Mosfet for PWM vape

Hi, I don't have a great knowledge of electronics and I'm struggling to find the correct Mosfet for my project.

I'm planning to build a PWM board for vaping. The conditions are the following: Battery voltage: 5V-8.4V (2S 18650) Logic level: 3.3V MCU: attiny85

With the MCU, I want to generate a PWM Signal to drive an N-Channel Mosfet which drives a vape atomizer (around 0.2 ohms). The desired frequency is 25kHz.

Through the power of maths, I found out that my current is around 40A.

It's there a fet for my application? What numbers do I have to look for?

Thanks everyone in advance for any help.

johannes_otto: Hi, I don't have a great knowledge of electronics and I'm struggling to find the correct Mosfet for my project.

I'm planning to build a PWM board for vaping. The conditions are the following: Battery voltage: 5V-8.4V (2S 18650) Logic level: 3.3V MCU: attiny85

With the MCU, I want to generate a PWM Signal to drive an N-Channel Mosfet which drives a vape atomizer (around 0.2 ohms). The desired frequency is 25kHz.

Through the power of maths, I found out that my current is around 40A.

It's there a fet for my application? What numbers do I have to look for?

Thanks everyone in advance for any help.

I'm a vape user myself ... I even make my own juice ... I'm not sure the attiny will work if your PWM needs to be 25k since the arduino PWM output is only 500 hz ... I just measured it yesterday on my scope.

I know how annoying it is when someone responds to your post and addresses something you never wanted to talk about ... and im gonna be that guy for a minute ... lol

Are you aware of the DIY vape controller boards that are readily available on ebay in the $10 range? I love all things Arduino and I spend a lot of time with them ... but I don't think I would tackle an arduino based vape board especially when there are so many good omes out there for do it yourself vape builds for around $10 ... but ... thats me ...

I'm just throwing some thoughts out there is all.

I did build an arduino / stepper motor controlled coil building tool that ive built several stapled staggared fused claptons with.

:-)

Thanks for the advice. In fact here in Germany there aren't many diy vape boards. And those that are, are not cheap. I'm building my own mods for some years now, but I'm not happy with the few board choices I have.

If I got the timer stuff right, I think 25kHz are possible on an attiny85.

If course in the big picture you are right. Even if a board costs 50 Euros, it is cheaper and better than to develop an own solution.

I would like to do this project anyway. Maybe not for profit, but for fun.

Can you help me to find the part I'm looking for?

johannes_otto: Can you help me to find the part I'm looking for?

Funny you should ask that... after reading your post I dug out a Smok Alien motherboard that i have that doesnt work ... the heat syncs on the mosfet are INSANE ... I think they welded the heat sync to the FETS ... lol ... anyways im trying to get at them to see if I can get a part number and if anything, we can use the specs to find something that should work for you.

There are plenty of MOSFETs that can handle 40A. It is possible to have the ATtiny produce a 25 kHz PWM signal - you indeed have to change some timer settings for this to work.

BUT! (you saw this coming, didn't you?) an ATtiny won't be able to drive that big a MOSFET by itself at that frequency. High current MOSFETs have high gate capacitance, and the ATtiny doesn't have the oomph to switch such a big gate super fast, and at 25 kHz you're bound to have your MOSFET a large part of the time half on or half off, causing excessive heating. You'll need a MOSFET gate driver to help you with this. An ATtiny can supply some 20 mA to charge the gate... a proper gate driver up to a few A (it uses a capacitor for this).

A gate driver basically is a small Mosfet to drive the bigger one?

So I applied heat - liberally - to the heat sync and then attempted to pry it off the mosfets … it lifted off … but it also split the mosfets in half (cross sectionally) down the side seam of the mosfets … the face of the fets still holding onto the heat sync like it was their last dollar… determined not to be outdone by some uppity mosfets, i cranked the heat gun up to max and applied more heat … eventually i broke em free and then found the datasheet which i attached here.

Looks like the smok alien uses FOUR MOSFETS…

aon6512.pdf (282 KB)

UPDATE: I noticed that the part I featured, below, is a P-Channel, and the OP asked for an N-Channel. Here's an N-Channel that looks promising, and might even function well at 3.3V [though, the "MCU" will, likely, not be able to drive it directly -- you would still need a gate driver, in which case, it makes sense to use the higher drive voltage a direct connection to the battery, would provide.

Part Number: PSMN6R0-25YLDX Datasheet: https://assets.nexperia.com/documents/data-sheet/PSMN6R0-25YLD.pdf

Check out the SISS67DN-T1-GE3. But, you'll definitely need a Gate Driver. And the Gate Drive voltage should be pumped up to at least 5V, but more reliable at 10V!

johannes_otto: A gate driver basically is a small Mosfet to drive the bigger one?

Nah! Much more than that. Usually some sort of Push-Pull affair, and even things like charge-pump voltage doubler. The voltage doubler will need to be run directly off the battery, and it will draw a lot of current. If you can come down on the PWM frequency, that would make the design a bit easier, and less demanding.

It's all about current drive because of the need to rapidly charge and discharge the hefty gate capacitance.

Thank you very much. I'll look into that's Fets datasheet and read into the topic of gate driving.

EasyGoing1: So I applied heat - liberally - to the heat sync and then attempted to pry it off the mosfets ... it lifted off ... but it also split the mosfets in half (cross sectionally) down the side seam of the mosfets ... the face of the fets still holding onto the heat sync like it was their last dollar... determined not to be outdone by some uppity mosfets, i cranked the heat gun up to max and applied more heat ... eventually i broke em free and then found the datasheet which i attached here.

Looks like the smok alien uses FOUR MOSFETS...

the processor : STM32F072CBT6 MOSFET: TDM3482 GATE Drivr: DGD0506A

hope this will help

johannes_otto: A gate driver basically is a small Mosfet to drive the bigger one?

No, its typically a control circuit with protection facilities, a logic input with significant hysteresis to prevent multiple triggering, an output push-pull driver (2 MOSFETs). The typical protection in a single MOSFET driver is undervoltage shutdown and gate-source overvoltage clamp. Gate drivers usually have output currents from 100mA to several amps so then can charge and discharge gate capacitance effectively.

With a high-low (half-H-bridge) driver, protection normally includes shoot-through prevention, undervoltage shutdown, and sometimes feature programmable dead-time generation (very handy). Some have in-built diodes or charge-pumps to assist in bootstrapping the high-side driver supply.

These things are designed to make using MOSFETs much less problematic and robustify them considerably. I'd always advising using them when controlling significant amounts of power as rolling your own is tedious and probably won't protect as well.

I will never be able to do this myself, I have seen so many tutorials on how to make this stuff. My battery broke and I even could not replace it myself. My boyfriend had to do it because there was something that had to be done before. I got lg hg2 brown battery for my vape which is great. He told me to get that one and to wait for him to replace it because I could not do it. It looked easy when he was doing it, but when I try it there is no chance that it will work. It is cool that you can buy online all that you need for vape, and I am lucky to have someone to do it for me.