ESR meter 200kHz

[tigr]

How do you answer the objection I raised about the 15 uF MKT capacitor that your meter showed has an ESR of .001 ohms? I said:
"I very much doubt that the Epcos 15 uF MKT capacitor shown here could have an ESR as low as .001 ohms.

I fully agree with you. This ESR meter was not developed for the measurement of ESR for film capacitors.

[Jay_Diddy_B]

Hi group,

Here is a message reproduced from this thread:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/

In this message I measure a Film capacitor with a HP 4274A LCR meter and compare the measurements against a relatively simple ESR meter that uses a phase sensitive detector. The ESR meter adapter is able to measure a result that is lower than the impedance of the capacitor but not accurate.

Hi,

I have also done some testing with a film capacitor. I do not have the 'official' 6.8uF film capacitor. I used a 4.7uF 250V Metalized polypropylene film capacitor.

Here is a picture of the capacitor used for test:



This is has an extremely low ESR. I first measured the capacitor using my HP4274A LCR meter with a HP 16047A test fixture. Here are the measurements:



Note: That the device impedance is very reactive, >89.6 degrees at 100 kHz

Here a couple of sample pictures of the measurements:







When I connected the capacitor to the ESR adapter using the 4W configuration, I zeroed the meter using the relative feature on  the DMM.
I measured 5.355mV which corresponds to 53 m Ohms.



This is 0.26% of the ESR meter adapters full scale, and significantly lower than the impedance of the capacitor 340.7 m Ohms.





Observations

1) The HP4274A with a four terminal test fixture, and short & open calibration, is able to read ESR values in the 1 to 2 m region.

2) The film capacitor ESR is very low. I looked for an impedance curve, but I couldn't find one.

Regards,

Jay_Diddy_B

[The Electrician]

Jay_Diddy_B,

   Do you have the manual for the 4274?  Looking in my old 1988 HP catalog, the specs for the 4274 show the measurement range for ESR is 100 milliohms to 10 megohms.  I wonder how much trust can we put in a measurement for ESR of 1 or 2 milliohms?

   I have been thinking for some time about how we could get a truly accurate measurement of very low ESR in the sub-MHz frequency range.

   For high RF the resonant line technique can give 1 milliohm accuracy, but how to do it at 10 KHz?

   The problem is that if the real part of the impedance of a capacitor is R ohms, when the reactive part is 1000*R the measurement becomes difficult.  For your 10 KHz measurement, the ratio of |Z| to Rs is about 3000.

   I notice that for ratios over 1000, the meaurement on my various meters becomes noisy.  I should dig out an appropriate capacitor, measure it very carefully, and mail it to you for comparison.

   But, what we really need to do is find a special technique making the measurement where phase angle of the impedance is very near 90 degrees.

   In the case of all these measurements tigr is making, he has no independent way of verifying his measurements--he is assuming that the ESR he measures is correct.  He should mail me some of his capacitors, or I should mail him some that I have measured.

[Jay_Diddy_B]

Jay_Diddy_B,

   Do you have the manual for the 4274?  Looking in my old 1988 HP catalog, the specs for the 4274 show the measurement range for ESR is 100 milliohms to 10 megohms.  I wonder how much trust can we put in a measurement for ESR of 1 or 2 milliohms?

   I have been thinking for some time about how we could get a truly accurate measurement of very low ESR in the sub-MHz frequency range.

   For high RF the resonant line technique can give 1 milliohm accuracy, but how to do it at 10 KHz?

   The problem is that if the real part of the impedance of a capacitor is R ohms, when the reactive part is 1000*R the measurement becomes difficult.  For your 10 KHz measurement, the ratio of |Z| to Rs is about 3000.

   I notice that for ratios over 1000, the meaurement on my various meters becomes noisy.  I should dig out an appropriate capacitor, measure it very carefully, and mail it to you for comparison.

   But, what we really need to do is find a special technique making the measurement where phase angle of the impedance is very near 90 degrees.

   In the case of all these measurements tigr is making, he has no independent way of verifying his measurements--he is assuming that the ESR he measures is correct.  He should mail me some of his capacitors, or I should mail him some that I have measured.

I am not sure about the accuracy. The nice thing about the HP4274A is that it can display the impedance as a magnitude and angle. This gives a good indication of how hard it is to do the measurement.

I have some low value resistors, single digit milliohms, that I could place in series with the film capacitor to check the calibration.

I also have a VNA and perform an impedance sweep.

Let me play in the lab and will report back with some measurements.

Regards,

Jay_Diddy_B

[Jay_Diddy_B]

Hi Electrician and the group,

I believe that this is the chart you are looking for:



This from the manual, available here:

https://www.doe.carleton.ca/~nagui/labequip/lcr/4274A_Operating%20Manual.pdf

Regards,
Jay_Diddy_B

[Jay_Diddy_B]

Hi group,

Here are a few tests made with the HP 4274A. This is a different capacitor than I used in the previous posting.

Packaging



The decoder



The test sample



A test sample was prepared with a 5m resistor soldered to one of the leads. The lead behind the capacitor will be cut in later tests.

Baseline Measurement



With the 5m resistor in series




And then two capacitors in series



This is a very good test, because the capacitance should be half and the ESR double. These capacitors when measured individually were about 17m each.

I had to the open and short cal several times, the instrument was warming up.

I believe that these are pretty good results.

Regards,

Jay_Diddy_B

[tigr]

Hi Jay_Diddy_B
Thanks for the table. It explains the whole point of the problem.
The main problem with all ESR meters is the ability to separate the active and reactive components from the capacitor.This depends on many factors,and primarily on the measurement method.In practice,most amateur circuits are not situable for measuring the ESR the film capacitors.After many experiments_the scheme of the universal ESR meter  was chosen for practically all types of capacitors.
I want to tell you that the main purpose of my work on the measurement of ESR is the ability to measure with maximum accuracy the ESR of the capacitors in the circuit and the analysis of the circuit in which the capacitor is located. The second task is to analyze the effect of other elements in the capacitor circuit. I'm glad to hear useful advice and thoughts from different specialists.

[Jay_Diddy_B]

Tigr and the group,

By far the best results I have obtained have been with a VNA (HP 3577A) and a homemade transconductance amplifier.

The setup is calibrated with a 1 resistor. This means all the amplitude reading are in dB .

Here is an electrolytic capacitor 100kHz to 1MHz



The marker is -17dB = 140m . You can see the ESL causing the impedance to rise at higher frequencies.


Here is a ceramic capacitor 100 kHz to 10MHz



The marker is at -44dB = 6.3m

Regards,

Jay_Diddy_B

[The Electrician]

Jay_Diddy_B,

   These measurements are what I would expect with a polyester cap.

   The older measurements where you were getting 1 milliohm values was with a metalized polypropylene if I remember correctly.  Do you still have that capacitor so the result could be rechecked?

[The Electrician]

I've set up an impedance analyzer (Hioki IM3570) to perform a frequency sweep from 100 Hz to 1 MHz, displaying impedance magnitude (|Z|) in green and ESR (Rs) in yelllow.  The first sweep is of a precision .02 ohm resistor; the top of the screen is at .1 ohms and the bottom at .001 ohms:



Both |Z| and ESR are right on .02 ohms until at the high frequency end of the sweep |Z| rises do to the inductance of the resistor leads.

Here's a sweep of a really good polypropylene 5 uF capacitor, using the ordinary settings for the analyzer; the top of the screen is changed to 1000 ohms.  Notice that the ESR curve at the low frequency end of the sweep is noisy; the ratio of |Z| to ESR is 4 orders of magnitude, making this a difficult measurement:



By using some averaging, the ESR curve is cleaned up at the low frequency end:



Here is the same capacitor, but with the top of the screen changed to .1 ohms:



Notice that there's a downward jump in the ESR curve (yellow).  This is due to the range change that occurs during the sweep; the instrument needs to go in for calibration.  The question is, which part of the curve is accurate, the left side before the jump or the right side.  I measured this same capacitor on a Wayne-Kerr analyzer and determined that the right side is accurate.

The lowest ESR is around 3 milliohms at around 3 KHz.  I get the same values on other analyzers.

I would like to mail this capacitor to you, Jay_Diddy_B.  If you're interested PM me a good mailing address.

Just for reference, here's a sweep of a Polyester (MKT) capacitor.  The lowest ESR is nowhere near .001 ohms.  It has been my experience that Polypropylene caps have the lowest ESR, much lower than Polyester caps:

[tigr]

Hi,

I have also done some testing with a film capacitor. I do not have the 'official' 6.8uF film capacitor. I used a 4.7uF 250V Metalized polypropylene film capacitor.

Here is a picture of the capacitor used for test:



This is has an extremely low ESR. I first measured the capacitor using my HP4274A LCR meter with a HP 16047A test fixture. Here are the measurements:



Note: That the device impedance is very reactive, >89.6 degrees at 100 kHz

Here a couple of sample pictures of the measurements:







When I connected the capacitor to the ESR adapter using the 4W configuration, I zeroed the meter using the relative feature on  the DMM.
I measured 5.355mV which corresponds to 53 m Ohms.



This is 0.26% of the ESR meter adapters full scale, and significantly lower than the impedance of the capacitor 340.7 m Ohms.

It would be interesting to see what your device will show if you connect these three capacitors in parallel.

[tigr]

Good film capacitors for testing ESR meters .
http://www.mouser.com/ds/2/427/mkp1848dcl-35623.pdf
It is possible to significantly increase the accuracy of the measurement of ESR in film capacitors by expanding the dynamic range of the measurement. For this we connect the meter to a voltmeter. The measurement takes place in millivolts. The value of 550 mV corresponds to zero ESR.In our case, the full scale is 550 mV. The values after calibration can be recorded, they are easily remembered with time.

[tigr]

Using an extended measuring range, we obtain an in-circuit circuit analyzer,capable of distinguishing low shunt resistance from abnormally LOW ESR, short circuit and ESR parallel connection of capacitors.
Motherboard.Capacitor 3300uF/6,3v.

[Jay_Diddy_B]

Tigr,

These images that show with the Amprobe DMM, what circuit is being used to drive the meter?

Regards,

Jay_Diddy_B

[tigr]

ESR meter 200kHz.

[tigr]

With this device it is also possible to check small coils for winding breakage. If the coil is OK, then the device can approximately estimate its inductive resistance. When the coil breaks, the device will show the maximum resistance.
All this can be done without removing the coil.

[Jay_Diddy_B]

Hi Tigr and the group,

I have done a little simulation on the circuit that is proposed by Tigr. There is no model in LTspice for the MAX253, so it was replaced by a simplification. I am stepping the value of the resistor through a series of resistors. I am plotting the output voltage versus Resr. This is gives you an indication of the meter reading (deflection).

LTspice model



LTspice modelling Results




In doing the analysis, I released that the MAX253 can be replaced by the 555 timer:



Results 555 timer version



Observation

The meter deflection is very similar to the 5 transistor ESR meter that I shared in this thread:

https://www.eevblog.com/forum/projects/5-transistor-esr-meter-design/msg171364/#msg171364

This is the results from modeling the 5 transistor circuit:





I have attached the LTspice models for those playing along at home.

Regards,

Jay_Diddy_B

[tigr]

Hi Jay_Diddy_B,
Your scale is 0.1-10 Ohm. My scale is 0.001-30 Ohm.
 Your meter will show that the electrolytic capacitor 1uF/450v is a bad capacitor.

[tigr]

For example.New capacitors.
Nichicon 1uF/450v.ESR-11Ohm.
1uF/450v.ESR-19,6Ohm.

[ocw]

You can increase many ESR meter's resolution by using the Q that it measures to then compute the ESR.  An example of that is below:

ECW-F4105HL  1uF 400V #2
   Measured capacitance:  1.0140 uF
   Measured ESR:  0.00 ohms
   Measured Q:  1832
   Calculated ESR:  0.0008568 ohms  (Xc/Q)

All of the above measurements were taken at 100 kHz.  Additional examples of this are shown on the attachment.

[tigr]

Hi ocw.
Thanks.
Tweezers MAX253 350kHz for in-circuit detection of faulty capacitors.
The full scale is 1000mV.
A good capacitor.

[The Electrician]

You can increase many ESR meter's resolution by using the Q that it measures to then compute the ESR.  An example of that is below:

ECW-F4105HL  1uF 400V #2
   Measured capacitance:  1.0140 uF
   Measured ESR:  0.00 ohms
   Measured Q:  1832
   Calculated ESR:  0.0008568 ohms  (Xc/Q)

All of the above measurements were taken at 100 kHz.  Additional examples of this are shown on the attachment.

Be aware that just because the meter gives you a number doesn't mean that it's accurate.

Here's the Panasonic info for the ECW-F4105HL: https://industrial.panasonic.com/ww/products/capacitors/film-capacitors/film-cap-electroequip/ecwfl/ECWF4105HL

On that page we see a link to the catalog page: https://industrial.panasonic.com/cdbs/www-data/pdf/RDI0000/ABD0000C245.pdf

The catalog page doesn't show the ESR directly, but it does give a maximum spec for the dissipation factor Tan delta.

On the first page linked above we see a link to "Temperature characteristics, permissible current and others": https://industrial.panasonic.com/cdbs/www-data/pdf/RDI0000/ABD0000COS29.pdf

That page doesn't list ESR directly, but there is a graph that gives us an idea what the ESR for a 1 uF cap might be:



The sharp bottom of the impedance curves is the ESR at the particular frequency.  Imagine what the curve would look like for a 1 uF cap.  It would be a little to the left of the .47 uF curve, and the ESR would be perhaps 20 milliohms.  It's very unlikely that the ESR of an ECW-F4105HL 1 uF cap would be less than 1 milliohm.

[ocw]

Be aware that just because the meter gives you a number doesn't mean that it's accurate.

That's true particularly when you are getting to the extremes of measurement.  However, my measurements of the other capacitors are not far off the measurements of the same capacitors by Jay_Diddy_B and his HP meter.  That is why I included those model capacitors.

As far as the ECW-F4105HL goes, Panasonic's information is generic, not for a particular capacitance value.  They say a DF of 0.05% at 1 kHz (Q = 2000) and a DF of 0.20% at 10 kHz (Q = 500).  I would assume that those are worst case figures for the capacitor value having the worst DF.  Given the other measurements, those for the ECW-F4105HL don't look too far out of line.  My calculated values might not be precisely accurate, but should be accurate as far as the ESR being definitely lower than the other models tested.

I also tested their ESR from 10 kHz to 100 MHz (I could have gone higher) using other equipment but didn't include that data since it didn't have much better ESR resolution.

[ocw]

I forgot to add, my attachment shows the ESR of one of those ECW-F4105HL capacitors by itself and then with a precision 0.2003 ohm resistor in series with it.

[tigr]

The main problem with all ESR meters is the ability to separate the active and reactive components from the capacitor.This depends on many factors,and primarily on the measurement method.In practice,most amateur circuits are not situable for measuring the ESR the film capacitors.

I will demonstrate using an example of a super-quality polystyrene capacitor 0.5 uF 250V 1%.