DC Fan driver ciruit. Driven by a microcontroller.

Discussion in 'The Projects Forum' started by woobis, Aug 24, 2012.

  1. woobis

    Thread Starter New Member

    Aug 22, 2012
    2
    0
    Hi All,

    I would appreciate a bit of help with a project I’m working on. I plan to use a micro controller to run a dc fan for my forge. The microcontroller I am using is a LV168 Pololu Orang-utan. I plan to eventually move it over to a dedicated Pololu Baby Orang-utan.

    The PWM output of the orang-utan is 10 KHz. PWM voltage is 5V on the LV168 and equal to V+ for the Baby O.

    My fan is a 9” car radiator fan. It is mark 12V 80W. I plan to initially run it at 7.5V (the fan draws 2.25A at this voltage) for my smaller forge but will eventually run it at around 12V (fan runs at 4A) when I build a larger forge.

    I plan on using the PWM output of the orang-utan to drive an IRL3716 N MOSFET (T1) via a UCC37324 MOSFET driver (IC2). The PWM output of the orang-utan will be protected by a MOX810X optocoupler (IC1).


    Here’s a quick sketch of the circuit.



    [​IMG]



    Sorry for the pen and ink. I've just noticed the fan isn't labeled it's the inductor by D1.


    My electronics has become a little hazy in the years since I left college. So could someone please let me know if I’m going about this the right way?


    I plan to run the optocoupler led at 30mA. So R1= 130 Ohms if PWM+ Max =5V. I’ve calculated this assuming PWM+ is 5V dc. This is only valid for when the duty cycle is 100%. Is this the best value to use?



    R2 is to tie the input of IC2 to ground. Is R2= 200 Ohms ok?


    I guess IC2 ties T1 to ground so I do not need R3? Does the same go for R2?


    I’m not sure what duty cycle the fan will be operating at yet. I have a good heat sink and the board can always be mounted in the airflow from the fan. That said I don’t want to waste too much energy as heat. Is T1 suitable? I hope the MOSFET driver will drive T1 through the ohmic region quick enough. It seems to have a suitable low RDS ON at 4 mOhms. It is also cheap and readily available.


    Do I need to add any capacitors to reduce noise?


    Also was I to run the fan off 12V. What size PSU should I consider? I tried running the fan off a 12V 10A switch mode PSU and the over current protection cut in before the fan started. Will a 20A supply start the motor? Might a 10A PSU start the fan if I program a soft start procedure into the controller? My guess is the rush current was only just over 10A.

    Thanks for looking over this. I hope I’ve got this right after going through other posts, text books and data sheets. Any help would be appreciated. I would like to get this right. I think it will work but I don’t know if I’ve missed anything important.
     
  2. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
    I haven't studied the spec's for the IC's, but here's my guess.

    If the LED in the optocoupler drops 2.5 V, then your 130 Ω resistor will limit the current to 19 mA, which seems better to me than 30 mA, but may still be unnecessarily high. Check the spec's.

    I think R2 is a pulldown resistor to keep the input of the driver from floating. As such, 200 Ω seems too small; I would try about 2k.

    I think you do need R3 to hold the gate of the MOSFET low, and it should also be about 2k.

    A bypass cap (.1 uF) across the power pins of the driver wouldn't hurt.

    More knowledgeable members will be along to provide better input.
     
    #12 and R!f@@ like this.
  3. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,749
    759
    First I would like to ask why do you need the isolation opto ?
    Is the PWM driver driven from a completely separate module or is the PWM power supplied from the same PSU tht you intend to run the fan ?

    As for ur q's!
    R2 should be chosen in such a way tht the current via opto's output transistor does not exceed it's rated load current tht is given in the data sheet of the opto coupler. Check the data .

    If the opto's load handling capacity is so 100mA, then chose R2 so that the load current is half for tht. You should use the opto's transistor collector voltage tht you are using. This way the transistor won't over heat and will survive longer. This current will add up to the current drawn from the Pin 4 of IC2 when it is active. This is also a factor tht you should take into account when choosing R2..

    But to make things easier for you I would suggest any value from 2.2KΩ to 4.7KΩ.

    Try calculating. You would learn something.

    Next is the gate drive of the Mosfet. U said tht the motor would run at 12V max. So chose a Mosfet that has a gate voltage of 10V max, see the data sheet. MOSFET Id should be rated at 50A are more. I suggest this cause you testing revealed that a 10A PSU shuts dawn at fan starts up telling you tht the start up current is well over 100% of the running current.

    An SMPS cannot be used unless it can supply transient load current of 200% or more of it's normal output current I believe.

    You would be better of with transformer rated at 12V, 6A secondary. It can definitely start the motor. These kinda transformer can be found quite easily in radio shack or ur local electronic stores. Any Is from 6 to 10Amps will suffice with a 25Amp stud mount bridge and a 10000μf 25V capacitor filter.

    I like to know why do u need IC2? Cause I dunno how the PWM is ?

    As for R3 it is not needed in most cases. But you could do better if u include a series gate resistor from output of IC to gate of around 10Ω to 47Ω.

    {ed}

    And as trace said a 100nf ceramic cap and a 10μf to 100μf electrolytic is necessary close to IC2 to suppress transients.
    @ thanks Trace for pointing tht out. I missed it.
     
  4. #12

    Expert

    Nov 30, 2010
    16,257
    6,762
    Just a generalization, mostly repeating what the other guys said. Your circuit is pretty good, except it works better to take your output signal off the collector of the opto and add in a small value resistor to the gate of the MOSFET. Your sense of proportions seems a bit off. There is no need to run the opto at the highest current it will survive. Better to consider the input requirements of the chip the opto drives and tone down the current in the opto until it works well, and that's going to be a long way from maxed out...and I'd bet you can run the MOSFET directly off the opto collector.

    I guess that about covers it.
     
  5. woobis

    Thread Starter New Member

    Aug 22, 2012
    2
    0
    I would like to be able to use a variety of inputs. I may use an IO pin rather instead of pwm output from the micro controller. The PWM is not going to come from the same psu (yet). I need control of the fan to finish developing the system.



    Funnily enough I had not just chosen random values but i had got the output current of IC1 wrong it is not 150mA it is 50mA! So if IC1 has a maximum output current of 50mA. i want a maximum I of 25mA (If V+ = 12V). So R2 needs to be a minimum of 480 Ohms. If I use 2.2K Ohms will pin 4 of IC2 pull down quick enough?

    IC2 will not draw current from pin4. Pin4 connects to a MOSFET internal to IC2.


    I'm using IC2 to drive the MOSFET cleanly. The fan will sometimes have to operate a low duty cycles. IC2 will always ensure that T1 is quickly switched through the on/off cycle. Also as i may be operating the MOSFET below 10V don’t i need the driver to open the gate properly?

    If i do this i guess i just use the inverting channel. What is the benefit?

    What is the resistor for? Will it increase the time it takes for T1 to change states? As it is increasing the RC of the gate.

    Thanks for your input.
     
Loading...