On Mon, 10 Aug 1998 19:04:16 -0400 Chris Eddy writes: >Stefan.. >The back EMF on a DC motor is a very good indication of speed. Of a permanent magnet motor it is. But the original poster was asking about a series-wound motor. With these, the magnetic field strength is not constant like with PM. It depends on the field current and any residual magnetization of the field core. If you were to measure the motor (field and armature since they are in series) current and armature voltage, the speed could be determined something like this: Speed * some constant = (V - I*R) / I [Series wound motor] I = motor current V = armature voltage R = armature resistance The motor would have to be operating near DC conditions (either DC drive or the end of a long PWM pulse). Also, the DC resistance of the armature circuit needs to be known (and as the motor wears out, it changes). Also assumed by the equation is that the field strength is entirely from the field coil current. If the field pole pieces can become permanently magetized, an additional correction factor in the denominator would be needed. > The >caveat is that when you use PWM to control the power to the motor, the >voltage reading on the motor at any randomly chosen time may have >nothing to do with the back EMF. With a permanent magnet motor, the DC voltage with zero current is directly proportional to the speed. The motor acts as a generator. With a constant DC current flowing, we can still consider that: Speed * some constant = V - I*R [Permanent magnet motor] This property is used in nearly all inexpensive cassette tape machines to regulate the speed of a permanent magnet DC motor with DC drive. An analog circuit built into the motor case is used. When the motor wears out, R increases, which causes the speed to be under-estimated. The feedback circuit actually causes the speed to increase, not what one would expect from a worn-out motor. > Therefore, you should time your >reading of the motor voltage to the point just before the on cycle on >the PWM output. This way, your motor voltage has settled fairly well >into the true back EMF value. Make sure that your overall period is >low >enough to allow this settling to occur. It is a little more efficient to use a fast period in the KHz range. You could use rapid PWM for a while, then drop it out every 10 ms or so to take a speed reading. _____________________________________________________________________ You don't need to buy Internet access to use free Internet e-mail. Get completely free e-mail from Juno at http://www.juno.com Or call Juno at (800) 654-JUNO [654-5866]