Sensitive RCD for workbench outlet

[unitelex]

I often work alone and on mains equipment.
I am considering improving the safety of my workbench AC supply.
Main consideration is minimising electric shock or fault currents in the event of the slip of the test probe/screwdriver etc or accidental touching of live terminals

My domestic consumer unit already has RCDs for each 32A ring main as typical. I believe the residual trip current for these may be 30mA. This being detected as a difference in current between the Live and Neutral conductors. I prefer not to rely on that because a) it is not very sensitive and b) the other family members are very sensitive about tripping and interruption to their online gaming or TV viewing!

For my workbench, for the UUT power outlet specifically I would ideally like a more sensitive RCD which trips before the domestic RCD. It would also be nice if it had an adjustable sensitivity and even a meter showing residual current levels.
I might even combine it with my Variac and lamp limiter into a single unit.

I found a commercial RCD which is adjustable down to 6mA:
https://uk.rs-online.com/web/p/rcds/5119121/
P/N ELR01PN 24-230V AC/DC

Maybe there is a homebrew solution out there?

BTW I am aware that an RCD cannot protect from electric shock from isolated secondary HT circuits, eg secondary HT and secondary chassis gnd.

Has anyone here done this already or has any specific experience with this kind of RCD, or any alternative suggestions to achieve the safety improvements without nuisance tripping the domestic RCD.

Thanks
Chris

[Station X]

My experience of RCD's in series is that you can never be sure which one will trip first. In some cases they'll both trip.

CU's fitted with dual RCD's generally have unprotected ways which I believe are intended for feeding smoke alarms and freezers. My solution was to fit a separate CU in the workshop fed from one of these ways. This CU has dual RCD's with the lighting circuits split between them so that I'm not left in the dark if one RCD trips.

This is one case where an isolation transformer could be really useful, the house RCDs won't trip.

[Chrispy57]

Are you sure about that 6mA figure? Looks like it's 30mA up to 30A for the sensitivity, and while the trip time says "instantaneous" a closer look at the small print reveals 25mS - so not much different to the basic 30mA, 30mS units?

When I worked in hospitals the electronics workshop, theatres, ICU etc were fed from 10mA, 10mS RCDs which were regularly tested. You may want to shop around for something similar?

Cheers
Chris

[unitelex]

Hi Chris
The product I linked to is available in a variant which is adjustable down to 6mA

I particularly like the adjustable feature, can set according to the UUT to avoid nuisance tripping.

[unitelex]

Yes, I have one somewhere. I think it is toroidal, around 100VA
Will consider that as an option also.
Thanks
Chris

Originally Posted by merlinmaxwell

This is one case where an isolation transformer could be really useful, the house RCDs won't trip.

[ajgriff]

Originally Posted by unitelex

The product I linked to is available in a variant which is adjustable down to 6mA.

Can you provide a link to information about the 6mA version? I'm intrigued. Thanks.

Alan

[Chrispy57]

OK - that sounds more reassuring, especially if fed separately from the house RCD circuits as mentioned above. Good luck with that project. I actually have an RCD Test Set made by "FDB" that has current settings from 3mA to 1000mA that you would be welcome to borrow when it's up and running if you wish, since we're fairly local to each other.

Cheers
Chris

[Red to black]

BTW domestic RCDs of 30 mA or less are meant to trip at within 40ms at Idn X 5 (I delta n) X5, ie. the nominal trip current X5, or in other words @150ma.
Depending on which standard they are built to (which I would have to look up) they are only meant to trip within 200ms or 300ms at their rated nominal trip current.
Bear in mind to that they are not supposed to trip at 1/2 Idn (15mA for a 30mA device) anything beyond and up to their rated Idn they are supposed to trip within the timeframes given above, typically a 30mA device will trip at about circa 22mA in practice.

Graham (station X) above is correct in that it is difficult to get RCDs in series to discriminate, ie.which one trips first.
The General rule is for the downstream RCD to have a 1/3 the nominal trip current as the one preceding it, however if the downstream one 'sees' more trip current than the Idn upstream device is rated for then in practice both usually trip anyway.

When you start with upstream devices of over 30mA, usually these are 100ma or 300mA or above then these are time delayed to provide the discrimination between the downstream device.
These TD devices are of course not suitable for the requirement of protection of final circuits and are primarily used for distribution circuits where either the Earth loop impedance is high (eg. TT systems), or where mandated for fire protection such as horticultural sites etc..

[Herald1360]

If you feed your whole workshop via an isolation transformer and tie its output "neutral" to earth at the transformer, a 10mA RCD on the output will offer its usual protection without any current through you upsetting the main board RCD. And you won't need to run another feed from the consumer unit either.

An unprotected feed could still be cheaper than a transformer though, depending on time, trouble and cost needed to run it in vs cost of transformer.

[Red to black]

Hi Chris, at least someone gets it ,
Usually these type of threads descend into urban myth and everything in between.

Another better way of doing it is not to continue the "mains Earth" through at all, connect your isolated "neutral" to the outgoing earth conductors only on the secondary side, this way is much preferred as you can tie all of your bench supplied CPCs (circuit protective conductors, ie. "earths") together as per BS7671 and still remain "earth free" and the outgoing RCD will still work without tripping the supply side (house one/s) although in this case the bench RCD won't provide true earth fault protection and is rather superfluous anyway.

Or do a 230V centre tap to earth isolation job, half the "shock" voltage albeit on both sides. 115V isn't a very shocking voltage.

[eddie_ce]

Just to add my 2 cents worth, as has been stated wiring standard RCDs one after another is not a good idea unless the RCD upstream has a time delay because there will be no selectivity.

On a slightly different subject, a trick I have used to increase the sensitivity of an RCD is to wire two pairs of inputs in a 4 pole RCD in series, turning a 30mA device in to a 15mA beast. However, I do realise that a 4 pole RCD is not as common in the UK as they are here, where almost all dwellings have a 3-phase supply, so consequently here 30mA 4-pole RCDs are the norm and much easier to source than 2-pole 10mA RCDs which are often fitted on bathroom circuits.

[unitelex]

It is Broyce Control type ELR01PN

Originally Posted by ajgriff

Can you provide a link to information about the 6mA version? I'm intrigued. Thanks.

Alan

[unitelex]

Ok thanks for your kind offer Chris, will consider that when I get around to this project.

Originally Posted by Chrispy57

OK - that sounds more reassuring, especially if fed separately from the house RCD circuits as mentioned above. Good luck with that project. I actually have an RCD Test Set made by "FDB" that has current settings from 3mA to 1000mA that you would be welcome to borrow when it's up and running if you wish, since we're fairly local to each other.

Cheers
Chris

[unitelex]

Thanks to all for your insight and suggestions.

I am still considering how best to do this bearing in mind various fault conditions and test scenarios...

Following good advice here I think I will most likely use an isolating transformer for the UUT (unit under test) outlet at least. I might use a second isolating transformer for my TDS224 oscilloscope to avoid risk of fault currents through the scope ground clip and avoid the scope ground clip tying the UUT ground back to mains earth.

The other equipment eg DC power supply has an isolated secondary anyway.
DMM is battery operated.

Best Regards
Chris

[unitelex]

Thanks to all for your insight and suggestions.

I am still considering how best to do this bearing in mind various fault conditions and test scenarios...

Following good advice here I think I will most likely use an isolating transformer for the UUT (unit under test) outlet at least. I might use a second isolating transformer for my TDS224 oscilloscope to avoid risk of fault currents through the scope ground clip and avoid the scope ground clip tying the UUT ground back to mains earth.

The other equipment eg DC power supply has an isolated secondary anyway.
DMM is battery operated.

Best Regards
Chris

[Bookman]

Upon reading this I wondered at how and why one would wish to detect 6mA currents in a domestics or industrial environment. In this I would refer back to research I had undertaken back in the early 1980s in relation to earth fault analysis and the protective devices available at that time.
Disregarding system stability as it does not apply and indeed nominal Voltage, let us consider the impact upon a hazardous environment; i.e. hazardous to human life.
It would seem that the heart of a man is fundamentally capable of withstanding a current flow in his body of values up to 59mA (anonymous) before the possible onset of ventricular fibrillation. A women on the other hand can withstand up to circa mid 70s mA.
These values, by the way disregard any implants that may be within the body that could act as a conductor.

It would seem however that the advent of electric car DC charging systems has created the need for such low currents but these are DC currents and furthermore are not classed as leakage but are fault currents (Western Automation and R&D).
This fact notwithstanding your wish to implement an RCD must not be in series with an upstream similar device in the same house hold unless of course you implement an isolation transformer as stated hereof.
You can then think to yourself that you are considering “Earth Leakage” only and not earth fault protection. Clearly, to this end all you would require is a current sensing device, a control circuit and an output contact be an alarm or isolation relay via a mechanical or solid state switch.
The current sensing device could be a core balance (zero sequence) or a ring type which you could make by using galvanised piping and wire as in the old days.
Good luck with your endeavours - a possible worthy project.
Joe