Voltage operated ELCB - what's wrong with them?

[davetheglitz]

For the first time ever I've had a chance to play with a Voltage Operated ELCB. Gas bonding was required - and the customer refused to upgrade to an RCD. Decided to proceed with the recommendation that they upgrade to an RCD ASAP. Also couldn't see an earth rod anywhere - assumed it was earthed via a water pipe - so put one in.

What I thought was a bond to the water pipe was in fact a concreted in rod (somewhere) helpfully wired in 16mm meter tail. Since found out that the water came in on plastic.

With the two earth rods on one side of the coil and the CU with bonding on the other, it tripped out OK when I switched on my 15W soldering iron with the neutral wire in the earth pin on the plug (the only quick and dirty test I could think of - no-one else in the house!)

I always understood that you couldn't test voltage operated ELCBs with an RCD tester so could never verify correct operation - but my Fluke 1651 tripped it on 30mA in 116ms on x1 and in 16.5ms on x 5. Seemed fine! Ze was around 55 ohms (very good for down here!)

I know these devices finished in the early 80's - but to my mind they seem simple devices that do the job nicely as as long as the earth is connected.

Note that the system I was working on had 2 independent earth rods - the old concreted over one - and the new one I put in - so the chances of losing the earth would be minimal.

By the same token you can virtually disable an RCD in a TT system with a N-E fault - something that this system would not be susceptable to.

Anybody got any thoughts on why they are not used anymore?

 

[binky]

Hi Dave, they rely on a voltage being present to operate, ergo if due to some strange reason the voltage drops to say for example 120V, they don't work properly, where as if you touch N to E on non-energised RCD protected circuit it trips every time. Surprised you managed to get a test result from it. Last time I tried it made strange noises and failed to trip, but might have been faulty, or doesn't work with meggers

 

[davetheglitz]

Hi Dave - can see your point that the performance would change with voltage - but my understanding is that they would operate at around 45v. Even if the mains voltage dropped to 90v you've still got 45v headroom for a potential divider.

As the coil would be high impedance I would hazard that the beastie would trip if there was 100mA of fault current rather than 30mA. Almost self scaling to the level of risk!!

Also how unlucky would you be to have a volt drop and a fault at the same time!

 

[sparkytim]

I have found that the bonding now required provides a parrallel fault path that bypasses the ELCB operating circuit

I was amazed to see the recommended PIR code in the ESC code of practice for a Volt Op ELCB

 

[batty]

My experience of them was a retired surveyor called me out to a fault which was getting a shock off a shower screen and when he was in the shower anyway to cut a long story short there was an uncapped socket under shower and as seal had gone around shower water had drenched socket and 220 volts was tracking back up wall making shower cubicle live. Now this property was protected by a voltage operated trip and it was not tripping. Now I assume it was not tripping because voltage was feeding back through metalwork ie parallel earth path although it may have just been fautly. These things don't seem to work so I would bin them. If it had been an Rcd it would have tripped a long time before you would have got a belt. They do not comply with regs and need replacing if you do work on a property.

 

[ProDave]

The crucial thing here, is when testing it, the imballance earth current flows down through the CPC of the circuit under test, and hence trips the ELCB.

So they might give you some protection if you touch a live wire with one hand, while holding the CPC of that circuit in the other.

But if you touch a live wire say while out in the garden using an extension lead and standing in damp soil, the fault current will NOT flow anywhere near the ELCB and it will afford you no protection at all.

I guess in their day they were the best available, but there are so many cases where they won't protect you that you really should ignore the fact they are there.

Best thing is just disconnect all the earth wires from them and just treat them as an isolator switch to make the circuit safe for fitting a proper RCD.

 

[davetheglitz]

I have found that the bonding now required provides a parrallel fault path that bypasses the ELCB operating circuitI was amazed to see the recommended PIR code in the ESC code of practice for a Volt Op ELCB

Confession time - I screwed this up when I connected it up first time - and the earthing on the boiler created a parallel path to the gas/water pipe so the beastie didn't trip.

Do all the problems go back to the old days where earthing meant clagging on to the nearest water pipe!

 

[davetheglitz]

My experience of them was a retired surveyor called me out to a fault which was getting a shock off a shower screen and when he was in the shower anyway to cut a long story short there was an uncapped socket under shower and as seal had gone around shower water had drenched socket and 220 volts was tracking back up wall making shower cubicle live. Now this property was protected by a voltage operated trip and it was not tripping. Now I assume it was not tripping because voltage was feeding back through metalwork ie parallel earth path although it may have just been fautly. These things don't seem to work so I would bin them. If it had been an Rcd it would have tripped a long time before you would have got a belt. They do not comply with regs and need replacing if you do work on a property.

Devil's advocate here - but would this have been any different on a PME ot TNS system? It's only since the 17th that bathroom fixed appliances needed to be RCD'd - and I can't see a water path taking enough current to blow a fuse.

 

[davetheglitz]

The crucial thing here, is when testing it, the imballance earth current flows down through the CPC of the circuit under test, and hence trips the ELCB.So they might give you some protection if you touch a live wire with one hand, while holding the CPC of that circuit in the other.

But if you touch a live wire say while out in the garden using an extension lead and standing in damp soil, the fault current will NOT flow anywhere near the ELCB and it will afford you no protection at all.

I guess in their day they were the best available, but there are so many cases where they won't protect you that you really should ignore the fact they are there.

Best thing is just disconnect all the earth wires from them and just treat them as an isolator switch to make the circuit safe for fitting a proper RCD.

I agree with you completely with respect to RCD's being brilliant safety devices - but within a house until recently TNS or TNCS was quite satisfactory with no RCD - so how is this any different?

To my mind the level of protection seems to be just as good as these non RCD systems.

These devices were abandoned in the early 80's - I'd just moved in to my house then and it had a 500mA RCD protecting with the earth being a cloth covered wire wrapped around a water pipe!!!! The inference is that Voltage Operated ELCB's must have been worse than that - so I think I must be missing something fundemental here.

 

[ProDave]

so I think I must be missing something fundemental here.

Yes I tried to explain it in my previous post. These old ELCB devices require a current to flow between their two earth terminals to trip them. So as I stated before, ONLY if the earth fault causes the fault current to flow back down the CPC of it's circuit will the device trip. Someone just touching a live wire, with the fault current taking any number of different paths will not trip them.

As such they probably provided false hope of protection which arguably is worse than no protection.

(that's a bit like my favourite airbag argument. The theory being if you replace the drivers airbag with a sharp spike in the middle of the steering wheel, the driver is likely to drive a little safer than without the spike)

 

[davetheglitz]

Yes I tried to explain it in my previous post. These old ELCB devices require a current to flow between their two earth terminals to trip them. So as I stated before, ONLY if the earth fault causes the fault current to flow back down the CPC of it's circuit will the device trip. Someone just touching a live wire, with the fault current taking any number of different paths will not trip them.As such they probably provided false hope of protection which arguably is worse than no protection.

(that's a bit like my favourite airbag argument. The theory being if you replace the drivers airbag with a sharp spike in the middle of the steering wheel, the driver is likely to drive a little safer than without the spike)

Love the airbag arguement!!

However I'm trying to put this in perspective of why these devices were abandoned so long ago. They are not RCD's in any sense of the word - but as far as I can see they bring a TT installation up to the same standard as a TN system without an RCD in that in the event of a L-E fault they will trip out pretty damn quickly. This should have meant that they were 16th edition compliant - but they're not. I still can't see why.

 

[sparkytim]

As discussed by previous they don't reliably trip in the event of a L-E fault do to parrellel earth paths that divert the current and hence do not operate the trip coil. They also depend totaly on the earth stake and were also prone to nuisance tripping from external sources.

I think we can trust the IEE who decided many years ago these were not reliable means for shock protection.

 

[batty]

I honestly don't think they do work, luckily that ex surveyor did survive although he still has his voltage operated trip but next time he may not be so lucky.

 

[Evans Electric]

I am sure I have read somewhere that ELCBs Volt op. are no longer acceptable in any instance and should be replaced with an RCD . I think ,as regarding the regs , they no longer exist . Don't ask me where I read that but I did.

Deke

 

[tom1]

they are in doc p page 36, as was pointed out to me last night

 

[davetheglitz]

As discussed by previous they don't reliably trip in the event of a L-E fault do to parrellel earth paths that divert the current and hence do not operate the trip coil. They also depend totaly on the earth stake and were also prone to nuisance tripping from external sources.I think we can trust the IEE who decided many years ago these were not reliable means for shock protection.

I think you've hit the nail on the head here. Parallel earth paths and one earth stake would do it. Both are avoidable however. Nuisance tripping I could believe as I should imagine these devices were more susceptable to upstream voltage surges.

I think we can trust the IEE to decide these devices are not a reliable means of shock protection given the a lack of competence of many electricians in the early 1980's to connect the devices properly, fit an earth rod and test that they are working!!!!

The N-E fault on an RCD in a TT system is one of the nightmare scenarios to me. Easily put on a system during routine maintenance. How long before the IEE legislate against the use of a TT system in total. How much better (not!) if everything was TNCS!!!!

I'm guess I'm just railing against a device being removed for the wrong reasons - as I don't necessarily think they were as bad as they are made out to be.

Never mind - back to work in my steam lorry tomorrow!

 

[sparkytim]

A lost nuetral on TNCS supply (or PME as the DNO calls it) is pretty scary, 230 vots on all metal work, hence to riskey to use for Construction sites, Caravans, Mariners and the like.

 

[Lurch]

A lost nuetral on TNCS supply (or PME as the DNO calls it)

No good trying to sneak em in, they're still not the same.

 

[sparkytim]

No good trying to sneak em in, they're still not the same.

Have you tried requesting a TNCS earth terminal from a DNO or Supplier?

Try it and see how You get on

 

[Lurch]

Have you tried requesting a TNCS earth terminal from a DNO or Supplier?Try it and see how You get on

Point missed completely.