Joeb,

Actually, you didn't contridict anything I said about servo torque. Torque is torque. Measured in in/oz. Holding power is how much Torque a servo can develop when stationary, or for small deflections. Which is what I was talking about when I mentioned that non-digital servos don't develop much torque for small deflections, and suffer from blowback. We're saying the same thing. I just didn't use the buzzword "holding power".

The pulse rate of a digital servo is higher. But for other readers, the pulse rate is between the servo board and the servo motor, not between the servo and the RX. Basically, the board in a digital servo updates the motor's location more often, which is part of the reason the motor develops full torque at small deflections. More importantly, a digital servo's control board delivers a higher voltage to the motor for a shorter burst of time, while an analog servo board delivers a lower voltage for a longer time for small movements. Lower voltage means less torque. The only time you see max voltage to the motor of a non-digital servo is when the board wants to move the servo far enough that max sero speed is reached. You said 20% of movement, and I'll buy that as about right. (never tried to measure it). Taking apart a servo and playing with the pot and motor with a voltmeter really tells you a lot about what is going on, btw.

However, I totally disagree about gear train makup. Torque delivered has zippo to do with gear train material. That is, up to the point where the gears fail. But short of stripping gears, nylon, metal, karnonite, unobtainium, whatever, gears are gears and follow the same laws of physics no matter what they are made of. Now, you can get more speed and less torque or more torque and less speed from the same servo motor by changing the gear ratios. And, in fact, many servo makers do exactly that. But if you keep the same gear ratio, the torque and speed won't change with material at all.

If you don't belive me, get a servo like an HS-81 (or any servo that comes in nylon and metal flavors). hang a weight from it and see what the max developed torque is. Now replace the gears on the same servo with metal and do the same experiment. You'll get the same result. (plus or minus friction from the gear train, but that's going to be almost unmeasureable).

Gear train also does not affect deadband. Gear train DOES affect backlash, which in some ways acts like deadband, but you really shouldn't confuse deadband with gear backlash. It is true that metal gears usally have more backlash, and when they wear, they develop even more. Nylon is great for wear and backlash reduction, but the gears are weaker, and sometimes can be stripped. Karbonite is an attempt to get metal-like strip restance and strength with nylon backlash and wear.

There's also no reason to go to a 6v battery with digital servos. Remember, a 6v battery will drain FASTER than a 4.8v pack at the same capacity. So, if you have a 1500 4.8v, you will get MORE flight time than with a 1500 6v pack. You do get more speed and torque (and smaller deadband on non-digital servos) when you go to 6v though, so people who are looking for every bit they can get from their gear find it worthwhile to use 6v. But if you aren't on the edge of your servo's abilities, there's no compelling reason to use 6v.

(The other reason to use 6v is that if your pack drops a cell, you still have enough voltage to land, or at least direct the crash to a safe location. But cells usually don't fail in the air, and if you check your packs carefully, you can usually detect a cell going bad long before you get in to trouble. But battery safety is beyond the scope of the thread)