Servo torque vs. moving surface.

[beto9]

I remember reading in some thread a rule of thumb about servo torque required for a given moving surface. It was something like "1 oz. of torque per sq. inch"...
Can anybody enlighten me. pls?

[Jeremy Cartlidge]

This site has an interactive calculator:

http://www.csd.net/~cgadd/eflight/calcs_servo.htm#formula

It is based on this formula:

Torque(oz-in) = 8.5E-6 * ( C2 V2 L sin(S1) tan(S1) / tan(S2)]

Where:
C = Control surface chord in cm
L = Control surface length in cm
V = Speed in MPH
S1 = Max control surface deflection in degrees
S2 = Max servo deflection in degrees

[beto9]

Thank you Jeremy... you are WIMAC's Wikipedia!

[Mick]

Excellent calculator, thanks Jeremy. I would apply a 50% factor to ensure enough margin for all the caveats.

One for the tools section, Greg?

[beto9]

It also says...

"Reducing the servo deflection from the default 60 degrees is similar to using ATV / Dual Rates to reduce the control throws. If you vary the servo deflection from the normal 60 degrees, you will see that using "Dual rates / ATV" to set the proper control surface deflection greatly increases the load on the servo."

So, if I understand it correctly, is better to set the maximum surface movement by choosing the right holes in the servo arm and the control horn and letting the servo travel its full deflection (60 degrees) rather than limiting the servo movement with "end point" or "ATV"....

Ah! All the things we don't know.....

I wonder how many people are over-servoeing their surfaces.... at $$$$$ cost!

[Mick]

Mechanical advantage - it is always good to use the maximum travel of the servo and then to use the arm lengths to get the right control surface movement. In addition, always set up the link length correctly to centre the control surface before using a sub-trim to fine tune it, this will maintain maximum travel availability of the servo.

In other words setup the radio like it is a non-computer radio and you will not be far wrong. The only exception to this advise is the throttle as it is a pain in the .... to do this manually, use the radio.

[Xavier]

This is a very interesting link Jeremy. I had seen it before but lost it. Thanks. It gives a starting point and then you add the safety factor you want.

Depending if you are looking for precision or speed, you may arrange the arms differently.

On my precision aerobatic airplanes, I setup the travel adjustement of the servos or ATV close to the maximum possible. (115-125% on the JR) and then use the proper holes on the servo arm and surface horn to achieve the required deflection angle. In some cases, the servos may be rotating 60deg each way and the surface only 20 deg for exemple. That means that the torque at the surface is actually 3 times the torque of the servo. It may allow the use of a smaller servo than we originaly think and save weight.

If I am looking for speed (for 3D), I will use an oversize servo, lower the servo travel and arrange the arm accordingly. The torque available at the surface is then lower. It is rare that the surface will move more than 45 deg, so with the servo also limited to 45 for exemple, the torque available will be the one of the servo. If the servo takes .18 sec for 60deg, the surface will move at the same rotational speed. In the first case, with the 3 to 1 ratio, the surface will move slower in angle speed.

You could also, see that with the first case, you move the TX stick to full deflection to have 20 deg on the surface. Moving to 5 deg will take 1/4 stick deflection (not including expo or linkages geometric effect). It the second case, 5 deg is on 1/9 of the stick deflection. It means that it the airplane will a lot more sensitive.

I mentionned the expo because it also has an effect on the surface speed. If It's like limiting the servo travel for the first part of stick movement and increasing it at the end. It's a good think but if you go too far (60 and more on JR, -50 and less on Futaba) you loose torque in the middle and precision at the end.

Mick is right about the trottle linkages also. That's another think easier with electric :-) Seriously, the trottle is the only (basic) control that could be set mechanicaly with expo. Arranging the arms correctly, you could lower the sensitivity of the trottle on the bottom half and increase the speed on the top half. It makes the slow flying and landing approach a lot easier. In the case, forget the 90 deg linkage geometrie shown on all radio instruction. I would need a sketch to explain it properly. Maybe later.

Xavier

[beto9]

A fellow flyer at RCUniverse downloaded for me a XCel spreadsheet to calculate torque.
Here it is.
I haven't tested it with the one proposed by Jeremy.... some rainy day....

[Jeremy Cartlidge]

It is the same calc as the one on the website (both by Craig Tenney). It is a really useful spreadsheet for understanding the loads and the importance of picking the right linkage configuration.

[beto9]

I tested the spreadsheet with the Web site and the results are the same. However the spreadsheet is much more sophisticated, I don't know if you noticed the different tabs in the bottom. There is even a (theoretical) speed calculation based on rpm, prop length and pitch... and for our large planes friends, Mach numbers (prop noise)!
It also confirms that allowing the servo 60 degree travel the torque required is much less.
Hurray! Now when we crash we can blame the math!

[Kevin Rochon]

All I have to say to all this is My Cap at 18 lbs which has big aileron I use reg. 3004 Fut servo! I can snap, maybe not as sharp has with maybe 100oz servo but it`s all marketing and getting you tho spend more.

kevin

[beto9]

I agree 100% with you Kevin. I made the math for all my planes and they go OK with 3004 or 3001, with one surface exception (out of 9). And I am talking about measuring for the large Great Planes Stearman (Saito 1.80), Aeroworks Yak 1.20 and my soon to be Seagull Yak (OS .91).
There is a great difference about how much you allow the servo to move. It develops its full torque when moving the expected 60 degrees each way. I made a few scenarios and the torque requirements increase substantially when you use less than the normal servo movement.
That means that you have always to try to let the servo move 100% on each direction (no ATV or END POINT) and then choose the proper holes in the servo arm and the control horn to achieve the required deflection.
Jeremy? Our Wikipedia! Comments?

And then, we can always say "Damn torque crashed me!"

The snap is more a function of the servo speed and not the torque... I think. Use 6V and see if there is a difference.

[Kevin Rochon]

I know that the torque they indicate is let say 40oz /in that means at one inch from the servo screw you have 40 oz if you bring it closer in at 1/2 you get more torque less movement, I don't know if it's double?

kevin