IRF7103 and 3.3V control with STM32

Anything not related to STM32
Post Reply
roniin
Posts: 7
Joined: Mon May 04, 2020 6:13 pm

IRF7103 and 3.3V control with STM32

Post by roniin »

I have used this solution for many years:
IRF7103.png
IRF7103.png (9.4 KiB) Viewed 1185 times
IRF7103 is controlled from the Nano v3 or Mega2560 pin with 5V voltage and PWM signal.
Mosfet does not heat up and I have no malfunction. They control 6W LED lighting per channel.
The scheme is simple, the layout is small and has few elements.

What happens if IRF7103 is controlled by 3.3V from STM32? Will this voltage be enough? Will STM32 withstand the switching current?
(unfortunately I do not have a workshop to test it right now, I have too little knowledge to deduce from technical documentation)

I am asking for help from an expert on Mosfet.

stas2z
Posts: 131
Joined: Mon Feb 24, 2020 8:17 pm
Answers: 8

Re: IRF7103 and 3.3V control with STM32

Post by stas2z »

im not an expert, but datasheet says irf7103 have vgs th min 1 max 3 volts so it should work properly with stm32

.rpv
Posts: 28
Joined: Wed Dec 18, 2019 10:19 pm

Re: IRF7103 and 3.3V control with STM32

Post by .rpv »

Hi, I don't think so, this mosfet has already too much resistance (rdson) on optimal conditions (130mΩ). Has a gate threshold voltage at max 3v, pretty close to the 3v3 signal, so it won't work at all or it will get pretty hot. You should look for other mosfets with lower gate threshold. You didn't mention the current or voltage at what you're using on this mosfet so you should be looking for something like the AO3400A (awesome for 3v3 signal and <30v and 1-2A).

EDIT. my bad :lol:, I didn't see that on the schematic you posted that it's at 12v, so it's 500mA, may work but may get a bit warm.

ag123
Posts: 475
Joined: Thu Dec 19, 2019 5:30 am
Answers: 1

Re: IRF7103 and 3.3V control with STM32

Post by ag123 »

yup the AO3400 and AO3401
http://www.aosmd.com/pdfs/datasheet/ao3400.pdf
http://www.aosmd.com/pdfs/datasheet/ao3401.pdf
are very attractive
searches like this returns lots of entries
https://www.aliexpress.com/wholesale?ca ... ext=ao3400
https://www.aliexpress.com/wholesale?ca ... ext=ao3401
but i've not really tried them out though i've stocked a small bundle of pieces.
getting 4-6 amps in a sot23 package smd and 3.3v logic level seemed too good to be true
earlier on i've been thinking about using bipolar transistors to drive them. these it seemed can go straight to the pin

stas2z
Posts: 131
Joined: Mon Feb 24, 2020 8:17 pm
Answers: 8

Re: IRF7103 and 3.3V control with STM32

Post by stas2z »

im not sure driving 2+ amps using sot23 is a good idea in terms of thermal stuff
powersop8 or equal is much better for heavy loads imho

and yes, fets with vgs(th) about 1.5v are better choise for 3.3v logic

.rpv
Posts: 28
Joined: Wed Dec 18, 2019 10:19 pm

Re: IRF7103 and 3.3V control with STM32

Post by .rpv »

ag123 wrote:
Fri Jun 05, 2020 9:25 pm
yup the AO3400 and AO3401
http://www.aosmd.com/pdfs/datasheet/ao3400.pdf
http://www.aosmd.com/pdfs/datasheet/ao3401.pdf
are very attractive
searches like this returns lots of entries
https://www.aliexpress.com/wholesale?ca ... ext=ao3400
https://www.aliexpress.com/wholesale?ca ... ext=ao3401
but i've not really tried them out though i've stocked a small bundle of pieces.
getting 4-6 amps in a sot23 package smd and 3.3v logic level seemed too good to be true
earlier on i've been thinking about using bipolar transistors to drive them. these it seemed can go straight to the pin

I use this mosfet a lot, but I agree, 1-2A it's ok, 3A o more it's asking for a trouble, maybe with lower duty-cycles, but still it's a gamble. I used their dfn3x3 models of this brand to switch 120w loads and they do just fine, the mosfet barely gets 10ºC above ambient, the package size it's not much bigger than the sot-23 (I think it was the AON7318).

I use the AO3401 mostly for reverse polarity protection.

stas2z wrote:
Fri Jun 05, 2020 9:56 pm
im not sure driving 2+ amps using sot23 is a good idea in terms of thermal stuff
powersop8 or equal is much better for heavy loads imho

and yes, fets with vgs(th) about 1.5v are better choise for 3.3v logic
Yep, at 3A with 35mΩ rdson the sot-23 it has to dissipate 315mW, I can't recall how much can dissipate this package but I think it's around 250-350mW, but when the temp rises the rdson also rises so it's a death spiral if someone over-estimates the mosfet capabilities.

roniin
Posts: 7
Joined: Mon May 04, 2020 6:13 pm

Re: IRF7103 and 3.3V control with STM32

Post by roniin »

I thought IRF7103 for 3.3V control is not a good choice. Thanks for the hint with the AO3400A transistor. I checked if it is available where I live (aliexpress is stopping the project for 1-2 months), I found it. I also found AO4800, it seems to have exactly the same pin distribution as IRF7103. This distribution of leads makes designing a single-layer board very easy.
IRF7103 board.png
IRF7103 board.png (27.67 KiB) Viewed 1110 times

Can I use AO4800? I need 500mA, but in theory what is the maximum, safe and constant load current?

http://aosmd.com/pdfs/datasheet/AO4800.pdf

.rpv
Posts: 28
Joined: Wed Dec 18, 2019 10:19 pm

Re: IRF7103 and 3.3V control with STM32

Post by .rpv »

roniin wrote:
Sat Jun 06, 2020 3:36 pm
I thought IRF7103 for 3.3V control is not a good choice. Thanks for the hint with the AO3400A transistor. I checked if it is available where I live (aliexpress is stopping the project for 1-2 months), I found it. I also found AO4800, it seems to have exactly the same pin distribution as IRF7103. This distribution of leads makes designing a single-layer board very easy.
IRF7103 board.png


Can I use AO4800? I need 500mA, but in theory what is the maximum, safe and constant load current?

http://aosmd.com/pdfs/datasheet/AO4800.pdf
Yes, the AO4800 it's slightly better than the AO3400A, has a bit better rdson and while it's a bigger package it's still a standard SOIC8/SOP8 package, so thermal dissipation isn't great, you can use it for 2-3A, maybe you can push it to 3.5A. Check the rdson graphic on the datasheet to see how much resistance can you expect at certain voltage:
rdson0.png
rdson0.png (18.74 KiB) Viewed 1083 times
Like this, at 3v3 you should get around 22mΩ at 25ºC and 34mΩ at 125ºC, so I'll take 30mΩ to give some margin, so I2R=P, 3.5A^2 * 30mΩ = 367mW
This packages that are all plastic can't dissipate much heat, so I'll keep this package under 300mW and the sot-23 at <150mW.

If you want to switch something like <=5A you should be looking for something with rdson <15mΩ and for ~10A an rdson <10mΩ, both cases with a package that has an exposed tab/pad that can be used for thermal dissipation (power SOIC/SOP8, QFN/DFN, TO252/TO220, ...). If you don't want to use heatsink and/or fan, of course, using heatsink/fan you can push a bit more on the all plastic packages and considerable more on packages that had exposed pad/tab.


*Take this as a very general guide, thermal calculations can get complicated and with a lot of variables in between.

You may want to check this guides: https://youtu.be/GW1GMbir6sA?t=355 (starts at 05:50), https://www.youtube.com/watch?v=ND8uJWlOgIQ and https://www.youtube.com/watch?v=GrvvkYTW_0k
The first 2 are related to 3dprinters so are focused on big loads, the third it's more general.

Post Reply

Return to “Off topic”