extremely non linear. Try a a decent film cap, and you'll see that the capacitance will measure essentially the same from 100 Hz - 100 kHz.

An electrolytic cap that has 2200uF printed on it really means "this cap has 2200uF at 120Hz". At any other frequency the capacitance will be different.

In any case, measuring capacitance of electrolytics is a pointless exercise, because, they're usually specced at -20+80%. What does measuring the capacitance tell you?

You can still measure the ESR of any DUT with the DE-5000. Simply switch to Rs mode. That will measure the real part of the impedance, i.e., ESR.

As Wytnucls already explained, measuring ESR to check if a cap is good or not isn't very helpful. Measuring DF is much more informative. Measuring ESR can be useful when you're comparing different types of caps for a certain application. Measuring ESR is a crutch, it's just better than nothing if you don't have a proper LCR meter.

So, the DE-5000 works fine and is very accurate. You just need to understand AC circuit theory better. 

[hgg]

dfmischler   Thank you for the link!  I will try it out.

mos6502   Thank you for your detailed explanation.  You might have a point about the MESR-100.
It seems though that it is accurate.  Resolution of 3 significant digits and repeatable measurements.
That is why I am looking for a way to measure absolute value accuracy.  Putting a resistor in series
with the capacitor is measuring the relative accuracy and not the actual calibration. 

Apart from the ESR,  I was also looking for a way to check the LCR meter absolute accuracy while
measuring capacitance.

[DBoulanger]

Hi there,

For your information, I also acquired the same DER meter.

Just for the fun of it, I did a simple RC circuit and evaluated the capacitance using a DSO.  For 5 electrolytics caps, ranging from .1uf to 1000uf, the readings with the meter at the different frequencies were fairly close to the ones I estimated via the RC time constant.

Yet, even with the scope, estimations were off at higher frequencies too, so that lead me to believe that the meter gave the proper reading.

I don't want to conclude too quickly, so I will do other readings with some friends LCR meters or DMMs, just to have some sort of peace of mind.

Will definitely get something interesting out of these readings.

[The Electrician]

dfmischler   Thank you for the link!  I will try it out.

mos6502   Thank you for your detailed explanation.  You might have a point about the MESR-100.
It seems though that it is accurate.  Resolution of 3 significant digits and repeatable measurements.
That is why I am looking for a way to measure absolute value accuracy.  Putting a resistor in series
with the capacitor is measuring the relative accuracy and not the actual calibration. 

Apart from the ESR,  I was also looking for a way to check the LCR meter absolute accuracy while
measuring capacitance.

Don't forget that the measured capacitance of electrolytic caps especially, vary with measurement frequency.  See reply # 36, 37, 38 and 39 in this thread:

https://www.eevblog.com/forum/testgear/mastech-ms5308-lcr-meter-with-esr-measurement-on-discount-at-the-moment/msg107489/#msg107489

[hgg]

Hi,

DBoulanger it looks that its doing a good job but the thing that made me question its accuracy in the
beginning, was that the UT61E gave measurement results that were closer to the component written
values.  Only at very high capacitance values the DE-5000 was more accurate than the Uni-T.

Electrician, that's what I saw indeed, higher frequencies gave lower capacitance measurement result.

[Samogon]

Got my new DER EE DE-5000 and it is spot on measuring inductance and low volume capacitance <1?F and i a m happy with that.
But as for electrolytics it measures less than marked value approx 20% which is in agreement of cap tolerances  . While my Fluke 289 gives me exact marked values on all ranges fro 1?F to 5000?F. Lets me assume that this LCR is not good at electolytics.
Since i am covered with both devices i am fine.
But still small devil in my head asks: should i return it as faulty?
Thank you

[_Wim_]

To measure a cap ESR, you can also do this with a function generator and an RMS multi-meter. Connect on leg of the cap to the output of the function gen (in 50 ohms output mode), the other leg to ground. Use a sine signal no higher then 100mv or so (at the frequency you want to test at), and connect the multimeter parallel to the cap to measure the voltage. We now have a 50 ohms resistor (output function gen) in series with the ESR of the cap => voltage divider network. As you know the voltage across the cap (Vcap) and the voltage of the source (Vs), the ERS can be calculated as follows:

ESR = (Vcap * 50) / (Vs - Vcap)

Edit: thanks rich, I indeed made a mistake when I wrote deducted the formula on paper. Now corrected

[Samogon]

Nice tip _Wim_, thank you will try that.
But what about capacitance? Should i just ommit this option from LCR meter and use Fluke 289 for it? Or should i consider it failure and return device?

[TimFox]

I am also happy with my DE-5000.
One important point is how to measure ESR of polarized capacitors:  The DE-5000 does not apply DC bias voltage to the device, which should affect its operation.  I am working on the design of an adapter that can apply battery bias, stealing the design from the old GR digibridge, but my lab is temporarily shut down for remodeling.

[rich]

ESR = (Vs * 50) / (Vcap - Vs)

really? It may be I didn't picture circuit correctly but equation suggests when Vcap tends to 0, the ESR tends to -50 ?

[_Wim_]

After some googling I found this one. He explains this alot better then I do: http://www.kerrywong.com/2015/11/23/esr-measurement-using-multimeter-and-function-generator/

[_Wim_]

ESR = (Vs * 50) / (Vcap - Vs)

really? It may be I didn't picture circuit correctly but equation suggests when Vcap tends to 0, the ESR tends to -50 ?

Thanks for pointing this out. I indeed made a mistake when I deducted the formula on a piece of paper. Best to follow the link I posted above, much more in depth discussion...

[Hendry Born]

Great tutorial!!

Anyway I will go for the DE-5000

[Chupacabras]

Got my new DER EE DE-5000 and it is spot on measuring inductance and low volume capacitance <1?F and i a m happy with that.
But as for electrolytics it measures less than marked value approx 20% which is in agreement of cap tolerances  . While my Fluke 289 gives me exact marked values on all ranges fro 1?F to 5000?F. Lets me assume that this LCR is not good at electolytics.
Since i am covered with both devices i am fine.
But still small devil in my head asks: should i return it as faulty?
Thank you

Hi, I got my DER EE DE-5000 today, and I see the same problem. 2200uF electrolyte is measured as 1950uF, and 4700uF is measured as 4500uF (at 120Hz).
When I measure those caps with UNI-T multimeters, they give me results very close to nominal values of those caps.

So I am quite disappointed by this meter - Correction: I am not disappointed. Later posts explained that. There is nothing wrong with DER EE DE-5000

[metrologist]

We must get to the bottom of these erroneous measurements. Is there a calibration/zeroing sequence for the meter that was performed?

[Electro Fan]

I measured some electrolytic capacitors with the DER EE DE-5000 and also a Fluke 179.  The DER, using the TL-21, seemed to read about 6% lower than the Fluke.  I did the calibration routines as shown on Pages 15-17 of the DER manual and the unit passed for both the open and short tests.  I haven't done a bunch of testing (either before or after the calibration routine) but after the calibration routine I found that the gap narrowed a bit.  For example, on a capacitor marked as 470uF I had a reading on the Fluke of 429uF (don't know at what frequency) and on the DER I had a reading of 418uF (at 100Hz).

Ran another test with a ceramic capacitor.  The Fluke says 108nF; the DER says 98nf at 100Hz, 94nF at 1kHz, and 89nF at 10kHz, and 77nF at 100kHZ.

Without something else to compare to kind of hard to say....