Tom: The LCD contrast control input is loading the RC filter. Since you measure 2V as the lowest voltage, this suggests the input to the LCD is a pull-up resistance to +5V about equal to the R in your RC filter. If you don't want to use an op amp to drive to the contrast pin - try a voltage follower PNP (base to the RC network node, emitter to the LCD contrast control, and collector to ground). Any common small signal transistor should work ( like 2N3904, 2N4403, 2N2907 etc.) This will get you to a base-emitter drop from ground or about 0.65V. If you need to get closer to 0V use a NPN inverter (with 10k to 47k RC node to base and 100 to 470 ohms emitter series resistance, collector to the LCD control pin) and invert the RC input duty-cycle. - Wayne Foletta BMI - Santa Clara > ---------- > On Sun, 4 Jan 1998 18:17:32 -0500 Tom Sgouros > writes: > >After noticing just such discussion last month about using the PWM > >module on a PIC to adjust LCD contrast, I designed a program to do > >just that. The problem is that the voltage output from the PWM > >through my RC filter only works properly on the oscilloscope. There, > >it cycles nicely from 0 to 5v (although I get the impression that the > >upper ranges are largely useless). However, when I connect it to my > >LCD module (run by a Hitachi whatever-it-is), it only goes from 2v to > >5v, leaving the displayed characters largely invisible throughout the > >range. > > > >The filter I am using is just an RC filter. I thought that since I > >want 0-5v, I could leave the op-amp off my integrator. Am I wrong?