Can anyone please tell me where I can find a P channel logic level mosfet (TO-220) in Australia that is good to use with an Arduino.
I did read a website that these fets typically start with IRL, not IRF (LINK).
I can't find any starting with IRL. The only one that I "think" is ok(???) is from "core electronics". The description says something about being compatible with 5V microcontrollers - but I don't want to buy one, then then wait a week, only to find it's not suitable.
(post deleted due to misinterpretation of datasheet)
If you don't intend to draw much more than 1 Amp, it should prove marginally acceptable.
Hi,
What voltage and current rating of the MOSFET you are looking for?
Thanks.. Tom... 
Hey Tom.
Sorry, I'm not very familiar with mosfets - other than I know they are switched with voltage allowing higher current (than transistors).
I'm currently trying to build a charger to charge a AA battery. I'm discussing the layout on another post (LINK). Since I had a general question about where to find a component, I started this as another thread - hope that's ok.
As was suggested to me, I'm looking at using a 5V supply to P channel mosfet to a resistor to a AA rechargeable battery to ground. An arduino PWM pin will turn on/off the gate. The AA batteries are max 2550mAh and I'd like to charge them between 0.5C and 1C if possible. A power supply will supply the 5V - not the arduino. Ultimately, I'd like to charge multiple batteries. I would also like to see if I can make a charger to charge Size D batteries that are 10,000mAh. Quite a bit of current with these!
However, after experimenting with an arduino mega, I'll be using an arduino nano, 3.3V unit to drive the PWM and measure voltage using the internal 1.1V ref voltage.
@runaway_pancake said that the FQP27P05 might be suitable, however can't draw more than 1 amp. Is there any that I can buy that can access more. Is there a downsize to getting a stronger mosfet?
@GigaNerdTheReckoning , I did read your reply before you deleted it - about the NDP6020P. I wouldn't mind having these in stock as well for future projects if they work better with 3.3V microcontrollers. How much current can they drive? I've found these in Australia online (LINK).
PS. Can someone please explain to me briefly what "values" I should be looking at and considering on spec sheets (eg. Rds, Vgs, etc). What specifies if is designed for 5V logic - which numbers. And if it is higher or lower, what effect does that have?
"Downsize" is right.
There's little available in through-hole packages. Higher current PMOS are going to be surface mount and inventory is in short supply.
Look at the drain current vs Vgs spec / graph
The NDP6020P was the last TO-220 P-channel mosfet with good logic-level performance even down to 3.3V. But it has been discontinued. What you've found online may or may not be genuine parts, but if they don't cost a lot, you might invest in a few and see how they perform.
But the little SOT-23 surface mount mosfets are actually very good, and their RDSon can be low enough to handle several amps without overheating. To breadboard them, you would need a supply of little SOT-23 - to - SIP adapter boards, which are a pain to solder.
One key value to look at is the gate threshold voltage - the maximum should ideally be a couple volts below the switching gate voltage. Remember that the threshold is just where the mosfet begins to conduct, not what turns it on fully. Try to find a "Transfer" curve diagram that shows how drain current varies with Vgs, and make sure that at 3.3V, or 5V, it will pass the current you need.
The second value is RDSon at somewhere near the gate voltage you will be using. It will determine the voltage drop across the mosfet when it's turned on, and how much heat that will produce. Generally, RDSon is higher for P-channel mosfets than N-channel.
Lot's of good example's here, provided by LarryD:
@ShermanP , (or anyone else), looking at the FQP27P06 the datasheet, it says:
Gate threshold voltage is min -2 and max -4V. Why are these values negative?
So given the Vgs values above and my 5V supply is "not" a couple of volts above the maximum, does that mean the fet may not turn on fully?
Is this the curve I should be looking at? If so, I don't get the values on the left. If my maths is correct, 10 to the power of zero =1, 10 to the 1 = 10 and 10 squared = 100, so my 5V is about 15Amps? Why don't they just as 1, 10 and 100 on the left (if that's what it is)?
If this is the chart, and I'm reading it right, then 3.3V might just be 1A.
As for Rds(on), it says Vgs=-10, Id=-13.5A. Again those negative numbers???
It says it's between 55 and 70mĪ©. So if I run at 5V, the current will be 15A, P=I^2R, so P=15^2*0.07 = 15W? Is that maths correct?
A P-channel mosfet is like an N-channel upside down. It's Australian.
So while an N-channel gate has to be higher than the source by at least the threshold voltage, a P-channel gate has to be lower than the source. And conventional current flows from drain to source in an N-channel, but it flows from source to drain in a P-channel. Datasheets often, but not always, give the gate voltages of a P-channel as negative for this reason.
Your interpretation of the vertical axis is correct. It is often shown that way when the axis is logarithmic, as this one is.
Such Transfer curves are usually presented for the "typical" part. But if you are unlucky and get one of those 4V threshold parts, then the mosfet may well not turn fully on at -5V. And I would say that your FQP mosfet isn't really logic level enough for things like battery charging. If you compare the Transfer curve shown for the NDP6020P, you'll see that it turns on at lower voltage levels.
The voltage drop and heat produced will depend on the the current you actually run through the mosfet. It won't be 15W unless you actually draw 15A.
I'm sure Einstein would have something to say about that - relativity and all that stuff 
I've ordered the NDP6020P online from "Tempro Systems" in Queensland - they had 7 left. Now they have zero 
I'll see how it goes - and handy to have for future projects as I mentioned earlier.
Thanks everyone for your help. Much appreciated.
Sorry, I just thought, if the voltage supplied to the gate is too low, can it be boosted with a simple voltage divider circuit? Or does that stuff up when driving it from a PWM signal?
I'm not sure what you're asking. Maybe you could post a schematic of what you have in mind.
I called it a voltage doubler, but also know as a "charge pump" (LINK):
What is the maximum expected load current? Which Arduino? What is "C" in Amps?
Quite a few of these are answered in post #5 above. At the moment I'm experimenting with an a 5V driven Arduino Mega. Down the track a 3.3V Arduino Mini. At the moment I'm just trying to charge a 2500mAh AA NiMh battery. I'd like to charge these at maybe 0.5C (1.2A). Down the track I'd be aiming to expand the circuitry to cater for 4 x AA. (4.8A). Not sure if this can be done over a longer duration but pulsed from one battery to the next.
I am also looking at maybe reproducing this then for D size 10,000 NiMh. I think charging them at 0.5C might be a bit hard, so maybe 0.2C (2A). How many? Minimum 4 again.
If you're still looking at a P-channel mosfet, then the issue is whether the gate voltage is LOW enough relative to the source voltage. I don't think a charge pump would do any good. The solution is just to find a mosfet that works in 3.3V systems, and then you won't have any gate voltage issues.
Edit: But see 6v6gt's post below if you are switching a higher voltage than 3.3V.
If you are mixing voltages in your circuit, for example powering the P channel mosfet at 5 volts, then to switch the mosfet fully off, the gate has to be brought near 5 volts. A 3.3 volt arduino Pro mini, for example, cannot do this without additional components. Usual is then to configure a P channel mosfet together with an NPN transistor (and a couple of resistors) as a high side switch. There are other methods.
You could also use a TO-220 PNP transistor for that charger.
Easier to find than a logic-level P-channel fet in TO-220 package.
Leo..
I'm still a bit confused about the basics of charging a AA battery. Not sure if I should continue this discussion here or in a new thread?
To charge the battery, I need a voltage higher than 1.44V. Literature tells me that the charge voltage should not exceed 1.6V. So I think I should try to maintain that voltage throughout the cycle.
This means that the overall serial voltage will be 1.6V + battery voltage (with a 5V supply).
I assume the resistance in the battery might change somewhat as it charges so the voltage may need to be "adjusted". If so maybe the code can do this via a mosfet.
But, to alter the "speed" at which it's charged, how could I maintain a constant current in that circuit?
@Wawa , I am aiming at charging the AA battery at around 1.2A if possible, and then down the track more than 1 battery at a time.
Hi,
Google;
constant current circuits
Tom...
