3-5V High Current 150A Power Supply

I have been asked to generate a 100Amp RMS precise sinusoidal current. It will have negligible burden/load.

Luckily I've done something very very similar in the past where I used a LEM sensor, a dsPIC33, an inductive load and made a current controlled PWM inverter based on a PI controller. It worked and I could inject any required current (sine, square, ramp etc) of desired magnitude by providing it with the desired reference.

So all I have to do is scale that system up for high current levels. Now getting high current MOSFETS is no problem. I'm planning to make a nice low resistance inductor with appropriate inductance so that a current of 150 amps will flow through it at 3-5V (preferably even less) so power loss will be minimal.

Only thing I need now is about 150 amps of current. Where to get that?

I'm caught between LiPo batteries that power RC planes or making my own SMPS that steps down 12V, 40A from an off the shelf SMPS to 3V, 160A.

Problem with batteries is that they would run out in a few minutes.

How difficult would option B, I.e. 12V to 3V, 160A push pull converter be?

Sounds like a job for a welding transformer. They step down mains AC, and put out a couple hundred A at low V.

I used to work with an AC welder that had a crank at the top, to change the Amperage. It may have changed voltage as well.

What is the expected load impedance and drive frequency? Even if the inductor has very low resistance, it's reactance will determine the necessary voltage. IF you need up to 5Vrms, then some 14V SLA batteries might be fine. The spwm inverter can scale the amplitude arbitrarily.

Using a welding transformer for AC current sensor test and calibration is basically a good idea. Unfortunately the waveform of a usual welding transformer has considerable distortions.

A lead acid battery (assuming you're talking about the 150Ah type battery used in UPS inverter systems) will be:
a) Too large
b) Too expensive
c) Not feasible

Yes the aim is AC Ammeter and Clamp meter testing. Hence power required is ideally zero. The design is basically ready. Just need the 150A source.

I'm inclined towards SMPS as this will be cheap, small and practically portable. The issue is that I've never worked at such current levels.

The most basic issue for example that hit me is what do I use to rectify the output of the push pull converter which will be putting out over 150 Amps. Parallel 20-30A diodes with heatsinks. Wouldn't that in itself require like 20-25 high speed diodes??? What do high current power supplies use to do this?

Can I use 4 high current MOSFETs without antiparallel body diodes kept always on? If so suggest suitable 100A+ MOSFETs with low Rds and without body diode.

How about using a "reverse" current transformer ?

A suitable toroidal mains transformer with one turn of welding cable looped through the hole might do it.

With that, you might only need an amp or less through the primary to generate 150 amps through the secondary.
A LEM sensor in the secondary, and a feedback system such as you already have should work.

A fairly high dc supply voltage, but at far less current should make things a lot easier.
If the toroidal transformer had two 110 volt primaries, a push pull drive system, may simplify things even more.

*edit*

Further thoughts.
I have done this myself to calibrate over current sensing in equipment. I just used a variac to drive my monster current transformer.

That should work for you too, but if you really need low distortion sine waves, perhaps the toroidal transformer could be made self resonant and only loosely coupled to the variac

The load power is not going to be that high, 150 Amps at perhaps 50mV to 100mV open circuit voltage is surprisingly little power.

chinuhark said:

Yes the aim is AC Ammeter and Clamp meter testing.

Click to expand...

Run 10A through a wire, and loop it into a circle 15 times. It mimics one conductor carrying 150A.

I cannot take credit for this idea. I saw it on a website and said to myself 'I have to remember that one.'

- - - Updated - - -

For a clamp meter it can work, anyway.

That also works very well for those LEM Hall current sensors which are typically scaled for +/- 75A dc or 75A peak ac.

That is often inconveniently high if you are only measuring a very few amps.

Wrap a few turns through the hole and problem solved.

Sometimes though you might need to calibrate something that uses a very large current shunt and then you are snookered, unless you can source the actual required current somehow.

There are crazy guys on the internet rewinding microwave oven transformers with welding wire and melting honking great steel bolts for fun...

chinuhark said:

A lead acid battery (assuming you're talking about the 150Ah type battery used in UPS inverter systems) will be:
a) Too large
b) Too expensive
c) Not feasible

Click to expand...

You don't need huge charge capacity, just large power capacity. A car battery would probably be fine, and I don't think the price is beatable.

I'm inclined towards SMPS as this will be cheap, small and practically portable. The issue is that I've never worked at such current levels.

Click to expand...

A FET bridge driving a coil basically is a SMPS. You just need an unregulated DC supply (battery), a PWM'd bridge, a coil, and a feedback loop.

Can I use 4 high current MOSFETs without antiparallel body diodes kept always on? If so suggest suitable 100A+ MOSFETs with low Rds and without body diode.

Click to expand...

Yes, a full bridge is a good candidate. It should be driven such that the body diodes only briefly conduct current during transitions.

- - - Updated - - -

Warpspeed said:

That also works very well for those LEM Hall current sensors which are typically scaled for +/- 75A dc or 75A peak ac.

That is often inconveniently high if you are only measuring a very few amps.

Wrap a few turns through the hole and problem solved.

Click to expand...

Was going to suggest this. For a current transformer or hall effect ammeter, there's basically no difference between one wire carrying 150A and fifteen turns carrying 10A. Obviously this won't work with a shunt ammeter.

The wrapping around several times thing is exactly what I had also suggested to my superior but he was afraid of whether that method is a hundred percent accurate.

Point being that if I use the best LEM sensor and ensure that precise 10 amps flow through a wire, will a super calibrated clamp meter show exactly 100amps when this wire is wrapped 10 times through its hole.

All theory points to the fact that there can be no issue in this method. Just asking if there can be any practical issue that hampers the accuracy of the arrangement. My money is on NO problem in this method. What say you guys?

A last resort I just thought of is I put those 10 loops through one LEM and instead of calibrating 1 loop for 10 amps, I calibrate the 10 loops for 100. Although I would do this if absolutely necessary (which I think it isn't) as I have 6 Nos of +-18Amp LEM sensors lying around and would like to avoid buying a 150Amp new one.

There is no introduced accuracy problem with using multiple loops.
The LEM itself has a 1,000 turn secondary winding.

It's decided then. That's what I'm going to do. Hell using the LiPo batteries I researched, the system can even be hand held/portable.

Just make an accurate ten amp RMS current source, and a bunch of plug in coils with suitable ranges.

In theory the distribution of current through the clamp's aperture can affect the measurement (if the flux in the core is not uniform), but this should be seen as a limitation of the clamp, rather than a problem with the test.

Specifically, you don't want the loops to be tight and bunched towards one side of the core. You want the loops to be large so that as they pass outside the core they stay far away from it. I would find a couple feet of multi conductor cable like this and solder its ends together in an interleaved fashion so that you get a nicely bunched coil contained in the cable's jacket.

Any calibration method has errors. You didn't yet mention your accuracy requirements.

AC current transformers have a perfectly symmetrical winding and core and are very insensitive to conductor position and surrounding magnetic fields. Clamp meters are less symmetrical, DC current transducers have an air gap and respective stray field. But as far as I'm aware of, you can still achieve 0.1% accuracy with multi turn configuration of DC current transducers.

FvM said:

But as far as I'm aware of, you can still achieve 0.1% accuracy with multi turn configuration of DC current transducers.

Click to expand...

If the turns are evenly distributed around the circumference of the core, I agree.

1℅ is OK. Basically this all started when one of our clamp meters showed 2A where a different one showed 5A. The frustrating thing was we had no idea which one to trust. I just wished I had a reference current to test which is correct and remembered that's exactly what I had made just a few months back.

These things keep happening now and again and our engineers have started losing faith in clamp meters altogether.

Thus a low cost solution to ensure that the clamp meters we take on site are not giving us total garbage. DIY clamp meter tester, sort of.

While we do send all our meters to professional calibration facilities from time to time, they still keep misbehaving. So this seemed like a good idea to me, as I already had what was required, just had to tweak it for higher current levels.

Shunts are easily available up to 150A & much higher (1%) all you need then is a good AC mV meter to read them, thus you can see which clamp meter is more correct.

While the multiple wire thing is OK, let us say I want to generate 100Amps RMS.

I like the short circuiting the secondary of a transformer and using it as a reverese CT thing but please help me with the math a bit.

Aim: To get 100A rms pure sinusoidal current through a conductor.

Ferrite core transformer: 100 turns primary and 1 turn secondary. So 12Volts applied to the primary will give 0.12V or 120mV output.

We must provide a very small impedance to the output (basically close to a short circuit) such that it draws 100A when provided with 120mV. i.e. 1.2mohm impedance. Most probably even a short circuit will not allow such a small impedance and so Primary voltage will have to be increased till we get 100A output.

Problem: As the primary will be excited by SPWM, to get a sinusoidal current at the output, the equivalent impedance of the secondary, primary, leakage reactances etc must have a dominating inductive component to filter out the switching frequency which I'm planning to have around 10-15kHz at the moment.

Do you guys thing this will naturally be the case? If not will providing small air gap which would increase leakage reactances X1 and X2' make the equivalent inductive without leading to other issues?

The SPWM will be bipolar (half bridge or push pull) so saturation should not be an issue.