Re: AC Resistance Measurements
Subject: Re: AC Resistance Measurements
From: Terry Fritz <twftesla-at-uswest-dot-net>
Date: Mon, 19 Jul 1999 19:43:46 -0600
Hi Jim and All,
I got my new meter today and Tesla coil AC resistance measurements are well
within it's abilities. It can measure about two orders of magnitude higher
than is needed. I could use even more digits if I used the fancy filters
and measurement averaging but that would require that I read the manual :-)
The DC resistance of my coil is 70.085 ohms. It has 1000 1/2 turns on a
10.25 inch diameter tube 30.0 inches long. The wire gage is #24 which is
0.0201 inches in diameter so the length is SQRT((1000.5 x pi x 10.26)^2 +
30^2) / 12 = 2687.40 feet. From the old ARRL chart the resistance should
be 26.17 ohms / 1000 feet or 70.329 ohms. Pretty darn close!
The AC resistance is 540.2 ohms. Subtracting the 70.33 ohms we get 469.9
ohms of pure AC skin effect (I guess) resistance.
The primary circuit with no secondary came in at 0.548 ohms.
The secondary resistance is higher than the 300 ohms I thought it was but
this measurement is far more accurate than looking at ringdown times. Of
course, the primary resistance does not include the gap. I think the
primary resistance is about 3 ohms so 2.5 ohms of that would be purely in
the gap. If the primary dissipates 340 watts, then the gap is burning ~280
watts while the primary cap (0.030 ohms) is burning 3.4 watts (my cap
calculations has that at 8.3 watts??). The primary coil and interconnect
wires are burning 56 watts.
I made the measurements at resonance so the coil and cap in both cases
should cancel each other just leaving the real resistance. I measured
across a 100.00 ohm test resistor in series with a low impedance signal
generator. The load of the meter and the "details" where modeled and
checked with MicroSim but nothing too dramatic there.
I am not sure knowing such things to high accuracy will change the universe
or anything but it is possible to measure...
For the record,
Lp = 121.2uH
Cp = 27.7nF
Ls = 75.4mH
Cs = 44.19pF
Fo = 86.9 kHz
At 06:16 AM 7/16/99 -0700, you wrote:
>> I just ordered up one of those fancy HP 34401A multimeters because "I
>> it" :-)) It will measure true AC RMS voltage up to 300kHz in the
>> 100.0000mV AC range (6.5 digits). If I use a resistor in series with a
>> primary or secondary coil, I may be able to put say 10mA of variable
>> frequency current through the coil. The meter should be able to read of
>> the voltage drop and give me the AC resistance. With 10mA and 0.1uV of
>> resolution, I should be able to read AC resistance down to 10 micro-ohms
>> normal TC frequencies. However, there will be radio interference and
>> resonances and such in the coil that would mess around with such a fine
>> measurement. I was thinking of putting some value cap across the coil to
>> swamp the RFI and lower the resonant frequencies down to where they would
>> not mess up the measurement. I was also thinking the same technique
>> be used to measure the series resistance of poly caps.
>I don't think you'll have any real problems with interference because the
>impedances are low. And, as for resonances, that's what you are trying to
>measure... Measure the voltage and current at a series of frequencies, etc.
>Of course, I think one of the new computerized antenna bridges, like the
>one from AEA, would actually be better for measuring RF impedances, except
>they don't go down low enough in frequency.
>> I think the AC impedances of the coils or caps will make such
>> impossible. However, someone may know all about this and could tell me
>> it could work or not. I would be interested in any hints anyone may have
>> on this.
>One standard technique is to combine your unknown device with a known
>impedance of the opposite sign, so that the impedance magnitude of your
>resistor and your unknown are comparable. Say you've got a coil with an
>impedance of 100K and you want to use a 1K measurement resistor... If you
>add a capacitor with a known impedance of 90K to the circuit, the net
>reactive impedance will be only 10K, so the voltage across the coil and
>resistor will be more similar.
>> If anyone has any other ideas for useful TC measurements this meter
>> do, let me know. I don't think I'll be measuring the output current of a
>> neon with it though :-))
>Figure out some way to make a computer controlled frequency generator so
>that you can make swept automated measurements (I assume the fancy HP has a
>RS232 or 488 interface?). Then, you'll have an impedance bridge.