I've purchased a used isolation transformer from a well known brand. Inside the case are two toroidial transformers (Youwon DV-130-1), on the back of the case are two duplex receptacles (each going to a transformer). There is a circuit breaker on the back and a power switch on the front. The transformers primaries are BROWN/BLUE, and the secondaries are BLACK/WHITE. There is a GREEN coming out of each transformer, and per this thread, I believe that this a shield internal to each transformer. Each receptacle has a ground wire leading to a lug on the chassis that has mains ground as well as the green secondaries (shields) on it. The neutral on one receptacle had a green wire also going to this lug. I thought I'd be clever and remove the receptacles ground wire that goes to the chassis ground (as well as that neutral secondary to chassis ground). I've verified there is no continuity between the hot/neutral on the receptacles and the green secondary wire. However, I'm still seeing continuity between the ground on the recep, and the chassis. It appears that the receptacle's tabs that screw to the chassis are also ground. So I really don't have an isolated ground. I think I'll need to purchase separate receps that don't have that trait.
The other worrying thing is that when I turn on the isolation transformer, put my meter on VAC, and stick a lead into secondary hot (receptacle hot) and to mains earth or neutral, I have about 70VAC. This seems very wrong, and I must be missing something. The referenced thread sounds exactly like the device I have purchased, down to the capacitors on each recep. There also appears to be a varister on one of the receps (mdc z131 20ul).
I really don't want to hurt myself, and I do want to learn. But having 70VAC between secondary hot and earth neutral can't be right. I'm not measuring anything else. I'd love to hear some thoughts on what I'm doing wrong. Thank you!
The other worrying thing is that when I turn on the isolation transformer, put my meter on VAC, and stick a lead into secondary hot (receptacle hot) and to mains earth or neutral, I have about 70VAC. This seems very wrong, and I must be missing something. The referenced thread sounds exactly like the device I have purchased, down to the capacitors on each recep. There also appears to be a varister on one of the receps (mdc z131 20ul).
I really don't want to hurt myself, and I do want to learn. But having 70VAC between secondary hot and earth neutral can't be right. I'm not measuring anything else. I'd love to hear some thoughts on what I'm doing wrong. Thank you!
NEVER isolate the ground!
Live and neutral must be isolated, to allow a safer repair on the live side of a power supply etc. If there is continuity between live/neutral output and earth, it is not an isolation transformer.
If you don't want the output to have an earth or ground, don't connect it in the plug.
All test gear must be earthed whilst using an isolation transformer to help avoid static high voltages building up, causing damage to your item being tested and of course your test gear and your fingers.
Live and neutral must be isolated, to allow a safer repair on the live side of a power supply etc. If there is continuity between live/neutral output and earth, it is not an isolation transformer.
If you don't want the output to have an earth or ground, don't connect it in the plug.
All test gear must be earthed whilst using an isolation transformer to help avoid static high voltages building up, causing damage to your item being tested and of course your test gear and your fingers.
@JonSnell Electronic I guess I have a misunderstanding of what is truly required. If having a continuous ground back from isolation side ground to true earth ground is needed, then I have that now. But you're saying I need to verify that ISO-neutral is truly separate from earth ground. What's concerning is that I'm not sure if I'm truly isolated. If I take a volt reading between ISO-hot and MAINS-netural, I see ~70VAC. Is that just a capacity effect of the secondary coils? I don't see continuity between ISO-hot and MAINS-hot or ISO-neutral and MAINS-neutral when the iso-transformer is powered off via the switch. Thanks for your help!
I forgot to connect the ISO-neutral and ISO-hot of both receps together. This drawing has that. In this drawing, we see the green wire coming out of the transformer secondary. This is not connected to MAINS at all. It appears to only be a shield.
I've not had a chance to wire a lamp between ISO-hot and MAINS-neutral.
I've not had a chance to wire a lamp between ISO-hot and MAINS-neutral.
Attachments
These kind of questions can be a minefield as they are very much safety related...
The 70 volt reading could well be normal and is due to the leakage inductance and inter winding capacitance of the transformers. You should not see a leakage current exceeding (what do we say for safety...) of lets say 0.5 milliamps.
There is no 'hot' or 'live' designation on the outputs as they are floating and isolated. Any such markings are notional only and have no meaning. If you connect a resistor of 1k5 between either output of each outlet socket and true mains ground you should not see more than 0.75 volts AC across that resistor.
Your drawing shows a center tap on the transformers. Is that actually the shield? Its also unusual to tie the secondaries of the transformers together. Floating vs. tieing the center tap to ground are both valid isolation options for noise. However the centertap is not OK for safety isolation applications. I would push for replacing the outlets with GFCI's for safety in any case. The 70V could well be from capacitive imbalance in the secondaries or any of a number of other leakage possibilities. The meter is a high impedance so it picks up the leakages. Paralleling a light bulb or even a 100K resistor will quickly determine if there is a real leakage or shorted turn issue. The caps and the MOV on the secondary are a good idea. The space is quite limited in there.
The issue is that the secondary is capacitive coupled to the shield, and this is probably at one end of the winding, so that the other end of the winding is hot with respect to it when unloaded. Disconnect the unit and do resistance measurements between things - that avoids being confused by capacitance. Or place 10k resistor between your meter leads when measuring voltage, which will short out any capacitive coupling.
A 70V reading on a 10Meg meter can be caused by a few microamps, but with 10k across it you'll need milliamps.
A 70V reading on a 10Meg meter can be caused by a few microamps, but with 10k across it you'll need milliamps.