[AMPS] Stupid transformer question
Sat, 15 Dec 2001 23:13:45 -0500
I would rig up a hi-pot tester to actually check the insulation after
lifting the grounded end. I was fortunate to get an old Bendix HI-Pot
tester at a hamfest years ago, but you can rig up something easily enough.
A quick&dirty test of the insulation might be to simply connect the (fused)
unloaded xfmr's primary to 220 briefly to see if it can stand twice the
voltage it will see when you actually use it. Then lift the secondary
winding from ground, ground the other end and re-test to verify the
insulation at both ends.
If it takes this OK, why worry? If it doesn't, toss it.
I would not float the frame. Just my personal choice. I ground them.
Think this through before doing it and make sure you are safe before
At 08:25 PM 12/15/2001 -0600, Jim Strohm wrote:
>I have this old but intact HV transformer from a microwave oven that
>(wonder of wonders) doesn't require knocking out the shunt to use.
>Predictably, it has one end of the HV winding tied to the frame of the
>transformer. Hoerver, the insulation between the HV windings and the
>laminations doesn't look good to withstand 2.5 KV.
>Am I better off to:
>1) Lift the winding to ground and attach it to my rectifier bridge, while
>grounding the transformer frame, and take my chances on the insulation.
>2) "Float" the transformer frame by placing it on suitable insulated
>standoffs, and attach the bridge to the frame and to the other end of the
>HV winding? I would provide a reasonable guard cage around the transformer.
>I favor #1, because if the insulation goes, I blow the (free) transformer,
>a fue or two, and maybe trip a breaker as the smoke leaks out.
>#2 seems unsafe, because if a fault condition developed, the HV could be
>present on the primary, and the potential against line neutral or line
>ground could be lethal and undetectable.
>If you don't hear back from me tomorrow, it's because I picked the wrong
FAQ on WWW: http://www.contesting.com/FAQ/amps
Administrative requests: amps-REQUEST@contesting.com