Simple PWM-motor controller

[Pax Writer]

Hello Experts

I'm back with one more, probably silly, question (although this time a little more thought-through).

I'm building a small PWM-motor controller based on an NE555 with adjustable duty-cycle on the primary output, which in this case is pin 7 due to the physical construction of the output and discharge pins.
Anyway, the NE555-parts works fine, but of course I ran into some trouble with the MOSFET controlling the GND-connection to the motor.
I can control motor speed to a very small degree, but the MOSFET doesn't seem to close properly. When I measured it with my old scope, it showed only very small voltage variations on the source-pin.
This is the first time I'm using a MOSFET, but I suspect that maybe D and S should be switched to get a clear GND-connection for the voltage on G to "compare" to. Am I off on a wild goose chase here?
As a note I might add that VCC is ~11.3V. D3 is in place because the battery will also drive a heavily loaded brushed DC-motor when the application is complete, and I was hoping that the occasional drops in supply voltage could be avoided by placing the diode in the supply path and let C4 and C1 stabilise the voltage on the cathode-side of D3 when the battery is on peak load every 5ms or so.
The MOSFET, as you can see from the attachment is an IRF3205 which should be able to conduct 20-25A of current even if VGS is only 10V... But again I might have read the data sheet wrong, so feel free to correct me.

Why do you think the MOSFET doesn't switch properly?

I stand by to be educated
Thanks in advance.
Pax

 

[Leftyretro]

Yes, D and S are backwards I think try reversing them.

 

[audioguru]

You have the Mosfet connected upside down. Its drain is supposed to be positive at the load and its source should be grounded.

The Mosfet has a very high input capacitance. The 10k resistor will spend a long time charging it. Then the PWM frequency must be very low.

Remove R1 and use the pin 3 output of the 555 to drive the gate through a 10 ohms resistor to prevent the Mosfet from oscillating.

Connect a free wheeling diode across the motor.

 

[Pax Writer]

Guru: Gotta give it to you: That looks very neat.
Its late here in Denmark now, but I'll try your circuit out tomorrow and post back the results. Thanks a bunch

 

[Pax Writer]

Guru: The battery on this application will be under heavy load in the peak periods, so there will be some ripple in the battery-supply when in use. Do you think this will cause problems for this circuit?

That being asked, I tried out our circuit and it works darn well. Thank you!

 

[audioguru]

Use a battery that is big enough to supply the max peak current without its voltage dropping much.
If the PWM frequency is high enough then C1 and C4 should supply extra current when needed.

 

[Pax Writer]

As it is now, the frequency is around 150Hz, but this is easily changed. Do you see any arguments to raise or lower frequency other than to facilitate decoupling?
The motor to be driven will routinely draw 15A, but currents up to 50A may occur for some systems. Moreover, the load on the motor varies, because it winds up a spring in a cylinder via a gearbox. A full cycle of winding and unleashing the spring takes around 5-7ms, of which the motor is under heavy load around 2/3 of the time.
The heavy and uneven load puts ripple on most batteries' supply, but I guess, I'll have to test this to be sure.
Thanks for your help.

 

[s_karbasi]

thanks ;
it is useful for me too;
in the fuature i will use in my design