Servos, is it me?

[Ron]

Hi All,

Is it me being fussy...?
I'm in the process of fitting a couple of aileron servos.
As normal, you plug them into the RX and make sure you've got them centred before fitting the nylon lever thingy so that is centred as well.

Now these two servos are different, even when plugged into the same RX output to centre them, with all controls and trims centred.
One servo you can fit the output lever exactly at 90 degrees to the servo, but with the other servo it's either just one way or just the other, only moving it one 'cog' on the splined output shaft at a time.

I've tried different nylon levers with no difference, as you would expect.

This is annoying as it means the trims on the transmitter have to be always off centre, or set up the push rods to compensate, but this will give different control surface movement at the extremes of travel.

This isn't the first time. I have other servos that are the same 'off centre'.

So is it just me being picky? Do servos all vary like this?
Should I just bung it together and stop moaning? :oops:

Ron

[Fred]

Hi Ron,
yep, that happen to me as well with different brand of servos...
I don't know why, that might be the way they mount the last gear in the factory. Maybe not well centered or so.. But sometimes I have 1 "notch" (?) difference between 2 same servos  :?:  Not all the time, but that happen...

Not a big issue anyway with the new radios full of processor thingy gadget, but I remember in my early days, that was more of a problem !  :wink:  :!:

[Ron]

Hi Fred,

Yes the engineer in me makes me want to take the servo apart and move a gear around one tooth to get it centred.
Don't think I'll be doing that as I've no idea what I'll find.

It's just a bit irritating really.
I start looking up which way I want more movement and which less to try to bias the servo output one way or the other.

Just me being fussy :lol:

What we really need is fully adjustable output arms, that can be clamped in the right position.

Ron.

[Fred]

 True.
A brand use to make servos with servos arms you can adjust in any position with a screw... Was a good idea ! Can't remember who that was  :?:

I still have one very old servo.. the last gear is square ! So you can move the arm only90 degrees at the time... Nightmare !  :!:

[Ron]

90 degrees only :shock:
Horror.

Just thinking aloud here now (waiting for the rain to stop so we can take the visitors out).
But if you had a 'sleeve' type bit that fitted on the servo output splines, with a sort of threaded groove around the outside, so that the arm itself fitted over it with a grub screw in it like a worm gear.
Then you could adjust the 'worm' and swing the arm.

Hmmmm...

Ron

[IceWind]

I hate that on servos too.
Despite some radios have a function like sub-trimm why don't they center the arm so it can be alwways in a 90degree angle perpendicular to the servo body. Most of the time that is the way the servo is going to work.

A solution could be also a variable resistor inside the servo for fine tuning.

[Ron]

Hi Icewind,

It's annoying isn't it?

I can see me stripping a servo apart....
Maybe I'll start with an old Futaba one.

Cheers,
Ron

[cvanscho]

Guys, remember that some cheaper servos do not "centre" electronically very well, some are better than others.  With "bad" ones, the servo never seems to return to the same spot every time..  So, even if the servo is mechanically spot-on, you might still find problems.. :!:  Only one thing to do with one of those: step outside, haul back into a nice grenade launch position, and send them into orbit :evil:  !
 
Chris

[Ron]

Yes that's right.
Having had a look inside a Futaba one, the pot is directly on the output shaft.
I know from experience in the electronics industry that pots are very poorly made. They have to be to keep the price down.
Normally the accuracy of the track resistance to position isn't any worry. No-one is concerned about that with a volume control on a radio.
But use the same sort of pot on a servo and you get position problems.

Also the gain of the servo amplifier is a big influence on the 'return' position.
But too much gain and you can get 'hunting' back and forth with the inertia of the motor etc.
Not enough and the servo will stop in a different position coming from one direction to coming from the other.

The good news is, as the pot is directly on the output shaft, fitting another 'preset' type pot on one end of the position pot, will give a mid-point adjustment.
In the one I took apart there are a couple of fixed resistors either side of the position pot, so replacing one of those with a small preset, perhaps with an access hole in the case, will give some adjustment.

Not sure if it's all worth the effort though :!:

Anyone know about digital servos?
Do they use a digitally coded output shaft system?
If so they should be more accurate.

That's my lot for now.
(Everyone breaths sigh of relief)
Back to taking the visitors out for the day... :roll:

Ron

[Fred]

Hi Ron,

Yeah, the pot lways was the problem, cheap servos or not.. A bit of dust  or xxx hours of uses and that's it...

I use a few digital servos now... I am not a specialist in that sort of things, but the movement is piloted by a chip, a bit like a sensorless esc, counting the number of steps as far as I understand.
You know you are using a digital servo by the noise of it... Very high pitch noise in idle as the microprocessor  is permanently correcting the center position, even to a few 0.0xx degrees and you touching nothing.