What a let down :/

[ped]

So, I thought I'd treat myself to a nice hefty PSU , 30v / 20A to power my ZVS driver.

However...

I think the switching PSU sends the ZVS driver haywire, on my 30v /2A linear PSU ,it works a treat, drawing exactly 100mA @ 5V.

Connect it to the big switching PSU and the current just runs away, I limited it but it's around 2200mA @ 5V.

Is there anything I can do apart for swapping it for a linear?.

 

[ARG]

If you're running it in CC mode the high current PSU wont be happy at lower currents.

The shunt resistance is lower in high current PSU's and when operating it at low currents the shunt voltage is too low.

You should be fine in CV mode for the high current PSU.

 

[ped]

So you're saying I should "open up" the current and just start at 0 V ?

 

[ARG]

Yeah, if you're able to supply the driver with constant voltage it should be fine.

 

[ped]

Bit of a risk though..If I open up the current and connect .

EDIT, just tried it...didn't work...currently uploading a vid of it, check back in 5-10 mins.

 

[ARG]

Don't have to open it up all the way, just to the maximum device operating current (or the minimum the power supply will reliably go down to).

If that doesn't work it sounds like a faulty PSU.

 

[ped]

Here is the vid..

 

[VisibleGreen]

SNES ftw! lol sorry, couldn't resist.
I need one of those psu's, how much do they usually go for?

 

[ped]

Iv'e had that SNES since 1994...21 years! . It holds a special place in my heart .

The PSU was on fleabay for £120 or BO.

I offered £100 and they accepted

/On topic.

Iv'e discovered that setting the PSU above 10v pleases the ZVS , I can then ramp the voltage. Just not go under 10v, or it becomes a current shunt.

 

[Cyparagon]

Yeah, ZVS will have trouble oscillating below 10V or so, simply because of the FET's Vgs threshold value.

 

[ped]

But it oscillates just fine @ 5V on the linear PSU? (see video) .

 

[Hiemal]

ZVS's also don't particularly like soft voltage rises when they're starting out, I've found. If you turn the voltage to what you want and then plug it in it should work right. I guess it has something to do with getting it to start oscillating; since the voltage slowly rises up it just turns one or both of the fets on completely and nothing happens, hence why you get high current draw for no oscillatey action.

 

[ped]

ZVS's also don't particularly like soft voltage rises when they're starting out, I've found. If you turn the voltage to what you want and then plug it in it should work right.

But it works fine on the linear? I can go from 0v to 5v on the linear and it handles it fine.

But as others have said, including yourself.. just set the desired voltage and connect, it does work fine that way.
I'm more curious than anything else as to why the linear can ramp it, but the switch mode cant. I suspect ARG was right.

Also that ZVS driver is a bit of kit, I can draw 2-3" arcs from that LOPTX.

 

[Cyparagon]

But it oscillates just fine @ 5V on the linear PSU?

I don't think that's possible. Put it on a scope and I think you'll see it's not.

It may "work" on the linear supply because the current limit is much lower, and the current regulation is much better than a switcher. Here's my wild guess:
A FET trips, shorts the supply through flyback primary, current limit kicks in, FET deactivates, repeat. I suppose you might technically call this oscillation, but not in the way the circuit was meant to operate.

 

[ped]

Maybe you're right , I don't have a scope so cant check your hypothesis.


I can 100% confirm that the circuit is definitely self oscillating and drawing a quiescent current, until I draw an arc then the current shoots up.

Iv'e always hated SMPSU's , but finding a 20A variable current linear in my price range is impossible.

 

[ped]

Double post

Drawing 5"+ arcs .