How to calculate site load and availability for expansion?

[ethanmiah]

I have been tasked with seeing if a site is capable of accommodating a large new installation.

I have done some working out but I'm not sure if I am going down the right lines, as this is not my area of expertise.

The site has a main fused switch disconnector on the main incoming supply, rated at 315amp. This then feeds a 300amp isolator, which provides power to the busbar chamber, where all the various DB's are fed from.

I have received usage data from the energy supplier. They have provided data from the previous 9 months. I have chosen the month with the highest usage. This was 13347.7kwh. This was over a 30 day period. I have then taken this value and divided it by the usage time over that 30 day period.

The site was running for 22 days out of 30, and for 10 hours a day. This gives me 220 hours running time. Therefore, I should have a kw value of 60.67kw (13346.7/220).

I have then used the 3 phase power formula, P=root3 x VL x I x Cos(theta), to find I. I have used a line voltage of 440 and an estimated power factor of 0.9.

I = 60.67kw/root3 x 440 x 0.9 = 88.45Amps.

Then comparing this 88amps to the total 300 amp supply. There appears to be large amounts of room for expansion and the site is no where near to overloading the supply.

Have I gone down the correct route here or am i completely wrong? is there a better way to do this?

Cheers

 

[davesparks]

For an existing site the best thing to do is connect a data logger for a few weeks and get actual data so you can see what is going on.

Your maths is flawed as it assumes that the load is perfectly balanced and constant. In reality the load will not be balanced and will fluctuate throughout the day/night

 

[snowhead]

As above a logger is the only way unless you can get hourly or less data from the energy supplier.
If it was an Office building it may have a fairly constant load, but if it's a factory or other production facility it could have wildly varying usage throughout the day and possibly a massive peak at start of day.

 

[plugsandsparks]

The only way i would do it is to use a logger as you will not only get sustained peak current but also power factor to help decide if some correction would increase capacity available. Get you skates on though as the period between Nov - Feb is typically (for my observations over time) the peak months. Measure all three phases as well as you may need a little jiggery to balance out the phases as you approach the limit of the supply. I have used VARservices in the past if you are struggling for a load survey

 

[Lucien Nunes]

As well as the bare numbers, it might be worth trying to identify a few of the largest contributors to the total in case there are varying usage scenarios that you don't capture on the logger. For example, we had a site where provided only one studio was in use, it looked fine, but every now and then both would be at full load and it was pushing the supply fuses 10% over their rating.

 

[pc1966]

I would go with the above. Hire a data logger to run for a week, or as @plugsandsparks suggests it might be easier and more reliable to get folks in how have a logger and know how to configure and save the data correctly from the start.

Also @Lucien Nunes point is very important: try and get some sort of a list/audit of what they think is in use and do a walk around with the head folks to confirm the list, and then try to get ideas of what sort of a current range they ought to be drawing, the typical usage pattern, and see if it matches the logger's results.

In some cases you will see less drawn then expected as often nameplates, etc, give max demands, in other cases you might find they have more loads in use than documented!

 

[plugsandsparks]

Lucien's point is a little raw for me as my last job was trying to work out why a plant that had been happy with 315a fuses, started to blow them. I went up and although could not work out why the load had gone up, it was not a fault just load and a lot more. After much head scratching we worked out some process heaters (which should only be used overnight) had started occassionaly coming on during a production run.... the reason ? - new chemical supplier delivering product to site at a lower temp than previous supplier and making heaters kick in. Solved by interlocking these heaters against process run.

 

[ethanmiah]

For an existing site the best thing to do is connect a data logger for a few weeks and get actual data so you can see what is going on.

Your maths is flawed as it assumes that the load is perfectly balanced and constant. In reality the load will not be balanced and will fluctuate throughout the day/night

Cheers mate! Appreciate the help.