Help with cable calculation please

GASC DAVID · Feb 16, 2023

Hi
Help please with the adiabatic
I am doing a small design for my course and I can't figure it out. Any advice will be appreciated.
Ring Final
Ib: 30A
In: 32A
It: 20A
Iz: 21A 2.5T&E method 100
V: 229V
Ze- 0.22 Ohm
L: 26m

So r1+r2 = 19.51*26*1.2/1000=0.6 Ohm.
R1+R2 = 0.6/4 = 0.15ohm
Zs=0.15+0.22=0.37ohm

Ia=229/0.37=618A
t=0.1
k=115
and then I get s=1.69mm2 for the cpc which doesn't work for the 2.5,1.5 t&e.

What am I doing wrong

This is quite a standard ring only 26m how can it be that the cpc isn't adequate.

Many thank

pc1966 · Feb 16, 2023

GASC DAVID said:
What am I doing wrong

This is quite a standard ring only 26m how can it be that the cpc isn't adequate.

You are using 0.1s for the disconnection time t, that might be the bottom of the graph but for a MCB/RCBO it is not the realistic disconnection time which is usually in the region of 0.02 to 0.005s.

For fuses you can rely on the current-time plot for computing I2t (in fact they are designed for that and 't' is known as virtual time as it matches the I2t let-through) but for MCB you really need to look at the manufacturer's data as sometimes they will give you plots of I2t to use for such a calculation.

For example, out of the Hager commercial cataloughe for the NBN series (the line at 10kA is the devices limit):

[ElectriciansForums.net] Help with cable calculation please

GASC DAVID · Feb 16, 2023

pc1966 said:
You are using 0.1s for the disconnection time t, that might be the bottom of the graph but for a MCB/RCBO it is not the realistic disconnection time which is usually in the region of 0.02 to 0.005s.

For fuses you can rely on the current-time plot for computing I2t (in fact they are designed for that and 't' is known as virtual time as it matches the I2t let-through) but for MCB you really need to look at the manufacturer's data as sometimes they will give you plots of I2t to use for such a calculation.

For example, out of the Hager commercial cataloughe for the NBN series (the line at 10kA is the devices limit):
View attachment 106239

Many Many thanks

GASC DAVID · Feb 17, 2023

pc1966 said:
You are using 0.1s for the disconnection time t, that might be the bottom of the graph but for a MCB/RCBO it is not the realistic disconnection time which is usually in the region of 0.02 to 0.005s.

Correct me please if I am wrong but from the diagram you provided 32A CB doesn't actually go bellow 0.4s

Pretty Mouth · Feb 17, 2023

The diagram @pc1966 posted shows current (I) against let-through energy (I²t). For calculating the adiabatic, it's much more useful than a current vs time graph: you just put the I²t data corresponding to the fault current into your calculation rather than calculating I x I x t.

This is the time vs current graph for the same devices:

pc1966 · Feb 17, 2023

That diagram is for I^2 * t and in kilo A2s units so the minimum you see if when it has just hit the "instant" magnetic trip (they don't show 3-5 * In but typical 4*In value).

The minimum I2t in that plot for a 32A device is about 0.25k = 250 and for a current of about 150A so the equivalent time is 250 / 150^2 = 0.011 = 11 milliseconds.

EDIT: Just beaten to the post by @Pretty Mouth !
EDIT2: Doh! Was looking at wrong curve for 32A - updated numbers...

pc1966 · Feb 17, 2023

If you re-calculate for max current, then let-through is about 50k at 10kA PFC, so time is 50,000 / 10,000^2 = 0.5 millisecond.

If is not actually opening that fast. It is a current-limiting device (both its own impedance and the gradually opening contacts and arcing) so while the circuit appears to be delivering 10kA for 0.5 ms in fact it is probably delivering something like a third of that for around the 3-5ms region.

If you look up fuse characteristics, or some other devices where peak current is a factor you need to design for, e.g. cascading to down-stream devices with more limited breaking capacity, then you will see plots of peak current versus the theoretical peak fault current.

pc1966 · Feb 17, 2023

Also that worst-case I2t of 50k A2s leads to a cpc size of just under 2mm, larger than 2.5mm or 4mm T&E provides, but within 6mm T&E. However, 10kA is a very high PFC for any domestic system to see and it does not take much of a cable length to get that down to manageable values for this sort of use.

For example, 10kA and Umax of 253 volts implies a Zs of 0.025 ohms, to bring that down to, say, 3kA PFC you would need Zs of 0.077 ohms, or R1+R2 of 0.052 added. For 2.5mm T&E at 19.5 mOhm/m that is 2.7m of cable.

See also Table B7 on page 146 of the latest OSG for example limits for generic energy-limiting MCB devices.

pc1966 · Feb 17, 2023

Finally for the domestic situation you also have the DNO's fuse in the 60-100A region. If you really did have a PFC of, say, 10kA at the DB and a fault occurred then that fuse is going to blow* along with the MCB tripping and the fuse itself has a current-limiting action, so in reality the MCB will not be interrupting 10kA but something like a third of that.

Domestic CU are certified for this sort of thing, so typical 6kA MCBs don't have a risk even for cut-out PFCs to 16kA.

[*] the Hager NBN 32A MCB is selective with a 100A BS88 fuse to 7.4kA, and with a 63A fuse to 1.3kA, so above those values and both will go. See the Fuse-MCB selectivity tables in the read of their commercial catalogue.

GASC DAVID · Feb 17, 2023

Pretty Mouth said:
The diagram @pc1966 posted shows current (I) against let-through energy (I²t). For calculating the adiabatic, it's much more useful than a current vs time graph: you just put the I²t data corresponding to the fault current into your calculation rather than calculating I x I x t.

This is the time vs current graph for the same devices:

Thanks so much for taking the time to explain it to me.
Great group thanks for your help.
Have a wonderful weekend

GASC DAVID · Feb 17, 2023

pc1966 said:
Finally for the domestic situation you also have the DNO's fuse in the 60-100A region. If you really did have a PFC of, say, 10kA at the DB and a fault occurred then that fuse is going to blow* along with the MCB tripping and the fuse itself has a current-limiting action, so in reality the MCB will not be interrupting 10kA but something like a third of that.

Domestic CU are certified for this sort of thing, so typical 6kA MCBs don't have a risk even for cut-out PFCs to 16kA.

[*] the Hager NBN 32A MCB is selective with a 100A BS88 fuse to 7.4kA, and with a 63A fuse to 1.3kA, so above those values and both will go. See the Fuse-MCB selectivity tables in the read of their commercial catalogue.

GASC DAVID · Feb 17, 2023

Wow . I wish you were my teacher at college.
Thanks so much for all the detailed answer. I need a couple of hours to understand them but I will. You are awesome many thanks

GASC DAVID · Feb 17, 2023

pc1966 said:
That diagram is for I^2 * t and in kilo A2s units so the minimum you see if when it has just hit the "instant" magnetic trip (they don't show 3-5 * In but typical 4*In value).

The minimum I2t in that plot for a 32A device is about 0.25k = 250 and for a current of about 150A so the equivalent time is 250 / 150^2 = 0.011 = 11 milliseconds.

EDIT: Just beaten to the post by @Pretty Mouth !
EDIT2: Doh! Was looking at wrong curve for 32A - updated numbers...

so I^2t/I^2=t . if I understood correctly. Do you know where can I get the I^2t chart for MK RCBO

pc1966 · Feb 17, 2023

GASC DAVID said:
so I^2t/I^2=t . if I understood correctly.

Exactly, though as already mentioned the "t" computed from let-through energy and PFC is known as "virtual time" as the actual disconnection time differs (since the fault current waveform is not a simple on/off block of the measured PFC).

The virtual time is useful and normally what you want to compute is I2t let-through and so from a curve and specific current you get a meaningful answer. This is more the case for fuses where they usually give curves that are readable all the way down, but even when they don't, fuses are normally very good at energy-limiting so the upper values of I2t are rather constant.

GASC DAVID said:
Do you know where can I get the I^2t chart for MK RCBO

You would have to ask MK, and hope they respond in a helpful manner.

Not all companies publish all of the data you might want to know about their range of products, and that is particularly true of consumer stuff, more so than industrial. The usual assumption is you just need to know the generic standard's limits and work to them, which is the basis of the OSG Table B7 and all of the "standard circuits" listed in Table 7.1(ii)

Hager don't (to my knowledge) publish curves for their RCBO either, the let-through must be a bit higher than the MCB (as selectivity with BS88 fuses, etc, is less for them) but still within the limits of its generic energy-limiting device class.

GASC DAVID · Feb 17, 2023

pc1966 said:
You would have to ask MK, and hope they respond in a helpful manner.

Not all companies publish all of the data you might want to know about their range of products, and that is particularly true of consumer stuff, more so than industrial. The usual assumpti