Fuse Eating Variac.

[Ptron]

Should there be an inrush of current when switching an unloaded variac on? It doesn't seem like there should be since it's an inductive load.

I've used my variac for a couple of years plug>variac>outlet with no problems. I recently decided to mount it in a box with a fuse, on/off switch, 0>120/0>140V switch, voltmeter and ameter. I decided to bring it up on a 2A fuse for starters, even though it's a 1.2kVA variac, and it pretty consistantly blows the fuse when I switch it on. I've disconnected everything except the fuse, on/off switch, and variac. I've tried a different switch. I've even taken the variac apart to look for signs of a short. Nearest I can figure is either there's a short I'm not finding or there is a current surge when it's switched on. Before I try a bigger fuse though I want to make sure that makes any sense.

[Robert M. Martinelli]

If you' re absolutely sure the variac has no problems, connect a DMM in series or use an amprobe to determine the inrush current, this way you' ll be able to properly size the fuse rating.
Hope this helps
Best regards
Bob

[Enzo]

The 1 kW variac sitting next to me has an 8 amp fuse.

WHy so small a fuse anyway? And are you using a slow blow fuse?

REmember the output is the variable part, the mains connection is across the full winding no matter what. As far as the mains side is concerned, there is no "bringing it up." You can only "bring up" the output.

[Robert M. Martinelli]

I figured you want to use the variac at a lower power, but to do so the fuse has either to be able to withstand a current greater than the inrush current or to be a slo-blo. The correct fuse rating for full power would be ( quite obviously ) a 10 Amp for a 120 VAC mains.

[Steve Conner]

Because of inrush current, big transformers need big slow-blow fuses, if they're not to pop randomly at switch-on. This page explains what inrush current is:

http://www.opamp-electronics.com/tut...nt_2_09_12.htm

Toroidal transformers made of modern core materials are especially bad, since they saturate really hard. Variacs too, since they're constructed in the same way: a tapewound toroidal core with no airgaps.

[Ptron]

I guess I could have used my light bulb limiter to check. One bringup device to check another. Amazing what a night's sleep will do.

And I should have said initial power up, not bring up. A bad choice of words, esp when a variac is involved.

I just used the 2A for more safety when I powered it on the first time with everything, including the meters and their 5V supply, hooked up. I assumed there was a short somewhere and kept using a 2A fuse while I figured it out. Partly for the safety margin. Partly because I have a ton of them.

I was thinking of the Variac as purely an inductive load that would limit current at turn on, not have a current rush. Apparently it's more complicated than that. Thanks for the link, Steve. I'll read it when I have some time.

[Ptron]

http://www.opamp-electronics.com/tut...nt_2_09_12.htm

Ah. I see. Good to know everything's probably ok. Yes, they were fast blow fuses too.

So out of curiousity, how does a transformer reach it's normal operating point? If the magnetic flux starts off with that "DC offset" (if you will) and you continue integrating it from the voltage, doesn't it theoretically maintain its offset?

[Steve Conner]

In an ideal, lossless transformer, yes, the offset would persist forever. In a real one, DC resistance of the windings causes the offset to decay away with a time constant of L/R. In a good transformer with high magnetizing inductance and low DCR, this can be several seconds.

This is the reason for the Thud, Brrrmmmm noise you hear when switching a large transformer onto the mains. As the DC offset decays, the noise changes from a spikey 60Hz buzz into a smooth 120Hz hum.

[Enzo]

Thud, Brrrmmmm

Are you sure there are four "m" in that?



Just beware when it goes Thud, brrraaaazzzzz....

[Ptron]

Now It's a Meter Eater

Thanks for your help guys. The variac is working fine.

Can't say the same for the meters. Check out the oddly worded warning, last line of the first paragraph on page 1 of this datasheet

Would you interpret this as saying the ground of the 5v supply and one of the AC ins (pins 3 & 4) are internally tied together? I sure didn't. Actually, I think it implies the opposite. Now I have two dead meters. I guess that's why they're so cheap on ebay; they skimped on the datasheet translation. That or the slave labor.

[Enzo]

I would interpret it to mean they are separate. If you want one side of the AC being measured to be at ground it is OK if it is the same ground as the 5v ground. But if you are measuring AC voltage not from ground - such as ACROSS a winding, then don't connect one side to the ground of the meter. ANy currents from that AC to ground would then flow through the meter circuits - as opposed to through its measurement port - and burn it up.

But from yuor post, then I am guessing they ARE tied together inside? Meaning the 5v supply has to float?

[Ptron]

I would interpret it to mean they are separate. If you want one side of the AC being measured to be at ground it is OK if it is the same ground as the 5v ground. But if you are measuring AC voltage not from ground - such as ACROSS a winding, then don't connect one side to the ground of the meter. ANy currents from that AC to ground would then flow through the meter circuits - as opposed to through its measurement port - and burn it up.

But from yuor post, then I am guessing they ARE tied together inside? Meaning the 5v supply has to float?

Yes, they are tied together internally. It wouldn't have been a big deal had I known that and hooked it up with the proper care. One side of the measurement is to neutral, as it's across the output of the variac, and the 5V supply is floating. I got into trouble because I arbitrarily hooked up the measurement leads. The ammeter was right and the voltmeter ended up backwards so I had hot going into the common gnd of the voltmeter> 5v supply "gnd"> ammeter 5v in gnd> which is internally tied to one of it's measurement leads> which is measuring current through the neutral side> *poof*

Neither the datasheet pinout or the meter itself indicate which measurement pin is tied to ground. Actually, there is a + and - on the meter by the pinouts but they have been written over with a sharpie, ~ and ~, so they're hard to see.

I complained to the company and asked for replacements but I don't hold out much hope as they're a mass, budget, ebay distributer from China.

[Tom Phillips]

Consider Analog Meter Movements

I like the traditional analog meters on my variable AC transformer unit. You might consider picking up some used analog meter movements since you need to replace your meters anyway. The way the needle moves (steady, oscillating, jumping around) gives you additional information that just looks like an unstable display on a digital meter. Very useful I find.
My humble opinion anyway.
Regards,
Tom

[Enzo]

I'd have to agree. You can watch trends on a digital meter, but a swinging needle I think gives a much more informative visual as to the relative steepness of the curent ramping.

I wouldn't think of any terminal on that meter as "ground." The 5v supply should completely float just as the 5VAC winding for a rectifier tube does. Then it doesn;t matter which side of the meter sees neutral and which sees hot. Unless I am missing something

[R.G.]

Big, fat inrush surge limiting thermistor...