I'm looking for a through-hole mosfet driver that can be driving from a 3.3v microcontroller. I'm not sure such a thing even exists. All drivers seems to SMT.

I'm not looking to light up Las Vegas with the mosfet and the on/off cycle can me measured in minutes not milliseconds. I'm looking to control power to things like small 5V fans, other microcontrollers - things in the 5-12v range at only a few amps or less. I'm not picky about the mosfet at the moment. There are lots of mosfets available that work at 5v (usually 4.5 volts, it seems), but a very limited selection for 3.3v.

 

Why do you think you need a specialized driver?
If you have a candidate MOSFET and availble supply voltages, ther is no reason why your 3.3V microcontroller cannot successfully work with them.
Help us to help you by showing us a drawing of how you imagine that the microcontroller will connect to the MOSFET and how the MOSFET will switch some "device" on and off. Pay particular attention to available voltages in addition to the +3.3V available to the microcontroller.

Something like this should do the job for example

 

I've spent hours research and reading forums, and there seems to be no consensus that any mosfet can be driven by 3.3V at the gate. I have FQP30N06L and IRLB8721 mosfets on hand and many claim they will work. However, I'm looking for a lot of margin of safety rather than pushing my luck. This sort of question (drive mosfet with 3.3V) comes up all the time, and most of the time, the answer is "use a gate driver" and the discussion ends. I'm trying to find that magic gate driver that everyone is referring to since no one ever mentions WHAT gate driver might work.

I'm trying to be as vague and unspecific as possible about the application so that I can apply the same part to many different situations. I'm trying to avoid a one-off mosfet (and driver if needed) for each project.

 

Assuming that the Switched-Load is running at a higher Voltage, lets say, ~12-Volts ...........
Just check the configuration of your Processor's Output, and the Maximum-Ratings in the Spec-Sheet.
It's very likely that an "Open-Collector-Output" can tolerate much more than ~12-Volts,
therefore, You can Switch a P-Channel-FET that is supplying a ~12-Volt-Load,
and therefore, be assured of achieving the minimum Rds value from the FET.

If the built-in Switch can't tolerate the higher Voltage, an external BJT certainly can.

There are also small FETs with a Threshold as low as ~1-Volt.
See the Attachment.
DigiKey has plenty in stock ........

TN0702N3-G-ND

.
.
.

 

Modern logic-level Mosfets don't need gate drivers unless you need level-shifting or triggering a p-channel Mosfet hanging from a rail that is not 3.3v.
Drivers were used when logic chips were delivering a few milliamperes or less and Mosfet gates needed 10v.
Your theory is kind of ancient crap - especially with logic-level gates widely available.

 

So I appear to a very old dog barking up the wrong tree. Clearly, a driver is unnecessary.

@LowQCab Thanks for the tip on that transistor. I hadn't come across it.

 

I'm looking for a through-hole mosfet driver that can be driving from a 3.3v microcontroller. I'm not sure such a thing even exists. All drivers seems to SMT.

I'm not looking to light up Las Vegas with the mosfet and the on/off cycle can me measured in minutes not milliseconds. I'm looking to control power to things like small 5V fans, other microcontrollers - things in the 5-12v range at only a few amps or less. I'm not picky about the mosfet at the moment. There are lots of mosfets available that work at 5v (usually 4.5 volts, it seems), but a very limited selection for 3.3v.

https://uk.rs-online.com/web/p/gate-drivers/2070180
They work with a 3.3V logic input, but they need a supply of more than 5V. Most MOSFETs need 5V gate drive or more. Some will switch at 3.3V, but give degraded Rds(on) performance.
If you need a driver that works on a 3.3V supply, put several sections of a 74AC125 or 74AC541 in parallel, or a few more sections of a 74HC125/74HC541 or all six sections of a 74HC04.

 

@Ian0 Thanks for the suggestion. I can supply the driver with 5V, but I need the 3.3V input. What you suggest is perfect.

 

So here is another suggestion. The 74VHC1GTxx series of parts will work on any supply voltage from 1.65 Volts to 5.5 Volts, and here is the cool part: the inputs have traditional TTL thresholds of 0.8 volts (maximum) for a logic low and 2.0 volts (minimum) for a logic high. That means your output from the 3.3V microprocessor is more than sufficient to drive the input of a device with a higher Vcc. I have used this part from Onsemi with considerable success in this type of application. In addition the output drive is substantial and symmetrical. That means it will pull down as hard as it pulls up.

https://www.onsemi.com/pdf/datasheet/mc74vhc1g50-d.pdf

 

A little late to the thread, but if it's any help, I've been successfully using DMN1019USN N-Fets for a while now, driving them with 3.3V straight out of a PIC16LF1823 MCU

 

A little late to the thread, but if it's any help, I've been successfully using DMN1019USN N-Fets for a while now, driving them with 3.3V straight out of a PIC16LF1823 MCU

Your link jumps straight to an SMT version. Is there a through-hole version that you know of? Thanks anyways. Someday I might be able to get better with SMT, but my vision isn't very good.

 

Your link jumps straight to an SMT version. Is there a through-hole version that you know of? Thanks anyways. Someday I might be able to get better with SMT, but my vision isn't very good.

I don't think many current chips are available in traditional packages. Your best bet is adapters. You solder the SMT part to the adapter and the adapter PCB fits into a DIP socket. We did our last thru-hole board just before the turn of the century.

 

Your link jumps straight to an SMT version. Is there a through-hole version that you know of? Thanks anyways. Someday I might be able to get better with SMT, but my vision isn't very good.

It’s not that difficult to solder SMT parts. In fact, I now prefer them. And for DIY PCBs fewer holes to drill is a big plus.

Bob

 

Here is the PCB I just designed, built by JLCPCB. Only through hole parts are the pin headers.

Bob

 

The first time I used smt parts in one of my designs was by accident. I was browsing through parts at the Digikey website when I clicked on the wrong link while placing an order.

When I received the parts a few days later I was scared out if my wits because the project I had been working on had become time critical and there was no way I could complete it on schedule if I were to place another order. So I was forced to use what I had in hand.

Things turned out to be much easier than I feared, thankfully. The only tricky part was learning how to set the proper temperature on my soldering station. I'm near sighted, with a prescription of -4.5 in my left eye. So the only thing I have to do when working with circuitry is take my glasses off... and wear proper eye protection, of course.

 

I see after images of everything. Sometimes it's hard to tell which is the real one, especially when there are identical small objects all lined up next to each other (ie the pins on an SMT chip). Magnifying glasses only make the problem larger, but no clearer. Even with through hole, my eyesight is only good enough to get the iron and the solder in the right approximate area, then I use touch (through the solder and iron, not directly!). I can also hear and smell the sizzling of the solder which tells me something is happening. I will never win a solder beauty contest, but I rarely end up with bridges, though occasionally I hit the wrong pins. The absence of sizzling is usually the hint that I have the solder and iron held against different pins. The tactile resistance of the solder and iron against the pins and board is what I need. So, in my very limited, very unique case, through-hole is the only realistic option, but I don't expect the world to conform to my needs.

 

I see after images of everything. Sometimes it's hard to tell which is the real one, especially when there are identical small objects all lined up next to each other (ie the pins on an SMT chip). Magnifying glasses only make the problem larger, but no clearer. Even with through hole, my eyesight is only good enough to get the iron and the solder in the right approximate area, then I use touch (through the solder and iron, not directly!). I can also hear and smell the sizzling of the solder which tells me something is happening. I will never win a solder beauty contest, but I rarely end up with bridges, though occasionally I hit the wrong pins. The absence of sizzling is usually the hint that I have the solder and iron held against different pins. The tactile resistance of the solder and iron against the pins and board is what I need. So, in my very limited, very unique case, through-hole is the only realistic option, but I don't expect the world to conform to my needs.

Bummer... I guess we all do what we can with what we've been given... I'll try and find a proper TH component for you if I can...

 

Here's a good TH alternative that I found at Digikey. It can handle 1/2 an Amp at 20VDC. And has a gate threshold of 2V, so it should do the job. There are nFets at Digikey boasting a much larger capacity that also work at such low gate voltages, but a minimum order of 500 or more units is required.

Here's another candidate that can handle up to 100V between Drain and Source.

 

Here's a good TH alternative that I found at Digikey. It can handle 1/2 an Amp at 20VDC. And has a gate threshold of 2V, so it should do the job. There are nFets at Digikey boasting a much larger capacity that also work at such low gate voltages, but a minimum order of 500 or more units is required.

Here's another candidate that can handle up to 100V between Drain and Source.

Thank you! I had stumbled across the first one after I started this thread and already have some on the way from DigiKey. However, I had not found the second one. Sometimes I'd swear that identical search parameters in DigiKey turn up different results at different times. I'm sure it's just my imagination.

500mA is just fine for most of my applications. Even half that would work. It could also be a first stage to a larger capacity FET if I need one.