Coil tester SRF/Inductance/distributed capacitance

[dbanner]

I made this little circuit today and I have to say that it's a very simple and accurate way of determining coil parameters.
I even used the BASIC program running in IBMpcBASIC emulator. The program is working good too. I like that a value for stray inductance and capacitance can be included in the program, so results are more accurate. Very good.
The circuit works very well.

Signal generator and oscilloscope not required. Only a meter which can measure frequency.
https://www.robkalmeijer.nl/techniek...e48/index.html

[green]

I made this little circuit today and I have to say that it's a very simple and accurate way of determining coil parameters.
I even used the BASIC program running in IBMpcBASIC emulator. The program is working good too. I like that a value for stray inductance and capacitance can be included in the program, so results are more accurate. Very good.
The circuit works very well.

Signal generator and oscilloscope not required. Only a meter which can measure frequency.
https://www.robkalmeijer.nl/techniek...e48/index.html

Tried the circuit in spice. Some error, maybe due to using J113 instead of 2N5245?

[Skippy]

Looks like it's working OK to me. What do you think is wrong?
It would be easier to interpret if both circuits used the same L C R for the coil, maybe fine tune the parameters so F_res was exactly 1 MHz.

[dbanner]

The BASIC program is deductive, meaning that two measurements have to be taken. First with 150pF in the tank, the second measurement with the capacitor switched out. The results are calculated algebraicallly with two resonant frequencies.

This circuit is used to measure an unknown inductor based on tank resonance. There is little that can go wrong.
I used NTE133 general purpose n-channel jfet. Changed 470 ohm to 1k. Seems to oscillate better. Don't know what will be the upper limit of these transistors, but I'm only interested in measuring PI coils in the range of a few hundred uH.

[green]

Looks like it's working OK to me. What do you think is wrong?
It would be easier to interpret if both circuits used the same L C R for the coil, maybe fine tune the parameters so F_res was exactly 1 MHz.

Tried your suggestion making resonance 1MHz. Right circuit looks good. Coupled circuit has error. Don't know what I did wrong yet. Thanks

Does anyone now how to display frequency with spice?

Tried again with 300uH, got same as yesterday. Tried with 338uH(1MHz), wasn't paying attention earlier. Cycles matched at 20us difference so right circuit looked good, off at 10us difference. Left circuit closer to calculation with spice.

[green]

The BASIC program is deductive, meaning that two measurements have to be taken. First with 150pF in the tank, the second measurement with the capacitor switched out. The results are calculated algebraicallly with two resonant frequencies.

This circuit is used to measure an unknown inductor based on tank resonance. There is little that can go wrong.
I used NTE133 general purpose n-channel jfet. Changed 470 ohm to 1k. Seems to oscillate better. Don't know what will be the upper limit of these transistors, but I'm only interested in measuring PI coils in the range of a few hundred uH.

Was using circuit to measure SRF. The circuit appears to have some error with spice. Have you checked the accuracy of your circuit? Maybe the circuit is accurate enough.

[Skippy]

I'm puzzled by your very slow oscillation decay in the left-hand circuit. Surely, with 970 Ohms across the coil, it's heavily damped, and the ringing will be gone in a few cycles ?

[green]

I'm puzzled by your very slow oscillation decay in the left-hand circuit. Surely, with 970 Ohms across the coil, it's heavily damped, and the ringing will be gone in a few cycles ?

Test coil is the Rx coil on the left. Tx coil excites Rx(test)coil.

Tx coil critical damped.

[Skippy]

OK, I assumed L1 in one circuit was also L1 in the other circuit. But it's actually L3. So you're comparing the self-resonant frequency of the RX coil, an making the assumption that the weak coupling between the heavily-damped TX coil and the RX will not affect the damping of the RX .

[green]

OK, I assumed L1 in one circuit was also L1 in the other circuit. But it's actually L3. So you're comparing the self-resonant frequency of the RX coil, an making the assumption that the weak coupling between the heavily-damped TX coil and the RX will not affect the damping of the RX .

Yes. Can change coupling in spice with little change in resonance if coupling is kept low. Measures closer to calculated than other circuit with spice. Wondering if anyone has measured a standard with either circuit to check accuracy.

[dbanner]

The accuracy of the circuit is quite good. The only thing that affects the measurements are stray capacitance and Inductance of the rig. This can be factored out by entering the values in the program. There is a procedure for calculating the values of stray Inductance. Stray capacitance is negligible if care is taken in construction. I left out the switch for this reason, I just put a jumper into a 2 pin socket soldered to the PCB.

I measured an inductor with color code red red brown (220uH). It measured 220.8uH with distributed capacitance of 15pF.

I don't have an Inductance meter, so can't compare results. But tank resonance is most accurate means of measuring air core coil inductance, provided care is taken to keep stray parasitics to a minimum.
Taking the output from the common source connection prevents loading the tank circuit with any stray capacitance from the probe or leads of the frequency counter.

Overall, it doesn't matter too much the accuracy of the Inductance measurements so long as they are within acceptable tolerances. What is important is the measurements of the distributed capacitance, which ought to be accurate to within a few picofarads.(The premise for construction of fast coils is for low distributed capacitance)

[green]

Does anyone now how to display frequency with spice? No replies


Would like to thank Skippy again for suggesting, calculate L and C to be resonant at a frequency that is easy to see if cycle time repeats(50kHz, 200kHz, 1MHz, one digit +0's). Was playing some more and it's a lot easier to see if simulation matches calculation.

[Polymer]

The circuit dbanner mentions here works excellently. There is a cool excel spreadsheet for it too, on Geotech1 !!!

It is the only SRF measurement I use. I test SRF of my PI cables with coils after use to see if they have sucked water.

[green]

Trying to understand how the circuit can work. Somewhere I'm missing something. Did another simulation. Coil by it self has a SRF of 1MHz. Both testers measure resonant frequency only. When I calculate inductance, I add a 10n or 100n capacitor in parallel with the coil. A lot higher than coil capacitance. Measure resonance, then calculate inductance using capacitance and resonance. The other tester does the same but uses 150p and resonance to calculate inductance. When I use 150p inductance calculates over 500uH. When I use 10nf, inductance calculates 341uH close to the coil inductance of 338uH. Don't see how to get capacitance or inductance for coil only just knowing resonance. Maybe someone could explain what I'm missing.

[6666]

Dbanner did you run it with 18 v ?