Unknown Filter Chokes; Making Them KNOWN

[W9LBB]

Howdy, Troops.

I'm guessing that everyone here has encountered the same problem numberous times.

I have several BIG power supply chokes that I can't find specs on, or can only find partial or intuitive specs.

First...  does anyone know of a bench testing way to tell a filter choke from a swinging choke (looking for the air gap is out; they're potted).

Next... how about determining inductance values, especially the range on swinging chokes? I DO NOT have an inductance bridge that will handle over very small values of inductance.

The chokes I have are from Wilcox transmitters, and all the information I have on them is rated current...   one is 600 MADC, and the other is 1.5 amps, so we're not talking about some insignificant stuff here. BTW, I also have Wilcox part numbers.


Mr. T., W9LBB

[Bacon, WA3WDR]

I don't know how it's done in industry, but running current through the inductor from a current source that you can modulate, and measuring voltage across the inductor will tell you a lot.  You can do this with a big tetrode (or several), but a power transistor is probably a better choice because this test doesn't have to involve a huge amount of power at lower voltages.  But if the winding resistance is 100 ohms, and you want to check it at one amp, you'll need 100 volts DC, plus maybe another 50V or so for the drop across the source resistor, and for AC swing.

A current source is easy to build.  Basically a moderate resistance in the emitter, source or cathode circuit will do it.  Then you bias the base, gate or grid, and it draws current.  The current produces voltage on the impedance which changes the bias, so the current produced is mostly a function of the current source, and it does not change much in response to the voltage drop across the device under test.

You can set the DC current level anywhere, and then superimpose a sine wave of a known frequency, and the AC voltage across the inductor will tell you the inductance at that DC current.  Y volts DC bias produces a certain current A, and Z volts DC produces another current B, and you set up AC sine wave that wiggles between these two points.  Peak to peak AC current is then (B-A), and RMS AC current is (B-A)/2.818.  Then you know what the RMS AC current through the inductor is, and the RMS AC voltage is pretty much a function of the reactance X of the choke at that DC current level (X=E/I), then L = X/(2*PI*f).  (Actually the inductance reading will be slightly high, because some of the AC voltage comes from the DC resistance of the choke.  Knowing the DC resistance, this can be calculated out. Using a higher audio frequency will reduce this effect greatly, but too high of an audio frequency (more than 1 or 2 KHz) can run into other effects caused by lamination thickness and stray capacitance.)  DC current can be measured by a DC meter.  Then you can change the current and repeat, and you will see the inductance-current relationship.

Make sure you aren't overdriving the current source circuit.  Too much inductance, too high an audio frequency, and too much AC will exceed the voltage swing of the current source, and your readings will be wrong.  Watch with a scope, and don't let the AC on the inductor clip.  The DC resistance error will be small as long as the inductive reactance of the choke is large compared to its resistance plus the resistance of the DC meter.  This is a function of frequency and inductance.  You can take the DC meter out once you read the current, and the voltage across the emitter or source resistor will confirm that the current is the same.

Other items of interest are impedance at higher frequencies, so you can check the eddy current losses and stray capacitance.  A hi-pot test will tell you if the unit seems capable of high voltage operation.

[WA1GFZ]

send some current through the inductor with a low voltage supply and series resistor to limit current. Then feed in 120 hZ through a series cap and fixed resistor like 100 or 1000 ohms. Measure the AC voltage across the resistor then across the inductor. The ratio of voltage is the reactance of the choke then find Xl=2 x pi x f x L
say the voltage across the choke is twice the resistor then the reactance is twice the resistor value.
Try this at different DC currents so you will know the L vs I if it is a swinging choke.

[N9NEO]

For small inductors you can take 20v or so from bench power supply and jam it directly into the choke.  Watch the current rise (assuming you have a current probe handy) and use V=Ldi/dt or L=Vdt/di to figure out the inductance.  When the rising current all of a sudden takes a turn upwards you have found the saturation level.

For larger chokes worth a hundred amps or so it takes a few big beer can capacitors charged up to 100v or so.   Object is to keep the voltage fairly steady so that the current rises in a straight line till it gets into saturation.

73
Bob

[WA1GFZ]

Bob
You might want to add the part about the inductive kick when you disconnect the probe. Don't want to stall any pumpers........

[N9NEO]

Yea Frank, I forgot.

Sometimes when we start talking big beer can caps and hundreds of volts it's just good damn engineering practice to get somebody else to run these tests for ya.  I like making the young kids do it.  Heheh.

73
Bob

[WA1GFZ]

yes and inductors have an interesting ability to turn current into a very high voltage
when you cut them off from flow. They get unhappy very quickly At least you know where a cap stands because the operator knows the voltage.