Odd readings on a bridge rectifier

Thread Starter

Just Another Sparky

Joined Dec 8, 2019
18
I'm observing an interesting phenomenon when taking current measurements on both the line and load sides of a single phase full-wave bridge rectifier. When I read 60 amps DC on the load side, I simultaneously read 73 amps AC on the line side. I doubt either of the two meters I'm using is that far out of calibration. My guess is that it has something to do with the way the AC and DC waveforms are being measured. It's just been so long since I've dealt with Vp-p, Vavg, Vp, Vrms, Vdc and all the others that the concise explanation escapes me. Could someone help me wrap my head around this?

For reference I am using:
A Klein CL600 for my AC readings. (Klein Tools website)
A Simpson cat. no. 2740 with corresponding 6713 shunt for my DC readings. (Simpson bulletin 2065, pages 2 and 21)
 

MaxHeadRoom

Joined Jul 18, 2013
19,777
Did you try using the shunt on both AC and DC sides and take a voltage measurement across the shunt and see if any difference in current calc.?
Max.
 

SLK001

Joined Nov 29, 2011
1,544
The 73 amps on the AC side is the DC equivalent reading. That means that your bridge is consuming 13 amps of current, so it might be getting hot. Assuming a 1V drop across the diodes, that means that 13 watts or so are being given off as heat. Confirm this.
 

Thread Starter

Just Another Sparky

Joined Dec 8, 2019
18
Forgive me, but doesn't Kirchoff's Law state that current in must equal current out for any given node in a system? How can a bridge rectifier 'consume' current? Voltage drop yes, but shouldn't current stay the same?

I understand how that often occurs in reactive loads like motors or transformers where magnetizing current is required to maintain the magnetic field on top of the real power transmitted to the work and lost in the laminations as heat. But no appreciable reactance should be present in a bridge rectifier. It's just a collection of 'one-way valves' for electrons.
 
Last edited:

SLK001

Joined Nov 29, 2011
1,544
The current doesn't disappear. It is converted into heat energy. So current in = 78, current out =60 + 13 into heat.
 

crutschow

Joined Mar 14, 2008
24,130
That means that your bridge is consuming 13 amps of current,
No, the bridge doesn't consume current.
The average (rectified) current in has to equal the average current out.

The difference is the way the meters are measuring the AC versus the DC current.
The AC RMS bridge input current can be much higher than the output DC (average) current, if the load is a large capacitor filter (nearly 2:1 for a typical bridge-capacitor supply).

Below is the LTspice simulation of a simple bridge-rectifier-capacitor.
The average DC output current is about 1A but the RMS current from the transformer is 1.77A.
That's why you need to need derate a transformer nearly 50% for a bridge-capacitor DC supply.

1575825206046.png
 
Last edited:

Thread Starter

Just Another Sparky

Joined Dec 8, 2019
18
The AC RMS bridge input current can be much higher than the output DC (average) current, if the load is a large capacitor filter (nearly 2:1 for a typical bridge-capacitor supply).
Inductance on the AC side wouldn't have that effect, would it? I don't have any filters on the output, but the rectifier is supplied by a 1.5kva transformer. I took those measurements under short-circuit conditions by dialing up the input voltage from zero until I reached 60 amps DC.

Would parasitic inductance on the DC side maybe be enough to cause that?
 

Thread Starter

Just Another Sparky

Joined Dec 8, 2019
18
Unfortunately the problem is that I don't have a hall-effect clamp meter in my inventory. I've got a few extra shunts lying around... it might be possible to insert one in series temporarily. I'll see what I can do.
 

Thread Starter

Just Another Sparky

Joined Dec 8, 2019
18
I threw a 50A, 50mV shunt directly between the set-screw lugs and the rectifier and came up with the following:

AC:
Clamp meter reads 53.8A AC
Voltmeter across test shunt reads 51mV AC

DC:
Panel meter with it's dedicated shunt reads 42A DC
Voltmeter across test shunt reads 45mV DC

All voltages were confirmed with an identical meter of the same make and model but they still seem a little suspicious to me. The Klein meters I used are built more for ruggedness than triple-digit precision. I want to scope it later on to be sure.
 

Thread Starter

Just Another Sparky

Joined Dec 8, 2019
18
Well, I figured it out. I put scope probes on the test shunt on both the AC and DC sides.

AC:
Clamp meter read 53.8A AC
Scope with test shunt read 54.64A RMS

DC:
Panel meter with it's dedicated shunt read 42A DC
Scope with test shunt read 54.75A RMS

I changed to a different test shunt to be sure and got essentially identical numbers.

So it's pretty definitive. The problem lies with the panel metering setup. I adjusted the current to 62.5 amps according to the scope reading (full load current for this piece of equipment). Panel meter read 50A DC. I put a voltmeter across the terminals and read 25mV.

So I removed the panel meter's shunt and compared it to two known genuine Simpson Electric shunts. Lo and behold it's a counterfeit. The genuine ones have #8 brass test lead screws and no calibration notches. The one I installed has #10s, a "calibration" notch cut into the resistive element and slightly shorter mounting centers.

Careful what you buy off of Ebay. Saving a couple dollars might cost you a few hours of your life.;)
 
Last edited:
Top