Originally Posted by fnnwizard
The Futaba S3150 is 1 microsecond deadband, which would be considered excellent. I won't buy a servo with more than 2us deadband anymore. At 3us you can tell the difference in the flaps at various flight modes if your linkages were correctly tight to begin with.
One of the reasons why the 378 buzzed so much more at neutral was because at 4us deadband and your tx/rx was giving it a command in between the 2 spots, it does not know where to go.
Most of today's subtrims are in increments of 1us and less, with trims also at 1us or less (user adjustable on most tx's), with pots on the sticks +- 1-2 all the time.
Speed is only really critcal on ele servo... ele resolution not as much due to "gearing" of the arms.
A 1mm error in today's te could mean you are not going up in a light thermal when others would or not make it home from downwind. It used to be we see slop of 2mm before, but most builders have gotten much better.
Anyway, for the wing, it pays to be able to fine tune the te to the smallest increments = smallest deadband possible.
Thanks for the offer Francesco. I'll be able to do some basic current, speed, deadband test on Joe's servo in AZ! If it passes intial tests, I'll order some standard ones to use on LiFe voltage. I don't like the HV ones on 6.6v as it looks like you are giving up 20% for speed and torque and I don't like to run Lipo in a sailplane. If you saw my servo bin you will understand why I am so relunctanct to buy servos just to test anymore
... maybe I go take a picture and post it
You should use mks servos then because the slop is far less than plastic cased servos.
I don't think I have had many, if any, servos less accurate than the graupner ds3288 in terms of double centring surfaces and I doubt they all had less deadband than it. They nearly all cost less too. Deadband isn't the sole measure of how accurate a servo is.