MOSFET design question..

Ed said

If this switch needs to be closed for long durations, you'll need a DC supply higher than the high-voltage-rail to provide required gate drive.

If this switch needs to be closed momentarily you can bootstrap the gate drive with a simple cap. Simpler yet, look into an IR2127.

Maybe easier: use a smartFet like Infineon BTS 728-L2, these devices are slow, perhaps you are intending to xmit data with these signals?

Your requirements are ambiguous. More detail required

That is what I was thinking, I cannot think of an easy way to do this.

Interesting - I dont know anything about these. I shall have a look.

I would like the main supply to be around around 40amps capable, with supply range from 12-28volts (either one or two lead acid batteries in series, assuming fully charged or indeed being charged etc). The data is not a problem now, this will simply be 5v and this I can do from there is plenty of room left for thid

Ok, well I dont know if this will help. I have a main control 'desk' which supplies power to various modules upto many meteres away using a 3 wire bus of Gnd +v and data. The modules are provided by power from the bus, and this supply needs to be controlled from the desk for safety reasons. Since these modules require each potentally large (ish) currents depending on configuration, I would like to make the supply rail as tough as possible - relability is an Issue and I dont want to use relays.

Can you guess what it is yet? :) (Clue: fireworks display..)

Thanks for your help.

Ed

Ed said

So, as I understand it, you are looking to switch 12-28V, 30A in a "DC fashion". This switched power source is "on" for many minutes. This seems to rule out bootrapping.

So, perhaps parelleling some PFETS is a solution. If'n you get some TO220's (or TO247's) you should be fine with a big heat sink.

So the "data bus" is supplied from this DC switch but controller via another circuit.

These devices have integrated charge pumps and hence can be controlled in a DC fashion. Given the currents, I'd opt for discrete MOSFETS tho'.

If these currents are transient in nature, ie only present when the firework is being lit, then the average current is likely MUCH lower. In this case a single PFET seems reasonable. A single TO220 might be more than enough?

Cool!

Yes this is exactly what I need to do.

I have found some SUP65P06-20 "MOSFET, P TO-220; Current, Id cont: 60A @

Its all controlled by a pic which I've got quite far with now, but needed to finish the hardware on the high current side, and I admit got unexpectadly stuck!

Yes definately, I dont think anything other than discrete components would suit here. I nearly settled for a high current BJT but really didnt want to do this as any voltage drop is bad in this application.

Thanks for your help by the way, I asked the same question on sci.electronics.design, and have been shown a diagram, that follows what you say:

| p-type FET | Power in ------o------, ,----- Switched power out | 15 - 50V | S_|__|_ | (65V max) | ,--- | | | | '- R1 -+ for 10V gate drive, use | 1.0k | R1 = R2 (10/4.3) = 2.3 R2 | | | |/ | 5V logic-control -----| 60V NPN BJT | |\V 2n4401 | | | R2 level-shift current-sink | 430 high-side FET switching | | | GND

I think this is what I'm going to go for, or very similar anyway.

I have ordered some fets which should arrive today (tues) hopefully ill have sometihng working within an hour or so of their arrival!!

Ed

Ed said

This looks very good.

Don't forget to provide a pulldown resistor from the base of the BJT to GND to insure it's default state is "off"... if the PIC is in reset or "off in the weeds" and it's output ports are in high impedance.

Let us know how this works and have a great 4th!

I shall do so! Unfortunately I am in the UK (its hald 2am here!) and not alot happens here on July the 4th, so this system is being built in preperation for our main fireworks season of around November and new year.

I'm about half way to having a completed prototype now, and I have to admit I cant wait for the first tests :) This project has combined so many things I have wanted to do for so long, but just never had the time or funds to do so :)

Ed said

Small world, I was working with a couple of your countrymen just this evening. They just arrived today. They don't seem interested in July 4th either? Odd. ;-D

Just hope the fireworks are sulfer-based and not silicon-based. ;-)

Cheers!

Haha - Some people in the UK celebrate this, none I know of however!

Well I now have a working P-FET current switching prototype which is great. I shall build this on to the board and finally can start programming and testing this part of the project!! I can see no other real problems left to overcome now!

Thanks again for the help!

Ed