# How to choose transistor for switching bigger MOSFET

Leading on from my last question: MOSFET choice for 12v LED Strip and 3.3v Logic I have decided to use the IRFZ44N (despite it not being ideal for the job) for being economical the cheapest option and now need a second transistor that allows me to switch the MOSFET. The idea is the following: Since the project already has a 12v lane in place for the RGB LED strips, I want to use another transistor in front of the MOSFET to raise the gate voltage to 12v allowing the IRFZ44N to be used for my project.

I know that I'm looking for a small transistor (nothing beefy) since all it will do is switch a MOSFET gate at 12v which is not current-intensive as far as I understand it. I want it to be fast-switching so that it does not interfere with PWM. Looking through a local supplier I found that I could get a BS170 for 10 cents. I wonder if this is a suitable transistor for switching a IRFZ44N at 12v Collector Voltage. The full plan repeated once again:

• wire the BS170 to the microcontroller (3.3v on Gate), PSU (12v on Collector) and the IRFZ44N (BS170 Emitter to IRFZ44N Gate)

• wire the IRFZ44N (12v on Collector, LED Strip to Emitter)

are any other components needed for this? Is the transistor a good fit? It seems to me that mostly any transistor would work as long as it is definitely conductive at 3.3v and can handle 12 Volts.

What I'm really unsure about is the amperage - how do I calculate how much current switching a MOSFET will draw? The BS170 cannot handle > 0.1A according to the datasheet (section 'transfer characteristics' for the Fairchild BS170.

Thank you for reading this - probably all too simple - question.

• Where did you read that value of 0.1A max for the BS170? The absolute maximum ratings table reports a 500mA max Id continuous. Moreover even fig.5 (transfer characteristic) has up to 2A on the vertical axis (note that the TC is not relevant for switching applications, since it is drawn for Vds values which usually put the MOSFET in saturation, not in the ohmic region). Jun 28, 2015 at 12:33
• I've read fig. 5 at Gate-to-Source voltage 3.3v (which, seeing it now, was probably also wrong to begin with) Jun 28, 2015 at 12:35
• Yep. You should look at fig.1, where all the 3 relevant quantities appear (Id, Vds and Vgs). The reasoning you did was not correct, but the results are sadly the same: the BSS has too high a threshold voltage to be useful (at 3.3V it is barely conductive), see Vgs(th), which in the worst case is 3V. Jun 28, 2015 at 12:42
• this answer (and the related thread) is relevant to understand the problem. Either you choose a MOSFET with much lower Vgs(th)(max), or you could use a BJT instead. Jun 28, 2015 at 12:44
• Thank you, I now know what I need to look for in terms of it definitely switching at 3.3v (at least I know how to read the graph properly) - what else do I need to look for in the specific case of driving a bigger MOSFET at 12v with a BJT and 3.3v logic? Jun 28, 2015 at 12:58

A single transistor will not provide an interface with your FET AND do it without inversion. You can do the job with two transistors, though.

simulate this circuit – Schematic created using CircuitLab

This will provide a decent drive for your FET with about a 1 usec added to on-time. That is, if you put in a 10 usec pulse, the FET will be on for about 11 usec.

Also note that R5 needs to be a 1-watt resistor.

EDIT -

It turns out I was using the wrong FET model, and my value for gate capacitance was too low. For a "real" IRFZ44, the added pulse width is about 3 usec, not 1. The problem is that, when you go to turn off the FET, The charge stored in the gate can only discharge through R5, and the time constant is in the 3 usec range. You can, of course, decrease R5, but then the power it dissipates when the load is driven on goes up. You can get a 2 usec delay rather than 3 by decreasing R5 to 100 ohms, but then you need a 2-watt resistor. If you need very narrow PWM cycles (low effective motor current) you're probably better off with a more sophisticated driver. I personally tend to go with the Maxim MAX4426/4427 (less than \$4 each, with 2 channels per IC), but that's just habit.

• Deemed best solution for being cheapest while also being practical, thank you! Jun 28, 2015 at 19:44
• @RobinHeller - See edit. Jun 28, 2015 at 20:33

I am actually working on a project with an RGB LED strip that requires 12V. I am using a 2N2222A (NPN BJT). A MOSFET will not draw current at the GATE, but a BJT will draw a small current through it's base. They are super cheap and super simple. I took this off the internet, but it is a similar setup:

Since LED's are current driven, this is a good example of "Turning" on and off the current. If the voltage at the base (Which is called gate for MOSFET's) is high, then current will flow. If the voltage is low at the base then no current will flow, the BJT will act like an open circuit. Keep in mind that you may still have 12V sitting on your LED, but because the BJT is open it will NOT allow current to flow, thus not illuminating the LED.

Also keep in mind, that since you are using an RGB LED you will need 3 BJT's for each LED strip. Good thing 2N2222A's are fairly cheap!

Hope this helps. Please let me know if you have any questions. I am going through a similar project at the moment so I might have some answers or advice as to why I made certain design decisions. Good luck!

Josh

P.S. Using a switching transistor to switch a switching transistor seems a bit overkill. The setup would need to be a bit tricky because a MOSFET will not drive by current only voltage. If you can, try to avoid that.

• Thank you Josh. The thing is I can get the IRFZ44N extraordinarily cheap and it seems "properly dimensioned" (as in: can take anything that you can throw at it with reasonable lengths of LED strips), only my 3v3 logic will not switch it. So using a second transistor seems to be reasonable, but I cannot figure out which transistor type, let alone specifics needed for this application, to use sadly. :( Jun 28, 2015 at 13:25
• Seems to be a pretty intense transistor for what you are doing. I would look into a voltage switching circuit. Or maybe an op amp to bump up your voltage. If you can find a simple BJT somewhere, that might be the way to go. Good luck. Jun 28, 2015 at 13:39
• Well they come really cheap at about 0,32€ per piece, just need to get my head around getting them switched properly now. Jun 28, 2015 at 13:41

This won't work as you want. For a start the BS170 is an N channel mosfet and does not have a collector or emitter. Secondly you are attempting to describe it being wired as a source follower - this will not amplify the gate drive voltage at all. In fact it's likely you'll get a smaller gate drive voltage onto the IRFZ44.

What you could do is have the BS170 wired with source to 0V and drain to the gate of the IRFZ44 - you will also need a pull-up resistor from drain to 12V. The action will be inverted i.e. when you formally produced a 3V3 gpio output, this will now turn off the IRFZ44.

If you can't live with the inversion then a PNP transistor can be made to work but the base drive to it is a little more complex. Given the speed you need to drive PWM I'd use a proper driver like this: -

Try googling "mosfet driver chip".

• What kind of transistor would I need to look for for the easiest circuit possible to switch the MOSFET? Thank you! Jun 28, 2015 at 12:34
• If you have a BS170 then use it as I described. Jun 28, 2015 at 13:18
• I'm fixed on the IRFZ44N but I don't have to buy BS170s - it's just what I found at a first glance and probably the BS170s will not turn on reliably at 3v3 (as pointed out by Lorenzo Donati in a comment at the original post). I'd prefer not to have inversion of logic - is there any fitting type of transistor that can reliably switch the MOSFET fast enough for PWM without logic inversion? Jun 28, 2015 at 13:22
• I'd look for a push pull mosfet driver. Linear tech and international rectifiers make them. Jun 28, 2015 at 14:01
• @Andyaka - Sorry for the comment. I realized as soon as I posted it that it was wrong, but I couldn't remove it. Jun 28, 2015 at 17:58