RCD Tripping on AMF switch-over to generator supply

[qwest]

Hi, I'm looking for some help and hoping I might find it here.

I am a house owner with a 10Kva stand-by generator (40Amp / single phase). I had the generator professionally installed by my local electrician but I supplied the Automatic Mains Failure switch and generator.

The set-up works perfectly with one issue - there RCDs trip. Given that the backup is supposed to be fail-safe this is far from ideal. The electricians that carried out the install have never worked with this type of equipment before so are not able to support the problem.

The generator is spiked to earth. The utility supply is PME. The PME earth is permanently connected to the generator earth spike so I believe the earth should be common to both generator supply and utility. There is no switching of earths.

The generator supplies the AMF panel through a 100mA RCD located at the generator.

The house is protected by a 100mA RCD at supply to the Consumer Unit (why it isn't 30mA I don't know). The consumer unit in this case is old (and may not actually be a consumer unit - I am not sure of the correct terminology) it doesn't contain RCDs - they look like RCDs but just break on current rather than earth fault.

The fault occurs on main failure (which my AMF panel can simulate). The generator starts up, stabilising for maybe 10 seconds and then the contactors switch over to generator supply. When this happens one of 4 situations occurs:

1. It all works as it should
2. RCD at gen trips
3. RCD at CU trips
4. Both CU & gen RCD trip

Some things I have noticed when trying to trouble shoot:

1. It isn't load related. There must be some load (it I turn all the breakers to off the fault never occurs), if I leave just one or two low current circuits operational then the frequency of RCD tripping remains the same as if full load.
2. I don't think there is an earth leak as once the RCDs trip and I flick them back on there is no recurrence of the problem.

I am now lost as to how I can trouble shoot further. I am not qualified as an electrician but have a reasonable understanding of electronic engineering. I don't particularly want to do this myself but I have struggled to find an electrician that can help.

I've seen a very similar thread here but couldn't find it a second time round for some reason - similar issue to what I have was discussed, alas with no actual diagnosis of problem.

So if I can get any indication of what to try next here that would be a huge help! Thanks.

 

[Darkwood]

Contact the suppliers of the gen' they will be aware of all issues customers will have had and possibly simple solutions to rectify such problems.....

Im a little saddened to hear your electrician has left this issue to you... has he tested the installation, monitored for earth leakage existing etc...there are many steps he can take to rule out several causes....but as an electrician doing the same job for you even without experience of the gen' system id personally put it on myself to ring tech' up and discuss possible issues they may know that cause such problems.... have you tried the mcb's one at a time to see if any particular circuit is the cause or whether this still occurs regardless????

 

[qwest]

Hi,

Electrician had a limited role. I did a bit of the cabling and donkey work. Just wanting to keep costs low.

The gen is about 6 years old and is being re-assigned this task. The background to this is that it isn't necessary - I had the gen spare and it was nice to have the back-up power as we have frequent power cuts 4 to 6 a year affecting things like heating, electric gates to the house and the fridge etc.

It isn't worth spending the £15k to £20k to do it with new kit specialist supplied and installed. But given I could get an AMF for £1,000 and the only other thing needed was some chunky copper I thought it would be simple - largely it was but this is a bit of a niggle I'd like to resolve.

I don't know what kind of earth I have on the gen - I think the neutral is tied to earth but not sure how to check - or even if that is accessible to check. The gen was originally in a truck and earthed to the truck chassis. That was professionally installed and it was used to power a range of things from lighting to computers for a mobile sales office.

I don't think I have MCBs - not the type that detect an earth fault. Just one RCD on the utility / gen supply to the bank of breakers for the individual household circuits and one RCD at the gen.

Switching any combo of breakers off in the "consumer unit" makes no difference. They all need to be off and then there is no issue - so I am confident there isn't an earth leak on the household circuits or appliances. I am also confident that this isn't an influx current from every circuit being switched on at the same time. Any load on any circuit could result in the RCD tripping on switch over.

The suggestions are appreciated though.

 

[Darkwood]

With so little info we can use here its going to be hard to suggest anything ....only an onsite visit by an Electrician experienced with Gen' systems is your best bet ..ive done a few myself but couldn't even contemplate throwing wild guesses at you ... Safety devices are tripping here and their may be a genuine reason maybe linked to the set-up...... without investigating first hand and testing the install i can't comment.

So my advice holds you will need to get in a competent Electrician or Engineer in to check out the install.... if its set-up incorrectly you or your family may be at risk.....

Your cutting down on costs here and doing a lot of the install yourself yet 'no offence meant' you seem unclear, vague to the gen set-up which in my experience is asking for trouble .. a little knowledge is a dangerous thing regarding electrics... and as you freely admit the electrician wasn't experienced with the generator he was in no position to be doing the job or advising you whether it was all done correctly.

I would be totally unprofessional if i wasn't to say disconnect it and get a experienced pro' in before it is used ....its your family at the end of the day and an area not worth the risk of taking shortcuts.

Try not to see it as a dig at your work or anything but your general vague responses means its crucial the work is checked over when using generators its crucial they are set-up correctly in line with the existing installation.

 

[Marvo]

Is the DNO neutral being isolated as well as the live before the contactors connect the generator supply to the CU on power fail? From you explanation I'm immediately worried about how this generator may have been connected and whether both poles of the DNO supply are being isolated.

Can you upload some photos to a photobucket account (or similar)? Photos of the AMF switch wiring and photos of the DNO head with the main earth terminal also photos of the CU with the cover on and the cover off.

I'd also emphasise Darkwood's view that there's lots that can be wrong with a back-up generator install, incorrectly installed generators can be lethal to the homeowner and also to DNO linesman so a professional electrician would be the way to go.

 

[qwest]

Not sure what a DNO is.

But from your question I guess you are asking what is switched within the AMF.

Within the AMF There is a relay that can put a PD across the energising circuit of one of two contactors. Never both. One contactor carries the utility L and the utility N to the CU via the household RCD. The other contactor carries the Gen L and Gen N to the CU via the same RCD. Each contactor is 100 Amp (I assume this is a continuous rating). The main RCD is 80amp and the gen is 40amp. So the contactors are more than adequately robust - and sound like a couple of bricks being clapped together when energised. They are both normally open.

The earth in both cases are common and not switched.

To this extent this install is competent. - No danger to linesmen since it is not possible to energise both contactors simultaneously.

There can be no risk to the household - that is double protected by RCD and the gen produces the correct voltage. At no time is anything unearthed and the house now has its own earth spike.

The system is all low voltage - there is no PD above 315 anywhere.

The only element I took care of was wiring the 12V system for auto start sequence on the gen and a couple of wires that determine PD on utility and gen circuits to allow the computer to determine the sate of supply and whether to start or stop the gen and what to switch.

Everything else was electrician installed but since I supplied the AMF and the issues could be with that then they won't support. I can accept that.

My suspicion is that on switching there is a momentary imbalance causing the RCD to trip or that the additional earthing is causing an issue - but in my view it is necessary as with PME it may be possible for a total loss of earth if there is a fault on N upstream somewhere. This would then leave my gen supply with no earth. So for that reason I took the decision to earth spike the gen.

I have measured the PD on L to E and P to E and L to N on the gen and there is nothing to be suspicious of. So I can only think that it is something momentary or some issue with the Earth.

 

[qwest]

^ DNO.... Distributor Network Operator.... just looked it up!#

Yes, any time I am making power they are isolated. Not interested in lighting up the rest of the street!

 

[Marvo]

Where is the power coming from that operates the contactors in the AMF panel? Reason I ask is maybe the initial inrush current of the contactor coil would trip the RCD if it's wired to a non-RCD neutral but once operated the lower holding current through the coil is below the RCD tripping threshold.

 

[qwest]

Utility powers contactor for mains.

Gen powers contactor for gen.

So this theory could be right. There is a chunky contactor operating with inrush current. 100mA is already below spec for a domestic supply. Is there some other type of RCD that may be less prone to this that I could try?

What can I do about it if I am stuck with current RCDs? That is in the design of the AMF panel so a little more tricky to resolve.

I think many AMF panels are a similar design. I thought this was an industry standard so I can't be the only one if this is the issue.

This could also explain why I see the same issue (RCD tripping) on return to utility supply though so the theory sounds good.

 

[Marvo]

Don't replace the RCD to work around the problem, that's treating the symptoms, rather find the cause. I wasn't questioning the general design of the AMF panel I was wondering out loud if the controls may be crosswired some way between an RCD live and a non-RCD neutral for example. To be honest you're unlikely to get your answer from a forum, it's the kind of problem you need an experienced electrician hands for an hour on to get it localised.