Nmos circuit for heating element

[gentlywiringit*]

I've had a Pmos diagram for a while (thanks Spec, it works beautifully!) but recently I've thought about changing the Pmos to Nmos for a new circuit, I've adjusted what I can but I'm a little confused about where to place the switch and the 470r.. does this look right?
Please excuse the really bad rendering :/

 

[Veraxis]

In this case R1 would function as a pull-up resistor, so the heating element would be normally on, and pressing the button would turn it off. I'm not sure if that is what you want or not. If you want it normally off, then I would swap the positions of R1 and the switch.

 

[gentlywiringit*]

As I thought, thanks JNLY.

There's one other thing i forgot to ask.. avelanche.. is the circuit at risk with a 7.8v source?

 

[Veraxis]

I would think it would be unlikely at those voltages, but depending on the current and the inductance of the heating element, it could potentially be a problem. The safe bet would be to add a Schottky diode or a snubber network across the heating element as a flyback diode to clamp any voltage spikes. It would have the "negative" side of the diode (usually marked with a line) facing the positive rail and the "positive" side connected to the drains of the MOSFETs.

 

[gentlywiringit*]

Current will be; 8v (allowing volt drop through circuit) to a ~0.2ohm element. I figure it as 40a 320w.
Seems overkill with the quad mosfet arrangement I know.. I need it to have the least internal resistance and volt drop to the element as I can get. It also needs to have scope for more powerful batteries in the future

Is this the circuit I've been suggested?

Is there a specific value Shottky diode I need?
*edit* I guess that would depend on the current? I can't say exactly what it'll be due to its 'one circuit fits all' nature, but the range will be from 20a right through to the full load the mosfets are capible of..
While on that subject, could anyone tell me the full range? Min to max current?

 

[Veraxis]

I think you have the polarity of the diode reversed on your schematic, but otherwise that looks fine. Switching the positions of the switch and R1 as you show will make it normally off and will turn on when the button is pushed, which I assume was your intention.

As for what Schottky diode to use, you will need one with a peak pulsed current rating of at least 40A to match the current being drawn by the heating element, but preferably a bit more for added safety margin. The reverse current spike will likely be short, and will only happen when the button is released, so the average current will likely be negligible compared to the peak current. Most kinds of power Schottky diodes, either the larger radial package ones or TO-220 package ones would probably be fine. You will need to look up the datasheets and check the maximum ratings for peak current.

Also, wow, I didn't realize the current was so high. I hope that your battery can pull 40A without any issues? Be sure to use heavy gauge wiring and keep the lengths to a minimum to avoid too much added resistance in the connections. The 4 MOSFETs in parallel should be sufficient seeing as they are being turned all the way on and have a pretty low Rdson, but I would keep an eye on them to make sure that they aren't getting hot or anything.

 

[gentlywiringit*]

Haha, I got it back to front.. sounds about right

40a would be easily attained by my battery (https://demonpowerproducts.co.uk/product_info.php?products_id=10033)
As a matter of fact it's supposedly capable of a whole lot more, 6.05kw for 21 seconds, though I've absolutely no intention of gunneapiging that claim!
I could use 4 Sony vtc4 18650 cells 2s2p or 4p to power it but I like the form factor of the lipo.

Can I piggy back each of the mosfets with its own Schottky if I match current handling of the two? Or do I need to use one for the whole circuit? I ask because I can't find one singular Shottky that has a rating equal to all four mosfet's combined capability. Size is also going to be an issue as this is a handheld device.
Also.. what is the third leg (centre?) on the TO220 style connected to?

 

[Veraxis]

I don't think that you would need a diode capable of handling the max current of the four MOSFETs. As you say, the heating element itself is going to be limiting the current to 40A, so that's all you really need to handle. You could definitely use multiple diodes in parallel if you need to, but I imagine a quick search of Digikey or wherever you prefer would probably yield many different parts able to meet the >40A peak current requirement with a single diode.

TO-220 packages are just a generic package type for various kinds of semiconductor devices, whether they are MOSFETs, voltage regulators, or diodes. Sometimes they will put a double-diode in a package with either a common anode (positive side) or cathode (negative side). I believe they also have 2-pin TO-220 packages available for single diodes. In some cases, the third pin might just be redundant, or just not connected at all. It depends a bit, but the datasheet should give the pinout details for any particular diode if you aren't sure.

If you are worried about massive current capabilities of the MOSFETs if the heating element got shorted or something, it might be a good idea to add a fuse in line with the battery. A simple blade-type automotive fuse for maybe 50-60A (or however much margin you want to have to avoid nuisance blowing) would be pretty cheap to add.

You will want to account for the fact that a cold heating element probably draws more current than a fully-heated one, so you will want your current ratings to be based on a cold element just as it turns on, or perhaps measure it for resistance using a four-wire resistance measurement for added accuracy at such low resistances.