I was given a 1:1 isolation transformer that I wanted to use for some measurement equipment that I have been using. I was going to use a common outlet and connect it directly to the primary and similarly connect the secondary to another outlet. From what I understand I need to pass the ground from the primary to the secondary.

I was wondering if I should or shouldn't also connect the ground straight to the neutral on the load outlet? It makes sense to me, but I don't know if there is a better way.

 

No, the common should never be connected to ground at the outlet. The AC common is a current carrying conductor. Ground should only have current in a fault condition.

 

If you connect the output neutral to earth you may as well not bother with the transformer at all - all you have done is reduce the available fault current due to the transformer impedance, there is no reduction of electric shock risk whatsoever!

The above is exactly how the normal public supply system works - the neutral and earth are connected together, the position of this connection varies and may be at your incoming supply position or at the local electricity substation but it is there.

The normal use for isolation transformers is to produce a zone of supply that has no reference to earth at all. The majority of electric shocks are those involving contact with a live conductor whilst standing on or touching an earthed surface - it is unusual to actually grasp the live and neutral simultaneously. Technicians are often taught to work with one hand in their pocket unless it absolutely essential to use both hands.

If the test bench supply has no reference to earth at all then the most common cause of electric shock is removed. If you grasp the live conductor there is no return path to the neutral via earth - hence you survive!

A great deal of importance is placed on the transformer inter-winding insulation to maintain this safe zone and the better transformers have metal screens between windings to ensure a flashover between input and output cannot occur.

As to connecting the input and output earths together this is also a possible hazard since it is possible during fault conditions for the incoming supply earth potential to rise - it is current uk practice to bond the electricity supply earth to the incoming water and gas services to create an equipotential zone to alleviate this problem hoping that all exposed metalwork will therefore be at the same potential.

If an earth reference is needed within the isolated zone to facilitate testing then this is sometimes provided by a completely seperate earth mat - you have to be careful that this is not sited near any earth connection associated with other power systems which as mentioned above can include other service pipes entering the premises.

Steve.

 

So will the neutral and hot in the secondary have no reference and be able to drift? If so does this create a shock hazard?

 

Yes that's precisely it - hence the interwinding screen on the transformer since if a short occurred between live incoming and the secondary everything would be at live potential to earth and that live would have an earth reference so shock hazards would definitely exist.

You could earth the output side by a high resistance if for instance lots of high-voltage work was being carried out and you were worried about charge build up. In fact the leakage current through the resistor can be monitored and the system shut down if it becomes excessive. But if a low impedance path exists to earth then there is no safety gain hence the resistor mentioned above would need to be a robust beast that wasn't likely to arc-over.

In designing these environments it is a matter of balancing the risks - for instance my first foray into the world of work was when I worked as a T.V. engineer as a Saturday job to help fund my way through further education. In those days - valve and early transistorised sets - it was usual to connect one side of the incoming supply directly to the chassis (called live chassis) if you were lucky it was the neutral connected to the chassis but it was often the live. Hence the workbench was fitted with an isolating transformer so the sets were fed from a completely floating supply - hey presto no more fried engineers!

If you search for details you will find information on configuring these environments on a commercial premises basis where the safety of employees is concerned.

Steve.

 

Just scanned the 15th edition UK wiring regs (yes I know the current edition is the 16th) and this is covered.

It specifies a transformer to BS 3535, voltage in secondary circuit not to exceed 500V, no connection to earth, cables run separate from other circuits, all sockets on the output side shall have their earth pins connected together by insulated but non-earthed bonding conductors to ensure metal cases of appliances are at the same potential. Any metal cased appliances must have 3 core supply leads, and it shall be verified that if a fault to 'earth' were to occur in 2 different appliances but from different poles of the floating supply the path from pole to case to bonding conductor to case to pole will operate the circuit protective devices.

Steve.

 

I have a LISN that requires connection to an 'AC Power outlet (with protective ground) according to VDE-0100'. The LISN has a high leakage current (800mA) so cannot be used with a leakage current breaker (RCD).

I have therefore bought an isolation transformer - the manufacturer states:

'...to securely earth the frame...both output connections from our transformer can then be connected to earth via resistors/capacitors/filters. The earthed metalwork will stay at earth potential and our transformer will be referenced to it. As long as you don’t touch our live connections then you will be ok'

Options:
1. Float the secondary as you mention above...but then may the LISN warranty be invalidated?
2. Earth the secondary Live / Neutral with a RLC filter plug - but where do I connect the filter plug earth connection to?

Finally, I see BS 3535 has now been replaced by BS EN 61558 but this standard costs over 100 GBP - which I would pay if I knew that the document would provide me with a solution...Does this standard clearly state how to connect isolation transformers for equipment of high leakage current that requires earthing?

Help.

 

I know there are many out there following this thread... So I am now answering my own question (and hopefully saving others the same problem):

Q. Do I need to earth the secondary side of the isolation transformer or not (and to where)?

Answer: Yes - you must connect the secondary earth to the primary earth connection (normally through a RLC filter) - in fact I have just made a direct connection but will look into buying a filter plug...

The two circuits are not connected even they 'share' an earth connection as the earth does not normally carry a current - so isolation is preserved...!!

Finally, if there is leakage current in the secondary it will NOT pass to the primary circuit (we have physically checked this) - because the secondary Live / Neutral are an isolated network.

This way you can connect a 30mA RCD to the primary side and have 800mA of earth leakage on the secondary - no problem. No current passes to the primary and the 30mA RCD does not trip.

The tests we have made are on a D3717 isolation transformer (240W) from Dagnall Electronics (UK) - who have given friendly and quick technical support - I can happily recommend them.