Powering 12v pump via Mosfet

I have a project that uses the arduino uno to control a 12v car windscreen washer pump, which I've tested at 3.8A under load. It works using a 10A relay, but I'd like to replace the relay with a mosfet.

Having done a bit of digging (I am new to electronics), I've found that a N-channel mosfet (RFP12N10L) would do the job. Looking at similar circuits on the web I've noticed that a diode is placed either across the motor connections or connected at the Mosfet. Can someone clarify what the best approach for this is please?

I'm assuming that I'd need a diode rated high enough to cater for the 3.8A load, so would a 6A diode like this: http://www.ebay.co.uk/itm/350819432740?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649 be a good start?

Dave

absolute0: I'm assuming that I'd need a diode rated high enough to cater for the 3.8A load, so would a 6A diode like this: http://www.ebay.co.uk/itm/350819432740?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649 be a good start?

Dave

5 X DIODE, ULTRA-FAST, 0.6A, 50V, Part # UG06A

0.6amps... not even close :)

Doh! Wrong link. These are the ones: http://www.ebay.co.uk/itm/390423686354?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649

I recall reading "somewhere" that the specs of the flyback diode are related more to voltage than current, and should be an order of magnitude above the supply voltage, since that spike back can be huge and it takes a hefty diode to keep it in check.

Disclaimer: don't quote me, I could easily have that wrong.

JimboZA: I recall reading "somewhere" that the specs of the flyback diode are related more to voltage than current, and should be an order of magnitude above the supply voltage, since that spike back can be huge and it takes a hefty diode to keep it in check.

Disclaimer: don't quote me, I could easily have that wrong.

No, the freewheel diode needs a voltage reading adequate for the supply voltage itself, and a repetitive pulse forward current rating appropriate for the current through the load.

You might be thinking of some other circuit where the inductive voltage spike is not suppressed.

Also you can find far superior MOSFETs to the RFP12N10L - aim for an Rds(on) of less than 0.03 ohms for that current level. And logic-level of course. The RFP12N10L will generate 3W of heat. Don't go for a high voltage MOSFET, high voltage and low on resistance are at odds with each other. A 30V part is the sort of thing.

OK, so I think this mosfet should be a better fit (STMICRO VNP14NV04-E 40V 12A TO-2203) http://uk.farnell.com/stmicroelectronics/vnp14nv04-e/mosfet-omnifetii-40v-12a-to-2203/dp/1739424.

Do you think this would be ok?

Anyone?

Thursday Bump

That's not just a MOSFET, its a device with built-in protection circuitry including current limiting. You motor is possibly going to trigger its limit circuit which may or may not cause issues.

You saw this warning in the datasheet?

When the current limiter is active, the device operates in the linear region, so power dissipation may exceed the capability of the heatsink. Both case and junction temperatures increase, and if this phase lasts long enough, junction temperature may reach the over temperature threshold Tjsh

Why not choose a straightforward MOSFET?