3
\$\begingroup\$

I am trying to figure out a way to create a high current ripple through and inductor, similar to that of the ripple the inductor sees in a SMPS. I would like to put a 2A peak to peak 400KHz current through a 10uH inductor. It is alright if it is a sinusoidal waveform rather than the saw tooth the inductor sees in a SMPS.

Ideas I have tried:

-Find a high enough power AC supply that operates at 400kHZ. I have not been able to find a supply with a high enough power rating that can generate 400kHz.

-Use a transformer with a very high turns ratio to step down the voltage/step up the current from a low current signal generator. The issue is that at 400kHz, a 10uH part has a relatively high impedance and I would need a very high input voltage to obtain 2A across that high of a load impedance.

-Use a H-bridge to drive current through the inductor in each direction. The issue here is I need a DC current through the part as well, and I can not vary the inductance or frequency so I am unable to vary the ripple at all.

-Design a buck converter for the DC and ripple current I want. This works, but I need to test many parts at the same time so I would need many buck converter circuits.

\$\endgroup\$
  • 2
    \$\begingroup\$ H-bridge sounds plausible. It is essentially a synchronous buck-boost converter with input and output shorted together. Continuous mode with DC current should not be a problem. \$\endgroup\$ – venny Sep 15 '15 at 22:48
  • \$\begingroup\$ Testing multiple parts at the same time will require separate drives for each part to maintain good precision, whether it's a buck or H-bridge or anything. \$\endgroup\$ – gsills Sep 16 '15 at 2:33
  • \$\begingroup\$ @venny I have played around with an H-bridge in simulation. It seems like the only way to vary the ripple current is by changing the switching frequency or the inductance. How do I set both the DC and ripple current if I have to stick with 400kHz and 10uH? \$\endgroup\$ – Brett Prudhom Sep 16 '15 at 13:19
  • 1
    \$\begingroup\$ When current, frequency and inductance are constant, the only variable that remains is voltage. It could be done by a buck converter before the H-bridge, which would be adjusted by sensed current ripple. \$\endgroup\$ – venny Sep 16 '15 at 13:34
  • \$\begingroup\$ @Venny I believe that changing the input voltage will vary both the DC and ripple current, but if I can't change my frequency or inductance and need a certain percentage of iripple compared to Idc, how do I vary the ripple and dc separately to get that ripple to dc ratio? As far as I can tell my only two parameters I can vary are input voltage and duty cycle, is that enough? Thank you! \$\endgroup\$ – Brett Prudhom Sep 18 '15 at 22:01
1
\$\begingroup\$

Make yourself a LR oscillator.The LR oscillator is the dual of the RC oscillator that everybody learnt about at colledge .The well documented RC osc that has a comparitor with hysterisis and a resistor and cap can have the resistor replaced with a current source giving a triangle wave of voltage across the cap .The triangle voltage runs between the upper and lower trip point on your voltage compariter.Now all you do is set up the comp with hysterisis for two current pionts and then use some transistors to drive the coil.You can choose your current thresholds to emulate your proposed SMPS .Your supply volts can control the frequency so you can get close to your situation .I did this to check SMD coils as a proposed replacement for thru hole coils for S TRAP buck convertors .

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.