Folks, I followed Joe Larson's article on SERVO LOADS in August RCM. Through the publlished equation, I have arrive to my version of table showing servo torque requirement vs. air speed in considering the control surface featured given by an open class ship (Prism), and the analysis is as follows: The data in table is divided into two sections, roughly upper left and lower right. The data split is driven by a predetermined threshold servo stalling torque of 35oz*in off a typical wing servo on the market. The border line is represented by junction between parenthesis ( ) and < >, for aileron and flap, respectively. Trying to be less specifical to the data presented here, servo-arm:control-horn radius ratio is 1:1. Displacement is in degrees represented by deg., Speed is in mile/hour as in mph and Servo torque as in oz*in. Table 1. Function: Aileron------------------Flap Servo : =<35oz*in >>>40oz*in Span : (23) in ----------------<26> in Length : 2.0 in ------------------2.0 in deg.=10 15 22 33 45 60 90 mph oz*in oz*in oz*in oz*in oz*in oz*in oz*in 10 0.5 0.7 1.0 1.4 1.8 2.3 2.6 15 1.0 1.5 2.0 2.8 4.7 5.1 5.9 22 2.2 3.2 3.9 16 8.9 11 13 33 5.0 7.3 7.8 29 (20) (24) (28) 45 9.2 14 20 (33) <42> <52> <60> 60 (16) (24) (35) <58> 75 92 106 90 <41> <62> <90> 130 169 207 239 127 83 124 179 261 338 414 479 Analysis: At 127 mph, most wing flap servo stall at 4.8 degrees. Given the above example and servo's stalling torque limit, it is clear that slick ships may risk flap serveo stressed failure if homing at steep terminal velocity and deploy flaps at the last minute, even at a speed as low as 45 mph. If terminal veolcity is not avoidable, there may be four solutions. First approach is to support/maintain tail up reaction due to flap down, that can transform more wing (tail+fuse) area into frontal area and thus drag, that would relax the servo torque required relative to condition when the tail is line up with the air stream. In addition, take more time to reach full flap configuration would help more in preventing servo stress. Second approach is to program two stage of 'butterfly' on each wing-halfs. i.e. +/-10 to +/-20deg. between time stages for (+)aileron and (-)flap, respectively. That should take the high torque demend off both seros. Third approach is to deploy spoilers on wing top. There should be no major concern on servo stress for hinged-spoiler type comparing to the gate-like types, because the air on top tends to lift it open. Remember the operating principle of high speed flow separation and the formation of separation-bubble? Forth but future approach is to develop a mass producible slow but high torque servo, such as one that can output 200oz*in over one second. Although there are technique and products which can utilize the trigonometrical transfer functions to our advantage here. But that can only comes out as supporting solution to the problem here. At this point, I begin to aware that preproduction servo stress test should be made with (fixed 60degree) sin-function torque-load curve with stall torque set at end of each different target displacements from 5 to 60 degree in multiple steps and tests. Because not all servos that are able to pass static stall test can also pass the dynamic stall test. That can be traced to the mechanical design dimensional tolerance error plus the stressed geometry deviation error of the gear and its box as a system. (haven't mention wear and tear yet) In particular to resin composition non-metal gear and housing. I thought that would be of interest to some of you. Comments are welcome. Regards, YK Chan Seattle area RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
