Bargain found for logic level power MOSFET

L

leftyretro

New Member
I'm sure many of you are aware of how useful logic level power N channel MOSFET transistors are for allowing a Picaxe to switch high power loads with minimum hassle. I found a nice buy on E-bay on ten 60 volt 50 amp devices and thought I would pass it along for those of you in the USA. It works out to be 55 cents each including shipping and I got them in 3 days from FL to CA. Tested one on my breadboard and it starts to conduct at 1.5 volt so they truly are logic level devices. Maybe I'm ready to design and build that Picaxe controlled arc welder as soon as I can find a heatsink ;)

http://cgi.ebay.com/10-LOT-P50N06L-60V-50A-28m-LOGIC-LEVEL-N-FET-TO220-PTH_W0QQitemZ270325566856QQihZ017QQcategoryZ4666QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

Lefty
 
M

moxhamj

New Member
What a bargain! Thanks for the link. Perfect for high current. 28 milliohms - that is not much more than the resistance of the pins themselves. These could drive 12V car headlamps directly off a picaxe and still not need a heatsink. I just bought 20 and the total cost including shipping to Australia was only $9.05. Quick - grab 'em before retrolefty and me buy them all!
 
M

Mookitty

New Member
Thanks for the tip, I ordered 2 lots. These just might save me from dropping a chunk on relays for a future project.

I'm not a fan of ebay/paypal, but after checking digikey prices ($1.04/ea.) it was worth the fleas from laying down with the dogs.
 
A

Andres Rodriguez

New Member
I bought 20 and have a question.
With Vgs(Max) 2.5V I assume I should use a voltage devider to connect to an output pin on the 28X1.
Is that correct?
 
Last edited:
A

Andrew Cowan

Senior Member
I read Vgs max as being +-15V. No voltage divider needed.

You were looking at the threshold voltage - the voltage at which it begins to switch (it will start to turn on somewhere between 1V and 2.5V).

A
 
D

Dippy

Moderator
Doh! No.

Have a read of this:-
http://www.fairchildsemi.com/ds/FQ/FQP50N06L.pdf
(Albeit a different manufafcturer, so get the proper one and READ it)

Maybe the Ebay Data Sheet is a bit dicky ;)
I dunno where you get Vgs(max) as 2.5V, can you tell us?

This same old MOSFET stuff comes around over and over.
Have a look at the spec to see the Rds versus Vgs.
That is a good start point to see how it will behave in your circuit.

That particular device seems to have a nice low-ish Rds at Vgs=5V.
Can you see what I mean?

I bet someone will PWM direct from PICAXE @100kHz soon and wonder why they get hot at 20 Amps.

BeanieBots --- don't do it!!!!
Yes, its OK, he HAS read the Data Sheet.
 
A

Andres Rodriguez

New Member
I read it on the Fairchild datasheet and now I realize that it is "VGS(th)"

VGS(th) Gate Threshold Voltage VDS = VGS, ID = 250 μA 1.0 -- 2.5 V

I intend to pulse 7.4V through a glow plug that is rated at 1.5V 3.5A.
 
A

Andres Rodriguez

New Member
I have not received the MOSFETs yet but I plan to use them to make an on-board glow plug igniter with a Picaxe 8M on the attached schematic.
I would appreciate any comments and advise regarding the circuit. The glow plugs are rated for 1.5V and 2.5 to 3 Amps. I think that a 5 to 10% duty at 8MHz will be the right pulses (I have many old plugs to play with)
My main concern is the circuit.
Comments please.

pwmout 2 , 249, 50 'Frq 8MHz 5% duty cycle

Thank you in advance
Andres
 

Attachments

A

Andrew Cowan

Senior Member
The FET may get a bit warm, but it shouldn't be too bad. As for the duty cycle, you will have to use trial and error. It is current that causes the heating effect - Let us know when you find out the best duty cycle. By the way, the lower the frequency, the cooler the FET will be.

I wouldn't have the diode though - if you want to keep it, you will need a 100K pulldown resistor on the FET gate.

A
 
Last edited:
B

BeanieBots

Moderator
Oh dear!
I'm afraid your circuit stands ZERO chance of working:mad:

You are driving the FET gate with a diode, it will turn on and stay on.
(nothing to remove the gate charge)
Even without the diode, odds are slim that it won't just fry with a high frequency PWM drive from a 'weak' PICAXE output.

This comes up so often and several times recenly too.
Please do a search on MOSFET and how to drive them.

While you're at it. Look up "buck converter". It would draw much less current from your battery.

As for duty, it should be reasonably linear but the resistance of a cold 'plug' can be quite a bit lower than a hot one so take care.
 
D

Dippy

Moderator
"nothing to remove the gate charge"
- what's a gate charge?
Fifty quid for Cold Play.

8MHz from a PIC PWM?
How?

Have a look around for recent threads on MOSFETs and drivers.
The subject crops up more often sub-prime lending.

BB... grit your teeth.
 
M

moxhamj

New Member
@marmitas: I have built glow drivers using mosfets and a few simple things: First, as said above, that diode is guaranteed to kill a glow within about 0.5secs. Nothing wrong with driving it directly from a picaxe but keep the frequency of the pwm low - eg 10hz (not 8Mhz) That means you don't use pwmout, but rather, use pause statements with different values. Next, I'd start with using a 1.2V single nicad supply for the glowplug and only progress up to higher volts once that works. I never managed to get anything over two nicads to work though - any slightest glitch and the glowplug was gone. Finally, with some smart thinking you can measure the temperature of the glowplug and adjust the pulse width accordingly. As the temperature goes up, the resistance goes up. V=IR and if you have a current sense resistor (which in practice can be 30cm of thick hookup wire), and you can measure the volts and the amps, you can work out the resistance and hence the temperature. You need three wires to the glowplug - one for ground, and two to the top of the glow - one for power and one for current sense. And I found you need to solder those wires onto the glowplug. Or you could do it another way and build a constant current switching regulator and (say) drop 4.8V to 1.2V but at constant current rather than constant volts. That involves a switching regulator (but not a picaxe) so scratch that idea...
 
Last edited:
B

BeanieBots

Moderator
Some good tips there Doc.
However, if an 08M was underclocked, it should be possible to make quite an effective buck regulator with a largish inductor using PWMout. (10's of mH).
 
M

Marcwolf

Senior Member
Thanks for the info

Thanks for that.. I've just grabbed 20.

They look perfect for what I need - switching power to banks of Servo's
 
M

Mookitty

New Member
Ordered Saturday afternoon, they shipped that evening. I got them today (Monday), this was via standard post from Florida to Arizona.

I'd do some testing, but my bench is being remodeled to add an "upstairs" (shelf) so that I have room for my new o'scope, freq gen, and bench power supply. $100 off Craigslist for all that and a bench DMM, and I got to see it working before purchase. It's old stuff, the power supply still has the NASA cert. sticker on it, but it's pure win in my book.
 
M

Marcwolf

Senior Member
Just an update. The MOSFETS's arrived today.

Nicely packaged complete with lable showing Part No, and what they are. Considering I am in Brisbane Australia, and they are in the States - thats pretty fast delivery.
 
M

moxhamj

New Member
Yes mine have arrived too, and like everyone else, I have been too busy to test them out. Will do soon...
 
S

sghioto

Senior Member
Mine arrived the other day. Only have a 3amp power supply but of the lot I tested all showed about .08 volts across the FET at 3amps driven directly by the Picaxe. This equals about a .027 ohm turn on resistance. Not bad.

Steve G.
 
M

Marcwolf

Senior Member
Just an update
I got time on Friday night to try out one of this units.

I have it pulsing a air solenoid directly from a PicAxe 08M. Its being turn on and off at 250ms. Solenoid is running at 12v.

Works well - no heating ot other issues. Scope on logic line looked very clean also.

Definately a bargain
 
You must log in or register to reply here.
Top