Max Zs For A Rcd

Talk Electrician Forum

Help Support Talk Electrician Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Joined
Mar 12, 2011
Messages
3,596
Reaction score
168
Location
Psychiatric ward??
Hi All,

Just been writing something for someone, and I thought I would post some of what I wrote here for constructive criticism from you lot!!!

One thing that causes no end of argument is the maximum Zs for a circuit with an rcd [i am assuming for the sake of argument a 30mA one here]....

At a first read, according to the regs, you have a CHOICE of device to achieve ADS, an OCPD or alternatively, an RCD, and, if you choose an RCD, then the regs VERY DEFINITELY say that the max Zs is 1667.

It is all in the table 41.5 [i only got the old red book] It is all there as clear as day....

BUT, wait a minute.... Look at the table before that... table 41.3........ Each heading reads for example, the first one; "Type "B" circuit-breakers to BSEN 60898 AND the overcurrent characteristics of RCBO's to BSEN 61009-1"

Why would they even care about the Zs to operate the overcurrent part of an RCBO if the RCD half of it will do the job for us??

The answer, it seems to me, is this,

SOME RCCB devices rely on the supply voltage to operate. RCBO's do for a start off, which is why i presume, they have included them in the table i refer to above...

Now have a look at this link. See how the RCD's are described as "line voltage independant" BUT the RCBO's as "line voltage dependant" VERY interesting eh!!

http://www.eaton.com/SEAsia/ProductsSolutions/Electrical/ProductsServices/PowerDistributionComponents/MiniatureCircuitBreakersProtectionAccessories/DINCircuitProtection/RCBO/index.htm

Now think of this scenario; You have a direct short L/E This will try to drag the supply voltage right down and, if the fault is close to the cutout, [so PEFC will be very high] might well do so. Now, in a TNCS system, the earth is connected to neutral at the cutout, they are one and the same as far as the "other" side of the cutout is concerned... Now, assuming the supply conductors have the same impedance, the voltage will be divided across our fault. The phase conductor voltage will be dragged down, and the voltage on our earth will be dragged up to meet it. As the earth is connected to the neutral, it will try to drag the neutral voltage up. The result can or could be, that there is not enough L/N voltage left to operate the thing.....

My verdict???

If you use an rcd, [if you NEED to and have no choice] use the 1667 figure, not literally of course, but you could take advantage of the increased latitude the use of the RCD will allow. If you use an RCBO use the figures in table 41.3 which are also the figures for mcb's... That seems best to me.

Constructive criticism please........

[SIZE=12pt]john...[/SIZE]

 
I agree, Failure rate of 7% apparently. That is why i would always only use them, [except where mandatory] where, for whatever reason, i could not achieve disconnection times with a proper OCPD in the PROPER way.....

john....

I agree with you too Steps, 100%..

john..

 
Also, the 1667 figure does most certainly NOT just relate to TT systems..

"In a TN system, where, for certain equipment in a certain part of the installation, the requirement of regulation [the bit about Zs] cannot be satisfied, that part may be protected by an RCD"

 
It does seem odd at first sight that the figures given for overcurrent protection use Zs which uses the path of an earth fault. It seems to assume that the pefc conditions would be more onerous than pscc, which would be the case for t+e cable with a reduced cpc csa.

AIUI, with RCBOs the mechanism for overcurrent and the mechanism for earth fault current are separate, even though an overcurrent may occur due to an earth fault. Table 41.3 is used therefore to ensure that the overcurrent mechanism will work regardless of the type of overcurrent. If the overcurrent turns out to be an earth fault current, then the RCD mechanism would trip sooner.

Incidentally, I found an eaton rcbo with voltage independence here http://www.eaton.uk.com/EatonUK/ProductsSolutions/Electrical/ProductsServices/CircuitProtection/ResidualCurrentCircuitBreakerwithintegratedOvercurrentProtection/CombinedRCDMCBforBritishStandard/index.htm

Still trying to get my head around VD and VI RCDs, and the TNCS scenario posted.

 
Hi There,

You say "It seems to assume that the pefc conditions would be more onerous than pscc, which would be the case for t+e cable with a reduced cpc csa"

Do not understand.. T+E has a reduced CPC csa, but this would give a reduced PEFC compared to "ordinary" cable. [you know what I mean] not a greater one owing to increased impedance... [increased R1+R2 for the circuit]

john..

 
the smaller the overcurrent, the bigger the max Zs would be. We are looking for the maximum Zs allowed, so it would be the smaller of the two overcurrents that we would be looking at, ie the PEFC

I might be wrong though, looks that way to me.

 
It does seem odd at first sight that the figures given for overcurrent protection use Zs which uses the path of an earth fault. It seems to assume that the pefc conditions would be more onerous than pscc, which would be the case for t+e cable with a reduced cpc csa.

AIUI, with RCBOs the mechanism for overcurrent and the mechanism for earth fault current are separate, even though an overcurrent may occur due to an earth fault. Table 41.3 is used therefore to ensure that the overcurrent mechanism will work regardless of the type of overcurrent. If the overcurrent turns out to be an earth fault current, then the RCD mechanism would trip sooner.

Incidentally, I found an eaton rcbo with voltage independence here http://www.eaton.uk.com/EatonUK/ProductsSolutions/Electrical/ProductsServices/CircuitProtection/ResidualCurrentCircuitBreakerwithintegratedOvercurrentProtection/CombinedRCDMCBforBritishStandard/index.htm

Still trying to get my head around VD and VI RCDs, and the TNCS scenario posted.

That RCBO has a restricted operating voltage though, so I would not say it was voltage independent.

184 - 264V ac is the rated operating range in the technical data.

 
Here is a link to a catalogue with another one [on page 4/7] they say this one is voltage dependent too.. It says in the blurb somewhere that it needs 85% of the supply voltage to work, at least that is what i think they are saying!!

http://www05.abb.com/global/scot/scot209.nsf/veritydisplay/c4e584f06cc6c4fbc1257ad800496193/$file/2csc420004b0201_rcds%20en.pdf

"Undervoltage of supply lines

According to the present standard, voltage dependent RCDs are required to work with a minimum voltage of 85% of the rated supply voltage.  Undervoltages can be caused by different reasons.

[SIZE=12pt]If the supply voltage falls to a value lower than the operating value of the voltage dependent[/SIZE]  RCD and greater than a dangerous level (50 V), the RCD will not trip but the risk of dangerous  faults is present. Voltage independent RCDs work properly regardless of the actual supply  voltage"
 
The one in their catalogue [top right page 4/7 is described as a "voltage dependant one"
 
All too clever for me!!
 
john...
 
That RCBO has a restricted operating voltage though, so I would not say it was voltage independent.

184 - 264V ac is the rated operating range in the technical data.
I saw that too, but then assumed the quoted voltage was relating to the supply characteristics (ie ~230V), not necessarily the actual (minimum) operating voltage, since there is a 50V limit involved as well.

bit too clever for me too. got to find some time for some more reading.

 
I saw that too, but then assumed the quoted voltage was relating to the supply characteristics (ie ~230V), not necessarily the actual (minimum) operating voltage, since there is a 50V limit involved as well.

bit too clever for me too. got to find some time for some more reading.
Where have you got the 50V limit from?

 
Hi Paul,

Yes, that is right, that is exactly what i mean. Turns out that RCBO's need a supply from the mains to operate, and what is more, I agree with you, if the thing needs a voltage of 184 - 264 it is most certainly NOT "voltage independant"...

It appears that, if you had a supply impedance of .18 ohms, [remember the RCBO is powered L/N and NOT L/E] you would only need a fault current of 255A to drag the voltage down that far... and the thing would not work!!

Is this why  the table in 7671 gives Zs for an RCBO in the same chart as "ordinary" MCB's as they KNOW full well, that given enough fault current, the thing will not work as an RCD and you would have to rely on the OCPD part of the contraption...

Hi Rob, the 50V is the "touch voltage" limit which is why Zs for a 30mA rcd is 1667 and NOT 7666 [which would STILL flow enough current to work the RCD]

Anyway, the 50V touch limit does NOT come into it, as the RCBO is powered L/N and not earth to "proper" earth [think about it!!] so the earth voltage can do what it likes.

Note also that if you have a leak to earth, yes, the RCBO will i assume, limit the touch voltage to 50V [given a max Zs of 1667] after which it will trip. But NO protective device ever invented can limit touch voltage during a "proper" fault. This is ENTIRELY dependant on two things...

1, The impedance of the R2 side of the earth loop.

2, The fault current flowing..

So, say you were holding an exposed conductive part during a fault and the total ELI was .18 ohms and the R2 part was .09 ohms and the fault current was therefore 1277A You would get a shock voltage of 1277a x .09 ohms = 115v As in this example the supply impedance across the phase and the earth were the same [as it would normally be with TNCS], you recieved half the voltage as a shock. For different values, [and currents] you would recieve different shocks, the system would follow the rules of a "voltage divider"

Without even thinking about that in a tncs system the voltage to earth and hence also the neutral might be dragged up [thus reducing the PD L/N], you can see that the supply voltage would drop to 115, so the RCBO would NOT operate as an RCD... You better hope that the magnetic part of it does!!!!.

john...

 
John,

I'm not sure I follow your science, if you have no voltage at the RCD then the circuit is dead, so where's the issue, it obviously needs a mains from the supply to operate or it's not required because with no voltage nothings on.

On your 1277A fault current you have your OCPD so that would trip within 0.4 sec.

On leakage to earth the RCD would trip within required time.

 
Yes, in the scenario I described, as you say, then yes, the OCPD would operate and all would be well. But imagine you have a supply with a loop impedance of .6 ohms. You have installed a 32A "C" curve device. This has a max Zs of .79 ohms, so ok up to now. Next thing, you have a earth fault with an impedance of 1.5 ohms. The fault current will be 153 amps so the OCPD will not operate for about 20 seconds. Meanwhile, with this current flowing the supply voltage has dropped to 138 V so you can forget the RCBO's rcd part operating as from the makers blurb it requires 184V… See what I mean… I might be incorrect in all this, but I know that others share this view point rightly or wrongly..

What I am getting at, is that this is why in the regs, RCBOS are lumped in with MCB's from the point of Zs, and just because the things incorporate an "RCCB" part, does NOT mean that you can have a ZS or 1667 or whatever.

A "proper" RCD is, as far as I know different, they DO NOT need a mains voltage to operate, and so, function as intended and expected, hence for them a far larger ELI is permitted, even if it is not exactly desirable…

As I say, I am no expert, and I may be wrong in all this. That is why I invited others to comment with constructive criticism [as you have done] and not just "that is crap!"

[SIZE=12pt]john...[/SIZE]

 
John,

But the RCD is not waiting for the voltage to drop, the instant the fault happens the imbalance between L&E or N&E the RCD will operate providing current is being drawn, and within the required time providing the ZS is lower than 1667, and the voltage should be at the acceptable level before the fault occurs.

 
only quickly read over this, but i think John is getting at if the Zs is higher (than MCB values), then it may not trip immediately, but the fault current may be high enough that the voltage at the RCBO is too low for it to trip on the RCD part, so its then relying on the MCB part eventually tripping, which could be a second or 2 later

 
Top