Should/can I Ground Reference HV transformer secondary?

[Qmavam]

I have the schematic below, for safety concerns I would like to ground one side of the secondary that is attached to the vessel. There is just a lot of exposed metal and I'm looking to make it safer. Does grounding the secondary cause any problem with the transformer since both are connected to ground, primary thru the neutral and secondary connection to ground.
                       Thanks, Mikek

[Ian.M]

Grounding any part of the secondary side will increase the risk of electrocution, as the transformer defeats any upstream GFCI (RCD) protection.

Tackle the root cause of your safety concerns by fully enclosing the apparatus with grounded or insulating guards with interlock switches to isolate the supply if any guard is missing or not properly closed.

[Marco]

Grounding any part of the secondary side will increase the risk of electrocution, as the transformer defeats any upstream GFCI (RCD) protection.

On the other hand you expect metal casings to be grounded, betraying expectations is dangerous in and of itself. There can always be an insulation failure or unintentional grounding of the other side of the secondary.

Your expectation drive you to see the casing as safe, not a floating death trap.

[james_s]

High voltage transformers for neon signs and xray machines have a grounded center tap, making the potential to earth on each side half the total voltage. In this case though I don't think that's practical, the outer metal casing should be earth grounded for safety. If you don't want to ground one side of the secondary then you should have an additional outer casing around the whole thing that is either grounded or non-conductive.

[Terry Bites]

Because there is a risk that leakage or fault currents may be condcuted from your exposed metal parts you need to earth them.
If this is a problem for your experiment, then insulate so that these parts cannot be touched in normal use.
Its good practice to bond and ground all exposed metal parts.
Even a small leakage curent could put your tube and a nearby grounded (or not) conductor at different potentials- that's a shock hazard.

As Ian M says, beware confirmation bais! You wouldnt expect to get a shock from a metalic object.

[Ian.M]

That was Marco's point, not mine.   My point was that grounding one end of the secondary makes the other end a severe electrocution hazard, with no GFCI protection, even if there is GFCI protection on the upstream supply. Presumably there is some conductive liquid in the O.P's centrifuge tube, contributing to the hazard if any leaks or spills.   
 
Also, be careful *what* you ground and and *where* you make the ground connection - get that wrong, and you are one broken wire / bad connection from making the exposed metalwork an electrocution hazard, 'live' with the HV AC.

Note the number of deaths that have occurred in the USA due to mishaps with home-built 'fractal woodburning' equipment, which is commonly based on a repurposed microwave oven transformer.  Like the O.P's proposed setup, that has one end of a HV secondary grounded, + presence of conductive liquids.

IMHO, interlocks and guards to block access to the cap with power on, and to prevent power being applied with the cap removed are essential.

[robert.rozee]

consider connecting the outer casing (aluminium tube) to ground via a high-value resistor, picking a value such that any current flow is non-lethal. for 1000v supply, this may need to be >2MΩ (for 500uA or less current). this ensures that the outer casing is not floating, but instead near ground potential, but with little opportunity for a fault current to flow.

OR, are you able to connect a pair of resistors in series with your high-voltage supply? a series resistor at BOTH sides of the winding may provide some additional protection if they do not interfere with the functioning of the apparatus. at this point you may then be able to safely connect the outer casing (aluminium tube) directly to ground.


cheers,
rob   :-)