one laat question hopefully, fan on all the time


 
P

Philip_G

New member
I thought I had all the bugs worked out and the last thing to do was connect a fan, my fans on all the time. Doesn't matter if it's in manual, auto, up to temp, not at temp, I get close to 12v to the fan leads on my ethernet cable and at the board where the rj45 solders on, all good connections there no bridges. I checked all three legs of the mosfet to ground and get 12v (well, 11.86ish) to all three, that can't be good, can it? Sorry to bug you guys with another dumb question but I have no idea where to even start with this one. Every time I put this thing back in the case for the "last time" something comes up.

I see in past posts it has been common to put the mosfet in upside down, I double and triple checked that mine is right, metal tab towards the board and good solder joints there too.
 
Last edited:
Is there any solder bridging the source & drain legs on the mosfet?

Note that the drain is also tied to the metal tab on the back/top.

ECzU6Sjl.jpg
 
no, no bridging. I poked around online for a way to test mosfets with my DMM, and I can't get a response but that's most likely because either A) I'm doing it wrong, or B) you can't do it with the mosfet on the board anyway.

I didn't know the tab was connected but I had already bent it off the board anyway just in case it was shorting or something.

I'm using the blue/white-blue wires from the middle two pins on the RJ45, I tested continuity from the fan through the connectors to the ground and blow pins on the board.
 
Last edited:
With the unit unplugged, you should be able to meter the mosfet across the drain and source to see if they're somehow shorted together, but I doubt that's the case since the voltage and current is pretty low on the heatermeter board.
 
Steve_M said:
With the unit unplugged, you should be able to meter the mosfet across the drain and source to see if they're somehow shorted together, but I doubt that's the case since the voltage and current is pretty low on the heatermeter board.
Click to expand...

no continuity between the drain and gate
 
I desoldered the mosfet, if I put my DMM on diode mode and put the negative lead to the source leg, then touch the + lead to the gate leg, I get an open circuit, no tone and OL on the screen, now that it should be turned on I leave - on gate and move + to the drain, I get continuity and a tone, low resistance. while holding those in place if I short the gate to the drain with my finger the tone drops and I get about 1.93 on the screen (I assume that M ohm no scale depicted though)
so the mosfet seems good. I have more on order so I won't put this one back on the board, I'll use a new one just to be safe.

Any thoughts? Is it worth reflashing the linkmeter somehow, maybe it's a software error?
 
Philip_G said:
I desoldered the mosfet, if I put my DMM on diode mode and put the negative lead to the source leg, then touch the + lead to the gate leg, I get an open circuit, no tone and OL on the screen, now that it should be turned on I leave - on gate and move + to the drain, I get continuity and a tone, low resistance. while holding those in place if I short the gate to the drain with my finger the tone drops and I get about 1.93 on the screen (I assume that M ohm no scale depicted though)
so the mosfet seems good. I have more on order so I won't put this one back on the board, I'll use a new one just to be safe.

Any thoughts? Is it worth reflashing the linkmeter somehow, maybe it's a software error?
Click to expand...

I assume that's a typo when you say "leave - on gate" as - should remain on the source and only move the + between the gate and the drain.

With the mosfet now out of the PCB, you should be able to test for voltage at the mosfet holes on the PCB. If the fan shouldn't be on, there should be no voltage on the gate hole and if you put the HM into manual mode or choose a low set point to trigger the fan to come on and off, you should see the voltage come and go on the gate hole. There should also be a constant voltage on the source hole and never any voltage on the drain hole with the fan on or off.
 
Last edited:
Steve_M said:
I assume that's a typo when you say "leave - on gate" as - should remain on the source and only move the + between the gate and the drain.

With the mosfet now out of the PCB, you should be able to test for voltage at the mosfet holes on the PCB. If the fan shouldn't be on, there should be no voltage on the gate hole and if you put the HM into manual mode or choose a low set point to trigger the fan to come on and off, you should see the voltage come and go on the gate hole. There should also be a constant voltage on the source hole and never any voltage on the drain hole with the fan on or off.
Click to expand...

Yes. That's what I meant. Per here: http://www.utm.edu/staff/leeb/mostest.htm


12v at the gate at all times. Looks like that trace just goes back to the atmega socket, I don't see any bad joints or solder bridges. I have the LED set to go on and off with the fan >0% and it's going on and off as I manually change the speed so the board thinks the fan should be on or off anyway, I am not good enough at reading schematics to see where it might be getting voltage from.
 
Last edited:
The drain (middle hole) should only be connected to the blower output and should only see voltage on it when the mosfet is connected to the PCB.
 
Steve_M said:
The drain (middle hole) should only be connected to the blower output and should only see voltage on it when the mosfet is connected to the PCB.
Click to expand...
ugh, this would make a lot more sense if I'd stop screwing up the terminology. I meant to say gate, right now with the fan at 0% I have 12v at the gate and source, nothing at the drain.
 
Looks like the gate is driven from the transistor beside the mosfet. Any solder there look bad?

Also, on the far right resistor in this picture, do you see any voltage on it on either side of it when the blower is supposed to be off?

AX4RKmx.png
 
the BC337 looked good, no voltage at the resistor. I retouched the BC337 just to be sure and now the screen doesn't work. I give up I'm just going to start over. When I retouched the BC337 I heated that diode by accident, that caused enough solder to flow and short with the far right resistor, I didn't notice until I applied power and it acted funny so I removed power and doublechecked. That's when I caught it. it must have damaged something but I don't know what. Think I'll just bin this thing.
looks like that leg of the resistor goes to the atmega so it probably applied voltage and blew that, I guess?
 
Last edited:
Philip_G said:
ugh, this would make a lot more sense if I'd stop screwing up the terminology. I meant to say gate, right now with the fan at 0% I have 12v at the gate and source, nothing at the drain.
Click to expand...
You think that's confusing, the terminology is even worse because that's a P-channel MOSFET which I sometimes see the Drain and Source labeled in reverse. I'll just call them pins 1 2 and 3 as looking from the side from Steve's photo.

A - No MOSFET installed, blower plugged in, blower output 0%
1 - 12V
2 - 12V
3 - Up to a volt or so.

You also need to check that BC337 that's right there. It drives the MOSFET so if there's current going through it, the blower will run


The GND is actually the top pin, I realized the image might be misleading after I already closed Photoshop. Pin 1 should be 0V with the blower off, Pin 2 should be 12V with the blower off.
 
Bryan Mayland said:
You think that's confusing, the terminology is even worse because that's a P-channel MOSFET which I sometimes see the Drain and Source labeled in reverse. I'll just call them pins 1 2 and 3 as looking from the side from Steve's photo.

A - No MOSFET installed, blower plugged in, blower output 0%
1 - 12V
2 - 12V
3 - Up to a volt or so.

You also need to check that BC337 that's right there. It drives the MOSFET so if there's current going through it, the blower will run


The GND is actually the top pin, I realized the image might be misleading after I already closed Photoshop. Pin 1 should be 0V with the blower off, Pin 2 should be 12V with the blower off.
Click to expand...

it's even worse when you don't know the difference between a P and N and are trying to find a place to sell you one locally so you don't have to spend $5 to ship a 25 cent part ;) matter of fact that's been my biggest irritation with the project, sourcing simple parts when you forget or break one or don't get enough. Remember when radioshack sold actual stuff instead of just cell phones?
can't test any further, without the LCD I can't tell what the fan SHOULD be doing. I'm sure I just fried the atmega but I've put so much heat to this board so many times fixing dumb mistakes I have no faith in it anymore, plus the mystery fan issue. Early on I got the MCP1700 mixed in with the BC337s and got one in the wrong spot. I fixed it before it was ever power up luckily but it's sloppy, one of the BC337's is on the wrong side of the board. I can do a better job this time around.
 
Last edited:
Steve_M said:
Looks like the gate is driven from the transistor beside the mosfet. Any solder there look bad?

Also, on the far right resistor in this picture, do you see any voltage on it on either side of it when the blower is supposed to be off?

AX4RKmx.png
Click to expand...

that's nice and tidy, I couldn't get my mosfet to lay that nice.
 
Bryan Mayland said:
Same image, except the one I posted today I mirrored horizontally because the one on capnbry.net is looking from the top down.
Click to expand...

Here's the one you posted: On yours, the mosfet gate is being fed from the 10k resistor ( blue line ), though I guess it's a moot point since that leg on the resistor is also attached to the emitter.

2mO4GP7.png


Here's the one on capnbry.net that I flipped. On this one, the mosfet gate is being fed from the emitter of the bc337.

aFWVcq9.png
 
Last edited:
You must log in or register to reply here.

 

Back
Top