by Simon Brown
A lot of users out there seem to be struggling with discrimination issues that arise for various reasons when creating your power distribution networks in our software. There are some very good sections in the manuals which highlight the issues with detailed explanations, but I thought I’d consolidate this information in a blog which should answer most of your queries.
There are 2 types of discrimination calculations:
Calculated (in the absence of discrimination tables)
For the current version, the overload discrimination is checked by ensuring that the time current co-ordinates of the CPDs in series that do not overlap. The calculation routine will normally ensure that this does not occur.
For Short circuit levels the fault is assumed to be at the far end of the circuit, and in the case of an MCB in series with a BS88 fuse upstream, we check that the I2t let through of the MCB does not exceed the pre arc energy of the fuse.
For example, if the short circuit at a board has already been calculated at 4.55kA. The fault at the end of the circuit will be given by:
1.5mm = 29mv/A/M. The phase to neutral short circuit will be the same value as the neutral is the same size as the phase conductor.
E.g. (29 x 10(M))/1000 = 0.29 + Zsc. Zsc = 240/4550 = 5.274 x 10−2.
Total Z at end of circuit is = 0.29+5.274 x 10−2 = 0.34274.
The SC current at the end of the circuit is now therefore:
240/0.34274 = 700.24A. The let through of the MCB at this fault is 1900 I2t
Therefore, no problems are encountered.
Where present combinations are set up by manufacturers and are present in the database the program immediately refers to these. The discrimination tables are either absolute discrimination where the max SC fault rating of the downstream devices is entered as below signified with a “T” on the tables, or where specific values are entered in kA.
Below is a typical example of an upstream NSX630N Mlogic 6.3 device with total discrimination with an NSX100S -TMD with a max SC fault rating of 100kA as below:
The manufacturers table is as below showing total discrimination “T”:
And below is the total discrimination entry of 100kA entered in the Cymap CPD database against the upstream device.
Below is a slightly different example citing the upstream device, where entries are made where 100% discrimination isn’t guaranteed based on the entries from the manufacturer’s table (on page 2) with an upstream evice of the NSX Mlogic 2.3 and the downstream NSX 250S TMD device with an entry of 4.8kA for each size/setting of downstream device.
If these combinations of devices don’t have entries, this is where the program refers to the calculation principles as opposed to referring to tables.
These tables are accessed from the CPD database you are using by selecting the upstream device, then entering the SC rating (if total discrimination “T”) or the tabulated values.
Remember manufacturers only test within their own ranges and generic devices such as BS88s or BS1361’s etc. If you are going to mix and match different manufacturers devices, then you would either need to set up the tables and seek advice or rely on the calculated method based on prevailing fault conditions on your particular job.
Also, it’s worth bearing in mind the combinations are infinite of which manufacturers only publish specific combinations so sometimes this is a 2-minute task you can undertake to enable your project, discrimination wise, to be completed.