Running power supplies of high frequency power, 80 hz and above

[VentMine]

Sup y'all.

Wondering if I can run my antminers off of higher input frequency, I have a genset and am thinking of overspeeding it to get more HP. E.g. instead of running it at 1800 rpm, how about 2400 rpm and boost it to 80 hz?

Can the AP3W++ handle that? Will I blow anything up?

[DaveF]

Most (all?) power supplies are setup for 50/60 hz and that is it.
You can probably figure in a 10% margin at best but that is REALLY pushing it out of spec.

If they are not rated for it, I would not do it.

Unless you like fire, if you like fire then you can try it :-)

I'm kidding, seriously if they are not rated for it don't do it.

-Dave

[VentMine]

I am planning to just do it and see if it catches fire actually, but certainly hoping someone can shed technical light whether these switching power supplies can manage a higher frequency.

[philipma1957]

I am planning to just do it and see if it catches fire actually, but certainly hoping someone can shed technical light whether these switching power supplies can manage a higher frequency.

how can you know unless you try.

just do one for shits and giggles


take proper precautions for fire or smoke or exploding caps.

make a video.



i will tell you this send 320 volts into them and they will fail.

[VentMine]

That's what I like to hear.

Yeah, not sure but I think the voltage regulator can adjust down the voltage at higher RPM.

[lightfoot]

It would be... interesting. The goal of the primary side of a switching power supply is to convert AC to DC, which is then fed to the chopper to make high frequency AC which then is rectified into low voltage high current DC. The actual regulation is done with the HF AC circuits, so that should not be affected.

The problem may wind up in the power factor conditioning circuitry in modern power supplies, which is used to balance the input AC voltage peaks more evenly across the sine wave. These are tuned for 60hz, so changing the input frequency much could result in a lower power factor (higher VA) which means less efficient, more stress on the input components, and a need for larger wires going in (as watts dissipated is based on VA, not simply amps).

I would think that an older ATX supply with a switch to select 120 or 240 volts would be more robust (and less efficient anyway) because the input diodes for the rectifier are rated to double the current anyway (since when you switch from 120 to 240 you're just cutting out a voltage doubler).

So... It would be worth trying, but I would get a Watts Up meter and watch what the power factor does with a 50% load at 120 volts for starters. Start speeding up the generator and see if the power factor drops like a rock and the Volt/Amps starts going up even though the Amps stays the same. That will tell you about how much extra heat the input stage is having to dissipate due to the PF difference.

(Interesting thought experiment, thanks!)

[philipma1957]

I had a German record player with  a drive motor that used 50 hz  Being in usa at 60hz   the motor ran slower  and a 33 ⅓ rpm record player played at 31 rpm

I got a ps audio power wave generator that  could go from 110 to 130 volts ac and from 50hz to 70hz  the motor ran great at 52 hz and 111 volts.

So I know there is some play if only hz is altered and that volts can drift abit.

but 80hz is more then I ever did as I did a 50hz motor at 60hz  I did not bother trying higher.

My guess is if he stays under 245 volts it will run at 80hz   without dying.

[mikeywith]

Below are the input specs for APW3++ from bitmain's official page
 

Code:

Voltage Range
100 - 240 V A/C

Starting Voltage
95 - 105 V A/C

Frequency Range
47 - 63 Hz
Power Factor

as far as I know,  the tolerance on frequency is not as generous as that of voltage , 80 hertz is way too much for this PSU to handle, I am not an expert in this field but, my bet would be against the success of your experiment.

[DaveF]

https://www.youtube.com/watch?v=XchwE9zVdnw

[VentMine]

Thanks for all the insight guys, I'll definitely report back once I do some experimentation.

[lightfoot]

Take pictures, can't wait to see where things go. As I say about 10 times a day:

"What's the worst that can happen?"

[NotFuzzyWarm]

I am planning to just do it and see if it catches fire actually, but certainly hoping someone can shed technical light whether these switching power supplies can manage a higher frequency.

Figured that the necro bump deserved a proper answer to the OP so...
If it was 'just' a switching supply with NO input power-factor compensation then yes 80Hz would work fine but it will not increase power per-se, the PSU just pulls lower current and it will certainly not increase power out of the PSU because that is set by the regulator(s) in the PSU.

That said, for the past oh, 10 years or so in the interest of increasing PSU efficiency and even more importantly greatly reduce/virtually eliminate the high power-factor that any capacitor input load presents to the power line all PSU's that can deliver more than 500W are required to use PFC (Power Factor Control) circuits and those are very sensitive to frequency. Go outside the 50-60Hz design parameters and yes, the Majik Smoke will escape...

I should add that one cannot just 'goose the throttle'... yeah you could fiddle with the carburetor to speed up the engine but - at a bare minimum the genset has a mechanical speed governor and most likely an electronic one you'd have to hack. Then there the circuits that regulate the voltage out of the genset. Bet they won't like it...