Someone made me a fan circuity.
The fans are 24V fans and I use a PSU with double 12V -24V output. 12V for Arduino 24V for fans.
Circuit is attached.
I think something might be wrong around this schematic or it may need to be extended.
The problem occurs when I power up the MC. As soon as I power it up the rear fan spins up for 1-2 seconds and goes off. It remains off until Arduino is up and running. When Arduino is fully booted then the fans start to work normally according to our code. This issue occurs before Arduino takes over the control over the shield so I don't think this is a software issue.
As I didnt make this schematic I dont really know if anything would be wrong with it or not.
My question is that is there any chance to prevent the fans spinning up right after I turn on the power?
You have these 24v motors, and you're driving them off a +/- 12v dual supply. So far so good...
You're switching them on and off with a MOSFET (still good), with the Source connected to the -12V (okay, still fine)... but then how are you driving the gate? Right off the Arduino's IO pin through a resistor it looks like... So apparently you've got Arduino ground connected to the -12v? And so ground is the positive side of the DC barrel jack?
I mean, I figure that's what you must have if it's essentially working (if arduino ground was connected to the 0v in between 12v and -12v, you wouldn't be able to turn off the MOSFET), so let's carry on with that (constantly remind yourself that you've tied ground to -12v not ground - you can easily forget and smoke something like that - I'd probably tie arduino ground to power supply ground, and use a PNP transistor to yank the gate up to 12v to turn on the fet)
In any case, to solve the problem of the fet turning on while the arduino is starting up, just put a pull-down resistor between the gate and source of the MOSFET, to ensure that the pin is never allowed to float. 10k to 100k is a good value.
Caps C26 & C28 reversed. (unrelated probably but still reversed)
In order to prevent that you need a 12V relay to switch the 24V fan power.
The +24V connects to the COMMON of the relay.
The N. O. (NORMALLY OPEN) contact connects to the fan power (X6-3). The relay can be put in series between X6-3 and the fan 24V by cutting the wire just before the X6-3 connector and connecting the cut ends to the contacts as described. One line of code is added to the s/w to turn
ON the fan power. In this way , there is absolutely no way possible for the fans to be energized until the FANS ON command is asserted because the relay coil has on power until the digital output that controls that relay turns on the transistor that controls that relay. The transistor can be a 2N2222 and the base resistor 1k. A 12V relay (small) should not draw more than 100 mA, which is well within the 2N2222 rating.
FYI, not that makes any difference to you since you are probably neither a H/W nor a S/W person, but ANYTHING controlled by a computer should normally be UN-energized UNTIL the computer ENABLES POWER for that subsystem. All professional systems work that way , with the exception of 24 hour systems that are always ON (manufacturing equipment, safety equipment etc etc)
A 10k resistor from Arduino pin to ground might solve your problem.
If only one fan is behaving like that, use another pin.
Some pins are used for other things, e.g. pin13 for the onboard LED.
Not sure about the IRF44N. Doesn't seem to be a logic fet.
If you can get the fans spinning @100% without too much heat, you're ok.
Leo..
Put a voltmeter on arduino pin-12 and pin-44 and watch the voltage from the moment you apply power to the uP, for the first 3 seconds. If you see a voltage > 1.8 V, the mosfet could be on.
The IRFZ24N is not a logic-level FET, so won't work properly from 5V.
Ignore graphs which show "typical behaviour" since actual gate thresholds vary a lot
between devices. If a FET doesn't have a guaranteed Rds(on) quoted for 5V or 4.5V it
is not suitable for 5V operation. The threshold voltage is normally quoted for 250uA
or something of that order, its the turn-off point, not the turn-on point.
