Don't sweat it, I wasn't bashing you and I didn't think you were bashing me. It's fine to disagree a bit, or correct something. I don't think I'm perfect, that's for sure. And while I try to be clear, I know I'm not always. Nature of these boards.

I didn't think you said anything about gear material and backlash. I was just going in to more details about deadband vs backlash as I thought you were mixing them. I could have mis-read what you wrote, however. Still, the info could be useful for anyone else reading. Anyway, the differnet backlash of the different gear materials was just something that came out "along the way", I didn't intend it as actually responding directly to anything you said.

As for cheaper digital vs non-digital, I agree, for many applications, digitals are worth the money. And there are few applications where digital would be a bad thing. Eventually, I think most all servos will be "digital" as the price comes down on them. It's just a better way to do things.

Your thoughts on 6v vs 4.8v are fine of course. As you say, it's really a matter of preferance. I've never needed to go to 6v, though I did try it out a little, and I didn't notice any difference in the plane I tried it on, so I went back to 4.8v. Obviously, in that case, the limiting issue wasn't the servo output, but other aspects of the plane. (note that I didn't say other folks wouldn't notice the differnece, or that no one should bother. I'm quite sure that in some applications, the boost on the servo would be more than worth while).

As for holding power, I disagree though.

I don't think it has anything to do with gear material. Can you explain how just changing out the gears in a servo will result in more holding power? It doesn't match my experience with servos that I've converted from nylon to metal gears (after stripping them out).

I mean, holding power is the amount of torque a servo can produce to try to resist movement from a given location, correct? It's just torque. And it's only become something talked about with the recent introduction of digital servos that produce full torque even for small deflections, which is in contrast to old style servos that don't produce full torque until the deflection gets to be beyond a certain point.

(in case I'm not being clear, what I mean by "deflection" is the differnece between where the servo is now, and where the servo wants to be, ie the commanded location. )

Now, if you're using "holding power" as a catch-all term to mean a combination of inital torque (ie torque for a small deflection), deadband, backlash and resolution to talk about how well a servo as a whole holds an exact position, then I don't disagree with gears mattering. But that's not what I think "holding power" means.

(another side note, servos are only part of the deal of a control system, of course. Things like pushrod flex, slop in linkages (such as a clevis pin being a hair smaller than the hole it's in), and so on, can make a bigger difference than any servo. I know joeb knows this, I'm just pointing it out for the origional poster and anyone else reading.