Pulse Width Modulator, with the MC14093 Schmitt Trigger NAND

Parts List:

 R1 = 330 ohm
 R2 = 1 MegaOhm
 P1 = 100K potentiometer (Lin)
 
 C1 = 0.1uF, (100nF) ceramic
 C2 = 0.01uF, (10nF) ceramic
 C3 = 0.01uF, (10nF) ceramic
 
 D1 = 1N4004 (or Schottky)
Led = Your choice
 Q1 = IRF513, T-MOSFet
 FB = Ferrite Bead (see text)
IC1 = MC14093, Quad 2-input NAND Schmitt Trigger

Couple Notes:
The IC used is a CMOS type MC14093, a quad 2-input NAND Schmitt trigger. If you wish, it can be directly interchanged with the CMOS MC14011 but this type is noisy.
The speed is adjustable from 0-max. Max rpm is 2/3 the supply voltage.
Supply voltage can be from 3 - 18volt, but I think around 12v works best for this application.
Input pins 8, 9, 12, and 13, need to be connected to Gnd. or 'V+'. Output pins 10 & 11 are left floating.
Maximum current draw, with the components shown, is approximately 220 mA max using a small type motor. Standby current at idle is about 88mA.

The way pulse modulation works is that it controls the motor by very short pulses. The longer the duration of the pulses the faster the motor turns. This method eliminates the excess heat associated with more conventional setups.
Depending on the motor, Q1 may need a coolrib, and C2 modified to eliminate the 'jerk' at the end of P1's adjustment.
P1, the 100K potentiometer can be a multi-turn type if your needs are towards specific rpm's.
For Q1, the IRF513, I experimented with several other types such as the IRFZ42, IRF511, IRF513, and IRF620. They all seem to do the job, although I prefer the IRFZ42 type for its very low R_dson.

To minimize RFI (Radio Frequency Interference), put a Ferrite Bead on the gate of Q1, or if you can't obtain one, wind some 5 turns of thick magnet wire on a 10-ohm resistor (diameter approx. 1/4") and insert that in between the gate and pin 4 of N2.
A Schottky diode of proper specs may be required for some motors which require faster switching.

Here is some information in regards to the motor I used. Remember those full-height, black floppy drives from those old early 1980's IBM pc's? That is exactly the motor I used. For those of you who remember that type, it was a belt drive model. The belt drove a little aluminum spindle. This motor is of excellent quality and made by the Buehler company in Kingston, USA.
A nice added feature is that those motors have a speed-sensor build-in and reads the rpm in AC volts. The yellow & green wires are the speed sensor and the red & blue wires are the positive/negative. It also has an extra aluminum shield around the motor-housing to keep rf interference to a minimum. Who wouldn't like a motor like that? If you can get your hands on them, take it!