Anyone ever do inrush current limiting with a MOSFET? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Anyone ever do inrush current limiting with a MOSFET?

1 2 next

Move Thread

LAN_403

jpsmith123

Wed Apr 22 2009, 08:12PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

I've seen an application note on the topic, but I'm having a hard time finding some "real life" examples of people using MOSFETs for this purpose.

I'm wondering, has anyone here tried it, and if so how did it work out?

Dr. Dark Current

Wed Apr 22 2009, 08:55PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Why would you want to do that? A MOSFET will always waste some power all the time. A resistor (~10ohm) shorted after a hundred ms by a relay is much better.

jpsmith123

Thu Apr 23 2009, 05:13PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Well, yes, but a pair of beefy MOSFETs in parallel for example won't waste much power, and it would eliminate an annoying electromechanical part.

Dr. Dark Current

Thu Apr 23 2009, 05:36PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Well it can be done, but I would leave the inrush limiting to a resistor, then short it with a MOSFET.

If you are using a bridge rectifier in your circuit, you can put the inrush limiter after the bridge so you can use only one FET to short the resistor.

If you are not using any bridge rectifier, you can put one across the resistor so the FET again switches only DC.

I would not mess with putting the transistor into linear region for the limiting part, because of secondary breakdown and unreliability, but maybe someone knows better.

Proud Mary

Thu Apr 23 2009, 05:42PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

I am famously old -fashioned and would have chosen the relay - which will produce good, reliable results - but am sure there are more modern solutions for those who don't like to hear clicking sounds from their equipment!

jpsmith123

Thu Apr 23 2009, 08:44PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

Well I just looked at the schematic for my Mastech 30V 20A power supply, and they used an SCR to short the charging resistor. The higher powered supply, the HY5020E, also uses an SCR. And looking at the data sheet for the SCR, it looks like they'll have 0.8 to 0.9 volts across it.

Sulaiman

Thu Apr 23 2009, 09:11PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I repair a lot of industrial electronics and I'd say the most common inrush methods appear to be;
-Ceramic wirewound resistor shorted by a relay contact
- As above but shorted by a triac (ac side of rectifiers) or thyristor (dc side)
-NTC inrush resistors for a few hundred watts and below

Newer equipment mostly uses some kind of pfc at the front end.

A neat 'trick' I've seen used for smps is to have a winding on the inverter transformer continuously 'fire' the inrush triac or thyristor.

I have not come across any equipment that uses a mosfet as the inrush switch,
triacs and especially thyristors are so much more rugged.
Low on resistance with high voltage rating has always been an expensive combination for mosfets.

GeordieBoy

Fri Apr 24 2009, 11:31AM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

As a designer of SMPSU's of various sizes i'd agree with Sulaiman's comments above. NTCs rule for low power inrush limiting up to a few hundred watts. Then at high powers use a power resistor (or NTC + power resistor in series) switched out by a relay or thyristors once the DC bus has been pre-charged and the inrush event has passed.

MOSFETs are un-necessarily lossy as an inrush bypass switch because they are always resistive. This would be particularly troublesome in a universal PSU where the MOSFET would need to be rated at 500V for safe operation at maximum line voltage, but would have to carry a massive line current down at minimum line voltage. The Relay wins hands down with its very low on-resistance, and the thyristor comes in a very close second with its almost fixed voltage drop.

As Sulaiman said, a common solution is to wind an auxilliary winding on the output transformer of a SMPSU, rectify this and use it to close the inrush-bypass relay. So when you first switch the supply on, the bus caps charge slowly through the inrush resistor, then the PSU controller enters it's soft-start phase and charges the output capacitors smoothly, then once the output is up to level, the bypass relay closes and the inrush resistors are shorted out to maximise efficiency. An advantage of switching out NTCs for inrush protection is that they have time to cool during normal operation ready for the next time the power is cycled. (NTCs remain hot during operation if they are not switched out and there thermal maxx means that they do not provide adequate inrush protection after brief loss of mains supply.)

Thyristors can work well to bypass an inrush resistor, but a few precautions need to be taken. The main weakness of thyristors in inrush-bypass duty is something called dv/dt induced turn-on. If the mains supply is switched on near the peak of a cycle the abrupt rise in voltage across the thyristor can cause it to spontaneously turn on. Such turn on at the peak of a mains cycle bypasses the inrush resistor exactly when it is most needed!!! The resulting huge current spike can cause the incompletely turned-on thyristor to fail due to hot-spotting of the die. Snubbers across the thyristors or "snubberless" thyristors with high dv/dt immunity are recommended for inrush bypass duty. Care also needs to be taken to ensure that the thyristors remain in conduction for the full mains cycles otherwise line-current distortion results.

For supplies with active PFC front-ends the same inrush limiting requirements exist, but it is desirable to disable operation of the PFC until the inrush event has passed. It is also desirable to use a dedicated beefy inrush diode to direct the inrush current around the boost cell rather than letting it pass through the boost choke and fast diode which is easily damaged by over-current.

-Richie,

Avalanche

Sat Apr 25 2009, 05:52PM

Registered Member #103 Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845

some old inverters I'm repairing at work at the moment have a variable DC link that uses a phase angle controlled rectifier arrangement. The inrush is limited simply by soft-starting the phase angle controlled bit. It's a fairly neat overall setup if the variable DC bus is of any use as well!

Other larger stuff we do just uses a large wire wound resistor, shorted by a whopping great contactor once the bus is up to voltage.

Inrush limiting with MOSFET just seems wrong, if it's in an application note it's probably just the manufacturer trying to show off their device

jpsmith123

Sun Apr 26 2009, 04:44AM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

I should have posted a link to the application note. Here it is:

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob