Inductor meter

"Jon Slaughter" schreef in bericht news:OPeIj.30975$ snipped-for-privacy@newssvr25.news.prodigy.net...

I build several inductormeters over time but no one measures inductances reliably. Last time I got two types of inductors which were supposed to be

petrus bitbyter

Also the DC current you've putting through the inductor, although this is usually much more significant for power supply designers (who think in pulses) than RF designers (who think in sine waves).

For small inductors and capacitors, I like to use the coax trick: measure the quarter-wave resonant frequency (f0) of a random piece of RG-58 (say 1 m long), and again with the inductor or capacitor soldered across the end (f1). Then calculate the component value as

L = -[Z0/(2*pi*f1)]tan(pi*f1/(2*f0)) or

C = [1/(2*pi*f1*Z0)]cot(pi*f1/(2*f0))

Works great. With a bit more work you can take the fringing fields at the ends of the coax into account as well.

Cheers,

Phil Hobbs

There are lots of ways to measure capacitance quite accurately. Problem is, most younger folks don't know how to work things like Smith charts anymore. Once I solved a matching problem at a client in about 15 minutes, thanks to Smith. No computer involved but I did have to buy new compasses because TSA considers those weapons (Walmart has plastic ones for around $1). The engineers had spent hours on it with MatLab and so on and couldn't reach a good compromise. Then they asked me how I dunnit and since I kind of knew this was going to happen I had copied a Smith chart onto overhead. Luckily they still had such a projector. Turned out none of the guys knew how to work the chart although some remembered it from their days at the university.

I've only used a Smith chart a few times, having never actually taken a circuits course myself. I did use a slotted line once in a physics lab class--I remember wondering how you could cut a slot in a waveguide like that and not screw it up. (A plot of the field pattern and a little thought brought enlightenment.)

Cheers,

Phil Hobbs

The purpose of the Smith chart is to allow you to not have to remember or manipulate the fancy equation for "input impedance of a transmission line with a given termination." :-)

I have more significant digits on my hands, though. Smith is still unbeatable for discovering solutions, and for finding the limits of different matching techniques.

Cheers,

Phil Hobbs

No, there's phase shift in there. You adjust so the voltage across both the L and the R are the same. Mag(Vl) + Mag(Vr) > Mag(Vsrc)

Mark Zenier snipped-for-privacy@eskimo.com Googleproofaddress(account:mzenier provider:eskimo domain:com)

Why not use your vector network analyzer? That is what I use.

On a side note, the free AADE filter design program is very good.

Good catch, thanks Mark. Looking at it a bit more closely, I believe the correct voltages are that Mag(Vl)=Mag(Vr)=.707*Vsrc.

---Joel

It's about the only way I could measure the winding inductance of some transformers with the Heathkit grade test equipment I've got. Boy, they can be real squirrely on a bridge. My guess was that the frequency was near the self resonance of one of the other (open) windings.

Mark Zenier snipped-for-privacy@eskimo.com Googleproofaddress(account:mzenier provider:eskimo domain:com)

I have just received a volume I+II UHF Unterlage. It contains a

Interesting reading for a 1980 book. Thanks for the pointer.

Joop

Beyond 250MHz separate probes are the only way to go. You can always upgrade when the need arises, for example should you ever feel the urge to fine-tune 2.45GHz ISM stuff. Nowadays we can buy 65GHz ft transistors for a buck. That was unheard of in the days these books were written.

Yes. To be honest I learned more from this and other ham radio books than at my university. At least when it comes to the know-how I need for my job these days.

Karl (the author of UHF Unterlage) passed away in 2001, AFAIK he didn't even reach 70:

Apologies for using Google Groups to post.

I built one of these and it is very good:

The main thing it is lacking is a low-battery indicator. For really small L and C, a different (faster) oscillator might be a nice addition, and to keep the same software, the frequency could be scaled by 10x, using 10x smaller L and 10x smaller C and a fixed divider- by-10 counter before the PIC.

Of course a VNA would still be nice :-)

Chris