Jump to content
Shutaro

Better DIY MOSFET

Recommended Posts

18775742948_7a82b07d96_z.jpg

 

After discussions with some of the more techical-minded people at some of the fields I've played at, I decided to try this (simple, hopefully) project. The background here is that I was hearing lots of comments on how the DIY MOSFETS that are popularized on the internet and youtube (using a 3037(?), a few resistors, and a TVS diode) are not ending up as reliable as hoped. In part, I suspect, due to poor assembly -- these ones would fail quickly or after only a little use. But the other part of this is just long-term reliability. People said their device had worked fine for a year or two, but then just one day suddenly stopped working on the field. Despite this, there is still plenty of interest in DIY MOSFETs over the commercial variety, so I said I'd look into it when I had some time.

 

Eventually this project might progress into something like a adding an arduino-based microcontroller (with USB program-ability) to get the features associated with many of the commercial MOSFETs, but for now, this is what I could fit -- just the basics. I had just a little bit of extra space on a miniboard for several other unrelated builds. The hope in having a solid circuit board is that this will help negate some of the assembly-related failures people have, help with general reliability and allow more complex designs.

 

Maybe the more technical-minded among you recognize this, it's similar in concept to the designs that are popular today, with a few reliability improvements. Namely, I've added some zeners and a TVS diode around the gate, as well as a zener power brake around the drain that should kick in before the TVS diode does. This is to mitigate concerns about TVS diode threshold 'creep' over time when exposed to many surges, somewhat common knowledge at least in the electronics design industry.

 

Also on this board, you can use any TO-220 MOSFET, rather than the 3037(?) I chose a IRFB7430 as it is still reasonably inexpensive but has a lower Rds than the more popular choice. As for the size of the board, this is maybe 7/10" high by 2" wide, and can be very thin as all the components lie flat on the board, especially if the capacitors are left off. Even if the capacitors are included, it won't be more than about 6/10" tall, I think.

 

While I would have liked to add a proper gate driver, that can wait, I think it's of the most benefit with a microcontroller.

 

More pics:

 

18940168296_a8940ab4d4_o.jpg

 

Bare Circuit Board (top)

For ease of soldering, this is 100% through-hole parts, no surface mount.

 

18345780303_242058a829_o.jpg

 

Assembled Circuit Board (top)

 

18966425185_ebae3a662c_o.jpg

 

Assembled Circuit Board (bottom)

 

18780208469_b8f6d07c15_o.jpg

 

Assembled Circuit Board (height)

 

So in actuality, the caps don't add a significant amount to the height of the device.

 

I did some preliminary tests using a power supply and voltmeter and it seems to operate correctly. I'll try testing it in a real setup next.

Edited by Shutaro

Share this post


Link to post
Share on other sites

So far my DIY mosfets based on the "Unconventional Airsoft" design has had a 0% failure rate, they've been perfectly reliable for me. I've put them through significant strain too, including setups pulling approx. 40 amps continuously. I use slightly better resistors than the ones suggested in the common guide, so that may be part of why they've been so reliable for me. I also assemble them as gently as possible, and make sure it's impossible for them to short out during usage. I also always install a PTC resettable fuse in the wiring, this does of course lessen the risk of failure. My oldest mosfets have been used for at least 3 seasons and have gone through at least 50k cycles, some are probably closing on 100k soon, and are showing no signs of performance loss. The mosfet assembly in my MP5K doesn't have the external TVS diode and even that haven't failed, it's my impression that the IRLB 3034 chip is quite robust.

Edited by Lefse

Share this post


Link to post
Share on other sites

Also use the "Unconventional Airsoft" design, have only had a few failures. Bought a few mosfets from ebay, those all did not live long. The mosfets from Newark have never failed.

 

Looking forward to seeing the results of tests of your design. Looks easy to assemble.

Share this post


Link to post
Share on other sites

For the record, one way to induce failure in that design is to put the gun in full auto mode, hold down the trigger, and then yank the battery out (with the trigger still down). Once the battery is removed from the circuit, the gate charge keeps the MOSFET switched on, the motor is still spinning for a time and that voltage has nowhere to go (with the trigger still down) but the gate. You might not cause the MOSFET to fail the first time you do this, but do it enough times and it will fail. The two resistors are not enough to stop this failure b'cos the limitation for the GATE-SOURCE barrier is voltage (+/- 20V), not current.

 

While this is not common behavior (who pulls the battery out of the gun while it is firing), a fuse being blown or other battery-related mishap mimics this behavior. This is one reason why it's important to protect the gate as well as the drain of a MOSFET.

Share this post


Link to post
Share on other sites

Nice board. Why on earth a 3 watt resistor? TVS not necessary IMHO.

Largest power rating I could fit. Also it claimed to have "excellent burst handling capability" or something like that.

 

If nothing else, the TVS diode is there as a safety measure. That's honestly what they're best for. Something like this is designed to work for "any motor, any circumstance". Short of something like a BLDC motor, I suppose.

 

"I buy all my components from Digikey"

Yeah I don't know what I'd do without Digikey. I've used them for as long as I can remember. I don't consider them a distributor on the level of, say, Avnet, but they are great for the "smaller quantities" that a lot of companies need. And while a lot of distributors have been having problems with counterfeit parts, etc., Digikey has a far better record.

Edited by Shutaro

Share this post


Link to post
Share on other sites

I want to say that one way to induce failure in that design is to put the gun in full auto mode, hold down the trigger, and then yank the battery out.

Once the battery is removed from the circuit, the gate charge keeps the MOSFET switched on, the motor is still spinning for a time and that voltage has nowhere to go but the gate.

You might not cause the MOSFET to fail the first time you do this, but do it enough times and it will fail. The two resistors are not enough to stop this failure b'cos the limitation for the GATE-SOURCE barrier is voltage (+/- 20V), not current.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...