Isolating transformer.

[Dick Glennon]

I would like to purchase an isolating transformer to give me extra protection when working on vintage valve radios. I do not want anything too extravagant but still adequate to power the radios. Any advice would be most helpful. Thanks, Dick.

[Herald1360]

A 100VA rated transformer should be adequate for most vintage radios, only really needed if you want to connect grounded test equipment to non-isolated sets. You might get lucky and find a 240:240 site type transformer at 500VA or more on offer for pocket money

Ensuring the chassis on non isolated sets is connected to neutral and having an RCD on the supply is arguably safer if you don't.

The whole subject has been exhaustively (exhaustingly?) discussed many times over if you have a search around the forum.

[Techman]

There's one at a bargain price in the 'for sale' section right now, but it's collection only from London. I've actually got one that I use on occasion that's the same and branded 'RS', they're good transformers.
https://www.vintage-radio.net/forum/...d.php?t=162964

[stuarth]

Without an isolating transformer, then on live chassis sets with the neutral connected to chassis, an RCD should give protection if you get the mains connection swapped, or if you touch HT and an earth at the same time.

With an isolating transformer, you have electrical separation, and you will not get a shock if the mains connection is swapped, or if you touch HT and an earth at the same time. You can also connect earthed test equipment to the set’s chassis.

Neither scheme protects you from shock if you touch HT and chassis at the same time.

Note that if you have a 3 pin mains socket on the output of an isolating transformer, the earth pin should not be connected to mains earth arth or the transformer secondary.

Stuart

[ajgriff]

Originally Posted by stuarth

With an isolating transformer, you have electrical separation, and you will not get a shock if the mains connection is swapped, or if you touch HT and an earth at the same time.

Really?

Alan

[stuarth]

Yes. If you feed a live chassis set through an isolating transformer, there is no return path from HT to earth as there would be if you just used an RCD.

By “an earth”, I meant something connected to supply earth, not the set’s chassis. Neither an RCD nor an isolation transformer can protect you from touching HT and the set’s chassis at the same time.

Of course, if you connected earthed test gear such as a signal generator to the set’s chassis (not possible without the isolation transformer), then you no longer have electrical separation between the set and earth, but if you are using such test gear, you are probably beyond the beginner stage, and would know that HT bites.

Stuart

[crackle]

Originally Posted by stuarth

Note that if you have a 3 pin mains socket on the output of an isolating transformer, the earth pin should not be connected to mains earth earth or the transformer secondary.

Stuart

Please can you explain that in more detail with reasons.

The reason I ask is my isolating transformer has the output earth connected to the output neutral.
Thanks
Mike

[ajgriff]

Thanks Stuart. I think it was worth clarifying your definition of 'an earth'. (Ref post#6)

Alan

[ajgriff]

Originally Posted by stuarth

Neither an RCD nor an isolation transformer can protect you from touching HT and the set’s chassis at the same time.

I'm not about to test out the theory but I think that there's a fair chance that an RCD would trip in these circumstances depending on a number of variables. However, the unfortunate victim would have no chance of avoiding a potentially lethal shock when using an isolation transformer.

I know that the subject of isolation transformers crops up regularly but I think it's worth stressing that they should only be used when the user fully understands the risks involved. Personally I only use one for specific work on SMPS powered equipment or a live chassis. In most circumstances an RCD provides far better protection in my view.

Alan

[See_Mos]

I have a 1KVA isolation transformer but rather then drag that from the garage when needed I also have a small box with two similar toroidal transformers with the same secondary voltages connected back to back, of course they don't have to be toroids but they take up less space than laminated transformers

[mhennessy]

Originally Posted by crackle

Please can you explain that in more detail with reasons.

The reason I ask is my isolating transformer has the output earth connected to the output neutral.
Thanks
Mike

This has been discussed many times on here and other places. In short, disconnect that earth ASAP.

It's hard to pick one single thread or post that covers this as there are many. Just a few:

https://golbornevintageradio.co.uk/f...d.php?tid=4593
https://www.vintage-radio.net/forum/...d.php?t=134263
https://www.vintage-radio.net/forum/...d.php?t=134177

If you don't want to wade through all those, get the BBC document that I frequently upload and jump to appendix 1 at the back.

Hopefully this direct link will work; otherwise it's in post #2 of the Golborne thread or post #18 of the second link.

https://www.vintage-radio.net/forum/...4&d=1488109603

The PDF is no longer on the BBC's website, but it's vitally important reading so feel free to distribute.

[stuarth]

Hi, Mike.

For an isolation transformer with no earth connection, the secondary is electrically separate from the supply and there is no concept of live and neutral, they are completely interchangeable.

If you connect the earth pin on your isolation transformer output socket to one end of your transformer secondary (but NOT to supply earth) you have defined one side as live, the other as neutral, but you should be mostly OK because you still have electrical separation from the supply, and most live chassis sets have a 2-core mains lead with no connection to your earth.

Some “live chassis” sets have a split chassis with an earth connection and a 3-core mains lead, for example I have an AC/DC Eddystone radio, where the live internal chassis is mounted on insulating bushes to the earthed outside casting; or you could have a record player with a live chassis amplifier and an earthed record deck. For such equipment, you would have to be sure the live chassis connection was to the correct side of your transformer, otherwise you would have 240V between the two parts of the chassis. Connecting a centre tap on the transformer secondary to the output earth pin (it has been suggested) means there is always 120V between the two parts of the chassis.

Connecting the earth pin of your output socket to mains earth has problems for a completely different reason. As pointed out by Mark Hennessy in a post some time ago, if you plug a piece of equipment which has the neutral and earth wires crossed in the plug into an isolation transformer which carries the mains earth across to its output socket, the chassis of such equipment sits at 240V wrt mains earth.

The safest thing is to have no connection to the earth pin on the output socket of an isolation transformer (puts on fireproof suit).

Stuart

Crossed with aforementioned Mark!

[Refugee]

It forms a regional earth at the point the earth clip of the scope is connected.
If you connect up to a telly with a bridge rectifier the secondary of the isolation transformer will float with the two ends of the secondary at 120V as if it were center tapped.
Nothing is completely shock proof on an isolating transformer unless it is a sealed wall wart.

[stuarth]

Hi Alan.

If you touch HT and chassis of a live chassis set connected through an RCD, the currents through the live and neutral of the RCD will be balanced, and the RCD will not trip. If the RCD tripped because of the current through you, then it will also trip from the normal load current through the set, the two paths are in parallel. I would strongly advise you not to "test the theory".

If there is a path through you from HT or chassis to a mains earth, the RCD will trip, but the isolation transformer will protect you from getting a shock in the first place. Note that RCDs are officially classified as a secondary safety device, not for primary safety. Not including the safety issues of getting to the RCD in a blacked out house or upsetting other members of the household who were just about to press "save".

Stuart

[kalee20]

Originally Posted by stuarth

If you touch HT and chassis of a live chassis set connected through an RCD, the currents through the live and neutral of the RCD will be balanced, and the RCD will not trip.

Agreed. You'll get a shock. Just as you would if you touch both wires of the AC coming from the isolation transformer. And no, the RCD will not trip.

Originally Posted by stuarth

If there is a path through you from HT or chassis to a mains earth, the RCD will trip, but the isolation transformer will protect you from getting a shock in the first place.

You won't get a shock, but why would the RCD trip? All you're doing is pulling 'something' on the fully-floating secondary side to earth. Whether it's a 0.1Ω length of copper wire or 100kΩ of human body, there's no complete path for current to flow, much less an unbalanced one.

If you manage to trip an RCD with an isolation transformer, whatever's connected to the output, there's something wrong with the isolation transformer.

[stuarth]

Ah yes, sorry I wasn't clear. I was assuming you had either an RCD, OR an isolation transformer, not both.

Perhaps I should have said "....an RCD will trip, whereas an isolation transformer will protect you from getting a shock in the first place"

There's no point in putting an RCD on the output of an isolating transformer with a fully floating secondary (what other types are there?) because, as you rightly say, as it will never trip.

Stuart

[ms660]

Drawing a schematic of the arrangement can help, for one example do one for an isolating transformer with the mains earth permanently connected to one side of the secondary winding, include the secondary windings connection to a bridge rectifier input DUT with a scope probes earth clip attached to the bridge rectifiers output....

I would ditch that permanent earth connection to the isolating transformers secondary as others have said.

Lawrence.

[stuarth]

If you have an isolating transformer with the mains earth permanently connected to one side of the secondary winding, the situation is little different to just using the mains without an isolation transformer. You have no separation of the transformer secondary from the incoming supply, one side of your supply is held close to earth potential.

If you feed that supply to a bridge rectifier, and then connect an scope probe earth lead to one side of the bridge rectifier output, then you have effectively shorted out one arm of the bridge.

For an isolation transformer to be effective there should be no connections to the secondary other than the two ends connected to the set where mains live and neutral would go.

There should be no connection to the earth pin of your transformer output socket to either mains earth or the transformer secondary.

Stuart

[ajgriff]

Originally Posted by stuarth

If the RCD tripped because of the current through you, then it will also trip from the normal load current through the set, the two paths are in parallel. I would strongly advise you not to "test the theory".

Agreed. However, the human body would not behave like a couple of simple resistors in most circumstances even if contact were to be precisely simultaneous and there would still be the likelihood of a current imbalance. My opinion only of course and I'm definitely not going to test out the theory nor would I suggest that anyone else should do so.

Alan

[Skywave]

Originally Posted by stuarth

Note that if you have a 3 pin mains socket on the output of an isolating transformer, the earth pin should not be connected to mains earth or the transformer secondary.

And the reason for that is . . . ?

Al.