Servo center?

Need some advice? Trying out a new idea? Fancy a beer?
Join us for some general banter and good times.
User avatar
ShowMaster
Posts: 4327
Joined: Thu Dec 29, 2011 3:44 am
Country: -
Location: Los Angeles, CA USA

Servo center?

Post by ShowMaster » Wed May 14, 2014 8:49 pm

I've always used 1500us/1.5ms as a servo center standard. I have several servo cyclers that have this as a preprogrammed feature. I have noticed but never dwelled on the fact, that with splined servo shaft to servo arm mating, 1500us will not give me a true 90 deg center?
It's always one spline off of center in either direction.
If 1500ms is a standard, why isn't the feedback pot tweaked at the factory to pit the arm at 90 deg? That or the splines offset designed to make up for this?
I of course know we can offset sub trims in our tx to make the arm at 90 deg with the main trims at center, but why.
I'm getting the feeling that 1500us isn't the true standard any more for computer TXs?
I'm hoping this post will open up some dialog to get the real story on what's up.
In the very olden days, one could use a jewelers screwdriver through the servo arm screw hole and adjust the servo center, of course no sub trims then.
What say about my observation?




Sent from my iPod touch using Tapatalk, small keys and a big finger.


User avatar
MikeB
9x Developer
Posts: 16732
Joined: Tue Dec 27, 2011 1:24 pm
Country: -
Location: Poole, Dorset, UK

Re: Servo center?

Post by MikeB » Wed May 14, 2014 9:03 pm

How many splines on the shaft altogether? I have noticed that some servos have an odd number, so rotating the servo arm 180 degrees may make it line up.

Mike.
erskyTx/er9x developer
The difficult we do immediately,
The impossible takes a little longer!

User avatar
GTiDon
Posts: 177
Joined: Wed Mar 28, 2012 6:15 am
Country: -

Re: Servo center?

Post by GTiDon » Wed May 14, 2014 9:24 pm

MikeB wrote:How many splines on the shaft altogether? I have noticed that some servos have an odd number, so rotating the servo arm 180 degrees may make it line up.

Mike.
This is true

Sent from my C6603 using Tapatalk
Er9X Flashed
Speaker Mod Done
Haptic Feedback Mod Done
Telemetry Mod Done with FRSky TTL Lite
Voice Mod Done with Emartee Board
HK Backlight Done
FRSky Telemetry DIY Module Installed
----All these mods removed and installed Sky 3.0 board with DJT Module.

User avatar
ShowMaster
Posts: 4327
Joined: Thu Dec 29, 2011 3:44 am
Country: -
Location: Los Angeles, CA USA

Re: Servo center?

Post by ShowMaster » Wed May 14, 2014 9:55 pm

That's probably it Mike. I'll check. I usually use the full arm and rotating it on the splines does have one spot that's center. I then cut the other side off if I don't use it. This time it's tgy90s servos and I was trying to use the single sided arm. I'm probably not using the servo in the correct orientation it was calibrated on at the factory.
I just thought I'd bring up the subject because it bugged me. Sub trims will come to the rescue.



Sent from my iPod touch using Tapatalk, small keys and a big finger.

Daedalus66
Posts: 1839
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Servo center?

Post by Daedalus66 » Wed May 14, 2014 10:12 pm

I've tested quite a few Spektrum, JR and other transmitters over the years, measuring end points and centering. I've been very impressed with the consistency of calibration: 1100, 1500, 1900 for 100% throw and centering. There's a bit of variability but as far as I can see no systematic bias.

On the other hand, calibration of cheap servos seems to be very rough. I don't think there is any attempt to do accurate adjustment. The servo centering is in approximately the right range, no arm is fitted, and it's up to the customer to find the best compromise position.

Maybe the higher priced ones come with an arm installed and are calibrated.

My experience with UMX BNF models is that they work pretty well with a DX4e which has centering at 1500 and trims centered.

Bottom line. I think 1500 is still the centering standard.


User avatar
ShowMaster
Posts: 4327
Joined: Thu Dec 29, 2011 3:44 am
Country: -
Location: Los Angeles, CA USA

Re: Servo center?

Post by ShowMaster » Thu May 15, 2014 3:55 am

Thanks for the servo details. Looks like sub trims are the best answer if I want them at 90 deg. Most of my servos are inexpensive.




Sent from my iPod touch using Tapatalk, small keys and a big finger.

Helle
Posts: 577
Joined: Sat Jul 21, 2012 7:08 am
Country: -

Re: Servo center?

Post by Helle » Thu May 15, 2014 5:27 am

Hy

centerstandard at the tx is 1500us

some (Multiplex) have/had 1520us

Helle

User avatar
kalle123
Posts: 788
Joined: Sat Mar 29, 2014 10:59 am
Country: -
Location: Moenchengladbach

Re: Servo center?

Post by kalle123 » Thu May 15, 2014 5:42 am

Helle wrote:Hy

centerstandard at the tx is 1500us

some (Multiplex) have/had 1520us

Helle
From my MPX Royal manual
Attachments
18.jpg

User avatar
Kilrah
Posts: 10416
Joined: Sat Feb 18, 2012 6:56 pm
Country: Switzerland

Re: Servo center?

Post by Kilrah » Thu May 15, 2014 10:41 pm

Futaba is 1520.

User avatar
ShowMaster
Posts: 4327
Joined: Thu Dec 29, 2011 3:44 am
Country: -
Location: Los Angeles, CA USA

Re: Servo center?

Post by ShowMaster » Thu May 15, 2014 11:34 pm

Looks like sub trim is the way to go. I like my main trims at center and adjust the linkage after trim flights to achieve this. I find that in the future I'm having to use a lot of trim, I've got a servo issue or something going on that's not normal.
I also like my arms to be at 90deg when centered to keep the throws linear. Old school but makes for a good predictable plane.
Not so much for foamies of course.


Sent from my iPod touch using Tapatalk, small keys and a big finger.

User avatar
Kilrah
Posts: 10416
Joined: Sat Feb 18, 2012 6:56 pm
Country: Switzerland

Re: Servo center?

Post by Kilrah » Fri May 16, 2014 7:43 am

On OpenTX the best way to adjust for that when you install a new servo is the PPM center setting as it shifts the whole sevo range, not only the center like the subtrim does.

User avatar
ShowMaster
Posts: 4327
Joined: Thu Dec 29, 2011 3:44 am
Country: -
Location: Los Angeles, CA USA

Re: Servo center?

Post by ShowMaster » Fri May 16, 2014 2:40 pm

Thanks Kilrah, I was wondering about that. Any downside? I usually use all the same brand servos so they all should be off about the same.


Sent from my iPod touch using Tapatalk, small keys and a big finger.

User avatar
Kilrah
Posts: 10416
Joined: Sat Feb 18, 2012 6:56 pm
Country: Switzerland

Re: Servo center?

Post by Kilrah » Fri May 16, 2014 2:53 pm

The only limitaiton is that you should not use extended limits at the same time or you might get outside of the travel range the servo considers valid.

Daedalus66
Posts: 1839
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Servo center?

Post by Daedalus66 » Fri May 16, 2014 2:59 pm

I should think the only problems would arise if you were using extended throws and needed to shift the centering a great deal.

Not a good idea if dealing with micro models using linear servos with hard stops, for example. Or with a stabilized receiver that expects 1500 centering. Otherwise it works for me.

User avatar
ShowMaster
Posts: 4327
Joined: Thu Dec 29, 2011 3:44 am
Country: -
Location: Los Angeles, CA USA

Re: Servo center?

Post by ShowMaster » Fri May 16, 2014 3:52 pm

I think we've got some good ideas as workarounds. Too bad a simple adjustable offset nylon servo arm isn't made. I do see $CNC versions that address this for Heli so I know it is a problem for them.

Thinking?
If I set my opentx servo throws to 80/90% and use sub trim to center the servo, and set all my servo arm pushrods to surface horn to give me the throw I need, I should have plenty of trim range without hitting a servo stop?
I'm not flying 3D.
I understand that opentx 80% is equal to Spektrum, Futaba 100%?

After a trim flight I would adjust the mechanical linkage offset to keep all required servo at center again.

I realize this is overkill and many just leave the arm offset and fly. Also in the good old days I actually adjusted the servos feedback pot to center. No other adjustment options except main trim available. The servos and tx would not be separated after the adjustment.
Too much trouble now with tx options.

I'm also just sport flying so it's not as critical, but still bugs me. In the past I think I used a wheel arm and drilled a new offset hole. No wheels with the small servos.

Maybe I'll bring it up at the next big hobby show I go to where the big RC manufactures have booths setup?
DuBro, Sullivan, Horizon?




Sent from my iPod touch using Tapatalk, small keys and a big finger.

Daedalus66
Posts: 1839
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Servo center?

Post by Daedalus66 » Fri May 16, 2014 4:34 pm

Spektrum and other mainstream transmitters use 1100, 1500, 1900 for 100% throw. So 125% gives 1000 to 2000 range.

Open source firmware uses 1000, 1500, 2000 for 100%.

So, yes, 80% on Taranis equals 100% on Spektrum, etc.

Some discussion as to best way to set up for ultramicro BNF models, which must not have throws over 100% (in Spektrum terms). I used to set the weight to 80%, but have been advised to use 100% weight and 80% limits or 80% for high rate. For simple setups I don't see much difference but for complex mixing it's better to leave weight at 100%.

User avatar
kaos
Posts: 3247
Joined: Wed Dec 28, 2011 1:15 am
Country: United States

Re: Servo center?

Post by kaos » Sun May 18, 2014 11:16 am

I am probably the least knowledgeable person here. Here is what i have been doing, looking for advise/correction:
When i need more than 100% and need to adjust the center, I use extended limit on both ends then adjust each end of limit so the servo will not be outside the travel range at the max 'limit' setting then use subtrim to center between these two 'limit'.
when I don't need extended limit, I do the same thing only the limit will be <= 100.

Anything wrong of doing this way?

User avatar
ShowMaster
Posts: 4327
Joined: Thu Dec 29, 2011 3:44 am
Country: -
Location: Los Angeles, CA USA

Re: Servo center?

Post by ShowMaster » Sun May 18, 2014 3:48 pm

I think if it works it ok. As long as you don't hit a hard limit and stall your servo I think it's more about the invved settings.
My original issue is not being able to have the same CW and CCW rotation and a 90deg center.
In practice not everyone worries about it.
Our radios have many offset options but that also makes that servo or servos require the offset mix every time a new model is setup. If the arm was able to have a secondary adjustment other than the normal spline hub, one would set tx center 1.5Ms, and adjust the arm for 90deg with its adjustment.
Ideally the new arm would have the courser spline hub to match your brand servo, and the arm on top would have a finer tooth setup to set 90deg and the screw would lock this all together.
No more worries with the universal servo arm accessory. As mentioned, I think there are CNC metal options but a lot of $ for a $5 servo.
So for now, use the many options opentx gives us.



Sent from my iPod touch using Tapatalk, small keys and a big finger.

Daedalus66
Posts: 1839
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Servo center?

Post by Daedalus66 » Sun May 18, 2014 7:09 pm

I'd think twice about using extended limits. The transmitter already has extended limits compared to most others (100% on Taranis or 9x = 125% on Spektrum, etc.). For basic controls it's usually best to stick to 100% maximum, unless you fly 3D and need huge control movements. Large angular throws can give you the effect of reverse expo (small throw rates at the extremes). I always start with a 100% setup on the transmitter and then set up the horns and push rods to give the maximum throw I think it needs. Then I dial it back using D/R.

Only where I deliberately want unusually large throw, as in certain flap setups, do I go to extended limits.

User avatar
jhsa
Posts: 19211
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Servo center?

Post by jhsa » Sun May 18, 2014 7:59 pm

I always fly with large throws on my planes.. I normally set the limits so the servos don't stall. then set the throw using the DR or mixer, depending on the model and what I want to do. Like that I won't break a servo by stalling it,..

There are different ways of achieving the same results ;)

João
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW

User avatar
kaos
Posts: 3247
Joined: Wed Dec 28, 2011 1:15 am
Country: United States

Re: Servo center?

Post by kaos » Mon May 19, 2014 1:33 am

Most of the time when I need the extended limit is when the servo is NOT centered very well. To get a 100% or large enough throw both ways, I end up doing what I described. By my observation without detailed electronic knowledge, the way I am doing seems to be doing it. I just want some confirmation that it works as it supposed to. I have found surprises after learning the details in the electronics, 'common sense' does not always work as it seems. ;)
the other use for extended limit that I want to try is to program a huge speed break when I land too fast. ;)

User avatar
jhsa
Posts: 19211
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Servo center?

Post by jhsa » Mon May 19, 2014 8:44 am

You can extend the limits as much as your servo allows. Set it just before stalling. That is the limit the servo will go without breaking.

João

Sent from my GT-I9195 using Tapatalk
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW

Daedalus66
Posts: 1839
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Servo center?

Post by Daedalus66 » Mon May 19, 2014 6:43 pm

The issue is not usually with servo throw per se. Most servos will go 60 degrees or more each way before running into any problem with the mechanism. And most will accept the full range of signal input that represents 150% on JR, Spektrum, etc. or 125% on the open source firmwares like ER9X or OpenTX for Taranis (888 to 2012 usec).

Rather, what I'm talking about are the mechanical issues, particularly slop and non-linearity, that can arise with large servo throws.

Note that a special case are the linear servos used in ultramicro models, which have hard stops that must not be exceeded. The maximum throw to be used with these is 100% in terms of Spektrum settings and 80% for Taranis, etc.

By the way, just what is the speed brake you use to slow things down?

User avatar
kaos
Posts: 3247
Joined: Wed Dec 28, 2011 1:15 am
Country: United States

Re: Servo center?

Post by kaos » Mon May 19, 2014 7:13 pm

this is for after landing air brake. both ail down and both ele up to create drag. with delta wing just the full ail down. amount needs to be adjusted based on the ail/ele surface and CG. ;)

User avatar
jhsa
Posts: 19211
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Servo center?

Post by jhsa » Mon May 19, 2014 10:16 pm

On a normal plane, right after landing, If you put both ailerons down, they will act like flaps increasing lift on the wing.. add up elevator to that and your plane might take off again and stall right after, crashing to the ground again.. But then it depends on the plane.. If I was going to do something crazy like that, I would move both ailerons UP to destroy the airflow on the wing and decrease lift..
But that's me.. :)

João
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW

User avatar
kaos
Posts: 3247
Joined: Wed Dec 28, 2011 1:15 am
Country: United States

Re: Servo center?

Post by kaos » Mon May 19, 2014 10:40 pm

I think you are right, should be ail up and ele down.
Physically back, mind is not. Result of away from RC for a whole month and a huge jet lag. :mrgreen:

Daedalus66
Posts: 1839
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Servo center?

Post by Daedalus66 » Tue May 20, 2014 9:52 am

Fully agree about ailerons going up. There's another reason it should be up rather than down and that is because ailerons full down are quite vulnerable to damage.

User avatar
tilmanb
Posts: 347
Joined: Thu Oct 11, 2012 9:36 pm
Country: Germany
Location: Karlsruhe, Germany

Re: Servo center?

Post by tilmanb » Tue May 20, 2014 9:59 pm

kaos wrote:I think you are right, should be ail up and ele down.
Physically back, mind is not. Result of away from RC for a whole month and a huge jet lag. :mrgreen:
I would use elevator up in that configuration actually.
Instead of deploying flaps down you change the whole angle of the wing. Which had the same effect.
But the before mentioned stall issue creeps up. That is why you kick the ailerons up, to stop the wing tips from stalling.

If you just kick the ailerons up your plane will go into a strong noise down attitude anyway. Even if you don't want to pitch your plane up, you will need to compensate hat with some elevator up.

User avatar
jhsa
Posts: 19211
Joined: Tue Dec 27, 2011 5:13 pm
Country: Germany

Re: Servo center?

Post by jhsa » Tue May 20, 2014 10:18 pm

It could be. that would have to be tested. The plane would, as far as I understand be already on the ground, And he want to use the existing surfaces as airbrakes.. Two main points. It must keep the aircraft on the ground, and produce drag. I think the plane's behavior will depend a lot on the aircraft itself..

João
My er9x/Ersky9x/eepskye Video Tutorials
https://www.youtube.com/playlist?list=PL5uJhoD7sAKidZmkhMpYpp_qcuIqJXhb9

Donate to Er9x/Ersky9x:
https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=YHX43JR3J7XGW

Daedalus66
Posts: 1839
Joined: Tue Dec 27, 2011 8:22 pm
Country: -
Location: Ottawa

Re: Servo center?

Post by Daedalus66 » Wed May 21, 2014 3:02 am

I did the testing on the Polaris XL, a 1m span delta flying boat with tail. I had it set up with separate aileron servos so I could test various flaperon, spoileron, elevon configurations.

As you might expect, flaperons were a bust, producing strong pitch down that took lots of up elevator to compensate but not really improving the flying. Elevons coupled to go down with up elevator, and up with down, certainly added control authority in pitch and made for very tight corners on square loops. Finally, the setup I liked best was spoileron (up aileron), which produced a steep and stable descent with the model in level attitude. It was so stable I could do hands off landings onto grass from a considerable height. The only requirement was to get the speed well down before popping the ailerons up or there was an instant loop.

I must try all this again now that I have sophisticated programming that can do things like slow and delay and easy adjustment of mixing.


Post Reply

Return to “The Pickled Gnu (The Pub)”