Author Topic: All isolation transformers must be safe!  (Read 5526 times)

0 Members and 1 Guest are viewing this topic.

Online ShockTopic starter

  • Super Contributor
  • ***
  • Posts: 4214
  • Country: au
All isolation transformers must be safe!
« on: May 05, 2016, 10:27:35 pm »
Quote
...you can touch either one of these, even if you are standing on the ground, and not receive a shock. The only way to get a shock from the isolating transformer is to touch both of the output connectors at the same time.

Good advice? Perhaps there are a few situations with a DUT connected he isn't thinking about.

Soldering/Rework: Pace ADS200, Pace MBT350
Multimeters: Fluke 189, 87V, 117, 112   >>> WANTED STUFF <<<
Oszilloskopen: Lecroy 9314, Phillips PM3065, Tektronix 2215a, 314
 

Online ShockTopic starter

  • Super Contributor
  • ***
  • Posts: 4214
  • Country: au
Re: All isolation transformers must be safe!
« Reply #1 on: May 05, 2016, 10:46:26 pm »
This video he starts off strong until 11:13 and talks about dual isolation transformers? Ok, but now site isolation transformers aren't actual isolation transformers? Hmmmm...

Soldering/Rework: Pace ADS200, Pace MBT350
Multimeters: Fluke 189, 87V, 117, 112   >>> WANTED STUFF <<<
Oszilloskopen: Lecroy 9314, Phillips PM3065, Tektronix 2215a, 314
 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3640
  • Country: us
Re: All isolation transformers must be safe!
« Reply #2 on: May 05, 2016, 11:06:30 pm »
The BS1363 socket and bright yellow housing suggests a UK worksite power transformer. They have a 230V primary and a 110V secondary with a center tap to ground. This means that any pole of the secondary-side cabling is less than 55V to ground, which is done for safety reasons in case someone saws through a cable.
They don't isolate the secondary side from earth, which means they won't protect a device in a test and repair situation. It is not the type of isolation that electronics technicians require. (Besides, you cannot test and repair a 230V apparatus if the secondary is only 110V.)
« Last Edit: May 05, 2016, 11:09:11 pm by helius »
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Re: All isolation transformers must be safe!
« Reply #3 on: May 05, 2016, 11:54:28 pm »
The thing about an isolation transformer is that depending on exactly what and why you are using in there are at least 4 or 5 different things that can make sense to do with the earthing, and some of these will not make working on a live chassis safe!

You can:
  • leave the output connector earth off entirely
  • Carry the input side earth thru to the output connector but leave the secondary floating
  • Earth one side of the secondary from the input earth, with or without carrying the earth thru to the output connector
  • Connect the output connector earth pin to one side of the secondary, but leave it isolated from the input earth
  • Connect the input earth to the secondary centre tap

And this list is by no means exhaustive.

All of these are useful in specific situations, and all of them are potentially lethal in other situations, you really cannot talk about an isolation transformer making something safe without going into a lot of detail about what exactly you mean by an isolating transformer, what the load is and what you are trying to protect against.

RCD (GFI) protection does not carry across a transformer for obvious reasons, and in the case of one where the secondary has a terminal connected to some sort of ground, you need to think about exactly in what order things hook up.

Now personally I am of the view that the actual need for an isolating transformer should come up so seldom unless you are hacking switched mode power supply designs that I personally don't even own one, if you need one often and are not a hardcore power engineer you are probably doing something wrong.

Stay safe all.

Regards, Dan.
 

Offline sokoloff

  • Super Contributor
  • ***
  • Posts: 1799
  • Country: us
Re: All isolation transformers must be safe!
« Reply #4 on: May 06, 2016, 11:19:56 am »
For someone who is occasionally troubleshooting and repairing power supplies on consumer electronics devices, is fairly price-insensitive, and is safety-sensitive, should I buy an isolation transformer for the bench? A variac as well?

If so, what am I looking for spec-wise? (US power standard)
« Last Edit: May 06, 2016, 11:34:30 am by sokoloff »
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 9497
  • Country: gb
Re: All isolation transformers must be safe!
« Reply #5 on: May 06, 2016, 12:08:59 pm »
The BS1363 socket and bright yellow housing suggests a UK worksite power transformer. They have a 230V primary and a 110V secondary with a center tap to ground. This means that any pole of the secondary-side cabling is less than 55V to ground, which is done for safety reasons in case someone saws through a cable.
They don't isolate the secondary side from earth, which means they won't protect a device in a test and repair situation. It is not the type of isolation that electronics technicians require. (Besides, you cannot test and repair a 230V apparatus if the secondary is only 110V.)

The one with the BS1363 socket has a floating 230V secondary. Incoming ground is routed to the ground pin (hole) on the socket, although there is normally an internal option to disconnect this too, depending on application.

The other transformer shown, the one with two round sockets, is a site transformer with 55-Gnd-55 as you describe.

Edit: I guess you could call them both 'site' transformers, one for 230V tools and the other for 110V.
« Last Edit: May 06, 2016, 12:15:30 pm by Gyro »
Best Regards, Chris
 

Offline mojoe

  • Frequent Contributor
  • **
  • Posts: 371
  • Country: us
Re: All isolation transformers must be safe!
« Reply #6 on: May 06, 2016, 07:20:28 pm »
The thing about an isolation transformer is that depending on exactly what and why you are using in there are at least 4 or 5 different things that can make sense to do with the earthing, and some of these will not make working on a live chassis safe!

You can:
  • leave the output connector earth off entirely
  • Carry the input side earth thru to the output connector but leave the secondary floating
  • Earth one side of the secondary from the input earth, with or without carrying the earth thru to the output connector
  • Connect the output connector earth pin to one side of the secondary, but leave it isolated from the input earth
  • Connect the input earth to the secondary centre tap

And this list is by no means exhaustive.

All of these are useful in specific situations, and all of them are potentially lethal in other situations, you really cannot talk about an isolation transformer making something safe without going into a lot of detail about what exactly you mean by an isolating transformer, what the load is and what you are trying to protect against.

RCD (GFI) protection does not carry across a transformer for obvious reasons, and in the case of one where the secondary has a terminal connected to some sort of ground, you need to think about exactly in what order things hook up.

Now personally I am of the view that the actual need for an isolating transformer should come up so seldom unless you are hacking switched mode power supply designs that I personally don't even own one, if you need one often and are not a hardcore power engineer you are probably doing something wrong.

Stay safe all.

Regards, Dan.

For bench use, to isolate the DUT and to keep from blowing up my 'scope, which of the four configurations would you recommend? Assume the DUT is a semi-modern unit with a three-prong plug.

Does this change is the DUT if an old two-prong plug AA5 tube radio?

I'm thinking either #1 or #2 would do.

 

Offline helius

  • Super Contributor
  • ***
  • Posts: 3640
  • Country: us
Re: All isolation transformers must be safe!
« Reply #7 on: May 06, 2016, 08:01:48 pm »
To a certain extent you need to know how your DUT is wired. If it bonds Vss on the board to its chassis, then transformer #2 ("Carry the input side earth thru to the output connector but leave the secondary floating") would not isolate the circuit from ground, allowing a ground loop through your scope. This is usually true whenever there are panel connectors like BNCs, or brass standoffs, or pins that are wired to a chassis stud.
Unless you have a good reason, it's safer to use differential probe techniques, or check that the probe ground clip is only connected to points that really are earthed.
 

Offline jitter

  • Frequent Contributor
  • **
  • Posts: 793
  • Country: nl
Re: All isolation transformers must be safe!
« Reply #8 on: May 06, 2016, 08:15:57 pm »
Quote
...you can touch either one of these, even if you are standing on the ground, and not receive a shock. The only way to get a shock from the isolating transformer is to touch both of the output connectors at the same time.

Good advice? Perhaps there are a few situations with a DUT connected he isn't thinking about.

What situations would those be?
 

Offline jitter

  • Frequent Contributor
  • **
  • Posts: 793
  • Country: nl
Re: All isolation transformers must be safe!
« Reply #9 on: May 07, 2016, 08:24:54 am »
Saying the output of a 240V isolation transformer is safe to touch is not something I'd want to say on camera.

With the necessary disclaimers, I would because I tested it on myself. But you are going to kill yourself with an isolation transformer if you grab one secondary output lead with one hand an the other with the other hand!

If you're not going to do Jackass stuff like that, then while troubleshooting a DUT, you're likely only going to brush one of the secondary outputs by accident. But the danger of that is that you won't notice it. Perhaps the UK building site transformers are better in that respect. A 55 V shock is enough to keep you alert :phew:, but doesn't seem enough to kill.

At work we have a variac that has insulated windings (warning: not all variacs are isolated!) and I tested it in increments of 10 V all the way up to 250 V, every time touching one of the secondary outputs. I had expected a slight tingle because of capacitive coupling between primary and secondary, but I could feel absolutely nothing. And to me that seems a danger in itself, but at least it's not lethal.
« Last Edit: May 07, 2016, 08:32:22 am by jitter »
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 4136
  • Country: gb
Re: All isolation transformers must be safe!
« Reply #10 on: May 08, 2016, 07:11:34 am »
Someone will probably pull me up on this as an example of deadly somehow but the benches I used to work on when I was a Telly repair tech were completely isolated from earth, all test equipment, DUT and everything else was run from the isolated supply.

It was possible to touch the live or neutral (relative terms as they were floating with respect to earth) wiring and feel absolutely nothing as long as you weren't leaking to earth (rubber floor covering on top of MDF or chipboard flooring and ply lined walls made it difficult to leak)

Nothing earthed was within six feet of the bench and even then it was enclosed.

It was entirely possible to 'scope the primary of a switchmode PSU in that environment and often very useful because at the time switchmode supplies were regarded as a dark art by almost all techs across the industry (personally, I loved em).

The transformer was a site transformer, I think around 1KVA.
 

Online ShockTopic starter

  • Super Contributor
  • ***
  • Posts: 4214
  • Country: au
Re: All isolation transformers must be safe!
« Reply #11 on: May 08, 2016, 09:21:43 am »
But, all this is just a waste of time if you forget to unplug the mains and stick your hands on the stingy parts :).



« Last Edit: September 22, 2017, 08:17:33 pm by Shock »
Soldering/Rework: Pace ADS200, Pace MBT350
Multimeters: Fluke 189, 87V, 117, 112   >>> WANTED STUFF <<<
Oszilloskopen: Lecroy 9314, Phillips PM3065, Tektronix 2215a, 314
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf