inrush current limiter circuit

[Holger Barske]

Hi guys,
going to hook up quite a massive power supply, need a simple but effective inrush current limiter.
There's an old Elektor circuit (relay, some massive resistors, no secondary supply) floating around here somewhere, but I can't find it anymore.
Can anyone help?

[janneman]

Probably downloadable from the Elektor website...

http://www.elektor.de/Default.aspx?t...On&SearchText=

Jan Didden

[Holger Barske]

Thanks Jan,
but that wasn't the one I was looking for. But I finally managed to find it: http://mitglied.lycos.de/Promitheus/...or_toroids.htm
Thank you all, issue solved.

[EUVL]

I thought about the same issue for a long time for a 100W Class A bridged amp with 8A bias. In the end, my decision went to NTCs. I know there are reliability concerns, but one of these rated for high current (say 20A) has less resistance at full load than a 35A relay, even at 0.4 Imax. And it has much less components.

The other solution I still contemplate using (also for high current) is to use a RC circuit to delay-switch a high current, high gm MOSFET (say IRF044 or even better 2SK2554 -- for the negative rail) to bypass a high-power throttling resistor in say 300ms. The time constant depends on how much time you need to charge the capacitors downstream, and how long your resistor would survive. (Unfortunately no schematics at hand, but I am sure you get the idea.)

Just my thoughts.


Patrick

[Holger Barske]

Hi Patrick,
using a Mosfet for switching mains AC doesn't seem to be a good idea... Surely works on the DC site.
As for relay contact issues: Oppermann sells 24V relays with 4 sets of 4A contacts for 61 cts - I think I will give them a try (contacts in parallel, of course).

[EUVL]

> using a Mosfet for switching mains AC doesn't seem to be a good idea

Sorry, didn't know you were thinking ac.

Patrick

[Holger Barske]

Sorry, didn't know you were thinking ac.

I am sorry, I should have been more precise .

[jamesrnz]

i like relays too, however with multiple contacts you must make sure they do not weld themselves if they don't make contact at the same time. for 100w i think ntc is better.

how about parallell ntcs???

i have generally used relays with large 50a or more loads.

i am going to try parallel ntc just for the heck of it..


jimbo

[Holger Barske]

make sure they do not weld themselves if they don't make contact at the same time

I absolutely agree. The resistors have to limit the inrush current to a value a single contact can take without problems.

[jackinnj]

Originally posted by HBarske
Hi Patrick,
using a Mosfet for switching mains AC doesn't seem to be a good idea... Surely works on the DC site.
As for relay contact issues: Oppermann sells 24V relays with 4 sets of 4A contacts for 61 cts - I think I will give them a try (contacts in parallel, of course).

you can make a "Quencharc" with a capacitor and resistor -- this really does prevent the relay contacts from blowing themselves apart. The Quencharc acts as a parallel circuit to the capacitance of the relay contacts and draws current away from the contacts as they draw into proximity. As you can imagine, when the relay contacts get to be micrometers away from each other the capacitance is very, very high indeed (a relay contact which is 2 millimeters in diameter has a capacity of a few nanofarads at 10 micrometers!) A lot of heat can be exchanged in a short period of time.

I have found the best inrush circuit to be a low value high wattage resistor which is shorted by the relay after some hundreds of milliseconds -- you can run this without a separate auxiliary power source -- I use a HexFET to control the relay and a PNP transistor to discharge the gate when power is removed. ESP's design is very easy to implement.

You can also use synchronous rectification -- but that's beyond the scope of the present thread --