Do you actually mean in series ? As in a 460V supply feeding two 230 windings in series, or did you really mean in parallel - as in the same 230V going to both transformers supplied by the same switch ?

In the first case the inrush current is the same, but the voltage is doubled.

In the second case the voltage is the same but the current is doubled. 

In either case there is twice as much core metal to magnetize, and as each core still sees 230V it will take the same time, so the transient duration is totally unaffected in either case.

Mike.

PS ' five times' sounds very rule of thumb and will depend on the actual transformers - is this a figure from the transformer makers or just well meant advice ?

Toroid for example have much higher inrush than conventional E-I ot T-C lamination designs as there is a lot less copper for the same KVA (but they are a sod to wind so end up more expensive despite being lighter for the same power)

Hi, the are in series the secondary of one feeding the primary of the other, step down step up, the transformers are specifed at 5 x inrush.

So double isolation - in effect the first transformer is loaded by the other.  Unusual, rather than a single 1:1 transformer, - I assume you have a high voltage platform to supply or something.
In any case, the situation will be similar to the parallel case - there is almost no delay - the first half cycle on the primary of T1 will be the first half cycle on transformer T2 as well. You may expect more or less twice the peak in-rush current, but probably not quite due to the conversion losses in T1.

The application is to run the cables under a track which has to be at 110V AC. this config is used on all grid networks ,(step up/step down?), I assumed there would be an increase in inrush current but the designers argued it would not, and so they are not planning on uprating the fuse. 

I have been unable to find anything published on the internet to confirm what current to expect through the fuse, other than modeling it which I am unable to do.

The fuse current  rating should not be based on inrush anyway - except that what you do not want is a 'fast blow' fuse. Now some of the fuses in BS88 are real slow  and 25A seems about right for 5KVA max load.

I have highlighted the 25 amp curve in the set below and unless you think the inrush will take more than half a second or so, you should be good for well over 100A.. and at that level the dip in the supply voltage may well help you.

Mike

They calculated the single inrush at 108.7A, so x 2 would be a max of 217.4A which would trip in under 300ms?

It is a BS88-2 system E

300 milliseconds is ages  in terms of inrush of a transformer, remember that a full mains cycle (50 per second) is 20ms. - you are looking a the best part of 15 cycles - but of course the  inrush waveform is not a constant - the first cycle is really badly affected in this case a couple of hundred amps,  and the second one less so etc, and  within a few cycles it settles down to a more sensible level.

The curve I showed may not be quite right for your fuse, but it won't be a million miles out and I'd not expect problems in practice.

A 25 amp B type  breaker on the other hand would be no use at all...

Mike

300 milliseconds is ages  in terms of inrush of a transformer, remember that a full mains cycle (50 per second) is 20ms. - you are looking a the best part of 15 cycles - but of course the  inrush waveform is not a constant - the first cycle is really badly affected in this case a couple of hundred amps,  and the second one less so etc, and  within a few cycles it settles down to a more sensible level.

The curve I showed may not be quite right for your fuse, but it won't be a million miles out and I'd not expect problems in practice.

A 25 amp B type  breaker on the other hand would be no use at all...

Mike