Need help with current limiting. / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Need help with current limiting.

first previous 7 8 9 10 next last

Move Thread

LAN_403

DerAlbi

Thu Jan 14 2016, 07:51PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Yeah hen, the safety margin would be at least 2 in your case.

Rectifiers allways need to withstand double the output voltage! (in push/pull)
Check Patricks topology image again. Its actually said directly there:
V_diode = 2*Vin*Ns/Np

Patrick

Thu Jan 14 2016, 10:09PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

I'm looking at 100 and 200 volt Schottky diodes. I'm needing 10 to 75 volts ouput. I can always wind the secondary to the voltage I want. Ill make at least 3 of these supplies.

DerAlbi

Thu Jan 14 2016, 11:13PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

if you want 75V, then 200V is minimum.
You dont necessarily need Schottky... if you find a nice soft recovery diode, i would allways prefer this. Remember.. schottky for 200V are 2 shottkys in series internally, so you do not have the low forward voltage drop benefit anymore. You only have the higher leakage and the really hard switching.

Patrick

Sun Jan 17 2016, 06:43PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

I'm putting in the order to Mouser tonight. ill use the ultra fast recovery diodes for 200+ volts. so ill have some margin at 75 volts output.

Carbon_Rod

Sun Jan 17 2016, 11:02PM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

I prefer the IXFB100N50P, and noted that "extra" parallel protection diodes can still cause thermal runaway.
At least try to bolt everything to the same heat-sink...

Cheers,
Rod

Patrick

Mon Jan 18 2016, 02:18AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Carbon_Rod wrote ...

At least try to bolt everything to the same heat-sink...

Yes I already have the heat sink drafted up. One for the heat sink for the mosfets. One for the diodes. Two total.

EDIT: components will be here in 2 or 3 days. Ill work on the heatsinks and board etch until then.

Carbon_Rod

Mon Jan 18 2016, 12:20PM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Patrick wrote ...

Yes I already have the heat sink drafted up. One for the heat sink for the mosfets. One for the diodes. Two total.

Recall that under load, the smaller protection diodes in the fets may heat up first, and therefore will want to draw more current through itself (heats up even more) rather than the larger/cooler-running external protection diodes on the other heat sink. It is something we keep in mind if paralleling FETs, and expecting the internal diodes to remain in SOA.
Note, the fet itself increases RdsOn under load so does not really have the same runaway problem. However, the external diodes should be placed on the same heat sink next to the fet.

DerAlbi

Mon Jan 18 2016, 07:36PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

Carboin_Rod... you got a good point there with the thermal runaway of the body diodes. (i dont think that its relevant for Patrick atm)
I find this specially interesting, because i never really thought about that before. Why? When i know my body diode is conductive, i will allways turn on the mosfet too! Its not a one-way device. The Rsdon passes current in both directions. So its an active diode then... a bit more efficient and its for free. why not.
Well.. it still depends on the application.. i know. But there arent many applications where reverse current is unexpected and you need to parallel switching elements.

hen918

Mon Jan 18 2016, 07:55PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

Well, the body diode is the MOSFET, it just acts like a diode in the reverse direction. They are always rated for the full load of the MOSFET because they are the MOSFET. It's only if you are expecting pulsed currents greater than the MOSFET can handle in the reverse direction that you will need them, and then you have to remember the current will be shared between them. Another thing to remember is that the body diode typically has high switching losses, so it's best if you don't let it get forward bias at all.

Patrick

Mon Jan 18 2016, 09:36PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

the FDP26N40 is capable of 26 amps, and ill only use 4-7 amps I think.

Given my specs they should run warm not hot, and given my space two heat sinks is easier than one.

first previous 7 8 9 10 next last

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