Re: Wire Sizing
tx2step, iwire -
I think iwire has a good point. When you're shooting in the dark, it's a good thing to have a conservative rating.
In the installation I described above, the intent was to start with a low power consumption and continue building with an expected peak of about 4000 Amps. All this was figured out ahead of time. The load did grow at the expected rate. It just couldn't get past the reality roadblock.
I would guess that when someone wants to have a 1200A service, they expect to use it. I'd size the underground for the full 1200A. There's a big difference between a 1200A service and a service for a given location.
If they just want a service, then what does anyone know. Using a code evaluation may be conservative (probably is most of the time). Basically, if you use the code evaluation, you did the best you could with no information. If I had to do that, I'd want to make sure I was legal.
I've never been in a spot where I used code load calculations. All I've done is add up the loads, use some load factors (I call them utilization factors instead of diversity factors), and size the installation accordingly. Basically, I size the installation for the load. I only missed once - had to add a new transformer. Typically, the load I estimate has been within about 5% of the actual load and I expect it to grow - it usually does.
In general, I oversize the primary feeders (usually 13.8KV or so) to allow some additional capacity. That way, the user can add some more switchgear if it's needed for expansion.
In one case, there was a 200A service at 13.8KV (about 5 MW of connected capacity). I installed two 4/0 CU underground. A part of that was for voltage drop and a part of that was for general conservatism. I didn't want another 3 MW added on the assumption that all was fine. With the nature of the user, they might want an additional 3MW.
Along with doing what the customer wants, you have to know your customer.
Hope this helped,
Dave