inrush current limiting for 150A at 45V

[Lovelec]

Does anyone have a suggestion on an elegant solution for limiting the inrush current going through a diode pack that is rectifying a 45V 150A current. The caps are very big and the initial draw is forcing me to rate the diodes at much higher than the nominal draw. If I can somehow limit the inrush, I won’t have to worry about burning diodes. I am thinking of using an inductive load in series with the diodes. Any thoughts?

 

[BobM3]

PTC Thermistors?

 

[Lovelec]

thermistors I have seen are for very low current. the hghest I could find can handle up to 30A. but my requirement is 150 nominal amps.

 

[fangas]

Might think about a large current limiting resister. Use a set of relay contacts to shunt it after the inrush.

I've seen this done in two ways;

A timing relay that shunt the resistor after a fixed period and having the relay in the DC curcuit after the filter cap. When the cap is charged, the relay operates, again shunting that resistor.

Ed

 

[Lovelec]

Good idea. Thanks.

 

[itsmoked]

That is the common method. Shunting a resistor.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

I think it would be an NTC Thermistor, not a PTC. A Thermistor works by virtue of the fact that the resistance changes as they heat up. With a PTC, the resistance increases with temperature, so when you first energize it the resistance is low, just like a light bulb filament. If you use an NTC Thermistor, the resistance drops as they heat up so they are only high resistance when first turned on and the caps are charging, then when they heat up, the resistance drops (hopefully) to a point of insignificance. The benefit over the resistor + shunt relay method is no moving parts.


"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

 

[Lovelec]

How hight a current rating can I find an NTC for?

 

[itsmoked]

Probably 149A...

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

LOL, how much can you afford? Somebody will make whatever you want for a price! But there is a cost/benefit crossover point with an active (shunt) system. It's been a while since I bought any, but seems to me that 50A is probably a practical stopping point.

http://www.ametherm.com/Inrush_Current/welcome.html

"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

 

[DHambley]

I've used several methods. Like "jraef" wrote, it depends on how much you have to spend. An NTC has worked for me, but not at the current levels you mention. Space allowing, you can tie a TO247 MOSFET to a huge heat sink. Slowly increase the gate voltage over a few millisec so that the MOSFET takes the energy. How may joules do you need to absorb?

DH