Can a Variac slowly power up an Inverter Welder? ..or -BANG!- ?

[Electro Detective]

Just a simple one folks (yeah, right.. )

Can a 0 to 240 volt Variac be used to slowly turn up a suspect (or Dodgy Bros. repaired  ) inverter welder (with no load attached),
be it MIG, TIG, STICK, Plasma Cutter, or a bells and whistles combo unit,

without trashing the onboard power supply ?

or is it a no go on certain type or all 'inverter' devices?


Up till now I've been rolling with an 8 amp Variac, jerry rigged fuse/s and clamp meter combo on just about everything for snafu checks

Apparently SMPS get snarly with Variacs, so I haven't tried it 

No point asking at a welding forum, as this is leaning towards electronics diagnostics repair

Any advice appreciated   

[SeanB]

Not good, you need a big dim bulb tester. Probably best is to go find some 300-500W halogen lamps as the current limiter, at least that will keep the noise down if there is still a fault, but will allow enough inrush current to start the unit.

[Electro Detective]

In the case an inverter welder was confirmed and tested as a 'no fault' device = working fine/no problems

would using the variac still do damage or compromise the electronics without a dim bulb tester ?



(hunting down dusty old school globes and sockets for a way overdue switchable wattage dim bulb array..  )

[mk_]

In the case an inverter welder was confirmed and tested as a 'no fault' device,

would using the variac still do damage or compromise the electronics without a dim bulb tester ?

Think in Power = U x I.

if U is low I trends to be high...

which allows - together with the internal resistance of the variac a lot of unexpected effects, even on "no default devices" with a switching regulator as one of the internal powersupplies.

[MadTux]

Not good, you need a big dim bulb tester. Probably best is to go find some 300-500W halogen lamps as the current limiter, at least that will keep the noise down if there is still a fault, but will allow enough inrush current to start the unit.

Big halogen lamps as PTC protection is a good way to test inverters without load.
If something goes wrong, the lamps will light up instead of the switching transistors blowing up.

[Ian.M]

At best, if the inverter control circuit has a properly designed and fully working undervoltage lockout function, and there are no faults in and around the inverter output stage, all using a Variac will do is severely stress test its powerup sequence (as mk_ has explained).

To actually bring up an inverter gently, you'd need to isolate the control circuit supply and power them at rated voltage then slowly bring up the DC bus voltage (feeding the output stage).  However, odds are, to avoid the control circuit going into a fault shutdown state, there are some monitoring circuits that would need to be tricked to avoid the control circuit detecting the power stage is not operating at the expected power levels, without defeating any overload protection function they may have.     Good luck doing that without a full component level service manual + a full set of factory Q.A. test jigs!

All it takes is *ONE* dodgy cap, resistor or catch-diode in a snubber circuit, failing in operation,  for the power output stage to destroy itself faster than a dim bulb tester can react.  Limiting the current to the output stage with a dim bulb tester helps limit cascading damage: e.g. if an output device shorts, it will prevent it blowing the s--t out of any associated current sense resistor or blowing traces off the PCB.  It wont prevent a failed output device going drain - gate short (or collector - gate short if IGBT) and frying its gate drive circuit, though if you are lucky, it will go three way short and the gate drive circuit may survive due to the dim-bulb tester limiting the energy. 

[edpalmer42]

In the case an inverter welder was confirmed and tested as a 'no fault' device,

would using the variac still do damage or compromise the electronics without a dim bulb tester ?

Think in Power = U x I.

if U is low I trends to be high...

which allows - together with the internal resistance of the variac a lot of unexpected effects, even on "no default devices" with a switching regulator as one of the internal powersupplies.

This will only be (possibly) true if the load is attached to the power supply.  That won't be the case for a welder.

But, even if the load is disconnected, it's still possible for a variac to kill a switching supply.  Control sections can power-up in odd states and incorrect order.  Power-on time delays don't work right.

Imagine if an 'output enable time delay' expires and turns on the output before the 'regulation control loop' is stable.  These are hypothetical examples, but you get the idea.

[Electro Detective]

Looks like the trusty Variac that's bailed me (and previous owner/s) from countless hidden dramas over the years
won't be hooked up to inverter based stuff any time soon,
perhaps for manual voltage fluctuations based around 240 volts, to test whatever percentage a DUT can tolerate as labelled.

bright idea/dreaming.. Would a pulsed/ramped/???  240 volt AC feed with series connected dim bulb improve the chances of not taking out electronics,
and show up any pending dramas on hooked up meter/s with a fast Min Max capture and or a logging function?


As far as dim bulb testers go, what bulbs are better, worse, faster, slower, better, (cheaper  )

incandescants or halogens? or a combo of both in a series/parallel arrangement etc

Thanks to all thus far  for the helpful and "no-o-o..don't do it! "  info 

[edpalmer42]

As far as dim bulb testers go, what bulbs are better, worse, faster, slower, better, (cheaper  )

incandescants or halogens? or a combo of both in a series/parallel arrangement etc

Thanks to all thus far  for the helpful and "no-o-o..don't do it! "  info

This isn't a precision test device so I wouldn't worry about incandescants vs. halogens.  What you want is anything that will limit the current if things go bad - both will do that.

What's more important is the wattage of the bulb.  In general, the wattage of your dim bulb tester should increase as your device power capabilities increase.  If your device has a normal startup surge, your dim bulb tester has to allow most of that to occur.  That will tend to increase the wattage that you need.

If you're working on a 5V@1A wall wart, you might not be able to find a bulb small enough.  Recently, I was working on a 33V@33A supply.  I needed a 100W bulb before it started nicely - but that still ensured that it couldn't draw more than ~1A so it was totally worthwhile.

I bought a selection of smaller bulbs so I can screw in whatever is necessary.  I now have 7.5W, 15W, 25W, 40W, 60W, and 100W.

Ed

[Ian.M]

+1. 

A tungsten filament is a tungsten filament, and you are using it for its PTC resistance behaviour, which at temperatures before it starts glowing is practically independent of whether its in vacuum, or low pressure inert or halogen gas.  Get whatever's convenient and cheap.