EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Tommy1984 on September 10, 2017, 02:36:25 pm
-
Hello everyone,
I have LCR400 Voltcraft http://www.conrad.com/ce/en/product/1301292/Component-tester-digital-VOLTCRAFT-LCR-400-Calibrated-to-Manufacturer-standards-CAT-I-Display-counts-20000?ref=searchDetail (http://www.conrad.com/ce/en/product/1301292/Component-tester-digital-VOLTCRAFT-LCR-400-Calibrated-to-Manufacturer-standards-CAT-I-Display-counts-20000?ref=searchDetail)
User manual: http://www.produktinfo.conrad.com/datenblaetter/1300000-1399999/001301292-an-01-ml-VOLTCRAFT_LCR_400_KOMPONENTE_de_en_fr_nl.pdf (http://www.produktinfo.conrad.com/datenblaetter/1300000-1399999/001301292-an-01-ml-VOLTCRAFT_LCR_400_KOMPONENTE_de_en_fr_nl.pdf)
As it's mentioned in the user manual the LCR meter can measure impedance from 0.001uH to 20KH....
I have tried to measure many different inductors with many different values (from 0.03uH to 0.440uH), but, it shows fail values!
As the user manual says, the measuring frequency for the rang 20uH must be 100KHz (there's nothing about less than 20uH).
Anyway, I have tried many frequencies, but, without any luck! Even in series or parallel, got nothing right or even a little close to the real values which should be |O |O |O
Anyone has an idea what's going on or why I got that?
Any help or tip is a very appropriated!
Tommy
-
The possible frequencies depend on the range, not on the actual value. The smallest range seems to be 20 µH.
Measuring small inductance needs low inductance test wires - e.g. have the wires close together and a zero adjustment. Down to about 1 µH things should work - lower values can be tricky, as the test probes / wires can have an influence.
In principle the inductance is a property of the whole circuit. Due to magnetic coupling defining an inductance value for an isolated part is difficult. At least one has to compare shorted probe / dummy wire to the part.
-
The possible frequencies depend on the range, not on the actual value. The smallest range seems to be 20 µH.
Measuring small inductance needs low inductance test wires - e.g. have the wires close together and a zero adjustment. Down to about 1 µH things should work - lower values can be tricky, as the test probes / wires can have an influence.
In principle the inductance is a property of the whole circuit. Due to magnetic coupling defining an inductance value for an isolated part is difficult. At least one has to compare shorted probe / dummy wire to the part.
Thank you for your reply!
So you mean that the minimum resolution is 0.001 uH , but it's not the lowest possible value which can be measured?
And for the probes, I'm using the SMD-tweezers which came with it!
PS. I calibrate the meter after every measuring and adjust the meter to zero by closing the tweezers before the measuring!
Tommy
-
The SMD tweezers are relatively good for low inductance. However it still needs an zero adjustment - the old adjustment might be off from measuring with loose cables. For best accuracy one should even try to keep the cables in the same way and away from the DUT. Have you at least done the zero adjustment for the tweezers ?
There is a resolution of 1 nH - however this is due to using a 4.5 digit display / ADC and chances are that in the lowest range the errors are larger than in the other ranges. So I would not give much on the last digit. Specs are +- 5 nH (= 5 digits) plus +-0.5 % of the value. For me this sounds rather optimistic as it's the same fixed error part for all ranges. The lowest range measurement takes more time and thus might be noise / EMI sensitive. A first check would be if the zero value is at least stable, e.g. with small movements of the cable.
Anyway with some inductors (e.g. ferrite beads) the inductance is frequency dependent - so no simple value. Also parasitic capacitance of the cable can be a problem - though more with larger inductance.
-
The SMD tweezers are relatively good for low inductance. However it still needs an zero adjustment - the old adjustment might be off from measuring with loose cables. For best accuracy one should even try to keep the cables in the same way and away from the DUT. Have you at least done the zero adjustment for the tweezers ?
There is a resolution of 1 nH - however this is due to using a 4.5 digit display / ADC and chances are that in the lowest range the errors are larger than in the other ranges. So I would not give much on the last digit. Specs are +- 5 nH (= 5 digits) plus +-0.5 % of the value. For me this sounds rather optimistic as it's the same fixed error part for all ranges. The lowest range measurement takes more time and thus might be noise / EMI sensitive. A first check would be if the zero value is at least stable, e.g. with small movements of the cable.
Anyway with some inductors (e.g. ferrite beads) the inductance is frequency dependent - so no simple value. Also parasitic capacitance of the cable can be a problem - though more with larger inductance.
Thank you again!
Yes I did do the zero adjustment to the tweezers after the calibration.
I have nothing more to add, I guess I have to live with it! I just can't measure inductors which less than 1uH and I just wasted my money on something garbage. It's not cheap by the way, I paid about €270 for it, maybe I have to stop buying garbage from Conrad :-DD
-
I've never had any luck with measuring small inductors (less than 5uH) with any kind of LCR meter. They all seem to be dicky, I suspect due to the parasitic properties of the fixtures and/or internals. Sounds lame and old school but I've been using a few 1% silver mica caps and soldering the inductor across one. Couple a DDS to one leg via a 1M resistor and to the scope 10X probe with a 1pF cap and clip the grounds to the other leg. The resistor and cap don't load the resonant circuit heavily and keep the Q high so you can scan the generator for a peak while observing the scope. You have F from the DDS and C from the cap so it's easy to work out L. Measurements are repeatable and within 5% of the correct values.
1950s style but it works. I have measured down to 100nH or so no problems. And I'm testing at the correct frequency unlike the standard fixed frequencies of your average LCR
No good if you want in circuit by that's a whole different kettle of fish.
-
I just can't measure inductors which less than 1uH and I just wasted my money on something garbage. It's not cheap by the way, I paid about €270 for it, maybe I have to stop buying garbage from Conrad :-DD
Hello,
presumably is nothing wrong with your LCR400.
As example 1cm wire with diameter 0.5mm can have about 0,02 µH.
So it is a art do measure so small inductors.
Best regards
egonotto
-
I think that is not art. That just needs a proper measurement method. In this case that is a four wire measurement and a vectorvoltmeter in principle.
EDIT: Like the Chinese XJW01, at least.
-
I just can't measure inductors which less than 1uH and I just wasted my money on something garbage. It's not cheap by the way, I paid about €270 for it, maybe I have to stop buying garbage from Conrad :-DD
Hello,
presumably is nothing wrong with your LCR400.
As example 1cm wire with diameter 0.5mm can have about 0,02 µH.
So it is a art do measure so small inductors.
Best regards
egonotto
Well, thank you very much amigo! That helps me a lot, I'm not upset anymore :-DD :-DD :-DD
-
I think that is not art. That just needs a proper measurement method. In this case that is a four wire measurement and a vectorvoltmeter in principle.
EDIT: Like the Chinese XJW01, at least.
Thank you for your reply!
I used the SMD-tweezers probes which came with the LCR-meter and which I suppose it must be low inductance. Actually, I do zero adjustment before any measurement.
Tommy
-
Most of the two wire only LCR meter use some doubtful methods to measure the value, like using the component as a part of oscillator, or whatever.
A proper vector voltmeter is the way to go, if one needs precision result or to measure lower values. If I am not mistaken, the SMD tweezers are also four wire vector voltmeter.
With today's electronics, it is not a very difficult task to make a cheap vector voltmeter up to some hundreds kHz. With a bit of help of some processing grunt on today's microcontrollers, one just can leave out even the synchronous detector (which was always the key element of vector voltmeters) and just sample the I, V signals with ADC and then process the data via software. Even the source sine-wave signal can be very easily generated by the MCU.
So the only think you need for a <100kHz vectorvoltmeter, is a cheap MCU with ADC and a DAC, anda bunch of OPams for the output driver, I-V converter and a diffamp for the V-sense accross the unknown impedance, probably a bunch of MUXes for range switching too. (in fact, that is what it is inside the tweezers).
(http://www.hardware-pro.com/wp-content/uploads/2016/05/Smart-Tweezer-SCH-ST5-S-6-Hardware-Pro.png)