TP 100amp submain on BS88-2 fuses, but the Zs is above the Max allowed...

[TPY765]

I am helping a local football club who are having some issues with Zs being too high on the submain. They have the supply Henley blocked into a 100amp TP switch fuse (BS88-2) which then feeds a 35mm 4 core SWA 95m to an isolator in another building which is being built. The Ze at the incomer from the DNO is 0.27 (TN-C-S) and the Zs at the isolator is 0.47. This exceeds the max permitted and from my calculations to get it within parameters, they would need to run a 70mm separate earth to supplement the current earthing. Given the higher Ze (although it is within limits), only leaves around 0.07 for an r1+r2 reading, and given the length of run doesn't leave much room to play with.

This will be extremely costly so I wanted to see if there were any other options we could explore first.

Any ideas and help would be much appreciated.

 

[Julie.]

I am helping a local football club who are having some issues with Zs being too high on the submain. They have the supply Henley blocked into a 100amp TP switch fuse (BS88-2) which then feeds a 35mm 4 core SWA 95m to an isolator in another building which is being built. The Ze at the incomer from the DNO is 0.27 (TN-C-S) and the Zs at the isolator is 0.47. This exceeds the max permitted and from my calculations to get it within parameters, they would need to run a 70mm separate earth to supplement the current earthing. Given the higher Ze (although it is within limits), only leaves around 0.07 for an r1+r2 reading, and given the length of run doesn't leave much room to play with.

This will be extremely costly so I wanted to see if there were any other options we could explore first.

Any ideas and help would be much appreciated.

Add a delayed RCD to trip in under 5s

Or change the fuses for a mccb which has integral delayed ef protection.

 

[westward10]

Does it need to be 100A.

 

[pc1966]

How about running a 16mm CPC in parallel with the armour? (Doh, just seen you looked at that)

Cheapest of all is dropping to 80A or 63A fuse of course, if that is adequate for the actual load max demand.

 

[pc1966]

Also allow for how hot the SWA is running, as there is a 0.8 factor in the likes of the OSG table B3 Zs value on the assumption the cable is up at 70C normally.

 

[pc1966]

A quick check has 80A fuse being met with 16mm CPC in parallel.

MCCB RCDs come in seemingly endless varieties, but something like this in place of the switch-fuse is a possible solution as Julie suggests:

LV426777 | Schneider Electric, Compact MCCB 4P 100A, Breaking Capacity 36 kA, DIN Rail Mount | RS

uk.rs-online.com

Not cheap either, but with up to 5A threshold and 1s delay you can get reasonable selectivity with some final circuits not on RCD protection but just smaller MCBs (e.g. any MCB that trips on magnetic threshold due to hard fault will be ignored, so long as you don't get many amps to earth without anything noticing).

This is probably cheapest option, as Julie suggested adding RCD to switch-fuse:

100A 4 Pole 300mA Type A Time Delay RCD

Insulated DIN rail clips

www.cef.co.uk

 

[Julie.]

A quick check has 80A fuse being met with 16mm CPC in parallel.

MCCB RCDs come in seemingly endless varieties, but something like this in place of the switch-fuse is a possible solution as Julie suggests:

LV426777 | Schneider Electric, Compact MCCB 4P 100A, Breaking Capacity 36 kA, DIN Rail Mount | RS

uk.rs-online.com

Not cheap either, but with up to 5A threshold and 1s delay you can get reasonable selectivity with some final circuits not on RCD protection but just smaller MCBs (e.g. any MCB that trips on magnetic threshold due to hard fault will be ignored, so long as you don't get many amps to earth without anything noticing).

This is probably cheapest option, as Julie suggested adding RCD to switch-fuse:

100A 4 Pole 300mA Type A Time Delay RCD

Insulated DIN rail clips

www.cef.co.uk

Personally, I would go for the mccb option, as down the line you can use mcbs etc to achieve ADS, without the need for rcd functionality. This could have significant advantages on particular loads or essential services, which are often needed in industrial or similar situations.

It also allows stacked discrimination, so further sub-boards could be installed downstream of this one under consideration.

I mentioned the RCD option of course because it is a cheap as chips in comparison, but doesn't have the flexibility that may be needed.

I would imagine a mccb would be a significant cost saving on running additional cables after the event.

In essence there are a number of options and which is best for the situation is impossible to ascertain without full knowledge of the overall system and usage.

 

[pc1966]

Personally, I would go for the mccb option, as down the line you can use mcbs etc to achieve ADS, without the need for rcd functionality. This could have significant advantages on particular loads or essential services, which are often needed in industrial or similar situations.

Are you talking about using a MCCB with the high current delay RCD feature (as above), or going just with one such as this:

LV429795 | Schneider Electric, NSX Range MCCB 3P 100A, Breaking Capacity 10 kA, Fixed Mount | RS

uk.rs-online.com

That could have the short-term pick-up set to around 300-400A so it can disconnect on earth fault currents of the expected Zs, but the short-term delay of 0.02s is enough to allow MCB selectivity on hard faults?

 

[Julie.]

Are you talking about using a MCCB with the high current delay RCD feature (as above), or going just with one such as this:

LV429795 | Schneider Electric, NSX Range MCCB 3P 100A, Breaking Capacity 10 kA, Fixed Mount | RS

uk.rs-online.com

That could have the short-term pick-up set to around 300-400A so it can disconnect on earth fault currents of the expected Zs, but the short-term delay of 0.02s is enough to allow MCB selectivity on hard faults?

The long term delay, you can get a pretty standard nsx type mccb, but add the earth fault module, this allows various settings for earth fault level from 30mA through 30A and delay from instantaneous to 4secs.

So quite flexible - many ways to skin a cat though

 

[pc1966]

The long term delay, you can get a pretty standard nsx type mccb, but add the earth fault module, this allows various settings for earth fault level from 30mA through 30A and delay from instantaneous to 4secs.

So quite flexible - many ways to skin a cat though

OK that looks even better.

That is one thing I hate about the Schneider (and other) MCCBs is the huge number of variants and the rubbish catalogue/web ability to find them or use any systematic product code to divine what it does, etc. I guess if I worked with them more it would become easier, but I don't!