Replaced Extruder Cooling Fan. Possibly blown (MOS)FET?

[Robert Mann]

I have the rev G mightyboard.  There is a Cooling Fan section with a LED.  That LED no longer comes on.  My thought is maybe I was dupped into a shorted fan even though it looked brand new.  My old fan would make noise until I stuck my finger in the fans to get it to run true so I replaced the fan with a brand new 24V 40MM fan that looks exactly like the old one.  But now I get nothing with the new fan or the old fan.

Can anyone confirm that the GREEN LED on the Mighty board should come on when the temp reaches 50c?  Has anyone ever changed out there FET or MOSFET have not looked up the part number yet.  Any other good troubleshooting tips?

[Brett Crawford]

Is your new fan 24V?  If it's only a 12V, that can lead to other damage.

[Jetguy]

The circuit is simple. The LED is connected to the 24V source (AKA 24+) as is one of the 2 output terminals for the fan. AKA 24V is constant hot or live. There is NOTHING between the main input and that output positive side to blow, fail or otherwise. It's about as direct as it gets. The GROUND is what is switched on and off by the FET. Only when the FET turns on, can current return through the circuit from Positive to Negative (ground).

The LED is wired in parallel to the terminals for output. So the LED is a true indicator of when the FET really is on or off and will work even if NOTHING is connected to the fan port.

As such, if you really want to make sure- disconnect the thermocouples. That should force the board into safety mode and not being able to read the thermocouples forces the extruder fan on instantly.

So again, just disconnect the TC if you think there is something wonky- that will validate the FET is blown and that's QUITE likely on a rev G board since the FET is rated around 200mA on a good day.

Any short or excessive load will blow them sky high. It's a suspect design with no margin for error. Sure it works most of the time but enough failures happen and it's dumb to kill a board over a stupid fan.

This is why I actually like older Rev E clone boards. They have 76A rated MOSFETs even for the fans. If you kill one of those MOSFETs consider me impressed.

[Jetguy]

Look for the part that let the magic smoke out (smoke is put into the chips at the factory and if the seal breaks and it leaks out, they stop working).

I've circled the MOSFETs in question in red. Specifically, Q11 and Q9 control the extruder fans, Q7 controls the main blower duct on a Rep-2

I found the specifics here:

Dan Newman posted in the Tips and Tricks group in reference to a user with a 2X obviously a Rev G or H board https://groups.google.com/d/msg/3dprintertipstricksreviews/Vgq2z2_x1zc/9PfZqflsbBIJ

"The MOSFET they use for the cooling fan is a 2N7002K which is good for 60V but 
only 0.3 A. So make sure that the two fans combined use do not exceed that 
current rating @ 24V. 

[Robert Mann]

Interestingly enough the LED's use the same FET.  So just for sake of it I put one of those in place of Q4 which BTW is the fan FET.  Not sure where Q11 and Q9 came from.  Still same issue.  No FAN.  But interesting enough I short out SOURCE and DRAIN and walla fan comes on.  So its like the FET is not being driven on.  I have not tried the disconnecting the thermocoupler but will do that now.

[Robert Mann]

Ok so disconnected thermocouple and nothing.  NOTE: I am running the sailfish firmware.  I will reflash firmware just to make sure that the EEPROM did not become corrupted some how.

Other ideas?  Anyone know what feeds Q4 GATE? 

[Robert Mann]

My bad!  Your right Q4 in my case is the big blower fan :(

I need to change out Q7 like you said.

Stand by for that change and response!

[Robert Mann]

Thanks JetGuy!  Q7 was the culprit!  Replaced that with Q11 since mine is a Rep2 not 2X so I only have one fan.  Will order some replacements from digikey to have on hand as well as replace Q11.

BTW: No magic smoke on this one.  It looked perfect would never have known from the outside it was bad.  I did how ever notice that the LED for that fan next to it was dim always.  Which should have been enough of a clue that it was blown.  

All is back to normal now.  21 cent fix from digikey!

[Jetguy]

Glad my info helped and can hopefully help others.

BTW, yes, FETs often fail on, or after continued abuse, slowing die from stuck on to barely on but no control.

It's just how they work.

[Ryan Carlyle]

By the way Jetguy, you might recall I DID blow one of those ridiculously-overspec'd RevE board FETs back in May. That was when I installed those "internally jumpered" screw terminal connectors and dead-shorted the PSU through the extruder fan FET.

[Jetguy]

I forgot about that. Man, that had to be one heck of a punch of current. I figured the traces and just connectors and wiring in general between the PSU would be the weak point.

But, likely that "not driving to 10V rated gate value" leaves the device in a higher resistance RDS ON state resulting in junction heating and thermal failure. Dead short to the extreme (really good solder joint, short path, ultra low resistance) and probably a PSU with some oomph (serious internal capacitance) behind it is what pushed her over the edge.

Point is, I know that reasonable loads of well into 15A can be used with those and they barely get warm in my testing. You normally smoke the screw terminals long before the FET breaks a sweat, let alone fails.

I do agree the 76A is not a valid rating in this board situation. I'm just saying, even extreme peak load under all but your dead short circumstance, they are pretty tough IMO.

[Ryan Carlyle]

Yeah, that kind of dead short is literally the worst thing you can do to a solid state device. It's actually pretty hard to short it that bad, too. Having the short "outboard" of the screw terminals (like in a HBP trace) probably would not have fried it.

[Robert Mann]

I am an electronics guy.  For anyone here that blows that FET that needs it replaced out of warranty I will replace them for 20.00 plus shipping.  If you send it to me with a return shipping label then no shipping.  Just putting that out there for anyone who might need it replaced without a hot air rework station or mad soldering skills.  Visit my website Stealth Micro to find my contact details.