How to measure small inductance? About 0,5uH or 2,5uH?
I tried to measure these with DMM, it is missleading. One time it reads 2uH. The next measurement 4uH, never steady. Try other DMM, also not convincing. For mH values, they are OK.
I tried to measure these with DMM, it is missleading. One time it reads 2uH. The next measurement 4uH, never steady. Try other DMM, also not convincing. For mH values, they are OK.
You need a LCR meter. There is also a way to measure inductance with a simple oscillator circuit and a O'scope or a freq.counter and a calculator. 🙂
You need a frequency up around 100Khz and a test setup to null your lead inductance.
A handheld will probably not be of use.
A handheld will probably not be of use.
I've tried a couple of LCR (handheld, DJK 😀), but no use, not consistent readings.
Using CLIO it reads "out of range"
I've got audio oscilator up to 1Mhz. Also got scope. How to do it?
From equation R=2.pi.f.L, I will need about 16Mhz signal for 1uH to get 100ohm, I don't have that high oscilator.
Using CLIO it reads "out of range"
I've got audio oscilator up to 1Mhz. Also got scope. How to do it?
From equation R=2.pi.f.L, I will need about 16Mhz signal for 1uH to get 100ohm, I don't have that high oscilator.
Hi, XEAGLEKEEPER
Thanks for the link 😀 I will try that method.
The opamp CA3100T, can it be replaced by NE5532 or other common opamps?
Looking at the datasheet, CA3100T is 38Mhz opamp with -3dB at 110khz.
110khz knee frequency? Interesting, how will it sound for audio?
Thanks for the link 😀 I will try that method.
The opamp CA3100T, can it be replaced by NE5532 or other common opamps?
Looking at the datasheet, CA3100T is 38Mhz opamp with -3dB at 110khz.
110khz knee frequency? Interesting, how will it sound for audio?
probably just about any general purpose op-amp will do.Instead of using a freq counter you can use a scope F = 1/T and T=1/F
maybe a bat could here it. LOL😀
maybe a bat could here it. LOL😀
you could build a wheatstone bridge circuit, and find where the inductor nulls a potentiometer at a known freq. you then measure the resistance of the potentiometer, and look up in a chart (or use the calculation) to find out what value of inductance has a reactance equal to the value on the potentiometer. so for instance if the inductor nulls a potentiometer at 47 ohms, you use the formula L= XL/2*pi*f so at 100khz it would be 47/628000, which would be 75uH. so you will need 1Mhz to bring XL up into the 10-100 ohm range for coils around 1-10 uH. it's best to use a 10 turn pot for your nulling element, as the nulls will probably be quite sharp.
When I have no suitable inductancemeter available, I use the resonance method, based on the Thomson formula:
http://en.wikipedia.org/wiki/Resonant_circuit
You need a good reference capacitor, say 100nF 0.5% or 1%, polystyrene or polypropylene, you make a tank circuit with the unknown coil and you feed the circuit via a resistor of some Kohm.
You connect a scope across the circuit and you sweep your generator until you find a signal peak; you then calculate the inductance using the reverse Thomson formula: 1/C*(6.28*F)².
With 100nF and 1µH, the frequency is around 500KHz.
The signal generator has to be preferably sinusoidal, otherwise you will get parasitic peaks at harmonic frequencies. Even if it is the case you can still get the correct value by selecting the highest and strongest peak.
The method is very accurate and only depends on the accuracy of the reference cap. In addition you can select a cap value giving a resonance frequency near the intended operating frequency of the inductance, giving an even better accuracy.
LV
http://en.wikipedia.org/wiki/Resonant_circuit
You need a good reference capacitor, say 100nF 0.5% or 1%, polystyrene or polypropylene, you make a tank circuit with the unknown coil and you feed the circuit via a resistor of some Kohm.
You connect a scope across the circuit and you sweep your generator until you find a signal peak; you then calculate the inductance using the reverse Thomson formula: 1/C*(6.28*F)².
With 100nF and 1µH, the frequency is around 500KHz.
The signal generator has to be preferably sinusoidal, otherwise you will get parasitic peaks at harmonic frequencies. Even if it is the case you can still get the correct value by selecting the highest and strongest peak.
The method is very accurate and only depends on the accuracy of the reference cap. In addition you can select a cap value giving a resonance frequency near the intended operating frequency of the inductance, giving an even better accuracy.
LV
bear in mind that the actual inductance will be influenced by nearby metalwork (including PCB foil) etc. - even the proximity of your body when measuring.
It's usual to provide some sort of variable adjustment, so that the circuit can be tuned under actual operating conditions.
It's usual to provide some sort of variable adjustment, so that the circuit can be tuned under actual operating conditions.
lumanauw said:
You say sometimes 2, sometimes 4, but not 2.00 or 4.00. Does this mean that your meter's resolution is 1 or 2 uH? If so, it is not possible to measure accurately what you wish.
djk said:
Yup, agreed on both counts. As for nulling lead inductance, if he's using a two wire set, best bet is a low z coaxial lead set, and best bet there is probably a dual braid one. If using a four wire set, I would hope the meter has an autozero function.
Cheers, John
dnsey said:
The human body will not affect the inductance unless there are steel head plates or replacement joints involved. However, if the test setup cannot distinguish confounding capacitance, then what you say is true.
As to proximity of other metals, that may be dead on, depending on many factors..best bet is to avoid interaction as you say.
Cheers, John
"If using a four wire set, I would hope the meter has an autozero function."
I use a four-wire HP every day with a null function.
The method Elvee outlined (a tank circuit) will be of the most use in this case.
I use a four-wire HP every day with a null function.
The method Elvee outlined (a tank circuit) will be of the most use in this case.
djk said:
Probably right, unless Elvee can get a hold of a good meter.
I use a 4263A, and have been able to resolve repeatedly to less than a nanohenry. But that would be quite difficult without the short correction function, especially with the 16089A lead set..
Cheers, John
I use an Agilent every day to tune a 14nH inductor to better than ±.5nH, the older HP is better for trouble-shooting.
The computer controlled Agilent has a cal routine for stray reactance that takes two minutes, and must be done every time I mount a unit on the test fixture. The HP has a delta funtion that only takes seconds to use.
The computer controlled Agilent has a cal routine for stray reactance that takes two minutes, and must be done every time I mount a unit on the test fixture. The HP has a delta funtion that only takes seconds to use.
What oscilator do you use? I only got variable, 1Mhz max audio oscilator.
To measure uH or even nH, would we need about 20Mhz oscilator?
To measure uH or even nH, would we need about 20Mhz oscilator?
djk said:
I love the HP set, it worked so much better when I actually read the instruction manual..(yes, yes, I admit it...I read the manual)
How does the agilent remove the stray? Don't you have to short the dut first at the probe tips to zero the loop?
Cheers, John
"..(yes, yes, I admit it...I read the manual)"
It's 480 pages for the Agilent, I read about a third of them that seemed pertinent. The HP seemed intuitive to me, I get along fine with it without the manual.
"How does the agilent remove the stray? Don't you have to short the dut first at the probe tips to zero the loop?"
I short out the DUT inside the test fixture with a relay that is part of the DUT.
It's 480 pages for the Agilent, I read about a third of them that seemed pertinent. The HP seemed intuitive to me, I get along fine with it without the manual.
"How does the agilent remove the stray? Don't you have to short the dut first at the probe tips to zero the loop?"
I short out the DUT inside the test fixture with a relay that is part of the DUT.
djk said:
Ah, thanks. So the relay contacts provide a reference.
When I correct the HP, I attach the dut to the clips, then short at the clips without disturbing their geometry. Then once correction is finished, gently remove the short. Woiks quite well.
480 pages....wimp...😉
Cheers, John
- Status
- Not open for further replies.