diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Power Supplies (http://www.diyaudio.com/forums/power-supplies/)
-   -   How to measure rectifier recovery current pulses (http://www.diyaudio.com/forums/power-supplies/216023-how-measure-rectifier-recovery-current-pulses.html)

merlin el mago 12th July 2012 08:42 AM

How to measure rectifier recovery current pulses
 
How to measure the rectifier recovery current pulses with a scope to calculate a snubber for transformer secondary to damp out the rectifier recovery current pulses that resonate with the inductance of the transformer secondary?

DF96 12th July 2012 09:50 AM

Current transformer. Essentially a step-up transformer with a single turn primary (possibly made using the actual wiring of your circuit), and a heavy resistive load on the secondary so the reflected resistance to the primary is very low so it doesn't affect your circuit. Inverse of a high impedance voltage probe.

PChi 12th July 2012 09:57 AM

The lots of money approach is to use an Oscilloscope Current Probe.
Low cost approach is to add a low value resistor in series to monitor the current.
If you have a differential probe then it's relatively easy to connect across the resistor.
If not if your Oscilloscope can add two channels and invert one of then a differential signal can be displayed. But common mode rejection could be a problem.
If the circuit isn't connected to ground the Oscilloscope probe can be connected across the resistor. But beware the capacitance to ground of the Oscilloscope. I have used this method on switch mode power supplies to trim snubbers. I have tried to calculate values but usually the stray inductance isn't well known.
I guess the best approach is to spend the money on fast recovery rectifiers to minimise the snubber dissipation.

RJM1 12th July 2012 11:47 AM

Because you don't need to measure the amplitude of the current with any accuracy and you just want to measure the frequency just wrap a couple of turns of wire around the transformer wire that goes to the rectifier and measure the voltage waveform across those turns of wire.I've even done it with the ground clip on the oscilloscope probe wrapped around the wire twice and clipped to the tip.

DF96 12th July 2012 12:24 PM

Quote:

Originally Posted by RJM1
you just want to measure the frequency

I'm not sure that is what the OP said. I think he wants to look at the current waveform.

RJM1 12th July 2012 01:01 PM

You will be looking at the current waveform on the oscilloscope, it just won't be calibrated to x volts per amp. And it will be a high pass filter, which in this case is something you want.I assumed he wanted to measure the frequency the waveform was ringing at to calculate the values for his snubbers. A couple of turns of wire around the transformer secondary winding is a cheap current transformer. Current transformers generally have more turns and are calibrated into a specific output resistor.

jneutron 12th July 2012 01:13 PM

Quote:

Originally Posted by RJM1 (Post 3089685)
Because you don't need to measure the amplitude of the current with any accuracy and you just want to measure the frequency just wrap a couple of turns of wire around the transformer wire that goes to the rectifier and measure the voltage waveform across those turns of wire.I've even done it with the ground clip on the oscilloscope probe wrapped around the wire twice and clipped to the tip.

That is not the proper way to couple to the current within a wire.

The magnetic field created around a wire will be field lines which encircle the wire. A wire wrapped around the current carrying conductor which is following the field lines will have little if any voltage induced by changing those field lines, the primary pickup mode using a wrap as this will be the loop area pickup between the wrap wire and the scope return lead.

If you make a multiturn coil of wire, and position it so that the flux lines around the wire will travel through the coil, coupling will be much better. Position the little coil so that the plane of the coil is in the plane of the wire. If you make the pickup coil with too many turns, it's inductance at hf may be too high to drive the measurement instrument, so it's a tradeoff between sensitivity and bandwidth.

jn

RJM1 12th July 2012 01:26 PM

Not Ideal but you will see the current waveform.

jneutron 12th July 2012 01:49 PM

Quote:

Originally Posted by RJM1 (Post 3089765)
Not Ideal but you will see the current waveform.

Agreed. The problem is the increase of capacitive coupling when wrapping the sense wire around the current carrying conductor. It makes it harder to discriminate between capacitive coupling of a wire with possibly large voltage swings, and magnetic coupling of high speed current swings.

By going multiturn coil and not wrapping it around the wire, you increase mag coupling without such a capacitive problem.

BTW, the only reason I know this, is I did the exact same thing you recommend back in the 80's to find one bad diode in an assembly of 3 parallel ones. Wrapping the current carrying conductor was of no use, but a pickup coil did work.

Conceptually, it's not the wraps that are doing the coupling, it's the return loop of the scope probe and it's trap area. If this is not considered, then changes made to the system and measured using your method may result in inaccurate conclusions if the loop area is disturbed during the changes.

jn


All times are GMT. The time now is 09:10 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2