Inrush Current Limiter

[bootspurs]

I have a 75kVA 208-480 transformer that is part of an off-grid solar system. The transformer has a minimum 3.5x inrush current, per the manufacturer. The power source is (6) SMA Sunny Island inverters, each rated for 5000W@120V, wired in two parallel 'clusters' of three inverters.

Each cluster has a max surge capacity of 150A @ 208V for 100ms, which is not enough to start this transformer.

The system has a max capacity of 4 clusters, so it could be brought up to a total surge capacity of 600A, but this is not ideal.

What other solutions exist? I know it is possible to make resistor circuits to soft-start the transformer; can anyone recommend a company that does this, preferably that has UL-listed products?

Thanks for any ideas here. I'm not an EE so this is a bit out of my range.

 

[DHambley]

I was looking at your schematic, particularly for motors which may require huge start-up currents and I didn't see any. I'm confused at the term "starting the transformer" because transformers don't need to "start". If it has a 3.5X inrush spec, I would think that this is that just the amount of current it can withstand, not how much it needs.

 

[bootspurs]

Maybe my terminology is wrong; saturate instead of start?

The transformer only starts when the entire power system is turned on, so this should be a rare occurrence. There is no utility grid, only the 'stand-alone grid' created by the off-grid inverters. Again, with my own non-EE disclaimer, I thought that for the transformer to saturate during this initial startup period, it would pull current up to 3.5x the nameplate rating (and that's because I asked for the smallest possible inrush from the manufacturer, it would have likely been more).

Once the system is on, you are right, no big motors that would have a large inrush.

 

[jraef]

I don't think you are thinking of saturation, you are thinking of magnetization; establishing the magnetic fields that are the core of what a transformer does. There is an inrush associated with that process. But that takes place in the first cycle (16ms) and a little beyond, not for 100ms. Generally, inverters also switch on at zero volts consistently, which minimizes the worst case scenarios of transformer inrush. Are you perhaps chasing a problem that does not likely exist? Are you having issues or are you anticipating what might happen? Because there are techniques that can be used to limit transformer inrush, but they are rarely needed for small systems like this.

If you are having issues though, investigate using an NTC resistor inrush limiter. NTC stands to Negative Thermal Coefficient which means as the temperature goes up, the resistance goes down, So when you first energize across the resistor, resistance is high and it serves to limit current. But as it heats up, resistance drops so that it gets out of the way (or in some cases it is bypassed).


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[bootspurs]

So the inverters need to be able to provide the inrush current for about 16ms. The max these inverters are rated for is what I mentioned previously, 150A for 100ms per cluster.

Just to clarify, these inverters work off of a 48V battery bank, and supply AC power on demand; not your typical PV inverter.

We don't have a problem yet, the system is being built over the next month or so. All this came about because as the designers, we were originally quoted a 2x max inrush spec from the transformer company. When the contractor went to build, the same transformer company then said that they couldn't build to 2x, only to 3.5x min, so we needed a new solution.

 

[waross]

Search this site. This subject has been discussed. Skogsgurra has described some techniques to reduce transformer inrush.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rockman7892]

Jraef

Wouldn't switching at zero volts as you mentioned be the worst case scenario since the current would be 90deg behind the voltage and therefore be at max when voltage is crossing 0?