power supply discharge

Larkin's power supply party !

There are a couple of dozen supplys in this lab, of all shapes and sizes - half of them never seen on a vendor's shelf.

Depending on what they're supposed to do, they'll have different final capacitive filter stage values.

You want people running around measuring adjustible bench supplies, to compare to your unique low power multi-output 'not yet in hardware' figment.

Get real.

The standard adjustible (or fixed) linear bench supply, will see between 47 and 220uF on their output terminals and a built-in loading of between 1/2 and 2W, depending upon that C and it's load rating.

Switchers are all over the map and may self-load as a mattwr of policy - but none like to regulate at zero load or zero volts - so again between 1/2 and 2W internal loading, if the mfrs actually give a damn - unless it's a DC-coupled bipolar job, in which case their'll be a zoebel network.

Have a nice day.

RL

There is no need to concede defeat and disable your switcher chips, just turn on a big load instead.

In variable speed drives for induction motors, the voltage of the DC rail and bulk capacitance can also rise when the motor is slowing down with a lot of inertia attached to the shaft. They have a switch built in, which you are supposed to attach a big load resistor to. When the DC rail rises above some threshold, it turns on your external load resistor. It cycles on and off to keep the bulk capacitor voltage in an acceptable range.

Due to the cumulative damage that many have warned of, you are better off with a power zener, a poewr-zener replacement made with a transistor, or a switchable load resistor.

I once had to repair a large CNC milling machine where the load dump resistor was on permanently because the controlling transistor had failed short-circuit. The resistor was mounted in a cage on the top of the control cabinet and the machine eventually switched itself off when the radiant heat from the red-hot resistor raised the internal temperature of the cabinet to an unsafe value.

Zeners, even official transzorbs, have a tiny fraction of the joule dump capacity of an mov. Ditto transistors.

A switched resistor would work but is more complex to implement than a single mov.

I'll test some MOVs and see how they behave.

The customer load could be a giant capacitor bank, or a battery. I don't want to short either. And I do want a supply to recover gracefully.

Same problem. My fix is to protect our cap with the MOV, and shut off the switchers when that voltage gets too high.

As I understand it, it's cumulative. Each time there is an event where the MOV operates (ie conducts), some damage is done. So for example a 10 joule rated MOV can operate for 10 1 joule events or 1 10 joule event, or any combination totaling 10 joules.

Ed

I suspect it can dump many 10 joule events suitably spaced, but that's not specified. Ten 1-joule shots within, say, 1 second would be thermally equivalent to a 10j shot.

The short-term thermal damage limit is likely different from the other, long-term failure mode.

I ordered a bunch to test.

“One good test is worth a thousand expert opinions.”

- Wernher Von Braun

Most electrolytic caps just draw more current as the voltage goes up. That usually gets serious at around 120% of rated voltage.

An elec may be as good as a MOV for absorbing joules. And I'll have caps anyhow.

I tested some polymer elecs that suddenly died shorted at over-voltage, without warning, but the Panasonics that we use now just leak and get hot. No big deal.

how is that different from a class D audio amplifier?

Audio doesn't go down to DC.

Audio amps are usually full-bridge, which doesn't reverse pump up the power supply under pathological conditions, like a half-bridge can.

And loudspeakers are very inefficient so don't store much energy and as you note don't go down to DC, again storing not much energy.

A big capacitor bank, or a battery, or a motor can store a lot of energy, so there can be a dilemma when the load voltage is higher than the supply setpoint.

I'm torturing an MOV, a 470KD14. It's rated for 47 volts and 0.1 watt and 10 joules.

At a constant 15 mA, it's at 58.1 volts, which is 0.86 watts. It's pretty warm. The voltage seems very stable after 4 hours so far. That's about 12K joules.

It's likely it could do that forever, but the data sheets suggest that high power shots can do cumulative damage. I might set up to try that somehow.

Now lower the voltage. At what voltage does the current drop to 0?

Ed

But class-D audio amps do. They aren't tightly specified at DC, but pulse width modulation does work down to DC.

I bet that the duty cycle affects the cumulative damage, with smaller duty cycles (more powerful pulses, but more widely separated) doing more damage than just the cumulative energy.

I looked at the Yageo 470KD14 MOV datasheet. It does not seem to mention any wearout effect. Perhaps they figured the mechanism out and remedied it, which would be a good thing.

But the "surge life" items under "Reliability" on page 9 only does ten surges and notes no visible damage, so we have no idea what happens beyond that simple surge test's parameters.

Joe Gwinn

On page 5, it doesn't say so but I think the curves are parametreized on the number of shots, 1 to 1e6.

I might have to cut over to using mosfets and resistors to dump my overshoot energy. MOVs may be too risky longterm. Pity... they are so simple.

I drops 48.2 v at 1 mA, about the same as always, after 62 K joules.

Yes, one can certainly read it that way. Probably have to ask Yageo how to read those plots, and the underlying physical mechanism.

How large are the surges and how long will it be to get to 10^6 surges in total?

Joe Gwinn