Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Load switch with high gate current drive?

Status
Not open for further replies.
T

treez

Guest
Hi
We have a 48vout, 12Kw output power supply powering a load.
We wish to use a load switch comprising paralleled back-to-back NFETs. The load switch must temporarily switch a battery in to supply the load if the power supply drops out due to a heavy load transient.
Also, if the power supply fails, then the battery must also be switched in permanently.
The load switch should also be switched off if the power supply starts charging the battery (the battery should only be charged via the charger).

As such , our contractor is thinking of using the LTC4368 Load switch IC...

..However, this IC only has 35uA of gate current capability. As such, the Minimum gate source resistor for the 8 pairs of paralleled back-to-back NFETs is 390kOhms. This I believe is way too high, I think we need at the most 22k gate to source to prevent the gates from suffering ESD damage….specially since we will be handling the PCBs a lot in testing for this prototype.
Also, the 16 FETs of the load switch have a combined Vgs of 240nF. As such, the LTC4368 would take about 50ms to turn these fets on after a load transient that made the PSU drop out…this is too long and means the battery wont be able to supply these transients.

Do you know of any Load switch ICs which have higher gate drive current capability? I looked everywhere but cant find one

LTC4368
https://www.analog.com/media/en/technical-documentation/data-sheets/LTC4368.pdf
 

Here is the Load switches as attached.
The FETs used will actually be IRFP4468.
Its an awful lot of TO247 FETs....and needs all the comparators and latchs etc as there appears to be no IC for this.
 

Attachments

  • battery management2.pdf
    32.9 KB · Views: 122

Have you considered using diodes to steer the current (or some of the current) instead of MOSFETS?
For example, Vishay VS-175BGQ045, two in parallel in series with another two in parallel. Gives you ~1V drop at 240A which is probably comparable with two MOSFETs with source resistors at that current.

Brian.
 

Thanks, yes we are toying with just using diodes...since the battery is only for use for brief transient dropouts, and also for when the PSU fails....so its disaster management...so yes.......diodes are looking good.

By the way, the back to back fets are just about 8mR (total) in rdson, (please forget the series resistors) so with 8 in parallel, the Volt drop is only some 0.24V...so using diodes gives us an extra 180W of dissipation...but then again, the battery is only for when the psu fails, and for transients...so i reckon you are wise to advise diodes.
Maybe a good few more diodes in parallel

...aaah woops...sorry, i forgot, we cannot use diodes, because the battery would simply supply the load when the psu voltage was slightly low....lest we can think of some kind of half diode, half switch situation.
 
Last edited by a moderator:

sorry, i forgot, we cannot use diodes, because the battery would simply supply the load when the psu voltage was slightly low
Isn't that exactly what you want - the battery supplying the excess when needed? Or, is there a risk of the PSU going under voltage for long periods?

Brian.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Thanks, actually we don’t ever want the battery to supply the load in normal case…the battery is solely for times when the PSU would drop out for a few hundred milliseconds due to a heavy load transient…or for times when the psu has totally failed.
So The battery is emergency only.

Actually, using diodes will be difficult, because LiPO batteries have a varying terminal voltage with temperatrure…as the following tells
https://www.gensace.de/blog/temperature-affect-lipo-battery-performance/
…so just using diodes will be a poor solution, because in hot weather the Lipo terminal voltage would rise and it would then start suppying the load
 

The PSU wont drop out for long periods no...it will stay at 47.5-50VDC (depends on loading as it has 'droop sharing) unless it fails...or a temporary transient occurs
 

Hi treez,

Any chance of an image of the schematics instead of pdfs of them? ;)

Maybe this part might fit what you're after? Ideal diode for reverse current protection and/or "Active ORing for redundant power". Page 21, figure 36 = or-ing for redundant power supplies. You provide the NFET and the enable signal.

LM74700-Q1

Ideal diode & ORing controllers

Hope the part fits.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Thanks,
The LM74700 is a good part, thanks for sharing it.
However, in our case, the “diode OR” function is not what we seek….the LM74700 would allow the battery to supply the load in cases of high battery terminal voltage (eg when temperature was high)…and we do not wish for this.
Also, the “through” current is clamped at whatever value gives 20mV across gate to source of the fet(s)…that’s inconvenient as FETs have varying rdson with temperature.
Also, the LM74700 will allow reverse current up to the value of amps that gives 11mv from source to drain…again….we do not want that…we don’t want any reverse current really

- - - Updated - - -

Also, the LM74700 can only supply a continuous gate current of some 10's of microamps...which means we cannot have say a 10k gate source resistor to prevent ESD issues, or to damp ringing at the fet gate....this is something unfortunate that we note in all of these high side fet driver type load switchs....their continuous gate drive current rating is ridiculously low.
 
  • Like
Reactions: d123

    d123

    Points: 2
    Helpful Answer Positive Rating
Again, if absolute output voltage stability isn't an issue and you can afford the 'dip' to reach 1V, consider high current SCRs as the switch. It isn't as complicated as it sounds, put one in the battery 'branch' and apply a trigger voltage when it is needed. As the battery and PSU reach equilibrium again there will be no voltage across it so it will turn itself off.

Brian.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Hi,

Sorry about that. Wasn't too sure of what you needed. I only looked at a few but noticed all load switches I saw for current over 6A are at most 20Vin.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
I've used the LTC7000 as a load disconnect in the past. Unlike parts like the LTC4368, it has a high current gate driver.

I kinda like betwixt's idea of using an SCR though. Probably simpler to deal with than a fancy IC.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top