Help with PSU (Temp control fan, load bank, & PWM circuit)

Status
Not open for further replies.
Chances are the fans will tolerate this but will heat up a bit. It's probably ok to let them stall so let me recalculate some values tonight.
Click to expand...

I have never had any problem with letting my controller turn the fans all the way off. My design slows the fans by limiting current. So even if they shut down and are not providing back EMF, the current through them still can't go more than what the controller will allow.

Still, that is with 12 Volt fans. It could very well be necessary to deal with stall issues for a 24v fan system. Never tested it.
 
Last edited:
Good information ()blivion...I am currently concerned about the 24v fans for the dummy load, but ultimately my primary concern is with 12 volt fans as this is what is in the PSUs.
 
I am currently concerned about the 24v fans for the dummy load, but ultimately my primary concern is with 12 volt fans as this is what is in the PSUs.
Click to expand...

OK, I understand now.

My original circuit will work 100% for almost any kind of fan you might see in a computer. Your PSU fans are no exception. However, my circuit *CAN* be quite touchy at times. You just have to play with it and come up with a formula that works for you. 80% of the way it will behave is determined by the FET's you use. The rest is almost entirely up to hand tuning it right.

I have never tried a high side version like what ronv posted. But I am sure he know what he is doing. It is an interesting configuration.
 
Last edited:
I wouldn't worry about the fans getting to hot because if they did they wouldn't pass UL. But you might want to make sure about the FETs. If they are just little guys probably no probem either.

Putting the motor in the source side gives a little negitive feedback to the FET threshold voltage so it should be a little less sensitive. The downside is you can't get the full supply voltage across the fan.
 
Putting the motor in the source side gives a little negitive feedback to the FET threshold voltage so it should be a little less sensitive.
Click to expand...

I figured that's what you were going for.
 
I think I would try it out first. Maybe with only one fan so it doesn't get to hot.

I'm guessing you already have a 25 k trimpot and themistor, so if you can just find a resistor close to 33k you should be able to try it without new parts?
 
OK, I can test that out later tonight...I got the TVSs in the mail today, so those just get soldered to ground and +12v on the dummy load to protect any over voltage surges so that it directs it to ground, right?
 
I don't have a 25k pot, just 5 and 10. Could I just add a 15k fixed resistor in series if I need to test it at 25k?
 
Last edited:
I am hooking something up wrong...is it basically the same as ()blivion's except the drain and source are flipped and there is another resistor added that goes from ground to one side of the pot?
 
It's a little different.

The drain now goes to +24 - no fan in the drain.
The fan goes from the source to ground.
The resistor across the thermistor is gone.
One side of the thermistor goes to +24 the other to one side of the pot
The other side of the pot goes to the 33K resistor. The other side of that resistor goes to ground.
Center pin on the pot goes to the gate of the fet.
Yes you can add 15 k to either leg of the pot it will hust have less adjustment range.

See #435 for picture
 
Last edited:
OK, who knows how to say schematic 101? Lol I really need to get better at reading these things. It's working now, thanks for spelling it out for me ronv . I can't quite get the fan to stall completely out it is going very slow. The fan isn't warm but the FET is.
 
So now I think the temp sensing is not quite sensitive enough. How do I make it more sensitive with the change in temperature?
 
We may need to come up with a different way to do this. But lets try one more thing and see how much trouble we get in with self heating of the thermistor.
Use your 5K pot and make the bottom resistor7.5K or whatever your closest value is.
This will have more range but maybe not as much as you would like.
 
Tried changing the resistors as above. It is better, still not much range though. I set the fan to come on at about 1/8 speed when cold, then I heated it all the way up (ran it for a while at 47amps with the thermistor in between two of the current sense resistors). The fans increased speed to maybe 1/4 speed. I want to make another tempurature sensor circuit like I have on the FETs to put on the current sense resistors to measure how much fan speed I really need. Maybe a quarter speed is enough, I don't know. Probably better for them to run faster though.
 
We will need to build a little circuit to amplify the sensor.
Maybe you could make a couple of measurements.

What is the fan voltage for the speed you like for the slowest speed?
If there is enough adjustment range maybe you could measure the voltage it takes to start the fan.
And the voltage at which it stalls - quits running.
We need to decide if it should be able to stall or if you want to use the pot switch to turn it off.
What temperature do we want maximum speed? Minimum speed?
 
I will try to get those measurements before I go to sleep tonight. Right now I am installing the on/off switch and it is turning out to be more of a pain than it should be, simply because of my own mistakes...I had to solder and un-solder one side of the switch 3 times because I kept forgetting to route the power cord properly through the right path inside the dummy load before I soldered it! I think I finally got it right, third times a charm .
 
Readings:

The slowest voltage I want it at (I will use pot switch to turn off if desired) is 6.39v

Fans start at 3.27v

fans stall at 2.91v

Need to wait until I build another temperature sensing circuit to get the max and min temp settings (hopefully tomorrow, if not I will for sure get that by Saturday). Now I need to get some sleep, it 2am here on the east coast .
 
Status
Not open for further replies.

Similar threads

F
  • Question
Automatic dummy load circuit
Replies
4
Views
2K
rjenkinsgb
  • Question
Help with boost converter circuit
2
Replies
25
Views
4K
audioguru
D
  • Question
LM3914 with a buzzing relay
2 3
Replies
52
Views
9K
eTech
E
I
  • Question
Battery Charger with Load Sharing + Current Sensor
Replies
4
Views
2K
iptvbrazar
I
P
  • Solved
Help on triac circuit to control fan from 2 mains signals
Replies
15
Views
4K
paultowlson
P
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…