Disconnection times not met

[Bradley6969]

Three questions haha

1. 160A MCCB the max Zs for this particular switch is 0.25 ohms all tested and I’m getting 0.44 ohms its a 95 SWA outgoing which feeds another panel which has its own outgoing circuits would I be right and say that would be a code 2? As I can’t meet disconnection time on it

2. Can’t meet disconnection times for various circuits 63A MCBs no RCD protection would you code 2? Max Zs should be 0.69 ohms and I’m in the 1.00 ohms

3. If I was to add RCD to question 2 could I achieve ADS - or would I need to test L-N also for my short circuit protection

 

[pc1966]

What sort of PSSC are you seeing relative to the PFC?

Those sort of numbers imply a significantly higher CPC impedance than the neutral or the voltage drop is going to be missed by a significant margin.

If it is down to the SWA armour impedance being too high (or rubbish connection to the glands/banjos), then to get the CPC impedance down a supplementary run of copper conductor along side together with decent connections at the ends might be a simpler/cheaper overall fix than changing to RCDs.

 

[ipf]

Where have you got your maximum Zs rating from?

 

[Bradley6969]

What sort of PSSC are you seeing relative to the PFC?

Those sort of numbers imply a significantly higher CPC impedance than the neutral or the voltage drop is going to be missed by a significant margin.

If it is down to the SWA armour impedance being too high (or rubbish connection to the glands/banjos), then to get the CPC down a supplementary run of copper conductor along side together with decent connections at the ends might be a simpler/cheaper overall fix than changing to RCDs.

Hi, thanks for the reply it’s a small site with some 20 caravan boxes so all SWA cable and underground, I’m just wanting to make sure my coding is right
[automerge]1594229615[/automerge]

Where have you got your maximum Zs rating from?

Manufactures data

 

[pc1966]

Where have you got your maximum Zs rating from?

That is a good point, just checked and a 63A B-curve MCB should be 0.56 ohms from OSG.

MCCB are more of a pain as you need to look up a specific model (and/or what settings are dialled in to it) but scaling a 100A B-curve MCB points to 0.21 ohm as being more likely for a 160A device.
[automerge]1594230137[/automerge]

Manufactures data

Interesting, as Wylex have this for their miniature 63A MCB:

• B-curve 0.729 ohm
• C-curve 0.364 ohm
• D-curve 0.183 ohm

 

[Bradley6969]

Would I be right for saying code 2 for questions 1 and 2? Is it just poor design for TNS system
[automerge]1594230270[/automerge]

That is a good point, just checked and a 63A B-curve MCB should be 0.56 ohms from OSG.

MCCB are more of a pain as you need to look up a specific model (and/or what settings are dialled in to it) but scaling a 100A B-curve MCB points to 0.21 ohm as being more likely for a 160A device.
[automerge]1594230137[/automerge]

Interesting, as Wylex have this for their miniature 63A MCB:

• B-curve 0.729 ohm
• C-curve 0.364 ohm
• D-curve 0.183 ohm

I have an app on my phone, plus I’m on the NIC certs so auto puts a maximum reading in for MCBs etc

 

[pc1966]

Would I be right for saying code 2 for questions 1 and 2? Is it just poor design for TNS system

Yes, that would be right. According to the Best Practice Guide #4 it is C2 for:

Earth fault loop impedance value greater than that required for operation of the protective device within the time prescribed in the version of BS 7671/IEE/IET Wiring Regulations or manufacturers’ published data current at the time of installation

 

[ipf]

You could work out the R1+R2 rating for the cable run and see how it compares with your measured Zs minus Ze. If it's similar it's a good reading and something needs sorting. If not, it could be a bad connection.

 

[pc1966]

I have an app on my phone, plus I’m on the NIC certs so auto puts a maximum reading in for MCBs etc

That is a neat feature!

 

[buzzlightyear]

If it is down to the SWA armour impedance being too high (or rubbish connection to the glands/banjos),

this what i was thinking on the same lines .

 

[Bradley6969]

I was just unsure as the 160A MCCB feeds the panel with 63A MCBs but the Max Zs is higher than the Max Zs of the 160A MCCB thought it was a C2

 

[pc1966]

As a carven park I guess most end circuits have RCDs anyway, so you could deal with the (probably) high R2 by going to delay RCDs and/or a delay RCD add-on to the MCCB, but first it would be worth finding out if it is simply too much cable/not enough armour, or if there is some crappy gland use to be fixed.
[automerge]1594230720[/automerge]

I was just unsure as the 160A MCCB feeds the panel with 63A MCBs but the Max Zs is higher than the Max Zs of the 160A MCCB thought it was a C2

Is it feeding just the internal bus of the panel? I.e. there is no cables longer than 3m coming from the MCCB?

If so then the issue is one of meeting disconnection times on the 63A outgoing circuit MCBs, as the MCCB is only needed for overall overload of the system if (as is likely) the sum of MCBs exceeds the supply capability. I.e. no C2 code on the MCCB as it is not providing earth fault protection anyway.

 

[Bradley6969]

As a carven park I guess most end circuits have RCDs anyway, so you could deal with the (probably) high R2 by going to delay RCDs and/or a delay RCD add-on to the MCCB, but first it would be worth finding out if it is simply too much cable/not enough armour, or if there is some crappy gland use to be fixed.

Yes I spoke to my boss and he’s on about putting a 300mA time delayed RCD inside panel with the MCBs as each box has its own 30mA additional protection as all readings are similar and get higher towards ends of circuits
[automerge]1594230810[/automerge]

Yes I spoke to my boss and he’s on about putting a 300mA time delayed RCD inside panel with the MCBs as each box has its own 30mA additional protection as all readings are similar and get higher towards ends of circuits

But he didn’t say anything about the 160A MCCB feeding the panel as that is 0.20 ohms higher???
[automerge]1594230931[/automerge]

As a carven park I guess most end circuits have RCDs anyway, so you could deal with the (probably) high R2 by going to delay RCDs and/or a delay RCD add-on to the MCCB, but first it would be worth finding out if it is simply too much cable/not enough armour, or if there is some crappy gland use to be fixed.
[automerge]1594230720[/automerge]

Is it feeding just the internal bus of the panel? I.e. there is no cables longer than 3m coming from the MCCB?

If so then the issue is one of meeting disconnection times on the 63A outgoing circuit MCBs, as the MCCB is only needed for overall overload of the system if (as is likely) the sum of MCBs exceeds the supply capability. I.e. no C2 code on the MCCB as it is not providing earth fault protection anyway.

So the 160A MCCB feeds the panel with 95 SWA to some bolts ons that feeds a 100A isolator and 63A MCBs

 

[pc1966]

So the 160A MCCB feeds the panel with 95 SWA to some bolts ons that feeds a 100A isolator and 63A MCBs

I think we are getting to the "we need a diagram" stage of the thread!

If the MCCB is feeding stuff beyond a well-protected cable less than 3m in length then it ought to meet fault to earth disconnection times.

I was thinking of a MCCB incomer to a board: in that case the MCCB need not meet disconnection on its own, as all circuits ought to be met by the outgoing MCBs' protection.

 

[Bradley6969]

I think we are getting to the "we need a diagram" stage of the thread!

If the MCCB is feeding stuff beyond a well-protected cable less than 3m in length then it ought to meet fault to earth disconnection times.

I was thinking of a MCCB incomer to a board: in that case the MCCB need not meet disconnection on its own, as all circuits ought to be met by the outgoing MCBs' protection.

Okkkk,
Yes so 160A MCCB fed from EB tails, 40 metre ish run of 95 SWA to the Panel with all outgoing 63A MCBs haha
[automerge]1594231588[/automerge]
So disconnection only applies to final circuits? Thanks so far not really done much with coding and testing

 

[ipf]

I've got an old table indicating that the csa of 95mm armouring may not be sufficient. I still go by it but it may be a bit dated.

 

[Bradley6969]

I've got an old table indicating that the csa of 95mm armouring may not be sufficient. I still go by it but it may be a bit dated.

View attachment 59330

Thanks for that
[automerge]1594232508[/automerge]
Spoke to another guy and he says C2 for the 160A MCCB as it still needs to disconnect in 5s if ever someone went through the cable??
[automerge]1594233780[/automerge]

Thanks for that
[automerge]1594232508[/automerge]
Spoke to another guy and he says C2 for the 160A MCCB as it still needs to disconnect in 5s if ever someone went through the cable??

But I have no idea haha

 

[davesparks]

I've got an old table indicating that the csa of 95mm armouring may not be sufficient. I still go by it but it may be a bit dated.

View attachment 59330

That table only shows the basic method of working out CPC size by selection from a table.
If you do the adiabatic calculation you will find that with almost all of those cable sizes the armour will comply as a CPC.
[automerge]1594234223[/automerge]

So disconnection only applies to final circuits? Thanks so far not really done much with coding and testing

It applies to all circuits, distribution and final.

What pc1966 is saying is that they thought you were talking about a 160A MCCB fitted as a main switch in a DB which is a different situation.
[automerge]1594234522[/automerge]

Yes I spoke to my boss and he’s on about putting a 300mA time delayed RCD inside panel with the MCBs as each box has its own 30mA additional protection as all readings are similar and get higher towards ends of circuits

But he didn’t say anything about the 160A MCCB feeding the panel as that is 0.20 ohms higher???

So the 160A MCCB feeds the panel with 95 SWA to some bolts ons that feeds a 100A isolator and 63A MCBs

What is the job you are doing?
If you are doing an EICR then you code it appropriately (C2 in this case) and submit the report leaving it up to whoever is quoting for the remedial works to come up with a plan to fix it.
If you are doing an EICR and also quoting for any remedial work that comes as a result of it then you still complete the EICR before planning the remedials.

If I was looking at the remedial work for this I would first be checking the Ze and checking all cpc connections for soundness. Then I would look at whether replacing the 160A MCCB with a fused switch would be a viable solution and similar with the 63A MCB's.

 

[pc1966]

Yes, to add to what @davesparks said - when you have the cable fed from the MCCB it should meet the disconnection time of 5s for the end Zs value just in case there is damage at any point along the cable. My earlier comment was just as he said - for a MCCB incomer.

Changing the MCCB to include RCD capability would be one fix. However if you get one of the fancier MCCB you might be able to configure the over-current trip curve to meet 5s for your cable-end Zs of 0.44 ohm (say setting it to trip at 5s for 0.95 * 230 / 0.44 = 496A) so you meet cable disconnection, your "instant trip" is still well beyond any downstream MCB for selectivity, and you are not having to find space for a MCCB add-on module.

I don't know enough about the details of MCCB selection to say for sure, probably @Julie. or @davesparks would have a lot more knowledge than me for this aspect.
[automerge]1594235027[/automerge]

What is the job you are doing?

Me? Talking nonsense on internet forums seem to be my current status.

At least I keep my clothes on for this forum

 

[Bradley6969]

Thanks for your help guys, yes my only concern or lack of knowledge was the MCCB and the Zs value I got which was exceeding the Max values allowed, now the MCCB has 3 settings 125A, 140A, 160A but I believe even setting on 125A I won’t be achieving the values I was getting,


So just to confirm it would be a C2 on report? Sorry I’m trying to understand incase I come into this situation again

 

[pc1966]

Thanks for your help guys, yes my only concern or lack of knowledge was the MCCB and the Zs value I got which was exceeding the Max values allowed, now the MCCB has 3 settings 125A, 140A, 160A but I believe even setting on 125A I won’t be achieving the values I was getting,

The simpler MCCB only let you adjust the thermal curve by 20-40% or so. They don't give you control over the "instant" trip that is from the magnetic aspect and it is that which usually gives you the sub-5s disconnection times.

The fancy electronic MCCB has the ultimate fall-back of a instantaneous magnetic trip at some high current (typically 10 or more times the max thermal limit) but they allow you to fiddle with the "thermal curve" as it is done electronically, so you can define constant I2t ranges, etc, according to your cable protection requirements, etc.

 

[Bradley6969]

The simpler MCCB only let you adjust the thermal curve by 20-40% or so. They don't give you control over the "instant" trip that is from the magnetic aspect and it is that which usually gives you the sub-5s disconnection times.

The fancy electronic MCCB has the ultimate fall-back of a instantaneous magnetic trip at some high current (typically 10 or more times the max thermal limit) but they allow you to fiddle with the "thermal curve" as it is done electronically, so you can define constant I2t ranges, etc, according to your cable protection requirements, etc.

Thanks for that

 

[Julie.]

Yes, to add to what @davesparks said - when you have the cable fed from the MCCB it should meet the disconnection time of 5s for the end Zs value just in case there is damage at any point along the cable. My earlier comment was just as he said - for a MCCB incomer.

Changing the MCCB to include RCD capability would be one fix. However if you get one of the fancier MCCB you might be able to configure the over-current trip curve to meet 5s for your cable-end Zs of 0.44 ohm (say setting it to trip at 5s for 0.95 * 230 / 0.44 = 496A) so you meet cable disconnection, your "instant trip" is still well beyond any downstream MCB for selectivity, and you are not having to find space for a MCCB add-on module.

I don't know enough about the details of MCCB selection to say for sure, probably @Julie. or @davesparks would have a lot more knowledge than me for this aspect.
[automerge]1594235027[/automerge]

Me? Talking nonsense on internet forums seem to be my current status.

At least I keep my clothes on for this forum

Firstly, based on the information, I would assign FI - this could be a very serious issue as I highlight below.

In respect of the mccb settings, that would be a no-go, if you take a measured loop of 0.44ohm, if this is measured cold then we need to correct it for hot running, and then 95% voltage, this would mean the mccb would need to trip in under 5s at circ 688A.

Since a typical mccb has the instantaneous around 10x this would mean a mccb of 68A(~63), or achieving 5s in the time-current curve; typically this would be circ 8x the overload setting, so you would need to set it at around 85A - this would be too low for the circuit I would guess.

The real issue though is why such a high impedance?

I don't know the number of cores, but if it's a 4core 70mm2 cable 40m long, I would guess a r1 of 0.011ohm, and a r2 of 0.048 ohm (The armour should be sufficient if it's 4 or more cores - if it's fewer than 4 core though, it wouldn't), anyway this means the Ze (or Zs at the feeder board) is around 0.4 ohm.

For something feeding 160A circuits this appears very high, and means either there is a bad connection, which could end up being c1; or the incoming fuses wouldn't trip in any sensible time. (C2 on the main protection, and on probably almost all other circuits).

This really needs further investigation, hopefully a loose connection and a cheap fix, but if it is a high Ze then the dno would have to sort it out as their own fuses wouldn't operate in time, only then can you investigate what needs to be done, if anything.

 

[davesparks]

So just to confirm it would be a C2 on report? Sorry I’m trying to understand incase I come into this situation again

In normal operation it is not dangerous, so it is not immediately dangerous, so it's not a C1.
This would be dangerous under fault conditions, so it is potentially dangerous, which is C2.
[automerge]1594243176[/automerge]

Firstly, based on the information, I would assign FI - this could be a very serious issue as I highlight below.

Why do you say FI for this?
We know that the Zs is too high, so there is a potential danger which is a clear C2 in my view.
[automerge]1594243233[/automerge]

I don't know enough about the details of MCCB selection to say for sure, probably @Julie. or @davesparks would have a lot more knowledge than me for this aspect.

I'm willing to bet you know more about it than I do.

 

[Julie.]

In normal operation it is not dangerous, so it is not immediately dangerous, so it's not a C1.
This would be dangerous under fault conditions, so it is potentially dangerous, which is C2.
[automerge]1594243176[/automerge]


Why do you say FI for this?
We know that the Zs is too high, so there is a potential danger which is a clear C2 in my view.

Because we don't know if its a poor connection, if it is, then under normal conditions this could be a dangerous - immediate risk hence c1 , but yet if it's only a high resistance somewhere c2 would be the case, it's a little academic as FI, C1, and C2 all need fixing.

 

[davesparks]

Because we don't know if its a poor connection, if it is, then under normal conditions this could be a dangerous - immediate risk hence c1 , but yet if it's only a high resistance somewhere c2 would be the case, it's a little academic as FI, C1, and C1 all need fixing.

No we don't know the cause, but we do know it is potentially dangerous so that is how it should be recorded.
If a loose connection is found, and if that is a C1 item then that would be recorded on the EICR as another item (or more likely just tightened up there and then if it can be isolated)
A loose connection in a live conductor would usually be quite obvious, especially at that kind of current so I doubt that this is the cause.
It could be a loose connection in the CPC but that too is a C2 item

 

[Julie.]

No we don't know the cause, but we do know it is potentially dangerous so that is how it should be recorded.
If a loose connection is found, and if that is a C1 item then that would be recorded on the EICR as another item (or more likely just tightened up there and then if it can be isolated)
A loose connection in a live conductor would usually be quite obvious, especially at that kind of current so I doubt that this is the cause.
It could be a loose connection in the CPC but that too is a C2 item

Agree, but I would use FI as it could be more serious than C2, and this is exactly what FI is there for

 

[davesparks]

Agree, but I would use FI as it could be more serious than C2, and this is exactly what FI is there for

This is not the way I understand the coding, FI is for items which require further investigation to establish whether they are affecting safety or not.
In this case there is a clear potential danger, no further investigation is required to establish this, so its a C2.

 

[bigspark17]

Have you confirmed that the tns supply has not been converted to a tncs? In my view any supply to a caravan site should be TT earthing arrangement with upfront rcd time delay.
How old is the installation?

 

[Julie.]

This is not the way I understand the coding, FI is for items which require further investigation to establish whether they are affecting safety or not.
In this case there is a clear potential danger, no further investigation is required to establish this, so its a C2.

Yes, it is at least a C2, but I would mark it as FI because we don't know whether it is more serious or not.

It has a big potential for being C1

However if I was on-site, I would have got the Zs at the feeder board, so would be able to narrow it down - perhaps the op did, but just hasn't posted this information.

The high figures stand out as something potentially dangerous given the circuit sizes, they are not just a little on the large size.
[automerge]1594245247[/automerge]

Have you confirmed that the tns supply has not been converted to a tncs? In my view any supply to a caravan site should be TT earthing arrangement with upfront rcd time delay.
How old is the installation?

Agreed, I assume it's a fairly remote site fed by wet string, these tend to have high loop impedance, but usually the dno would have taken issue with the lack of earth protection when first connected or upgraded given the current rating.