The main reason for using teflon pads sliding on a textured plastic laminate surface is not to reduce friction to a very low level. The purpose is to exploit the special nonlinear coefficient of friction which teflon on textured plastic laminate has under these conditions (loaded at about 15lbs/sq in).
For teflon, the coefficient of sliding friction is very near the static coefficient of friction. Most materials exibhit a sliding coefficient of friction which is much less than the static coefficient of friction. Thus with ordinary materials you must push hard to get it to move, and once it starts moving, it continues to slide using a much smaller force. So if you push with a constant force, after it starts to move it quickly accerates because you are pushing too hard. Thus it is hard to make very small adjustments by sliding with ordinary materials (a very bad condition for a telescope mount). Teflon on textured plastic laminates, on the other hand, has a sliding coefficient of friction very near its static coefficient of friction. So this sudden jerk of acceleration is much reduced. The sliding coefficient of friction of teflon on textured plastic laminates also increases with velocity. For ordinary materials this is nearly constant with velocity in the ranges under consideration. This increase in the sliding coefficient of frictin creates a nice self-damping effect making the motion very smooth.
Nylon does not exhibit these effects and should not be used.
Replacing one or two of the teflon pads with low friction rollers will perserve the desirable teflon effects while reducing the total friction. I am aware of a few telescopes that have done this. But replacing all three pads with rollers makes it much harder to do fine pointing adjustments.
There is a fine line between too sticky and too slippery. Once you have used a telescope that has the properly adjusted "buttery smooth" feel, nothing less than that will be acceptable.
Once you have rollers, there are problems (which may not be insurmountable) in using one of the rollers for the drive in slow motion control system.
1, the rollers (and pads) are connected to the ground board, so the location of the slow motion control would probably be fixed, with respect to the ground board, which would be inconvient when pointing the telescope in different directions. There are structural and practical reasons why the rollers (and pads) are connected to the ground board, and not the rocker box, so this is not simple to change. It wouldn't be too hard to do this for a small, light telescope, but for a big, heavy telescope where any ground board flex is significant and adding any height to the base is a major problem, then it is difficult.
2, For ease of use the slow motion knob is probably geared down so you can make fine adjustments. Slewing the telescope would spin this knob rapidly introducing additional friction with undesirable dynamics. A clutch may be needed to make hand slewing reasonable.
You might investigate Rick Shaffer's (sp?) dobsonians with lever/cam slow motions. An interesting design, but I found the mount too "springy" for my taste when I played with at RTMC once.
Chuck Grant