Test Bench - Isolation Transformer grounding question.

[bdunham7]

That is true but you can also make sure to connect the chassis to neutral (in case that is equal to ground).

That often won't work as planned as the chassis may not simply be connected directly to one prong of the cord.  In some designs, the chassis is connected to the negative of a bridge rectifier or full-wave doubler, resulting in there being a complex but exciting voltage present no matter which way the plug is inserted.  In others, like the OP's Zenith radio, the chassis is connected to one prong, but through the power switch.  The result is that if you insert the plug so that the chassis is at neutral during operation, it will still become live (through the filaments and other circuitry) when you turn the unit off. 

[m k]

The drawing in the original post has been updated for clarity, and it now shows a 2-way toggle for the secondary side ground lift.

Why there is a switch, why you want to lose the protection?

[IanJ]

Here's my input.

I think it's good practice to call the two conductors on the primary side LIVE & NEUTRAL for obvious reasons, but on the secondary side of an isolation transformer I think it's best to call them L1 & L2. Thats what is done in the offshore industry anyways.

I use my own isolation transformer all the time.
I do have the option to connect over the earth connection to the secondary side 'earth' connections of the output power sockets, but I don't use a switch....too easy to have it in the wrong position.....rather I have two banana sockets where I need to physically add a short jumper to achieve it.

My isolation transformer is originally a medical unit which I modified. It had the incoming earth connected right through to the outputs permanently.
I also added a fuse to the output L1 which it didn't have originally.
It also has 240VAC/110VAC output voltage selection.....which is handy.

Extra precaution as you'll see in the pics is that the UK output socket that you can see plugged in is a metal case (usual for UK), but it's insulated with a 3D printed cover. Why?.....imagine sitting the metal socket on top of a metal cased bench test instrument.....instant connection of incoming to outgoing earth.

[m k]

Explain situation or two where Protective Earth is dangerous.

Counter example.
Faulty device has L1 live case and other case has L2.

I didn't understand metal case (usual for UK) part.

[IanJ]

Explain situation or two where Protective Earth is dangerous.

Counter example.
Faulty device has L1 live case and other case has L2.

I didn't understand metal case (usual for UK) part.

Hi,

If you are referring to my post, I never said anything was dangerous. To me it's just about having options, to know exactly what's going on at all times and being safe to ourselves and safe to our bench equipment.

Regarding the metal case on my flying dual socket:-
So, the flying metal case is ultimately connected to the output earth banana, and lets say I am using the isolation transformer without the earth jumper (bananas) in place, but then sit the flying metal case on top of a different instrument (not the one I am testing) in my workshop that has it's own case that is earthed, then the output earth would be connected to the incoming earth.....as if I had the banana jumper in place. So, to avoid this I 3D printed a cover for it. I want to be sure that what I set up is what I get at all times.

Ian.

[robert.rozee]

just to add my 2 cents worth from the lands down under...

note: we use the term ACTIVE for 'hot' here (both in NZ and across the ditch in Australia), but i will use the term HOT so as to not confuse things further. we also have NEUTRAL connected to GROUND via a single link at the distribution board.

1. a cardinal rule of mains wiring (here in NZ) is that swapping HOT and NEUTRAL should never create an unsafe situation. if any piece of equipment were to contain an internal connection between EARTH and NEUTRAL, that would violate this rule if said equipment were plugged into an outlet with HOT and NEUTRAL reversed.

2. with the one exception of double-insulated equipment, all exposed metal parts must be connected to EARTH. this ensures that you can not receive an electric shock via contacting said metal parts.

the above are enshrined in law.


applying the above two rules, (a) your isolating transformer should have the metal core along with any metal casing components connected to EARTH. and (b) the incoming HOT and NEUTRAL must be kept isolated from EARTH. if these conditions are not met, the equipment has the potential to create a hazard, as well as being illegal.


as for the secondary, well the terms L1 and L2 are an apt choice here, indicating the secondary 'knows' nothing about HOT and NEUTRAL. L1 and L2 are not a part of the mains supply system, however do need to be treated with care as a hazardous voltage exists between them.

EARTH, while advisable to be connection to the DUT, is optional as your isolating transformer creates a safety barrier. while current can flow between L1 and L2, it can not flow from the DUT down the EARTH wire.

having said this, there is no sensible reason to connect either L1 or L2 to EARTH that i can think of, as there is simply no path for the secondary current to flow through apart from between L1 and L2. as others have said, with L1 and L2 kept floating, you are free to work on 'live chassis' equipment with that chassis connected to the same EARTH as your oscilloscope or other expensive/fragile equipment. said 'fragile equipment' includes the service-person!


cheers,
rob   :-)

[RacerR]

I think it's best to call them L1 & L2.

I've updated the labels on the drawing in the original post.

rather I have two banana sockets where I need to physically add a short jumper to achieve it.

I like this idea - I was thinking about using a switch with a "warning" lamp, the jumper is simpler.

[m k]

You're still live and kicking, good.

More meaningful would be that fuse IanJ have added.
Bad area for your gear is between out from isolator and any means to limit power.

My room is, I guess, fused by a single 10A and I can't always turn on my variac.
But all smaller stuff are still cooked without a notice.

For test purposes the isolation transformer is allowed to have a single output only.
Medical area is different, there generally everything must be isolated.

[RacerR]

More meaningful would be that fuse IanJ have added.
Bad area for your gear is between out from isolator and any means to limit power.

Thanks - yeah, I still need to decide on fusing, power and standby switching, indicator lights, and the actual device layout.  I'm using a 250VA transformer with a secondary current rating of 1.7A, and I think a 2.5A fast blow fuse is typically used on the secondary side - though maybe I could get away with something smaller.

-Rob

[TheBay]

Earthing is often a topic on this and a lot of confusion, it totally depends on where you are in the world, what rules apply, use cases and what type of mains sockets/plugs are used too.

I'm in the UK and on a PME system (TN-C-S) so the Neutral and Earth are combined and then split at my property.
I have an isolation transformer in the workshop, I only use it when I need to use it.

The mains earth is connected to the metal chassis of the isolation transformer as it is a Class 1 appliance, the mains earth is also connected to the socket that is on the secondary side of the transformer. In the UK this is how they are sold. Neutral and Earth are never connected together at the appliance.

 I have no reason to remove the mains earth connection to the secondary. It is beneficial for me to leave that connected for various reasons, such as filtering etc.
There is no risk of additional danger by having that connected here, the only condition that could arrise which could cause an issue is having an appliance with the plug wired wrong such as a N/E or L/E reversal.

As I am fully aware of this scenario it is extremely unlikely this would ever happen here. Firstly the UK BS 1363 plug is polarised and cannot be put in to the socket on the transformer incorrectly. If I encounter something that does not have a moulded plug I always pop the lid off and check that the wiring is okay, if it's a moulded plug I check the pins back to the internal wiring of the appliance. Every IEC C13, IEC C5 lead that enters here is checked, regardless of where it came from or even if it came with a product brand new, that is something I have always done regardless of using an isolation transformer or not.

Lastly if I personally do need to use the isolation transformer all the covers will be removed on the DUT and everything throroughly checked before hand. I cannot think of a scenario here where I would want to use an isolation transformer without doing so.

Isolation transformers can have their place and should never ever be taken for granted, they can give a false sense of security and should only be used when absolutely necessary. They are quite often purchased and used for the wrong reasons where other test equipment would be more suitable, such as differential probes etc or methods for testing and correct safety precautions.

Regardless of whether you have a mains earth passed through to the DUT or not, there are many situations that they can become dangerous. And the RCD/GCFI on the primary side will not save you, it is completely oblivious to anything connected to the secondary side.

[m k]

More meaningful would be that fuse IanJ have added.
Bad area for your gear is between out from isolator and any means to limit power.

Thanks - yeah, I still need to decide on fusing, power and standby switching, indicator lights, and the actual device layout.  I'm using a 250VA transformer with a secondary current rating of 1.7A, and I think a 2.5A fast blow fuse is typically used on the secondary side - though maybe I could get away with something smaller.

-Rob

Fuse here is a bit different.
Normally it's preventing fire hazards, like it still is in the primary side and in DUT, but when wall side can give practically what is wanted this secondary can't give more than it can, you can think it as a power plant of its own.
So secondary fuse is like power company protecting their distribution network.

For transformer the main problem is heat, and literally melting insulation, so it's not very picky.
More picky it can be if you're switching it off and on when it's already hot.
Transformer can also heat up quite a bit and still be totally happy.
Class A, first insulation design temp is 105C and if core is 30C below it's still pretty warm.

If you use things that are overloading your gear much then the fuse must stand that.
Means that voltage goes down and current goes up, for a moment, and fuse must be happy all the time.
So slow fuse is possibly better.

For other nuances I'd say that no switches, no lights, just a plug and a jack.

http://www.smartgauge.co.uk/iso_wire.html
https://www.eevblog.com/forum/beginners/isolation-transformer-102532/

[bdunham7]

I'm using a 250VA transformer with a secondary current rating of 1.7A, and I think a 2.5A fast blow fuse is typically used on the secondary side - though maybe I could get away with something smaller.

You typically don't need a fast fuse for a transformer.  A slow-blow like the MDL 2A would probably work best.

[TaylorD93]

I think it is always a tricky one with Isolation transformers.

The Incoming Earth should ideally be connected, and it is in my unit, but It does complicate matters sometimes. Having a few Analog CRT Scopes im in the middle of repairing, the Chassis (Earth) is also the 0V of the Power supplies etc. So probing any of the HV circuits referenced to Earth can be difficult as most probes arent rated for the high voltages.

Even some Differential probes arent quite man enough.

My "work around" as my Isolation transformers Earth connection is hardwired, is to have a separate mains lead with Red heatshrink sleeving along its length to clearly identify it. That way its much clearer than usual.

[dannybeckett]

How bizzare... I've just finished building myself an isolation transformer too!

Learned a lot about inrush current and remanent magnetization. Earth will be passed thru to the socket but not conected to the iso transformer(s) in any way.

Check the thread for simulation and schematics of inrush limiter: https://www.eevblog.com/forum/chat/1000-va-230-vac-50hz-how-much-inrush-current/