RCD Types, Ticking time bomb

[John-]

Juts been catching up on some prep work for my EV install course next week (Tring, 8th December) anyone on hear doing it?

I digress. Delving deeper into RCD types. Hell of a mine filed now! I am guessing most homes installed or upgraded with RCDs, RCBOs with type AC need now to be replaced for lighting circuits and Kitchen socket outlets?

Great thread hear form a while back i see:-

(11) RCD types and DC current | ElectriciansForums.net

I don't see anything in the latest proposed amendment, nor from the IET about the importance of selecting the correct type, or the growing risk in peoples homes as more kit that could saturate the coils gets used. Makes sense to me to make them mandatory or at least the checking of the connected devices and type for 'at risk' circuits in the kitchen and lighting.

Looking at my Sentry board with all RCBOs, they are all type AC. I have LED lights, washing machine and an induction hob... Doesn't look like i can get type A RCBOs for MK anyway. So a massive upgrade cost :-(

My washing machine is old, so unlikely to be an issue - but the rest could quit easily be compromising my RCD as i speak. I guess some testing with the circuits at various loads could tell me if i have an issue now.

Anyone do any of this testing as art of their inspections?

Cheers

John

 

[freddo]

look like i can get type A RCBOs for MK anyway.

The funny thing is the older version of MK RCBOs were type A!

 

[pc1966]

While I fully agree that type AC should be consigned to history and type A the norm for all "normal" installations, there is a need for some perspective on the matter.

There are millions of home (I guess?) that still rely on EEBADS without any RCD for additional protection, other than maybe for an outdoor socket, and they are not causing deaths left, right and centre. So the first line of defence, that of proper earthing of parts that might become line in the event of a fault, is still a very good approach.

The second point is that type AC will often trip on type A tests, just not meeting the desired specifications (current and/or time). Now they might fall in to the region where life is at risk, but in many cases if you don't have a simultaneous loss of earth and a fault with DC making that critical you don't have a death as the earth fault current will either trip the RCD or take out the OCPD (for many faults).

Fixing the choice of RCD should be done, but for many folk and properties good care and PAT testing would cover the most at-risk issues.

Or would it? Discuss...

 

[ipf]

While I fully agree that type AC should be consigned to history and type A the norm for all "normal" installations, there is a need for some perspective on the matter.

There are millions of home (I guess?) that still rely on EEBADS without any RCD for additional protection, other than maybe for an outdoor socket, and they are not causing deaths left, right and centre. So the first line of defence, that of proper earthing of parts that might become line in the event of a fault, is still a very good approach.

The second point is that type AC will often trip on type A tests, just not meeting the desired specifications (current and/or time). Now they might fall in to the region where life is at risk, but in many cases if you don't have a simultaneous loss of earth and a fault with DC making that critical you don't have a death as the earth fault current will either trip the RCD or take out the OCPD (for many faults).

Fixing the choice of RCD should be done, but for many folk and properties good care and PAT testing would cover the most at-risk issues.

Or would it? Discuss...

Of course it would. Just as upgrading a CU is an improvement, even if not every aspect of 7671 (2018) can be accounted for. We're talking possibilities.......or maybe, more like impossibilities.
We talk about CU fires, metal CU's and the London link. Before we started installing circuit breaker CU's regularly, in the 80's, I would have thought the risk of fire would be higher....with consumers having a pack of various size fuse wire available for repairs, for instance, or just sticking a bit of copper in carriers. In fact, it's the opposite, fires due to bad connections, faulty breakers, breakers incorrectly fitted, who knows?
I've seen a good few old Wylex CU's with holes burnt in the front due to overheat/burning...they never seemed to catch fire, though, even under such duress.

 

[Lucien Nunes]

if you don't have a simultaneous loss of earth and a fault with DC

This has to be the key point. Most loads don't produce significant DC leakage even if theoretically they could. And it's not strictly a new phenomenon. Many TVs from the 1970s had the chassis connected to the negative of rectified mains, so almost any possible shock fault from such a set was DC. Many hairdryers from the 1980s have a diode in series with the mains lead for low heat, so dropping one of those in the bath (if the diode is in the line) will produce pure pulsating DC leakage. Etc.

 

[pc1966]

One aspect that is not being addressed it seems any time soon in the regs is the issue of the RCD being additional protection and yet some folks are now ignoring the Zs requirements for the OCPD side of things assuming the RCD will disconnect.

I know the regs allow for RCD disconnection (more or less essential for TT, for example) but I personally would like to see 30mA RCD being only for additional protection:

• On a TT setup you have both an incoming RCD and 30mA additional protection RCDs so should one of the RCD's electronics fail then you don't have a complete loss of ADS
• On a TN you have to meet both the OCPD disconnection requirements and have the 30mA RCD (or if Zs too high then treat it as TT and cascade two selective RCDs).

To me that would be a less contentious change than mandating AFDD, being generally cheaper and potentially safer then depending on an all signing and dancing complex AFDD to deal with every problem.

 

[SparkySy]

I have to agree with what you are saying @pc1966, AFDD's are like putting all your trust in one device to protect every aspect of fault conditions, what happens if that device becomes defective but is unknown to the end user. We all know RCDs have a tendency to stick if not tested regularly what about AFDD modules yes they self test every so often but again is that something that is reliable for the lifespan of the unit.

 

[pc1966]

AFDD modules yes they self test every so often but again is that something that is reliable for the lifespan of the unit.

But who notices if they fail? And the self-test is not going to actually trip off the power to see if the mechanics are stuck or not!

The same issue applies for RCD, they should be trip-tested at least twice per year but I suspect in most cases they are tested only when EICR are done!

 

[SparkySy]

That's it very few people would notice, some people still don't know where their consumer unit is located so cannot be expected to check the light is still the correct colour or press a test button every 6 months.
Don't get me wrong I am all for improving safety but putting all your eggs in one basket doesn't seem very safe to me!

 

[Gavin John Hyde]

Juts been catching up on some prep work for my EV install course next week (Tring, 8th December) anyone on hear doing it?

I digress. Delving deeper into RCD types. Hell of a mine filed now! I am guessing most homes installed or upgraded with RCDs, RCBOs with type AC need now to be replaced for lighting circuits and Kitchen socket outlets?

Great thread hear form a while back i see:-

(11) RCD types and DC current | ElectriciansForums.net

I don't see anything in the latest proposed amendment, nor from the IET about the importance of selecting the correct type, or the growing risk in peoples homes as more kit that could saturate the coils gets used. Makes sense to me to make them mandatory or at least the checking of the connected devices and type for 'at risk' circuits in the kitchen and lighting.

Looking at my Sentry board with all RCBOs, they are all type AC. I have LED lights, washing machine and an induction hob... Doesn't look like i can get type A RCBOs for MK anyway. So a massive upgrade cost :-(

My washing machine is old, so unlikely to be an issue - but the rest could quit easily be compromising my RCD as i speak. I guess some testing with the circuits at various loads could tell me if i have an issue now.

Anyone do any of this testing as art of their inspections?

Cheers

John

Circpro now make the MK stuff under license and Type A devices are available. give them a message online and they will give you a product number and can then order through numerous suppliers.

 

[SparkySy]

I think it would be safer for us to install RCDs that can work for general protection of the fixed wiring installation and consumer so for instance a type A then it should be down to the manufacturer of XYZ product to protect the product with say a type b rcd or Afdd ect. This way you ensure that the consumer will see it every time they use the XYZ product and if it does trip they know where it is to re-set it. Take a washing machine for instance they get used every day, if it was built into the controls it would be easy to find and reset and test. And it would also give the user a secondary protective device in case of fault.
Just an idea can't see manufacturers stumping the bill for that when they can ultimately make it the consumers problem and expense!

 

[ipf]

Don't worry about it. We'll be resetting them from our phones by 2030, when I start thinking about getting rid if my then old diesel van.....honest!

 

[AJshep]

I agree with @pc1966 about the AFDDs.
AFDDs are great but in terms of making installation safer I'm still failing to see the real benefits. I know they will pick up parrallel arcs but most likely a OCPD or RCD would take care of these, its rare that you get a high resistance joint between L-N that doesn't turn into a dead short.

My issues:
1) At the moment they only go up to 32amps ( no 40amp shower protection which is where they would benifit IMO )

2) only rated for domestic use

3) Need a series arc greater then 2.5amps to operate (Electrium claim 1.5amps I think) so not so good for Led lighting circuits, which they are possibly adding to the new DPC... Even if you had a lighting circuit with a load of 2.5amps the AFDDs would only operate if the arc was near the origin of the circuit.

4) will work on RFCs but only on the connected load to the appliance, still need to remember that a series arc of 2.5amps is needed downstream of the fault so again it depends on where the arc develops.

 

[pc1966]

Wylex do a 40A AFDD so that can be covered, but I don't know about others.

Hager have an AFDD add-on to a MCB that could be used in commercial setups, but a major PITA compared to the neat Wylex solution, and some DP ones rasted at 10kA in EU but not for typical CU seen in the UK.

My biggest concerns with them are:
1) Cost. Just now it would add about £1k (or more) to a board change so if it puts off people from doing a CU upgrade that is likely to put more lives at risk than if more CU are upgraded to RCD protection.

2) Testing. there seems to be no standard test for them or MFT that allow you to diagnose problems. AFDD trips, what do you tell the customer? Change it only to find no improvement? Test for faults in the wiring - how? Test for sparking in appliances causing it - how?

 

[stevethesparks]

Wylex do a 40A AFDD so that can be covered, but I don't know about others.

Hager have an AFDD add-on to a MCB that could be used in commercial setups, but a major PITA compared to the neat Wylex solution, and some DP ones rasted at 10kA in EU but not for typical CU seen in the UK.

My biggest concerns with them are:
1) Cost. Just now it would add about £1k (or more) to a board change so if it puts off people from doing a CU upgrade that is likely to put more lives at risk than if more CU are upgraded to RCD protection.

2) Testing. there seems to be no standard test for them or MFT that allow you to diagnose problems. AFDD trips, what do you tell the customer? Change it only to find no improvement? Test for faults in the wiring - how? Test for sparking in appliances causing it - how?

As said before, a standardised testing port should be included. All to rushed forward in my opinion. More equipment to buy in the future, more calibration costs.

 

[AJshep]

I agree, I don't think there will ever be an effective way of testing these devices either, We just have to rely on the microprocessor and a flashing LED.

I got it wrong earlier they can be used in Commercial, but the pollution degree is only 2 so consideration is needed.