"Paul Burridge" wrote in message
...
Hi all,

I've got quite a huge stash of inductors in my parts bin. The colour
codes don't always seem to relate to the values I've been able to
measure, with my multi-function DVM, however, and I can't accurately
measure any inductor about 10uH. Is there a circuit anywhere that
would enable me to get a reasonably accurate idea of the values I've
got down to say 100nH or thereabouts?


"splat" test

V=L*dI/dt

charge up a large cap to some voltage V

short the cap thru the inductor, and measure the current with a scope
(digital is great, as long as the single-shot sample rate is high enough - I
did this once with an HP 56000 series scope, and got 1 data point - very
easy to fit a line to, very hard to fit the right line to!)

knowing dI/dt and V, calculate L

then, you can see saturation from where dI/dt increases rapidly (invaluable
for power electronics)

also, ensure that 0.5CV^2 0.5LI^2 at the current of interest - this
ensures that V is approximately constant

I have a couple of 33mF 35V caps in parallel, with 10 x 1R MRS25 resistors
paralleled as a current sensor. It takes about 45s to set up for a
measurement. I often use the shaft of a screwdriver to make the "splat" - a
nice hard material is good.

I have used this technique successfully to measure inductors ranging from
the tiny ( 500nH) to huge (3mH 2,000A chokes). Actually, big chokes tend
not to read true on LCR meters, as the magnetic material permeability is
often much higher at very low currents - especially true of iron powder.

if you dont have a digital scope, make a tiny, LF oscillator with a 555.
Drive a grunty FET, with a small (I use 0.1R - 10R) source resistor. Then,
you'll get a repetitive waveform, which works nicely on any analogue scope.
If Rsense is nice and low, use a 50R BNC cable to connect to scope, with a
50R terminator at the scope - lovely clean waveforms.

I built a little tester like this to test air gaps in planar cores for a 55W
smps that lives inside an LED video screen (actually, thousands of them live
inside). works a treat, and cost $5.

One of my techs once built a splat tester for big chokes - he had several
1200V 600A IGBT's in parallel running as a series pass regulator, with about
100mF of capacitance (charged to 700Vdc thru several lightbulbs) to control
V, even at currents on the order of 5,000A. A honking great LEM DCCT
measured the current, and the "switch" was purely mechanical - he used the
pole faces from an HV contactor, as the material is astonishingly arc
resistant, so no pitting/welding

On Mon, 15 Mar 2004 10:46:53 +0000, Paul Burridge
wrote:

Hi all,

I've got quite a huge stash of inductors in my parts bin. The colour
codes don't always seem to relate to the values I've been able to
measure, with my multi-function DVM, however, and I can't accurately
measure any inductor about 10uH. Is there a circuit anywhere that
would enable me to get a reasonably accurate idea of the values I've
got down to say 100nH or thereabouts?

Thanks,

p.
Another idea which not seems to be mentioned, but won't let you
measure down to nH, is to put an accurate 10µH coil in series with the
meter, and you should at least measure with reasonable accuracy down
to 1µH or lower. You may buy ready made coils down to below 100nH
which can be used to check the calibration.

My over 50 years old Radiometer multimeter uses this technique and
measures well down to 20nH. I made my own meter with 18MHz oscillator
to meassure 0.01-1uH

LA8AK

----
Jan-Martin, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/

On Mon, 15 Mar 2004 10:46:53 +0000, Paul Burridge
wrote:

Hi all,

I've got quite a huge stash of inductors in my parts bin. The colour
codes don't always seem to relate to the values I've been able to
measure, with my multi-function DVM, however, and I can't accurately
measure any inductor about 10uH. Is there a circuit anywhere that
would enable me to get a reasonably accurate idea of the values I've
got down to say 100nH or thereabouts?

Thanks,

p.
Another idea which not seems to be mentioned, but won't let you
measure down to nH, is to put an accurate 10µH coil in series with the
meter, and you should at least measure with reasonable accuracy down
to 1µH or lower. You may buy ready made coils down to below 100nH
which can be used to check the calibration.

My over 50 years old Radiometer multimeter uses this technique and
measures well down to 20nH. I made my own meter with 18MHz oscillator
to meassure 0.01-1uH

LA8AK

----
Jan-Martin, LA8AK, N-4623 Kristiansand
http://home.online.no/~la8ak/