Amps
[Top] [All Lists]

[Amps] Resonant choke measurment

To: <amps@contesting.com>
Subject: [Amps] Resonant choke measurment
From: "Bob" <bobt@iafrica.com>
Date: Sun, 23 Dec 2012 15:09:14 +0200
List-post: <amps@contesting.com">mailto:amps@contesting.com>
Hi Peter I tracked down what I said originally see below:-
 
First there isn't really a circuit..I usually use whatever is to hand at the
time !
The idea is to connect the choke across a DC supply and measure the current
through the choke and the voltage across the choke as the current rises.
Since this happens very quickly a digital scope setup to store a once off
event is required.
The current required is usually more than is convenient to supply from a
power supply so I usually use one or more big electrolytic capacitors and
charge them from the supply via a charging resistor which is small enough in
value, that it will take 5 to 10 seconds to charge the caps. It should also
be large enough in wattage and value that it can left across the DC supply
for a few seconds without burning out the supply or the resistor.
The voltage of the supply will depend very much on the current range that
you are interested in and the internal resistance of the choke to be tested.
Generally I aim for the IR drop in the choke to be about 10% of the supply
voltage at the maximum current that I am interested in. I.E. if I had an HT
choke that had a 1 ohm resistance and supplied up to 5 amps then I would aim
for  5 X 1 X 10 voltage supply of 50 volts.
The most convenient way that I find to connect the setup is to have the DC
supply with the capacitors connected to it with the charging resistor
connected in the positive lead to the capacitor. A switch of some sort is
then connected to the positive terminal of the capacitor. The other terminal
of the switch is connected to one terminal of the choke under test. To the
same point one voltage probe of the scope is also attached.
The other side of the choke is then connected to the negative terminal of
the capacitor via some kind of current measurement device.
To do the test, charge the capacitors, then with the scope setup to trigger
on a positive going transient from the probe connected to the voltage at the
switch, close the switch !
The current will rise rapidly and discharge the caps, next go and switch off
the supply, and then open the switch, and have a look what has been
captured. 
 
You should see two traces, the voltage one, should show the voltage across
the capacitor slowly discharging (well several milliseconds maybe). The
current one should start off from zero and rise at a rate depending on the
capacitor voltage. As the current reaches the saturation level of the choke
the curve will get more and more vertical as there is less inductance left
to slow the rate of rise of current.
You then can calculate the inductance at any value of current by looking at
the slope of the current at that point ( so many amps per second) and
dividing that into the voltage across the choke. If you want to get picky
then you need to subtract the voltage dropped across the internal resistance
of the choke from the capacitor voltage first. I.E. V= I X Rint where I is
the value of current at the point you want to make a measurement.
I usually use a shunt to measure the current with the shunt connected
between the terminal of the choke and the negative terminal of the
capacitor. The second scope probe  then is connected across the shunts
measuring terminals ( and usually measures a few tens of millivolts ). 
 
If I can I make a special kind of shunt called a "hairpin" shunt that has a
much higher bandwidth than normal shunts. They are very easy to make, and
you can find their description on the net. I use a strip of brass shim stock
that has the correct millivolts drop for the current that I need for the
measurement. I then double it over and glue it together with a thin strip of
insulation in between the two pieces. I solder two power terminal wires at
the two ends. For the scope sensing I drill a hole in the shunt close to the
end with the terminals. The insulation and inner of the coax goes through
the hole and the inner of the coax is soldered close to the hole on the
other side, while the outer of the coax is soldered on the other side. The
other end of the coax is then made off into a bnc connector and connected to
the scope input.
The switch is anything that will carry the pulse of current and make cleanly
without too much contact bouncing. My favorite is actually a mercury 20 amp
switch which does not bounce at all ! but of course must be used carefully.
I hope that this hasn't confused you too much  ! 
 
P.S. I also use this design to test DC circuit breakers up to 10kA ! 
 
Regards Bob ZS6BXI
 
So while a Hays bridge will work (I seem to remember the bridge that I came
across was a version of that, and you need the large inductance to isolate
the choke under test) the method described above has been used over many
years and used from inductances of a few micro henries ( at several thousand
amps DC !) and henries at a couple of amps. It works very nicely for
measuring the chokes in switch mode power supplies and also the  so called
"swinging" chokes that are normally used for LC smoothed power supplies.
This is one of the bits often missed when designing LC supplies where you
really need your choke designed for two values of inductance at two
different values of current (say 20 H at 0.1 amp and 2 ~5 H at 1 amp). The
advantage of this is that you get a smooth supply, but the minimum load
current at which the output DC voltage will be stable at is a much lower
value. In any case one of the design maxims that I always worked to was that
the optimum size for the output choke was a similar VA core size to that of
the supply transformer, after that it was tuning the core air gap to get the
best output ripple but lowest critical load current. So if it's a key down
"brick" supply you want look for a suitable LC combo. The problem is that
the design of such chokes has always been more art than science (too big a
variation of steel core values from batch to batch) and it has been very
difficult for the average transformer manufacturer to actually measure the
critical values on test. So hence the "demise" of the LC power supplies (
and the weight and cost of the chokes, as well as the cost of the HV caps.).
Of course the availability of  cheap high current rated 450 volt
electrolytic caps for switch mode primary supplies may have had something to
do with it as well.
 
The next real step is already here.. Using a switch mode  PSU for the HV
supply. It can be done, it has been done, but the costs and complexity are
still up there.
 
So having "lit" the fuse stand well back.
 
Regards Bob ZS6BXI
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>