Author Topic: Building isolation transformer?  (Read 8585 times)

0 Members and 1 Guest are viewing this topic.

Offline rexxarTopic starter

  • Frequent Contributor
  • **
  • Posts: 439
  • Country: us
    • Forever Tinkering
Building isolation transformer?
« on: October 18, 2013, 02:59:43 am »
I've got a couple of dead UPSes here, and I was thinking that I'd rip out the transformers and turn them into a mains isolation transformer. I figured since they're identical transformers, I could just connect the secondaries together, but I'm not sure which one to use. There's a high-current center-tapped winding at 12V and another lower-current one at about 16V. I guess it would make sense to use the higher current winding, but do I connect all three taps, or just the outer ones? For that matter, would something bad happen if I connected both secondaries?

When I go to connect it to outlets, what do I do with the isolated ground? Do I connect it to mains earth or?

Thanks guys
 

Offline Circlotron

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: au
Re: Building isolation transformer?
« Reply #1 on: October 18, 2013, 04:18:21 am »
Just connect the high current secondaries - that will account for 99%+ of the transformer capacity. Use all three connections and put the centre taps to ground. That will help reduce transfer of mains-borne noise. Not a big deal in this situation but if you can do it, why not.
 

Offline ddavidebor

  • Super Contributor
  • ***
  • Posts: 1190
  • Country: gb
    • Smartbox AT
Building isolation transformer?
« Reply #2 on: October 18, 2013, 05:13:32 am »
The  ground wire will be  isolated in a isolation trasformer. You must not connect it to ground
David - Professional Engineer - Medical Devices and Tablet Computers at Smartbox AT
Side businesses: Altium Industry Expert writer, http://fermium.ltd.uk (Scientific Equiment), http://chinesecleavers.co.uk (Cutlery),
 

Offline Kremmen

  • Super Contributor
  • ***
  • Posts: 1289
  • Country: fi
Re: Building isolation transformer?
« Reply #3 on: October 18, 2013, 06:01:07 am »
The  ground wire will be  isolated in a isolation trasformer. You must not connect it to ground

That is correct in principle but the ground they are talking here is the one between the 2 transformers. The first xformer is a step-down and the second, inverted one is a step-up. Both isolate galvanically, so the connection between the xformers is by default not galvanically connected to either the input or the output; it floats. If you now connect the center tap to ground, it grounds neither the input or the output - only the circuit between the transformers. That is the reason it is safe to connect to the supply side ground to minimize passthrough EMI.

It is another matter that the transformers were not originally designed as safety transformers and probably lack the shielding foil or winding between primary and secondary, and likely do not have the extra insulation and creep distances required for true safety isolation transformers. They would still work though, but the combo would not qualify as an "official" safety isolation transformer. That said, i would not hesitate to use it as such in a hobby lab if the voltages turn out right.
Nothing sings like a kilovolt.
Dr W. Bishop
 

Offline rexxarTopic starter

  • Frequent Contributor
  • **
  • Posts: 439
  • Country: us
    • Forever Tinkering
Re: Building isolation transformer?
« Reply #4 on: October 18, 2013, 06:29:53 am »
So, use the high current secondaries, but connect both center taps to mains earth? This seems intuitively wrong to me, but I guess since the secondary is isolated from the mains, there's no real current flow from there to ground?

And I should leave the isolated ground floating? This seems wrong to me, but again, it's isolated so there shouldn't be any real issue, right?

One more question: There's already a circuit breaker in the UPS case I'm going to use, so I'll go ahead and use it. I assume it'll be best to leave it on the mains side, right?
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4278
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: Building isolation transformer?
« Reply #5 on: October 18, 2013, 07:28:23 am »
It is another matter that the transformers were not originally designed as safety transformers and probably lack the shielding foil or winding between primary and secondary, and likely do not have the extra insulation and creep distances required for true safety isolation transformers.

I'd need to check the spec to be sure, but I'd have expected a transformer designed to isolate a primary circuit from a SELV circuit to require a greater degree of isolation than one designed to sit between a primary circuit and a TNV3 circuit.

Offline Circlotron

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: au
Re: Building isolation transformer?
« Reply #6 on: October 18, 2013, 12:52:10 pm »
With the high current in-between windings grounded, not necessarily by the centre tap, one end of a non-tapped winding would be okay, if the insulation of the first transformer does fail it gets no further than the grounded winding before a fuse etc blows.
 

Offline geraldjhg

  • Regular Contributor
  • *
  • Posts: 61
  • Country: ar
Re: Building isolation transformer?
« Reply #7 on: October 20, 2013, 12:09:59 am »
if the combo manages to light a 60w standard bulb (that afterwards could be used as
a series "fuse" for low power output testing) it will be usefull

try and make some current transformers from little small toroid transformers
they really work at smps frecuency

G E R A L D
 

Offline johansen

  • Super Contributor
  • ***
  • Posts: 1123
Re: Building isolation transformer?
« Reply #8 on: October 20, 2013, 12:53:32 am »
So, use the high current secondaries, but connect both center taps to mains earth? This seems intuitively wrong to me, but I guess since the secondary is isolated from the mains, there's no real current flow from there to ground?

That's a really good idea. "good enough" to replace for all intensive purposes the emi shield from medical grade isolation transformers. you'd have to have both transformers fail to get anything through to the tertiary winding (two transformers cascaded)
i have such a system set up with a 13.8 vac secondary from a 1.5KW sinewave inverter feeding the 14.x vac secondary of some other sine wave inverter. so i get 114vac out for 122vac in but i can't complain. i have the 13.8 vac running out the front panel through a 50 amp apc connector, but its floating. i suppose i could wire in another switch to reference it to ground.

anyhow, OP, is the high current winding square copper or copper foil? the 1.5KW transformer i have is copper foil, 5mohm resistance including the 8 inches of 10 awg wire. 12 volt secondary might be appropriate for a sine wave inverter off a 24 vdc battery but 16.5 vac is more likely a MSW inverter.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12356
  • Country: us
Re: Building isolation transformer?
« Reply #9 on: October 20, 2013, 01:53:22 am »
When I go to connect it to outlets, what do I do with the isolated ground? Do I connect it to mains earth or?

I think I would pick one end of the isolated secondary as "neutral" and connect the isolated "ground" to that (but still floating, not connected to mains earth). That way if there is any short circuit between hot and ground in the device under test it should blow the device fuse.
 

Offline dr_p

  • Frequent Contributor
  • **
  • Posts: 309
  • Country: ro
Re: Building isolation transformer?
« Reply #10 on: October 20, 2013, 12:27:10 pm »
When I go to connect it to outlets, what do I do with the isolated ground? Do I connect it to mains earth or?

I think I would pick one end of the isolated secondary as "neutral" and connect the isolated "ground" to that (but still floating, not connected to mains earth). That way if there is any short circuit between hot and ground in the device under test it should blow the device fuse.


Doesn't this increase your chances of getting zapped since there's no GFDI to protect you? You're likely to touch the "grounded" chassis while working on the DUT, so you only need to touch the isolated "hot".  :bullshit:
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12356
  • Country: us
Re: Building isolation transformer?
« Reply #11 on: October 20, 2013, 03:17:11 pm »
Doesn't this increase your chances of getting zapped since there's no GFDI to protect you? You're likely to touch the "grounded" chassis while working on the DUT, so you only need to touch the isolated "hot".  :bullshit:

I suppose so, but Rule 1 is that you should not work on any DUT while it is powered up. You always disconnect the supply before taking the lid off or touching the insides.

If you must probe a powered item you should follow the one hand rule and keep one hand behind your back.

Isolation transformers on the workbench are not used to protect you from electric shock. If you think an isolation transformer makes you safer you have a misconception about the use of isolation transformers.
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 19919
  • Country: gb
  • 0999
Re: Building isolation transformer?
« Reply #12 on: October 20, 2013, 06:02:07 pm »
With an isolation transformer, I prefer to leave the output floating.

You can build an indicator using a couple of Y2 capacitors and a neon lamp to indicate an earth fault.
 

Offline Circlotron

  • Super Contributor
  • ***
  • Posts: 3319
  • Country: au
Re: Building isolation transformer?
« Reply #13 on: October 21, 2013, 03:49:01 am »
If you are probing inside an SMPS anywhere after the input bridge rectifier with a grounded scope probe you MUST have both ends of your isolation tranny floating.

Also an extremely good idea is a pair of safety goggles. A blown electro or the sharp bits of plastic from an exploded TO-220 or TO-247 can be, shall we say, hard on the eyes.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf