Carrying out a refurb on old panel boards, worried about discrimination.

[johnwicklover]

I am carrying out a refurbishment job on this old panel board and fitting in a new panel board from Schneider. The current supply to this building is a 100amps BS88-2 Gg (assuming). I am planning to protect the outgoing submains with the NSX100 ML 2.2 set at 100 and 63 respectively. I don't know the full load but the current submains have 100Amps MCCBs and 63Amps. For some reason the Schneider quick selectivity tool says there is no selectivity between the upstream 100amps fuse and ML 2.2 but shouldn't it be fine? Here is the picture : ML 2.2 @100 and BS88-2 Fuse.


ML2.2 @63 and BS88-2


Looking at this graphs, the protection device for the submains would trip first before the main fuse.

Any help would be appreciated, thank you.

 

[Julie.]

It depends on the fault level, sure at low currents up to circ 800A there is coordination, and most likely discrimination, however once you get into the current limiting phase of the fuse operation the I^2 t let-through of the mcb/mccb will be more than the pre-arcing I^2 t operation of the fuse.

To check, you need further curves/tables the I^2t pre-arc will be easy to find, but the let-through on a mcb/mccb is usually harder to find for some reason.

 

[johnwicklover]

It depends on the fault level, sure at low currents up to circ 800A there is coordination, and most likely discrimination, however once you get into the current limiting phase of the fuse operation the I^2 t let-through of the mcb/mccb will be more than the pre-arcing I^2 t operation of the fuse.

To check, you need further curves/tables the I^2t pre-arc will be easy to find, but the let-through on a mcb/mccb is usually harder to find for some reason.

Sorry for the delay in reply but thank you Julie. I will investigate this further.

 

[pc1966]

The Schneider on-line tool has combinations where it gives no selectivity results and times when it does, based on combinations of stuff but with no obvious reason why.

As @Julie. says for the plots you have to realise the fuse continues down from 0.1s smoothly but gradually reaching the constant-I2t region with a slope of 2 decades time per decade current on the log-log scale.

The two regions (extrapolated fuse & breaker) cross at around 0.8kA so you have overload selectivity but no practical fault selectivity. Unfortunately that is generally the case with OCPD close in ratings of any types (except fuse-fuse at ratios of 1.6 or more where you generally get total selectivity).

But I am struggling to understand what you have as a 100A supply fuse and multiple sub-mains off that with 100A and 63A MCCB does not appear to ever had much chance of providing selectivity!

 

[pc1966]

Also what sort of loads do you have?

In the example config for the Schneider MCCBs you show above the "instant" trip is at about 140A / 90A. That is similar to a 50A/32A B-curve MCB, or 32A/20A C-curve MCB so any high inrush load like a medium DOL-started motor, etc, that might trip that sort of MCB would al;so be able to take out the sub-main feed!

 

[johnwicklover]

The Schneider on-line tool has combinations where it gives no selectivity results and times when it does, based on combinations of stuff but with no obvious reason why.

As @Julie. says for the plots you have to realise the fuse continues down from 0.1s smoothly but gradually reaching the constant-I2t region with a slope of 2 decades time per decade current on the log-log scale.

The two regions (extrapolated fuse & breaker) cross at around 0.8kA so you have overload selectivity but no practical fault selectivity. Unfortunately that is generally the case with OCPD close in ratings of any types (except fuse-fuse at ratios of 1.6 or more where you generally get total selectivity).

But I am struggling to understand what you have as a 100A supply fuse and multiple sub-mains off that with 100A and 63A MCCB does not appear to ever had much chance of providing selectivity!

It is an old school and looks like things were added on over the years, so I don't think they cared about selectivity that much as long as the final circuits tripped.

 

[johnwicklover]

Also what sort of loads do you have?

In the example config for the Schneider MCCBs you show above the "instant" trip is at about 140A / 90A. That is similar to a 50A/32A B-curve MCB, or 32A/20A C-curve MCB so any high inrush load like a medium DOL-started motor, etc, that might trip that sort of MCB would al;so be able to take out the sub-main feed!

Yes, that's another problem. I don't know the loads but the highest breakers they have are 32Amps type B. I guess this problem can be worked around by using 160 MCCB ML 4.2 or higher and setting the tripping point (ir) at 100 and then setting the instantaneous tripping at 200A. I guess full selectivity won't be obtained but as long as the submains are adequately protected, and discrimination between final circuits and the submain protective device is there, it should be okay.

Any other suggestion on how to get around this would be appreciated.

Thanks.

 

[pc1966]

Yes, that's another problem. I don't know the loads but the highest breakers they have are 32Amps type B. I guess this problem can be worked around by using 160 MCCB ML 4.2 or higher and setting the tripping point (ir) at 100 and then setting the instantaneous tripping at 200A. I guess full selectivity won't be obtained but as long as the submains are adequately protected, and discrimination between final circuits and the submain protective device is there, it should be okay.

Any other suggestion on how to get around this would be appreciated.

How many sub-mains, are they single or 3-phase, and what loads do they actually need?

It might seem very old-school (pun intended!) but how about a busbar chamber and a couple of 63A fused-switches for each sub-main?

They would have total selectivity with the 100A incoming fuse, and moderate selectivity with final circuit MCB/RCBOs.

I recently looked at that for another question and, at least for the Hager one I had a table of selectivity values for, the 63A fuse feeding 32A B-curve RCBO was OK to 1.5kA faults on the final circuit which could well be above what you see at the end of the sub-main and wiring to appliances, etc.

 

[pc1966]

This was the post:

Relocating Consumer Unit to new location | Page 2 | on ElectriciansForums

Find Relocating Consumer Unit to new location Advice and Help. How-to Relocating Consumer Unit to new location in the UK Electrical Forum advice boards on ElectriciansForums.net | Free Electrical Advice - Electricians Forums: Electricians' Talk Forums. Relocating Consumer Unit to new...

www.electriciansforums.net

 

[pc1966]

The electronic MCCB are not cheap, so you might find that a couple of decent fused-switches such as Schneider SQB0633K, or maybe better the 100A SQB1003K with 63A fuses fitted so upgrading is possible in the future, will work out about the same cost.

Of course replacing those fuses is a job for an electrician, not like re-closing a MCCB, but ideally they won't be blowing unless you have a serious fault anyway!

 

[pc1966]

If it really has to be a MCCB board then you are probably as well giving up on the idea of selectivity on the sub-main to incoming fuses, and just got for the basic model rather than trying to fiddle with the fancy electronic settings to try and walk the line between the MCCB/fuse on the supply side and the MCB/MCCB on the sub-DB side.

 

[johnwicklover]

If it really has to be a MCCB board then you are probably as well giving up on the idea of selectivity on the sub-main to incoming fuses, and just got for the basic model rather than trying to fiddle with the fancy electronic settings to try and walk the line between the MCCB/fuse on the supply side and the MCB/MCCB on the sub-DB side.

Thank you for suggestions and the links, I will have a look through them.

There are 6 single phase submains and 2 three phases - submain. Unfortunately I don't know the exact loads at this time.

The bus bar chamber idea sounds good but they want an upgrade to 6 way MCCB panel board. Like you said the fuses probably won't blow but circuit breakers are easily switched on. The MCCB will be more expensive but the school will have to swallow it.

 

[pc1966]

You might want to check the main fuse, as that is quite a lot of sub-mains coming off 3*100A for example, so if they turn out to be 160A or more then you have an easier job.

I had a go at the Schneider on-line tool but it seems broken for me today

I would concentrate on getting tolerable MCB/MCCB selectivity using the short-delay aspect of the fancy MCCB, that way in most cases the selectivity falls back to incoming fuse and final MCB, and that is probably no problem.

Yes, a sub-main fault might not open the MCCB before the incoming fuse, but that is a lot less likely than some final circuit/appliance giving them grief.

If you can find out loads and the PFC & PSCC at each DB just now it would help to decide how well the whole system would work.

 

[johnwicklover]

You might want to check the main fuse, as that is quite a lot of sub-mains coming off 3*100A for example, so if they turn out to be 160A or more then you have an easier job.

I had a go at the Schneider on-line tool but it seems broken for me today

I would concentrate on getting tolerable MCB/MCCB selectivity using the short-delay aspect of the fancy MCCB, that way in most cases the selectivity falls back to incoming fuse and final MCB, and that is probably no problem.

Yes, a sub-main fault might not open the MCCB before the incoming fuse, but that is a lot less likely than some final circuit/appliance giving them grief.

If you can find out loads and the PFC & PSCC at each DB just now it would help to decide how well the whole system would work.

Yeah, that's a good point. I will get the main fuse confirmed.

I might be able to get the loads confirmed as well.

Thanks.