Bored yet? I've always found electrical fundamentals fascinating, and the better you understand them, the better you'll understand the more complex concepts. OTOH, maybe I like fundamentals because that's about all my pea brain can grasp.
This time I wound the same 10 ohm resistor, but as a twisted pair. That is, I measured out the 50" of wire, doubled it over, then chucked up the end in the electric drill and gave it a two-turns-per-inch twist. That twisted pair was then wound over the mica card.
Performance was much better, though I think the actual model might be more complex due to the higher internal capacitance. The value was about 10.5 ohms at 1MHz, with a phase shift of 14 degrees. That works out to an Ls of 400 nanohenries. As important, the Q is about 0.25. I wouldn't hesitate to use this in any crossover, signal divider, level control, or anywhere but the feedback network of a high speed op-amp.
There are a variety of other published winding patterns, but this one is a good trade-off between performance and easy.
This time I wound the same 10 ohm resistor, but as a twisted pair. That is, I measured out the 50" of wire, doubled it over, then chucked up the end in the electric drill and gave it a two-turns-per-inch twist. That twisted pair was then wound over the mica card.
Performance was much better, though I think the actual model might be more complex due to the higher internal capacitance. The value was about 10.5 ohms at 1MHz, with a phase shift of 14 degrees. That works out to an Ls of 400 nanohenries. As important, the Q is about 0.25. I wouldn't hesitate to use this in any crossover, signal divider, level control, or anywhere but the feedback network of a high speed op-amp.
There are a variety of other published winding patterns, but this one is a good trade-off between performance and easy.
Nope, an inductor always has resistance, so it's also an inductive resistor. There will be unavoidable capacitance between the terminals and the windings too. The goal for a resistor is to keep both the stray inductance and stray capacitance as low as possible. That's the downfall of wirewound resistors. If they'll handle the power, nothing beats the bulk metal foil resistors like Vishay and such. For an inductor, the goal is usually minimum resistance, to keep the Q high, and minimum capacitance, to keep the self resonant frequency high.
Forget inductive issues, they are so low any way and almost identical between cheap and 'audiophile' resistors. 'Identical' in the way they fall under insignificance threshold for the bandwidth of a speaker.The stuff we hear is PPM cycle and material used. It is economically tempting to think that a sand cast one and a Litz silver wire on a wooden tube one should sound the same because the inductance is about the same. Alas, the assumption is busted when we alternatively install them before a tweeter and listen.
I have done so.
I have done so.
I would think about measuring some inductors, with known (or at least labeled) values that are in the same inductance range as you measured for the power resistors.
I don't know if this would be helpful or not, but, it might also be interesting, for comparison, to then measure a known inductor with one of your previously-measured power resistors in parallel (and also try it with the R in series), and also with a non-inductive resistor in parallel and in series. And, of course, I'd probably also try to measure the inductance of a non-inductive resistor by itself, if your meter is suspect.
I would also think about measuring the inductance of the power resistors is some other way, so you could compare the results to your meter readings. You should be able to get some estimate with a square wave generator and an oscilloscope. And I'm sure that Conrad will be able to give you a much better method, involving a simple bridge setup.
(I think you might also be able to do it by making a parallel (R)LC circuit, with a known C, and sweep the frequency of a sine wave applied across it, and look for a resonant peak, and then calculate the L.)
I don't know if this would be helpful or not, but, it might also be interesting, for comparison, to then measure a known inductor with one of your previously-measured power resistors in parallel (and also try it with the R in series), and also with a non-inductive resistor in parallel and in series. And, of course, I'd probably also try to measure the inductance of a non-inductive resistor by itself, if your meter is suspect.
I would also think about measuring the inductance of the power resistors is some other way, so you could compare the results to your meter readings. You should be able to get some estimate with a square wave generator and an oscilloscope. And I'm sure that Conrad will be able to give you a much better method, involving a simple bridge setup.
(I think you might also be able to do it by making a parallel (R)LC circuit, with a known C, and sweep the frequency of a sine wave applied across it, and look for a resonant peak, and then calculate the L.)
It wouldn't surprise me if a general purpose inductance meter couldn't measure an extremely low value, extremely low Q inductor. Heck, my Digibridge won't do it. After all, it looks pretty much like a resistor ;-) To see what's going on you need either a very sophisticated device like a network analyzer, or you have to make a conventional measurement at a high frequency, say 1MHz or higher. One reason I wound 10 ohm resistors was because I trust my vector impedance meter more there, than at the bottom of the scale near 1 ohm. Also, any small inductance at the 1 ohm level is almost certainly insignificant, i.e., it's similar to the lead inductance. If anybody's really interested, we can probably come up with a bridge for this sort of measurement. I don't think it can routinely be done by making a resonant tank and sweeping because the Q is so low that the "peak" will be impossible to see. For my first resistor, the inductance was high enough that the Q was around 3 at 3MHz, and that should be easy to see, though not locate with great precision. As the resistors get better, you have to go to frequencies that I'm not sure are relevant to audio, to see anything useful.
Salas- I really did try to hear the difference between silver and brass, but it was for naught, at least for a low level interconnect. Once I removed all dielectric materials so the only variable was the metal suspended in air, they sounded exactly the same to me. No doubt people will tell me something similar about my resistors, but I still believe that if there's a difference in resistor materials, Manganin might well be at the top of the list. A bit of trivia- if one is concerned about PPM level effects, Manganin, in spite of it's normal stability, changes resistance under extreme pressure. Apparently it's used as a sensor for measuring the pressure levels in explosions. So, if you're playing the music really really loud...
Salas- I really did try to hear the difference between silver and brass, but it was for naught, at least for a low level interconnect. Once I removed all dielectric materials so the only variable was the metal suspended in air, they sounded exactly the same to me. No doubt people will tell me something similar about my resistors, but I still believe that if there's a difference in resistor materials, Manganin might well be at the top of the list. A bit of trivia- if one is concerned about PPM level effects, Manganin, in spite of it's normal stability, changes resistance under extreme pressure. Apparently it's used as a sensor for measuring the pressure levels in explosions. So, if you're playing the music really really loud...
Suppose your meter only took a magnitude measurement at one frequency. This would not be enough to determine magnitude and phase, it would need two measurements. Going a little further, suppose it assumed it was measuring an inductor, because you set it to think it was an inductor, and took the measurement at about 8 kHz. Then it would perform the following simple calculation to back out the inductance value.
20 ohm / (2 x 3.14159 x 8000 Hz) = 0.398 mH
12.5 ohm / (2 x 3.14159 x 8000 Hz) = 0.249 mH
8.2 ohm / (2 x 3.14159 x 8000 Hz) = 163 uH
5.1 ohm / (2 x 3.14159 x 8000 Hz) = 101 uH
3.3 ohm / (2 x 3.14159 x 8000 Hz) = 66 uH
Maybe that is the simple explanation for the results. You were using the meter incorrectly and it was giving you a bogus result.
20 ohm / (2 x 3.14159 x 8000 Hz) = 0.398 mH
12.5 ohm / (2 x 3.14159 x 8000 Hz) = 0.249 mH
8.2 ohm / (2 x 3.14159 x 8000 Hz) = 163 uH
5.1 ohm / (2 x 3.14159 x 8000 Hz) = 101 uH
3.3 ohm / (2 x 3.14159 x 8000 Hz) = 66 uH
Maybe that is the simple explanation for the results. You were using the meter incorrectly and it was giving you a bogus result.
I have a Mastech inductance/capacitance meter - let's see what it says:
0.006mH: With supplied 7" leads shorted
0.378mH Ohmite wirewound 20R (2 measure the same)
0.049mH Dayton white block. 2.4R 10W
0.158mH No Name white block 8R 20W (2 the same)
FWIW, it seems to be spot on with store bought inductors.
I thought the meter worked at 400Hz, will have to double check it.
0.006mH: With supplied 7" leads shorted
0.378mH Ohmite wirewound 20R (2 measure the same)
0.049mH Dayton white block. 2.4R 10W
0.158mH No Name white block 8R 20W (2 the same)
FWIW, it seems to be spot on with store bought inductors.
I thought the meter worked at 400Hz, will have to double check it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.