Variable autotransformer abuse

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

crochambeau

Active member
Joined
Feb 14, 2012
Messages
26
Location
PNW
I believe I have a part built for 400 hz operation.

Okay, here's the back story: I'm building a very simple variable DC power supply for a weirdo project I am building. I know it's going to be noisy, that's part of the charm.

Power supply consists of AC mains, switch, fuse, variable autotransformer, 120 to 24 volt step down transformer, full wave rectification & a cap (3300 @ 35v).

So, the cap is old and I figure I may as well burn it in slowly. There's a little electrical buzz coming from the variac (actually a Standard Electrical Products Adjust-A-Volt, but variac is easier to type), but nothing major and no cause for alarm. About half way through the travel my nose/gut tells me it's time to temperature check everything. Remove mains connection and feel around. Everything is nice and cool until I get to the variac.

It's uncomfortably warm enough to conclude that long term use is not going to work out for me. I run through the standard stupidity checks and everything seems OK. No shorts, wiring looks good. Then it dawns on me, I think the variac was harvested from some mil spec survey equipment which was built for 400 hz AC.

Naturally, the part in question has no labelling, so I'm reaching a little bit. My cheap inductance meter operates at higher frequency, so I don't think it's measurement is conclusive. The more I think about it, the more sense it makes, perhaps 60 hz was working on cooking the winding and there's simply no way to safely deploy this part in 120 volt 60 hz mains?

So. My options now seem reduced to wiring up the variac in the 24 volt AC portion of the secondary, which also "suffers" from low frequency but at a fifth of the voltage, or just using the part as a variable inductor for passive EQ.

Would 24 volts at 60hz be that much safer, or is it six of one half dozen of another?
 
JohnRoberts said:
Danger Will Robinson...  Using an auto-former for the connection to the mains supply does not provide galvanic isolation. This is not safe.

JR

True, the variable unit I am speaking of does not provide true isolation, but connectivity is confined to the fused primary mains circuit and contained within an enclosure that is earth referenced. I'm under the impression that this is not courting catastrophe, based on a history of use of similar constructions. Can you elaborate on the safety aspect?

For what it's worth, I have several unisolated variable transformers (and one that does float the variable AC output via separate windings) that I've used numerous times without a heat issue, one saw daily use for years to buck my line down to 110 volts to keep an old amp happy.

Does the 400 hz theory hold water?

 
crochambeau said:
JohnRoberts said:
Danger Will Robinson...  Using an auto-former for the connection to the mains supply does not provide galvanic isolation. This is not safe.

JR

True, the variable unit I am speaking of does not provide true isolation, but connectivity is confined to the fused primary mains circuit and contained within an enclosure that is earth referenced. I'm under the impression that this is not courting catastrophe, based on a history of use of similar constructions. Can you elaborate on the safety aspect?
While you didn't say what kind of circuit you would power from this auto former the PS "ground" with this transformer is connected directly to yout mains neutral, if all is well. But perhaps to mains hot if your outlet is mis-wired, or if the neutral wiring back to the drop fails open or high impedance neutral can rise to hot voltages.. 

I don't know where you live or what you outlet wiring looks like, but mine unfortunately is not to the highest standard.  ::)
So far roughly with half the outlets I've checked, my hot and neutral lines have been wired backwards, and I haven't found one yet with anything at all connected to the ground lug. It would be wishful thinking to speculate that my house is some rare outlier with shoddy wiring, I fear it more common than we would like to beleive. I recently replaced two outlets with GFCI so I will be safe even without a safety ground.

Perhaps if you had a GFCI on your primary the autoformer would be OK.  8)
For what it's worth, I have several unisolated variable transformers (and one that does float the variable AC output via separate windings) that I've used numerous times without a heat issue, one saw daily use for years to buck my line down to 110 volts to keep an old amp happy.

Does the 400 hz theory hold water?

It seems a 400Hz model would be stressed all the time at 60Hz.

Another consideration, if the malfunction appears related to a specific setting, suppose the wiper that is supposed to pick off only one turn on the auto-former is worn and manages to touch two adjacent windings at the same time.  This would in effect create  a shorted single turn winding across the magnetic field. I would expect this to make the magnetic circuit very unhappy, if this is what is occurring.

Just a guess.

JR
 
The circuit I'm building is an array of noise oscillators (Chua circuits, relaxation osc, etc) to use as odd source material.

JohnRoberts said:
It would be wishful thinking to speculate that my house is some rare outlier with shoddy wiring, I fear it more common than we would like to beleive. I recently replaced two outlets with GFCI so I will be safe even without a safety ground.

Perhaps if you had a GFCI on your primary the autoformer would be OK.  8)

Yes, good point, thanks John! I know in my audio room there's a miswired outlet (please don't ask me how I found this out)  :eek:

So, I took the auto-former apart, and it became apparent that - while I had wired it similar to the way it had been wired from the factory, it was not a good match to my application (two inner taps were wired, not the outer legs as I want).

Cleaned & reassembled, I stuck it in the 24 volt secondary and everything runs cool now. So I think it was user error, not frequency mismatch. Whatever it was, I'm not pulling a bunch of current now, which makes me happy.

That said, has anyone ever deployed a variac core as a variable inductor for a passive EQ?
 
The "400CPS" part should work OK on 60Hz *at 0.15 times the voltage*.

Though if you don't know the rated CPS, are you sure this is a 120V part? Not impossible, but aircraft systems also had 48V and other variants.

If known to be 115V 400CPS, then 17V 60Hz is the same stress.

Mounted *after* a 120V:17V transformer, the lack of isolation is no added problem.
 

Latest posts

Back
Top