Maker Pro
Maker Pro

PWM motor control circuit-transistor issue

coloradoxc

Jun 28, 2011
4
Joined
Jun 28, 2011
Messages
4
Hello. I am having trouble figuring out how to select the right type of transistor for a project I am working on. The project involves running a small pump (link below) off a rechargable lead acid battery. The pump will be connected to a filter sampler and must be able to run continuously for 5 hours minimum. The pump will be connected to a 555 circuit that will allow the flow rate of the pump to be adjusted using a potentiometer that changes the on/off Duty cycle, not the frequency. My problem, the NE555 chip I used can only output a maximum of 250 mA of current, with a max voltage of 15 V. The pump below requires at standard operating conditions 1A at 12 V with a 12 Watt power dissapation. I assume that when I adjust the on duty cycle to less than 50% I could be expected higher currents, double, tripple? What kind of transistor should I use to amplify the current without wasting to much energy to heat, and work within those regions. I got the circuit to work with a different pump (Hoover hanyvac motor) using a IRFZ46N mosfet, but it generates a lot of heat. Any ideas. Thanks a lot. I don't have any other information on the pump since I'm still waiting on the shipment. Also, the dc signal is curertly set to oscillate at 130 Hz. (Changable if necessary)

http://www.datasheetcatalog.org/datasheet/philips/NE_SA_SE555_C_2.pdf
http://www.sparkfun.com/datasheets/Robotics/Other/spec sheet.jpeg
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The IRFZ46N should be able to handle your current (about 1A) without breaking into a sweat! It needs to be turned fully on to minimise the heating. Does the gate get up to 10V when it is turned on? It needs to be turned on and off rapidly but the 555 should do that.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Do you have a reverse biased diode across the motor? If not, the inductive kick as the motor turns off may be being dissipated in the mosfet. As long as each spike is not too large it will not damage the mosfet, but it may heat it up.

Show us the circuit you're using. It may also be that you're using the N channel mosfet as a high side switch (same problem arises using a P-CH as a low side switch) thus reducing the Vgs available and therefore operating the mosfet in a mode where the Rgs stays quite high. This would also show up as not being able to get full power out of the pump even when you approach 100% duty cycle.
 

coloradoxc

Jun 28, 2011
4
Joined
Jun 28, 2011
Messages
4
Attached is the circuit I am using. As for testing the voltage, I can't anymore and it would be pointless. The motor I oringinally tested was to see if the circuit even worked. The pump I mentioned will behave differently. But, I would much appreciate a brief explanation of what things I should be measuring, looking for, ect when I get the pump. Basically, I know I can get the motor to run with my limited knowledge, but the fact is it will be running on a battery for 5 hours meaning the whole circuit must be as efficient as possible.

Thanks for the help
 

Attachments

  • 555.JPG
    555.JPG
    21.4 KB · Views: 2,025

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The circuit looks OK to me. Do you know the current demand of your Hoover motor. Was this taking many amps. Try it directly on your 12V supply and measure the current.
An interesting measurement would be to measure the average voltage at the output of the 555 using an analogue meter. It should vary from a low voltage to almost the supply voltage as you vary the control.
 

coloradoxc

Jun 28, 2011
4
Joined
Jun 28, 2011
Messages
4
At 7.2 Volts, the factory voltage, the motor ran at 3 V for a few seconds then down to 2.5A. (NOT connected to the pwm circuit) I won't be using this motor by the way. When I receive the pump that I am ordering I will measure the current of the motor over varying voltages, core resistance, ect... I wish I could give better numbers, but as of now all i know is the motor is rated at 12V, 12 W. So I assumed 1 A motor at 12 Volts. Another question?, the pwm will statistically appear to the motor as a voltage lower than the 12V based on the on duty cycle duration, how do these circuits extend battery life if the current will statistically increase with a decrease in voltage? I have no experience with PWM prior to a few weeks ago. Please correct me, and explain.

thanks.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Your Hoover motor may have been sufficiently thirsty to drop the voltage so that there is not enough voltage to fully turn on the fet (you need about 10V) explaining the high temperature.

The average current of the motor will not rise as the speed is lowered, since the demanded power will be less. I do not know what the peak current will be.
 

coloradoxc

Jun 28, 2011
4
Joined
Jun 28, 2011
Messages
4
Ok the biggest concern I have using the IRFZ46N is the fact that on the datasheet it says the junction temperature will be over 150 degrees C at a drain current source of less than 10 A. The drain source current is the current the motor will pull at the voltage is sees, correct? So, do mosfets function most efficiently when they are ran at loads that pull close to their maximum current. The datasheet says at 40 A the case temperature is less than 100 degrees C. I also read somewhere that the frequency at which the voltage is switching also plays a critical role in effeciency losses. Can someone clear this matter for me, and or steer me in the right direction. I assume my application won't pull for than 5 A stalled, but I can't be sure until I get the pump.

Thanks
 

daddles

Jun 10, 2011
443
Joined
Jun 10, 2011
Messages
443
If you're interested in low power dissipation, take a look at the IRF3205. I used it in a 1 A current source; it can run continuously and not need a heat sink (whereas the IRF540 in the TO-220 package got quite hot in the same application). I used it with a 9 V battery driving an op amp, so it doesn't even reach the "normal" 10 V gate voltage that is usually recommended.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
By my calculations, the IRFZ46N will have a channel resistance of around 0.02 ohms because you will get an Rds of about 10V with this circuit -- it may be a good idea to add a small gate resistor due to the fairly high capacitance of the gate though, say 47 ohms)

At 10A, the voltage drop is 0.2V and the power dissipated is 2W.

Given the junction to ambient thermal resistance, this will cause a rise in junction temperature to around 120C above ambient, which is around 140C to 150C.

I would call that too high, and I would attach a small heatsink to it.

We've had an almost exact duplicate of this in another thread (I'll try to find it in a minute) and the addition of even a small heatsink meant that the temperature rise was almost imperceptible.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Here is a thread where heatsinking is discussed. I think the issues are very much the same as yours.

The practical upshot is that if a TO-220 device gets close to it's max junction temp without a heatsink (and 2 watts will do it) then even a tiny heatsink will reduce that dramatically.
 
Top