Re: [RCSE] Icon 2
Get in line folks - this is gonna be a biggy, both literally and figuratively. Designed by the intrepid Dr. Mark Drela, this new high aspect ratio F3J/Thermal Duration plane has a wing span of 150 (3.81 m) and a VERY slender 2.4 friendly pod and carbon boom. Just for the record, I really didn't design the Icon 2. I was more of a consultant. Don Peters and Phil Pearson have been doing all the detail design work. MD RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
[RCSE] Top end speed in MPH ! Math quiz
First of all, if the glider is being pulled level, the winch doesn't see the weight, only the drag. The drag is tiny compared to the winch's pull capability, so it will be spinning at nearly its zero-load speed. I think a typical Ford LS 6V has Kv=350 rpm/Volt, so at 12 Volts it will be spinning at rpm = 350*12 = 4200 rpm = 70 rps With a 3 diameter drum, the line speed will be 3 * pi * 70 rps = 660 in/s = 55 ft/s = 37.5 mph Seems about right. In any case, you can check it by timing how long it takes the winch to pull the chute some measured distance. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Super AVA airfoil
The intended airfoils for the AVA center panel are AG24 in the center, transitioning to AG25 at the joiner. I don't know how close the actual wing is to these. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
[RCSE] re: Rudder Stalls and Turns
There is a tendency for R/E models to pitch up when a turn is initiated due to gyroscopic precession. Yes. But this precession effect is very very weak. In my simulation of a BD being given a fast 30 degree rudder input, the nose pitches up by only 0.03 degrees. Nevertheless, in this ruddering maneuver of a r/e glider, there is indeed a sudden large AoA increase of about 3 degrees (more than enough to cause a rudder stall). But the cause is simply due to the linear inertia of the glider, not to precession. Imagine this sequence during a rudder turn, with the rotation angles exaggerated to show the effect: 1) Glider is flying level in a straight line. 2) Glider yaws to the right 45 degrees (while still moving along the straight line) 3) Glider rolls to the right 45 degrees along its now-yawed fuselage axis (while still moving along the straight line) You will see that the glider's belly now faces the oncoming wind along the original direction of motion, which constitutes an Angle of Attack increase. In reality, the glider's yaw, roll, and turn motions will all blend together with some lags between them, but the effect to cause the AoA increase will still be there. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] 2M Rambles
I've written on this topic before, but I think it might be good to repeat it. The reason that most 2m gliders fly like crap (quoting one CRRC flier), is that they are WAY too heavy for their span. Consider a log-log plot of weight versus span. The following three points lie roughly on a straight line: 1.5m, 10 oz 2.0m, 19 oz 3.4m, 60 oz All three gliders will have good flying characteristics, and are able to work very light lift. A 40 oz 2-meter is off the chart, and will fly like a lead sled in comparison. Adding wing area via lower aspect ratio helps with the minimum turn radius, but it worsens the max L/D and doesn't help min sink all that much because of the increased induced drag. As was recently pointed out, the main problem with a 19 oz 2-meter is building it strong enough to take a winch sized for 3m+ gliders. Compounding the problem is that a ~20 oz 2-meter ship wants airfoils which are more like those on a DLG rather than on the 3m ships. Such thin airfoils make the structural problem even more severe. If we were all launching on 12v winches with correspondingly lighter lines, the light 2m gliders would likely be competitive with the 3m ships, because the 2m ships would launch higher. Their maneuverability and lower speeds on landing is another advantage. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
RE: [RCSE] re: Rudder Stalls and Turns
When a rudder stalls due to AoA deflection, what would a pilot see? Is there less rudder response? I'm not sure what the original poster (Jay Hunter) meant by rudder stall. It was either 1) deflecting the rudder causes the glider to stall, or 2) the rudder itself stalls. Effect 1) can easily happen if there's enough roll-due-to-yaw coupling, which of course is strongest on a r/e glider. So if your poly ship is flying straight at minimum sink on the edge of stall, and you simply slam the rudder hard over, the wing will almost surely stall. As Blaine pointed out, an experienced poly pilot will unconsciously apply some down elevator along with the rudder to prevent this stall when rolling into a turn. On a flat wing airplane this effect doesn't exist, and there's no need to apply this down-elevator correction. Effect 2) will happen with enough rudder deflection and/or sideslip, but it's almost certainly inconsequential. When a low-AR surface like a vertical tail stalls, its lift typically will just level off rather than dropping sharply. So you probably won't even notice it. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Push Rods
I've done quite a bit of soldering with stainless steel, since most of my pushrods and my RDS hardware is SS. I simply use acid flux, aka tinner's fluid, together with ordinary soft solder and a normal soldering iron. Nothing fancy. The acid flux is hydrochloric acid with zinc chloride salt -- not really toxic, just corrosive. I thoroughly clean the SS with clean 320 grid sandpaper, apply the flux, and touch with the molten solder. About half the time the solder doesn't grab completely, just in some spots. No problem... I just keep applying flux to the piece (while hot), until it tins completely. Two or three tries usually does it. If you're worried about corrosion, a good trick is to first tin the SS using the acid flux, and clean it completely. Then you can use regular rosin flux to make the actual joint. I never use high strength solders which require a torch, because a torch will turn hard piano wire or most hard SS into soft annealed coat-hanger material, which usually defeats the purpose of the strong solder. This is not a problem with soft solder. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Supra #42 problem
It looks to me like the bolt anchor mounts extend down into the pylon, but nowhere near as far as they could. A sound fix would be to switch to longer bolts which go all the way through the existing anchors, and into epoxy/glass-flox threads potted in between the pylon walls. This may require tapping through the bolt anchords, dunno. If the bolt extends down below the point where the pylon flares out at the bottom, then the bolt will be mechanically trapped, so you're not relying on the epoxy's peel strength. Potting longer bolts should be straightforward even if the platform plate is still intact. Just use a syringe to inject epoxy/flox splooge through the bolt holes. Or maybe through the pylong access hole? Then screw in the well-waxed bolts. BTW, the best way to prepare the CF surface for bonding is to wet sand it with wet epoxy. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Gluing Servos v.s. Servo Frames
If keeping a few grams out from around the servo holes improves yaw inertia by 5% then how can loading your tube spar with ounces ballast be a real benefit? First of all, the contribution of any added weight to the inertia depends on the square of the distance of the added weight from the CG. So adding mass at the tips is 4x worse than adding mass halfway out. Here's how much 10g of added mass (5g on each side) adds to the percentagewise inertia of a 60 oz Supra, versus spanwise position of the added mass: y=67: +4.0% at tip y=36: +1.1% at outer servo y=10: +0.1% at inner servo Second of all, what really matters for handling quality is not the inertia, but the inertia/mass ratio. This makes adding mass near the center even less consequential. Here are the effects of the 10g mass on the inertia/mass ratio: y=67: +3.4% at tip y=36: +0.6% at outer servo y=10: -0.5% at inner servo So adding mass near the tips is still quite bad. But adding mass near the center (inner servos, wing ballast, etc) actually makes things better, not worse. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
RE: [RCSE] Carbon Supra #48 flies - weight 61 oz !
Did anyone else notice that the calculated flap servo torque requirement (62 in-oz) is higher than the specs on all commonly used servos other than the Volz Micro Maxx HP. The JR DS368 is spec'ed at 53 oz-in. The thin wing servos, JR DS-168 and Hitec equivalent at 46.6 oz-in. Two comments: 1) What Daryl said. The assumed 200 lb load is kinda huge. 2) If the flaps and ailerons don't move significantly during the last few seconds before the zoom where the load maximizes, then the torque requirements I quoted are for holding torque, not working torque. AFAIK, digital servos like the DS368 have way more holding torque than their spec'd torque. In that case they should be OK for the flaps. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Carbon Supra #48 flies - weight 61 oz !
What about elevator loads? And elevator servo moment? How would that be calculated? The all-moving Supra stab is aerodynamically balanced, so the stab servo loads from the airload are minimal. The only loads the stab servo will see is due to the inertial load when the boom cracks the whip in the zoom. I have no idea what the associated accelerations of the tail might be here, but let's assume the tail is accelerated vertically at 100 G's. The 1 oz stab then weighs 100 oz. Since the stab's CG is about 0.5 behind the pivot, the hinge moment will be 50 in-oz. With the typical 2:1 horn length ratio for an all-moving tail, this translates to Stab servo moment = 25 in-oz But again, the crucial assumption here is the 100G vertical acceleration of the tail. No idea what it really might be. It could probably be estimated from a high-speed telephoto movie of the zoom seen from the side. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Carbon Supra #48 flies - weight 61 oz !
Molded Supra launch hinge moments are listed below, for a 200 lb tow load (hard F3j launch). For smaller tow loads, hinge moments will be proportionally smaller. wing camber = +10 deg rudder defl = 30 deg wing CL = 1.2 lift = 200 lbs q = 0.16 psi = 1104 Pa V = 95 mph = 42.5 m/s Flap hinge moment = 4.87 in-lb = 78 in-oz = 5.62 kgf-cm Ail. hinge moment = 3.15 in-lb = 50 in-oz = 3.60 kgf-cm Rud. hinge moment = 2.12 in-lb = 34 in-oz = 2.45 kgf-cm To get the servo moment (torque), this hinge moment must be multiplied by a linkage geometry factor, which is approximately by the ratio of servo/surface horn lengths. This is between 0.7 - 0.9 for most installations. Assuming a horn length ratio of 0.8, the servo moments are: Flap servo moment = 62 in-oz Ail. servo moment = 40 in-oz Rud. servo moment = 27 in-oz RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Molded Supra Raw Weights
The next thing, I am trying to understand why some people are talking about having to add on the order of 4 oz of nose weight to balance. Was this due to variations in fin, stab, boom, or pod weights, or is this inherent to the molded design? Most likely it's because the stuff in back is heavier than on my bagged Supra: tail feathers boom pushrods V-mount fin attachment I anticipated this and specified the molded pod to be longer in front, but apparently not long enough. Anyway, it's easy to check the weights of all the bits above and see where most of the added tail weight is. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
RE: [RCSE] Supras done
could someone explain to me the balancing method and tools used to place a CG to 0.5mm accuracy? To measure the CG: I turn the assembled glider over on its back, and find the level-balance point by supporting it on a wood-pencil eraser which is slightly rounded. I might add some tape to a wingtip if necessary to get perfect lateral balance, so the eraser point can stay exactly on the centerline. I mark the eraser (CG) position on the wing, and measure its distance from the LE using a ruler which has a balsa stick taped onto its end to make an L. The balsa stick rests against the LE while the ruler is held parallel to the wing bottom or the chordline to make the measurement. The recommended 3.6 CG position is what I measured off my Supra in this way. It's a relatively far-aft position, which is just on the verge of tuck-in in a steep dive test. It recovers very slowly in a shallow dive test. The best CG and hook positions may depend somewhat on the chosen dihedral. To measure the hook location: I tape a short balsa stick or dowel into the hook, parallel to the wing. I then place a drafting triangle against the bottom of the wing, and touch it to the stick. This gives the hook position on the wing chord, measured perpendicular to the wing bottom surface. The recommended 3.8 hook location is what I measured off my Supra this way. This is rather far aft hook position which requires special action during launch. On my Evo program the left slider is an elevator (speed) trim with a small gain. For the launch throw I push the slider all the way forward which feeds in a slight amount of down-elevator. Once the glider rotates and settles down in the climb, I pull the slider back to its normal center position for maximum load just short of stall. I have 100% Ail.Diff and lots of Ail-Rud coupling during launch, so I can steer with might right thumb, leaving the left thumb free to work the slider. PS I've done some sims of the initial pitch dynamics immediately after the throw. There is a very significant CL overshoot at the end of the initial nose-up rotation. So if you trim the glider for maximum pull during the climb and zoom, and throw with this trim, then you are guaranteed to stall at the top of the nose-up rotation. So for maximum launch performance, it is necessary to add some initial nose-down trim to safely get past the initial pitch transients after the throw. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
RE: [RCSE] Long music wire????
You can get dead-straight hard stainless steel wire from www.smallparts.com in 60 lengths. Not quite as hard as piano wire, but pretty close. The drawback is that the 0.06 size is about $12 a piece, plus an extra oversize shipping charge. They also have loose coils of hard stainless wire, but only up to 0.043 diameter. This stuff is vastly cheaper at $4 for 30 feet. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
RE: [RCSE] Thermal duration V or + tail
Many European moulded models are available with either style of tail. Yes, but the effective tail sizes between the two version are always very different. From a recent RCGroups post: During a stall with the v-tail they like to do the snap spiral. Its quick and you have to be careful. I think you've been flying too many V-tailed Euro-moldies. :-) All the ones that I've looked at typically have badly undersized V-tails, although the X-tail version of the same glider typically has an OK tail size. For example, here are the horizontal and vertical tail volumes I calculate for the Europhia from the 3-views at www.aerodesign.de : V-tail version: Vh = 0.36 Vv = 0.0115 (yikes! that's real small, and likely to be snap-prone) X-tail version: Vh = 0.41 Vv = 0.0205 (barely OK) The effective vertical tail size on the V-tail version is roughly HALF that on the X-tail version, which isn't all that big to begin with. Go figure. In reality, there shouldn't be much of a difference in aerodynamic behavior or in weight between different tail types, provided each one is sized appropriately. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Tip on bagged wing twist removal.
Here's another method without the towels: 1) Tape over any servo bay openings. 2) Set wing on edge, leading edge up. 3) Twist the wing to remove the offending warp, overtwisting by maybe 50%. 4) Slowly pour boiling water on the LE so it runs down and heats the whole wing. 5) Hold twist until cool. Having a helper to pour the water makes this very simple. The heat will probably cause cause the cloth weave texture to become more visible in the surface finish, but this is not aerodynamically significant. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
[RCSE] Re: Wing Twist
Warm up the wing with a hot air gun (monokote variety) and gradually heat it (top and bottom) until you have about twice the amount of final twist you want. This works, but it's risky. You want to heat up the epoxy to at least 200F for a nice permanent set, but the foam will start to melt at 250F or so. That's not a big spread, given the rather poor temperature control of a heat gun. Using boiling water is totally safe by comparison, since it cannot heat the wing above 212F. The water also heats up the skin almost instantly, while the heat gun requires considerable patience. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
[RCSE] Laser Arts retriever plans
Ken Antonellis and John Hayes at Charles River RC have built these. I also copied the frying pan spool design when I built my retriever. They work well, but I would make the following mods: 1) Make the rear spool plate as large as possible. This will discourage the line from jumping behind it and into the V-belt. My plate is over 12 diameter, and it could be bigger. 2) Use a large number of bolts to clamp the spool together. I used 18 bolts, 10-32, spaced uniformly on a circle which is as large as possible (just a bit smaller than the spool core disk). This was just barely adequate. The packed retriever line exerts tremendous bursting pressure, and without the closely-spaced bolts being adjacent to the wound up line, the soft cast-aluminum flying pan will deform outward like taffy. The LA plans show only the 3 U-bolts holding the spool together. This is OK for attaching the pulley, but it's grossly inadequate for clamping the spool together. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Teaching a kid
First of all, the Spirit is not a particularly good-handling r/e glider. It's wobbly in yaw, and prone to tip stalling because of the saggy unsheeted tip panels. An AVA-type glider is much easier to fly well I think, and of course it would be a huge step up in performance for him. I'd avoid the full-house-RES converted gliders like the Victory. They're heavy by comparison. But if he really wants to go all the way to a full-house glider, I'd go with something light and forgiving. The Aegea/Mantis is an obvious choice at a reasonable cost. A DLG would be even better, unless of course he's set on TD. If you do get an Aegea, I suggest bending the rods for more dihedral. This really helps in thermalling, especially for an aileron newbie. There is no real downside to it. My Supra has 2.5 deg on the inner panels, and 7.5 deg on the tip panels, and it can almost fly hands off as a result. For the Aegea, I'd shoot for around 8-9 degrees on each tip panel. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
RE: [RCSE] soldering problem
The acid flux did the trick. What he said. Use regular solder with acid flux (aka tinner's fluid). Works on stainless steel, too. A big problem with using silver solder on piano wire is that it requires a torch. The solder may be strong, but the high heat turns the hard piano wire into soft annealed mush. This often defeats the purpose of the strong solder. A soldering iron is cool enough so the piano wire is unaffected. If the solder joint needs strength, there are often tricks to get that. Increase the solder area, or wrap the joint with fine wire before soldering. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Still using aileron differential?
1) still use aileron differential in 2 and 3 meter planes and 2) still couple the ailerons to the flaps (so they act as flaperons) Here's what I use on my Supra: 1) During normal flying: * 10% differential, which is almost nothing in practice * 20% Ail-Rud * 40% Ail-Rud when airbraking The Ail-Rud mix is to make the aileron stick nearly a pure roll control, and it's in no way a replacement for independent ruddering. I still use the rudder stick a lot during circling maneuvers. During launch: * 100% differential * 100% Ail-Rud I use about 12 deg of full-span camber on launch, and trim close to stall. I don't want any downward TE deflection in this condition. 2) Yes, I make the flaps respond to the aileron stick, about 60% as much as the ailerons do. This is a compromise between: - least drag in roll maneuvers (wants flaps moving 33% as much as ailerons) - maximum tip stall resistance (wants flaps moving 100% as much as ailerons) RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
RE: [RCSE] Re: Programming tricks help
Shouldn't the flap and ailerons move the same amount to keep drag to a minimum where the flap and aileron meet? No. This would create an even bigger discontinuity at the center between the two flaps. Repost from a previous message some time ago: I'd like to more about your thoughts on aileron to flap mixing. A small percentage of mix, one to one, somewhere between? Why is good too. Deflected ailerons deform the load distribution away from the ideal near-elliptical shape, and hence increase induced drag. Partially slaving the flaps to the ailerons can alleviate this load distribution deformation, and thus mitigate the ailerons' CDi penalty. The question is what's the optimum amount of ail- flap mixing. The lowest-drag aileron system is wing-warping as used by the Wright Brothers -- the wing is linearly twisted from tip to tip. When such a twisted wing reaches its steady roll rate, the load distribution returns to its optimum level-flight shape, and the drag penalty is zero. With a finite number of hinged control surfaces such a linear twist cannot be achieved. But it can be approximated as close as possible if each surface's deflection is made proportional to its distance from the aircraft's centerline, measured at the surface midpoint. If the four control surfaces have equal span, we then have: surface mid_span_loc. deflection --- - -- L.aile -3/4 -100% L.flap -1/4 -33% R.flap +1/4 +33% R.aile +3/4 +100% So for this wing the flap motion should be 33% of the aileron motion. Using AVL I've verified that this mixing ratio produces very nearly the smallest induced drag penalty. If the flap span differs from the aileron span, the table above can be adjusted accordingly. Longer flaps will have larger travel and vice versa. BTW, this distance-proportial deflection rule strongly argues against stopping the ailerons short of the tip. The resulting unhinged tip portion should in fact have the largest deflection. The distance-proportial deflection rule can be fudged if there is a tip stall problem in a sustained turn, where some opposite aileron must be held. By increasing flap travel over its optimum amount, the flaps can carry a greater share of the roll power, which reduces the required downward deflection of the inside aileron, and thus delays tip stall. So if your TD glider has insufficient tip stall margin, I suggest increasing the flap mixing and you should see some improvement. The extreme case would be 100% flap mixing, which mimics full-span flaperons. Flaperons give excellent tip stall resistance, as is obvious to anyone who flies a DLG with a good 2-servo wing. A 4-servo TD wing with decent planform should not need to go to this extreme, especially if it has some washout like the Aegea wing. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Are you using long servo arms?
| Bearing/casing/mount/hinge/pushrod loads depend only on the surface horn length, | in inverse proportion. Doubling the surface horn length will reduce these | loads by half. But pushrod loads would be in porportion to the horn length, not inverse porportion. Double the horn length, double the load. No. The horn length and pushrod force are related to the servo torque by horn_length * pushrod_force = servo torque For a given servo torque, doubling the horn length will halve the pushrod force, and vice versa. Example: If you mount a 10 horn on the servo, how much force would you have to apply at the tip of the horn to stall the servo? Not much. With a 1 horn, stalling the servo requires 10x more force. Not sure what you mean by hinge or mount loads. The hinge, servo casing, bearings, mounting, etc. all see the pushrod force. Reducing the pushrod force via longer horns will proportionately reduce the deformation of these items, and give an overall stiffer system. Longer horns will also decrease any slop due to oversize horn holes or loose clevis threads. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Are you using long servo arms?
Simultaneously Increasing the length of the surface horn by a similar amount will not increase the load on the servo. Servo linkage should always be designed to use the maximum amount of servo travel to give the desired control surface throw. We should distinguish between different types of loads: servomotor/gear loads bearing/casing/mount/hinge/pushrod loads Servomotor/gear loads depend only on the ratio: surface_horn/servo_horn Doubling or halving both horn lengths has no effect on these loads. Bearing/casing/mount/hinge/pushrod loads depend only on the surface horn length, in inverse proportion. Doubling the surface horn length will reduce these loads by half. The following linkage design procedure will minimize all the types of loads: 1) Determine the maximum servo travel range. Set the horn length ratio so as to get the desired resulting maximum surface travel range. Assuming the travel of the servo and surface are symmetric, the equation for the horn length ratio is: surface_horn/servo_horn = sin(surface_throw/2) / sin(servo_throw/2) Example: 100 deg servo throw (+/-50) 70 deg rudder throw (+/-35) rudder_horn/servo_horn = sin(100/2) / sin(70/2) = 1.33 2) While keeping this horn ratio fixed, make the horns as long as possible. Let's say the longest horn I can fit on my rudder servo is 12mm. The corresponding rudder horn should then be 12mm x 1.33 = 16mm Nonsymmetric throws require a more elaborate analysis. Blaine Rawdon's spreadsheet does the job well: http://members.cox.net/evdesign/ RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] Wing design ....
I have been under the impression that a thinning of the section and a slight increase in camber as you progress out to the tip was preferred. There's no unique way to do this. I normally set a combination of taper and aerodynamic twist (i.e. washout) to give a good cl safe distribution in a trimmed sustained slow turn, and also in a fast upwind dash. Some tradeoff decisions are needed here. In the slow turn one can use a very broad tip with no twist, or a highly tapered tip with lots of washout twist, with the same cl distribution. The slow-speed case favors lots of taper and twist, while the high-speed case favors little taper and twist. An intermediate compromise choice is often effective. Also, poly, 2-servo, or 4-servo wings will differ substantially here. Poly (RES) and 4-servo (TD) wings generally want significant washout twist in the intermediate panel, while 2-servo wings (DLG) can get by with much less washout. For the 4-servo wing, the amount of washout also depends on the amount of aileron-flap linkage. Then I pick or design airfoils to match the cl range and Reynolds number. This usually means a reduction in thickness and a slight reduction in camber towards the tip. Some iteration with the specified cl, planform, and twist distributions may be needed. At the end I set the geometric twist as the sum of the earlier specified aerodynamic twist and the airfoils' zero lift angles. For building convenience I round off the twists to the nearest 0.5 degrees or whatever. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
Re: [RCSE] How to use Klass Kote
I would like some recommendations on spray equipment On recommendation of Denny Maize, I got the following touch-up spray gun from Harbor Freight: Item # 86-7VGA It's fully adjustable in spray pattern and paint volume output. I notice that it's on sale right now for only $20. You obviously also need a compressor with an adequate flow rate capacity. One of the little compressors for airbrushes probably won't work. Mixing the paint and cleaning up is a lot of work, but this is less of an issue if you spray many mylars with one paint batch. Is it OK to just spray the white paint with catalyst on the mylar and then go directly to bagging the wings I didn't use any primer on my Supra wing. The paint by itself stuck very firmly to the MGS epoxy. How long should I wait after painting the mylars before bagging the wing? I let them fully cure for many hours at 120F. It certainly didn't hurt the adhesion to the epoxy. I assume that the paint will continue to cure normally even after the wing is bagged. Am I right about all this? Yeah, but bagging too early may give some print-through of the Kevlar weave. A test is probably in order. My white paint coat was extremely thin and light -- only about 12g for the wing top surface, but the coverage and adhesion was still good. Print-through is less of an issue for a very thin coat. You might want to consider painting your tails like this, since less than 1g of paint on the tail will give great coverage. Once the gun is ready, spraying the tail mylars is negligible extra work. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off. Email sent from web based email such as Hotmail and AOL are generally NOT in text format
[RCSE]
You are doing it wrong. You cut the bottom lug off and use a flat head screw (through the top lug) to mount the servo to the block, which is about 3/4 of the thickness of the servo. Also the mounting block would block the servo from moving. We can get more specific here. In general, the ideal place to attach any servo is as close to the pushrod as possible. This minimizes the torque that the pushrod exerts on the wing or fuse structure which holds the servo, and thus minimizes servo squirm. This is why I now try to mount standard-lug servos by putting the screws up from the bottom, rather than down from the top (e.g. Supra wing and fuse servo mounts). The flat-mounting lugs on the Airtronics servo would have been better placed above the standard lugs, not below them. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] All flying vertical stab...
I'm building the new tail for my solar plane and I'm thinking I'll make a Drela style carbon V mount for the vstab. The original carbon V-mount was on the Daedalus HPA, on both the stab and the rudder: http://trc.dfrc.nasa.gov/gallery/photo/Daedalus/Medium/EC88-0059-002.jpg http://trc.dfrc.nasa.gov/gallery/photo/Daedalus/Large/EC88-0059-002.jpg (hi-res) http://trc.dfrc.nasa.gov/gallery/photo/Daedalus/Medium/EC87-0014-8.jpg http://trc.dfrc.nasa.gov/gallery/photo/Daedalus/Large/EC87-0014-8.jpg (hi res) Putting the rudder on a V-mount works fine, except striking the rudder on the ground may cause it to break. I suppose you could have the V-mounted rudder mostly above the boom, which would minimize the torques on the V-mount from a ground strike. But rudder airloads will torque the boom more. Then there's the agony of deciding whether to put the rudder on the right side or the left side of the boom... RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] Battery food-Charger
Are there automatic / peak detection chargers that will ramp down enough to put a small enough charge (C/10 or so) on the tiny 200 mAh NiMH (yep, quad A) batteries I'm using in HLG? I use the Sirius 200 for the batteries. After the peak charge it switches to a maintenance mode which gives an occasional charge pulse every 10 seconds or so. If possible, I leave it in this mode for some time. It should be equivalent to a very slow trickle rate which will eventually equalize the pack. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] Harley, Gordy, RDS and molded wings
I used the typical euro, top actuated, bottom hinged system for the flaps. It kept everything inside and a smooth wing bottom, but Gordy's right it is a bit squishy as a system. I tried to stiffen it up with 4-40 threaded rod, soldered to heavy links. It's the nature of the geometry I guess. I looked closely at the linkage stiffness issue. I conclude that a conventional horn and pushrod system is inherently more flexy than a properly-executed RDS system. These are the sources of compliance/slop common to both systems: gear backlash servo shaft twist These are the additional compliance/slop sources in RDS: driver arm torsion driver arm bending wiper slop in pocket These are the additional compliance/slop sources in a conventional system, due to the large pushrod force which the servo and hinge see: servo arm bend servo output shaft flex servo output shaft bearing slop servo mount flex surface horn flex hinge flex pushrod/clevis slop in holes The big difference is the large pushrod force in the conventional system which causes compliance all over the place, while the RDS servo sees a nearly pure torque. The RDS servo does see a small force from the shaft acting like a lever, but this force is typically a small fraction of the pushrod force in the conventional setup. The pushrod force and the associated compliance can be reduced by using longer horns on the servo and the surface, but there's a practical limit to this, and of course there's a drag penalty. In contrast, it's easy to make the RDS driver arm as stiff as needed. It this is done, the RDS system will be significantly stiffer. On my Supra I wanted a pinned shaft attachment, so instead of the Kimbrough solid wire shafts I used a large-diameter hypodermic tube. This is almost infinitely stiff for all practical purposes, and metal servo gears and shafts do not flex much either, which leaves gear backlash as the only significant source of slop, with little compliance anywhere. The bottom line is that a properly-built RDS gives the most solid linkage possible. I only wish it was as quick and easy to build as a horn/pushrod system. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] What is the specific AVA Towhook Location and CG?
Here are my Bubble Dancer locations. Should be very similar to the Ava's: CG is 3.75 behind the center LE. Rear face of the towhook is 3.50 behind the center LE. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] New differential question
How do you deal with adverse yaw caused by a V-tails use as a rudder? I think you mean adverse roll, not adverse yaw. Yes, a V-tail has adverse roll which forces you to deflect the ailerons a bit more while rolling. On the other hand, for a given amount of yaw moment, the V-tail has less of its own induced drag than a conventional vertical tail of the same height. I think the conventional tail slightly wins here overall, but that's a guess. I haven't run the numbers. It's surely configuration dependent to some extent. If you're holding rudder into an established turn, the adverse roll of the V-tail has an unexpected benefit -- it unloads the inside wingtip and helps to suppress tip stall. In practice the effect is very small, but then we're talking very small differences here in any case. I see the V-tail vs X-tail choice as mainly a structural, mechanical, and construction issue. The aero differences are puny by comparison. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] Mapleleaf Encore wing damage, need help fixing
1) Heat damaged area with boiling water to get it to mostly spring back to the original shape. Let dry. 2) Apply thin resin, blot off excess, and clamp lightly between two 1/4 balsa planks (or whatever) to force it into the correct flat shape. Use thin polyethylene for a nonstick film. RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] Surface flutter revisited
But it increases the rotational mass without increasing the surface area, so the increased mass dampens the movement. That's backwards. Mass is not the same as damping. Adding rotational mass (inertia to be more exact) actually reduces the damping forces relative to the inertial forces, so it aggravates flutter. But flutter is vastly more complicated than this simple consideration. Mass balancing suppresses flutter by reducing the inertial coupling between wing bending and aileron deflection. This more than outweighs the unfavorable effects of adding to the aileron's rotational inertia by adding the balancing mass. But in any case, mass balancing seems very impractical in foam-core RC glider wings. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] Max L/D vs Min Sink, and the effects of aircraft weight
Jack Womack writes: The fact is that some full scale ships have a better L/D or glide ratio when fully ballasted. It's a function of the airfoils and configurations used. They also have a higher sinking speed in the ballasted condition, and that's why they have dump valves. They are now optimized for fast cruise, with thermaling a secondary consideration. There have been 300 mile tasks flown where the pilot never circled in a thermal, and achieved better than 100 mph in average speed. Full-scale sailplanes are fast and heavy because that's what the soaring contest rules favor. The mammoth Eta sailplane is the logical endpoint of this development. http://www.leichtwerk.de/eta/en/project_eta/ The unfortunate outcome of this development is that top-level competition soaring seems like a $$$ arms race, and this most advanced sailplane is a very rich man's toy. Perhaps it's not a cooincidence that there has been a steady decline in full-scale soaring participation over the years (according to an article in Soaring or SailplaneGliding I saw a while back). The Eta is accessible only to those who have too much money and too much hangar space. Partly in response to this trend, there is a significant grass roots movement back towards ultralight gliders which give up speed and range for exceptional thermalling ability. The 13.5m span, 145 lb Carbon Dragon is one of the earliest examples of this. See http://www.isd.net/sadkins/index.htm I've seen video footage of Gary Osoba pulling off a save from 200 feet in this glider, in an RC glider-type thermal, called microlift by soaring pilots. A more recent development is the 15m span, 150 lb Light Hawk (about 1/3-1/4 the weight of normal 15m gliders). See photos and test pilot report at http://www.glidersport.net/ OSTIV is on the verge of sanctioning a new Ultralight Sailplane category, if they haven't done so already. http://www.anla.gr/greek/library/ULSailplanes.htm This class should make competition soaring more attractive to a larger number of people. Sorta reminds me of the RES saga. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] Max L/D vs Min Sink, and the effects of aircraft weight
Can you give us an idea of what magnitude of performance change we would be looking at if the weight of an airplane varied say plus or minus 10% (20% total weight change.) Are we talking sink rate improvements of 1/2 of 1%? Or more on the order of 20%? A +10% change in weight will roughly produce changes of... +1% in L/D +4% in minimum sink rate +5% in min-sink speed +5% in best-L/D speed +10% in turn radius at a given bank angle - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] This winter is getting long ..
Winter? Whats that? Oh I remember - that is when it rains a little down here, and gets down to 15C for a month or so. Klaus K Weiss Sydney, Australia So I can assume the skiing sucks too, eh? :-) Mark at least one toy for any kind of weather Drela Boston, MA RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
[RCSE] Blaine Rawdon in the news
Check out what Blaine does in real life: http://www.washtimes.com/world/20021007-34366738.htm RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED] Please note that subscribe and unsubscribe messages must be sent in text only format with MIME turned off.
Re: [RCSE] re: Problems with RD6000
In a nutshell - when his aerial is fully/partially extended his RX doesn't respond at all. As soon as you touch the extended aerial with the bare hand or fully collapse it, the RX responds. My RD6000 once had similar behavior -- very flaky RF output. The problem was a poor connection with the antenna attachment bolt. Tighten it firmly, or better yet first clean off the bolt, bolt platform, and antenna threads with alcohol. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] one more comment about zooming, and then I swear I'll stop
More important, all of the good energy built up in the line and in the airframe will be mostly if not entirely lost in the transition to the steep dive before any reasonable gain could be made by the winch pulling the model down the line. Only if you push the elevator in a rather leisurely way. The nose-over should be as quick as your guts/reflexes allow. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Sliders
Profi users: what do you use the right slider for and how do you set it up to do that. On my 3030 I select one of the three flight modes via the memory switch: 1. launch 2. fly 3. land For each of these flight modes, I use the left (throttle) stick to control camber or flap, not the sliders. For launch mode, the throttle lever sets camber from +20 to -3. So I launch with the left stick all the way back to get +20, and at the start of the zoom dive I push it all the way forward to get -3 for max acceleration and coast-up. That way I don't have to grope for the slider during this stressful maneuver. The left thumb stays where it is there the whole way up. For flying mode, the throttle lever sets camber from +2 to -3. Most of the time it's at one of these limits, so it doesn't interfere with ruddering. When I put the lever in the center to get 0 camber I'm typically ranging in a straight line where ruddering is unnecessary. For landing mode, the throttle lever sets the center flap from +65 to -3, and ailerons from +5 to -3. That way the throttle stick works just like a brake/throttle during approach. I do use the left slider as elevator pitch trim to allow changing trimmed speed without reaching WAY over to the other side of the 3030 pizza box to get to the elevator trimmer. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Spoilers-Elevator
Computer mixing should not be required for properly designed spoilers. This is an artificial requirement. One could just as well argue that elevator mixing should not be required for properly designed flaps, since this just might allow you to use a somewhat simpler radio. It could be done... just move the landing flaps sufficiently outboard. For RES I always use one very big center spoiler, and mix in whatever amount of elevator is needed to minimize the change in pitch trim. The mixing is really a convenience, and could be skipped if manual elevator compensation is used. On landing you're always closely controlling the elevator anyway, so the manual mixing is hardly onerous. There are many advantages to having only one center spoiler: * There is only one spoiler to build and set up, not two. * Only one wing servo is needed. * It's near the CG, so it doesn't add yaw inertia (a handling-killer for r/e). * Outer wings are clean so there's no degradation of roll control on landing. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Bill John's, light weight tail groups
I thought a stab has to pull DOWN, not lift (or if you want: lift down) in order ot counteract the tendency of the wing to roll-over ? The stab on a typical RC glider with a good CG location can lift either up or down, but only slightly. In particular, the stab's Cl is rarely outside the range -0.05 ... +0.05, so the stab airfoil camber, as long as it's small, doesn't matter all that much. Note: Smallish stabs will typically have a very slight download (negative Cl). Largish stabs will typically have of very slight upload (positive Cl). Jerry uses rather large stabs, so his slight positive camber actually makes some aerodynamic sense. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Spyder Foam?? High Strength??
Dose anyone know the difference between CST'S SPYDER (Blue) FOAM 2.3 lb./cu. ft and there High Strength (Blue) Foam 2.3 lb/cu ft. They are both rated at 60 psi compressive strength. Recently I measured both. I cut 1/2 thick, 1 x 1 blocks of each type of foam. The specific Dow sample I had is also called Plazamate. The 1 square faces were carefully fine-sanded perfectly flat. I weighted them on a precision indoor scale. The densities were identical at 2.30 lb/ft^3. Each block was placed on a bathroom scale, sandwiched between two small metal plates. A drill press was then lowered onto the top plate to slowly crush the foam, while noting the indicated load on the scale. The load leveled off when the foam started to flow in axial compression. The max loads: Spyder Foam : 85 lbs (85 psi) Dow 60 Hi-Load: 75 lbs (75 psi) The loads at the onset of deformation surely were slightly lower, but this onset was impossible to detect reliably. In another test I placed two blocks face-to-face, and crushed them together. The Dow started to flow before the Spyder did. This confirms my subjective observations that Spyder has slightly more dent resistance than the Dow Hi-load. This may or may not be worth Spyder's higher cost. Both foams are vastly more dent-resistant than the lighter Dow grades. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Monocote v.s. Ultracote
The Allegro Lite is another case in point. It is designed for Monokote on the center panel for torsional stiffness. This design choice was made before I had any real torsional stiffness numbers. After I built the center panel I measured the torsional stiffness before and after covering with transparent Monokote. The D-tube was about 5x stiffer per weight than the covering. The conclusion is that on any D-tube wing, the clearly most efficient choice is to use the lightest covering, and put the weight saved into denser D-tube balsa. More bang for the buck ... er ... gram. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Somebody has to know the answer to this question
For instance if you stretch a 1 foot section to 3 feet will it have the same tension as a 100 foot section stretched to 300 feet? Yes, of course. The % stretch is the same in each case. If you stretch a piece of highstart tubing we will assume latex to a certain percentage of its initial length does the resistance have any relationship to the initial length. Yes, but this is more complicated. Rubber does not act like an ideal linear spring they teach about in Physics 101. For a spring the force/stretch relation is a straight line: twice the stretch = twice the force. Rubber used for rubber FF planes has a distinctively nonlinear S-shaped force/stretch relation. As you gradually stretch the rubber the force rises quickly at first, then more slowly, then quickly again as the rubber reaches the breaking point. Typically the rubber can be stretched 6-7 times its rest length before breaking. Old FF rubber like Pirelli was basically latex, so I expect histart rubber to have similar behavior. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] How steep of an approach with RES?
I try to approach with partial spoiler and some excess speed. This way I can either steepen or stretch the approach with more or less spoiler as needed. How steep this type of approach is depends on how powerful your spoiler is. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] pitch stability question
Because the dive test does NOT indicate anything about pitch stability. It ONLY indicates elevator trim/decalage angle effect during increasing airspeed. The second sentence is in fact one definition or at least one direct consequence of pitch stability. A pitch-stable aircraft must see a decrease in trimmed airspeed with up-elevator and vice versa. A pitch-unstable aircraft will do the opposite (if it's artificially sustained in trimmed flight via a gyro or whatever). To answer the original question... Any given real airplane does not have a fixed stability margin at all airspeeds. Normally the stability margin decreases with airspeed. There are typically three reasons for this: 1) Cm of low Reynolds number airfoils is not really constant. It typically becomes more negative (nose down) at small AoA. This looks and smells just like a reduced stability margin. 2) If the CG is below the wing's center of area, like in any poly glider, the CG will effectively move aft at lower AoA and reduce pitch stability. 3) Flexibility of the wing, tailboom, tail surfaces, and linkages tends to have the same effect as imparting down-elevator trim at increasing airspeeds (reducing AoA). Again, this looks and smells like a reduced stability margin. If the glider has slight positive stability in a slow glide, it can easily go unstable in a fast dive, resulting in a tuck-in. The dive test, if done correctly, is useful in that it represents a worst-case scenario where all three effects above gang up. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] wing tips: a peculiar experience
Perpendicular to the LOCAL LE ?? Wow! I had no idea. Personally I've always favored squared-off tips. Should I re-think this? The local LE shape is significant as long as the local sweep is not excessive, so the velocity component normal to the leading edge is still significant. The square-tip edge has a zero normal velocity component, so its shape is less important. One reason I like rounded tip LEs, and try to get a half-decent LE shape all around the tip is that a glider will sometimes sideslip. A RES ship will definitely sideslip, possibly quite a lot. A swept LE will then see an almost perpendicular flow, and hence its shape will matter in keeping the profile drag low in the slip maneuver. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] What I found out about tooling costs.
Just the aluminum @ 2.20/lb for a set of 120 wing molds would be around $2000 I assume this is for super-rigid molds cut from 1.5 plate. There's a better alternative. The molds can be intentionally floppy, and then held against a flat surface when closing the mold during the final bond. Slight bending or twisting of the mold halves when molding each surface doesn't matter much. Such a mold could be machined out of 1/2 plate for the bottom, and 3/4 plate for the top. A 124 x 12 mold then works out to about 180 lb, or $400 for the aluminum. Plus you don't need a crane to move the thing around, since each mold half will be less than 40lb after machining. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] (spoiler size)
Is there an optimum spoiler size - say percent of chord? Working on a low aspect ratio 2M with a root chord of 15.375. Center panel (where the spoiler will be) is flat and 20.5 span. I'm not sure what you mean by optimum. The bigger the spoiler and the farther forward it is, the more sink rate you get. As you probably know, I favor putting the spoiler very far aft where it can't hurt the airfoil significantly even with an imperfect fit. Such a spoiler should be very big to compensate for loss of effectiveness. For your big wing area, I'd make the spoiler 2.0 to 2.5 wide, and the full 20.5 panel span for simplicity. Putting the spoiler's TE at 65% chord will place the spoiler over one of the AG3x airfoil's flat facets, so the spoiler wants to be flat and hence will be easier to make. Make sure you have at least 10 deg of additional up-elevator travel to compensate for the spoiler's upwash via mixing. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] (spoiler size)
Currently I have a 2 spoiler drawn in with the leading edge at 45% which I believe is the beginning of the 1st flat facet. Will that be ok? I'll go ahead and lengthen it to 2.5. OK. Your comment about the elevator is interesting. Although I would like to locate it just off the boom per the Allegro Lite, I had assumed it best to locate the elevator higher up, perhaps a T-tail, to get it out of the turbulence caused by the spoiler. Would you care to comment? The stab sees two things from the spoiler... turbulence (maybe), and upwash. The turbulence is fairly compact vertically -- some fraction of the wing chord. But I've never seen any adverse effect of this turbulence on the controllability, at least on the Allegro-Lite. I don't think it's a problem in general. The upwash extends vertically over a greater distance -- some fraction of the spoiler's span. So mounting the stab higher up on a T-tail is almost futile. You won't reduce the upwash it sees by all that much. I would do the mechanically simplest stab mount, just above the boom, and use spoiler-elevator mixing to deal with it. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
[RCSE] re: Optimum Spoiler Size
Blaine writes: You can see from this that you want to locate the inboard edge of the spoilers someplace between the centerline and the stabilzer tips in order to balance the two effects described above. I think you're unduly worried about the spoiler's effect on trim. The spoiler-elevator mixing works quite well, and it's adjustable! The mechanical simplicity of having one center spoiler wins here. A second factor in this is the pitch stability of the airplane. I don't think that's the case. I fly my Allegro-Lite close to neutral stability -- 10% margin at low speed, less than 5% margin at high speed. When the spoiler is deployed the glider becomes more stable in pitch, not less -- it solidly locks on to a particular pitch trim, and the elevator's influence on pitch trim is diminished. These are sure signs of an increased stability margin. This increased stability also makes the spoiler-elevator mixing gain not so critical. I'm not totally sure why the spoiler increases pitch stability. One possibility is that the spoiler decreases the wing's dCl/dalpha because of the massive separation. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] RDS drag reduction, speed increase.
This did stiffen the surfaces to an acceptible level, but there was still some flex right where the rod bends 45 or 90 degrees. An all carbon rod would be ideal! Uh.. actually, a carbon rod would be horrible. Unidirectional carbon has a puny shear modulus compared to steel, even aluminum. The stiffest solution would be a steel tube between the servo and the hinge-line bearing -- the biggest one you can fit. SmallParts has thinwall hypo tubing up to 0.238 diameter. Steel rods could be soldered into the tube at each end to use the standard hardware, with a suitable steel or brass sleeve for a concentric fit if necessary. Someone else mentioned using hardened RDS torsion rods. This is not a solution, since hardening has little or no effect on the stiffness of a metal. It only increases its strength. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
RE: [RCSE] FLIGHT 587 and R/C MODELS
These articles are indeed very interesting. The more I read, the more it does not make sense. Could the tail coming off, also cause the engines to fall off? And what about reports that fire was seen coming out of an engine. Here's one feasible scenario... Vertical tail falls off. Airplane becomes directionally unstable, reaches a very large yaw rate. Precession (gyroscopic) moments on the turbine engines break the pylons. Fuel spewing from broken fuel lines ignites. Jetliner engines are specifically designed to break off cleanly when loads at fuse points in the pylons exceed specific levels. If the engine seizes up or whatever, you want the whole pod to come off cleanly rather than take part of the wing with it. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] squishy glider
My electric glider doesn't penetrate well (omega 2M). The CG is just slightly ahead of what the manual says though I have tried CG a bit back and a bit up and the prob is the same. Penetration capability has virtually nothing to do with the CG location. It depends only on the airfoil, the wing loading, and how clean the glider is. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] DS discussion
I have been trying to work through the concepts and simple physics of their explinations to gain a grasp of how DS works. Well, try as I may I can not make the theories work. Perhaps you can help me out. This sketch shows how the airspeed changes during one DS orbit: http://www.charlesriverrc.org/articles/flying/markdrela_ds.pdf Here's some accompanying text: http://www.charlesriverrc.org/articles/flying/markdrela_ds.htm - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Omega 1.5M
Anyone out there have any suggestions for someone just acquiring one of these, be nice! I've flown one and seen one break. The pros: * Nice slippery HLG * Qood aileron response The cons: * Work-intensive to fly in rough air. Could use a few degrees of dihedral. * Tail is too small and elevators are too narrow. Elevator power is especially anemic in tight thermalling maneuvers. Add 3/8-1/2 TE extensions to the elevators. * Fuse is quite fragile around hatch. Add carbon rod or 3/16 square basswood siderails to the inside of the fuse, extending 1 from the nose to wing spar location (taper ends). If using basswood, can use these rails to mount servos on crosspieces. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
[RCSE] Channels 6,7,8 from 555 RX
The following page shows how to convert the channel 5 port into a channel 6 port on the old Hitec 535 RX: http://www.northlandflyer.com/Tips__n__Techniques/slickraft2_5a/body_slickraft2_5a.htm I just did the same modification to a 555 RX and it works! The modification allows using the 555 on a 4ch HLG, with a radio which wants left aileron to be on channel 6 (like the RD-6000 for instance). The page also shows which chip pins are channels 7,8 if you want those. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Allegro Two-Meter
I am very interested in building an Allegro two-meter for next year. What would be the easiest way to assemble the materials needed to build this plane? I assume you're talking about the built-up Allegro-Lite. You could try to talk one of the Allegro-Lite Yahoogroup members into selling you his kit. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
[RCSE] Web shear loads (was: spirit elite)
The forces on the balsa shear web are not acting on the end-grain of the balsa as you suggest. They are shear forces that are acting through the neutral axis across the grain. Actually, you're both sorta 50% correct. Shear stress in the web material is the same in any direction. Counterintuitive perhaps, but it's true. So for withstanding shear stress, endgrain balsa is as good as lengthwise-grain balsa. The great advantage of endgrain balsa is that it is vastly better at restraining the sparcaps from buckling. The endgrain web is also necessary if the spar is wrapped with a +/-45 deg cloth, which puts a large compressive load on the web if the cloth's compression fibers buckle. To belabor the point... A shear web can also be considered to play the same role as the diagonal members in a truss beam -- it prevents parallelogramming of any given chunk of the beam. In a truss, the parallelogramming is restrained by the diagonal members, which are either in compression or tension depending on whether they are angled at +45 or -45 degrees. In a spar, the parallelogramming is restrained by filling the spar with a homogeneous web material (e.g. balsa), which is then in compression along one diagonal, and in equal tension along the other diagonal. This equal and opposite tension/compression along the diagonals is what's commonly called shear. This interpretation of shear loads suggests that an ideal shear web material is balsa plywood oriented at +/-45 degrees, so that the diagonal tension/compression loads associated with the shear are along the grain, just like in a truss. Such a web is indeed about 2.5x stronger in shear than an endgrain web of the same weight (which I've verified by busting spar samples). However, if such a web is pierced by ribs like on the Allegro-Lite, all advantage of the diagonal grain is lost, since the ribs will fail in shear well before the web. A plain endgrain web is the easy and appropriate solution in that case. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Wing rod
I am planning a 7/32 rod, and about 3 inch insetion length in each panel. A 3 rod is OK provided the joiner tube is solidly embedded in the web between the sparcaps. The shear load taken by this web (and the cap/web bonds!) is equal to the sparcap load, which is about 300 lb max for a typical 0.5 x 0.125 spruce cap. A lite-ply web will suffice. If you add a 0.5 x 0.014 carbon strip to it, the max load will be more like 1500 lb, in which case using regular ply will be more prudent. The web's shear load drops dramatically outside the joiner rod. The drop factor is approximately S_outside/S_joiner = 4 h / b h = spacing between sparcaps b = span For a 100 wing with a 0.75 sparcap spacing, the factor is only 4*0.75/100 = 0.03 So if the joiner web sees a shear load of, say, 500 lb (sparcap load), the web just outboard of the joiner sees a shear of only 500*0.03 = 15 lb. There's no reason to continue the ply web outboard of the joiner tube. Balsa will certainly suffice. I'm running out of thickness between the spar caps with the foam cores and I don't want to cut into the caps if I can help it. You can safely taper the top sparcap over most of the length of the joiner, down to near zero thickness at the root rib. This will give the cap's thickness worth of additional depth to work in. This assumes that a) The tapered sparcap is solidly bonded to the ply web which is then solidly bonded to the joiner tube all along its length. b) The joiner tube is strong enough to resist bursting at its end. A KS 0.015-wall brass tube will be quite adequate. A soft 0.015-wall aluminum tube maybe not. Wrap the end with Kevlar if in doubt. Doing both a) and b) is sound practice in any case. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Joiner sizing
When I plug these numbers and assume a 200lb loading and using piano wire (affordable) I came up with a 0.32 piano wire joiner. a = 0.45 (gross interpolation of joiner at 0.33 span) F = 200 (load number frequently tossed about on RCSE) b = 100 sigma = 18 (piano wire) I seriously doubt you will see 200 lb pull on your winch. (I wanted to work back to the load given a 0.25 radius wire, but I've forgotten how to rewrite an equation that is raised to a fractional exponent.) 200 lb * (0.25/0.32)^3 = 95 lbs Probably more realistic. You might still put a small bend in the 0.25 wire on a very hard launch, but I doubt you'll fold it completely. Further design questions - If I were to make the polyhedral by putting dihedral at each joint (except the root) would it be hard to bend the piano wire, or would it be better to put the dihedral outside the joiner area? Why not just use a straight joiner wire and embed the tubes at an angle in each panel? The 9/32 joiner tube should fit angled like this in your 9 chord. A straight rod is easy to check for a permanent set and easy to replace. What is the rule-of-thumb for how long a joiner should be? Depends on the quality of the shear bond between the joiner tube and the sparcaps. The better the bond, the shorter the rod can be. Embed the tube in plywood filler between the caps, and wrap 2-3 layers of 2 oz glass at 45 degrees around the whole sparcap/ply/tube assembly. A short 3 tube will be OK with this kind of joint (6 rod). If you're not using pre-cured sparcaps, the structural sizing of the joiner rod attachment system gets very uncertain. Can't give advice in this case. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] C/F Boom Molding Question
Why not just use a mandrel and CF braided tubing? That should give you 45-degree fiber orientation, a very nice outer surface, and no need for mold building or bagging. Or am I missing some important fact? This is ideal for a torque tube, but not a tailboom. The 45 degree tube will have only 15% of the bending stiffness and even less bending strangth as a unidirectional tube of the same weight. One compromise solution is to use an oversize braided tubing like the 1.5 circumference stuff from CST. When stretched out to neck it down to the 1 mandrel, it will have roughly 30 degree fibers. This will have 40% of the bending stiffness of the uni tube. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Apogee (Sort of)
what percentage of MAC do you recommend for the CG on the AG03 foil? The Apogee plans give the location, but it looks like percent of root chord and not MAC. My 36 Apogee has the CG at 29% of MAC, which is at 45% of center chord. This results in a very slight tail download in cruise (CLt = -0.05 or so). For zero tail load, the AG03 wants the CG around 31% of MAC, but for the Apogee this would require a larger horizontal tail volume to still have adequate pitch stability. Slightly downloaded tails in general can be made smaller, lighter, and usually less draggy. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
[RCSE] Bending music wire
Is it necessary to heat music wire to get a good 90 bend and if so what's the best way to do it? Don't do it! Heating music wire sufficiently to soften it will also anneal it and destroy its strength. Once annealed there's no way to get the strength back other than cold working (i.e. by destroying the part you just made). Heat treatment won't work -- wrong kind of steel. This is what I do to make clean bends in hefty music wire (between 3/32 and 5/32): 1) First decide on which side of the bend you want to be as straight as possible. On RDS this will be the part of the wire which sits in the pocket. 2) Clamp that part in a vise, tightening just short of forming flat spots on the wire. Vise-Grips or very hefty pliers may also work for the smaller wire sizes. 3) Bend the protruding part over, using a large piece of metal. Apply the bending force as close to the vise jaws as possible to keep the bend localized. Wear eye protection in case things go SPROING! accidentally. Hammering on the metal piece may be necessary for really heavy wire. 4) Bend about 5-10 deg PAST the intended angle. Then bend back to the intended angle. Note 1: The reason for going too far and then returning is to relieve the huge residual stresses (about 1/2 of yield stress!) in the steel which would otherwise remain. The bend will be much stronger then. It's easy to observe that these stresses are there: It takes a lot of force to get from 90 to 100 degrees bend the first time, but starting the backward bend from 100 to 90 takes much less force. The residual stresses at the surface are helping you when you go back. Note 2: If the wire cracks at the vise edge, put a softer-material block between the wire and one vise jaw face, and bend against the edge of the softer block to get a larger radius. I've used copper, soft aluminum, even maple. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: Fw: [RCSE] efficent wings for pylon turns
There is still something going on with the wing design of the JW and my flying wing. The wing planform and not so much the airfoil changes how the wing performs. I use the same airfoil on both my DS Raptor (straight leading edge) and my Combat Raptor (swept back wing) and the CG, control throws and flight performance is much different. The DS Raptor has a CG that is at 13% of chord rather than 28%. The DS Raptor needs less than half the control throws that the Combat Raptor needs. Finally the DS Raptor corners like it is on rails. The Combat Raptor will not fly if it is set up like the DS Raptor so the planform has an affect. What I am trying to figure out is why. If I can figure out the Lift Roll program I will set up both wings and see what the difference is. Are you accounting for the zero-lift angle changes from the surface deflection? The Zagi's inversely-tapered elevons will produce more loss of lift outboard. Also, there will be a pitch rate effect on the lift distribution. On a swept wing, different spanwise locations see a different freestream angle if there's a substantial pitch rate present. This in effect modifies the effective twist angle distribution, or the effective zero-lift angle distribution, whichever way you want to look at it. In practice, this pitch rate effect cancels some of the up-elevon trim power near the tips of a swept wing, so more up-elevon is required to get the same effect. That's probably what you're seeing in flight. I'm not sure if Liftroll takes this pitch rate effect into account. Up-elevon significantly reduces the CLmax of the section, so the larger deflections required by the swept wing may induce partial stall, or at least a large profile drag rise. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] More Euro Cruciforms
The ability to share the elevator servo load between two lessor torque servos, rather then having one giant servo of the elevator and a smaller one for the rudder. You got that backwards. A full flying elevator hinged at its mean quarter-chord puts negligible airloads on its servo. The rudder servo will see larger airloads. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] ? paint for a HLG tip visibility
FWIW, I find that using two bright contrasting colors at the tips really helps to see the HLG's orientation when far away and/or with a cluttered background. I use fluorescent red or magenta on the left tip, and fluorescent blue or green on the right tip, both top and bottom. Even when I can't make out the outlines, I can still see the tips as two colored dots. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Rx Recomendations
I am setting up an e-powered model for the Elexaco event and the trusty Hitec 555 is begining to look a little heavy. The Hitec Feather I just bought seems light enough but as I expect the model to spec out I am worried about its range BG I'd go with the dual-conversion FMA Extreme. It's 9 grams if you remove the shrink wrap, foam pad, and antenna (use thinner wire). It's also reasonably priced: About $60 for RX + crystal. Works on 3 cells. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
Re: [RCSE] Re: F3J model glide ratios?
Wouldn't the ground effect have a large affect on your measurments in the field? No. Ground effect is significant only when the altitude is less than about half-span. - Mark RCSE-List facilities provided by Model Airplane News. Send subscribe and unsubscribe requests to [EMAIL PROTECTED]
[RCSE] Spiral stability
A number of people have asked for clarifications or references on the vertical tail vs spiral stability issue. There aren't any that I'm aware of. Unfortunately it's a very complex subject. Here's a writeup I did which expands only on the basic effects. It's still kinda complicated, but I don't see any way to simplify it further. - Mark - Whether an airplane is spirally stable or unstable depends on a delicate balance of forces. After a yaw or bank upset, some of these forces oppose the bank (resist spiralling in), while some act in the direction of the bank (promote spiralling in). Although the forces are varied and numerous, on a typical glider there are four dominant ones, all comparable in importance: Resist spiraling in: roll moment due to yaw angle Cl_b (dihedral effect) yaw moment due to yaw rate Cn_r (vertical tail's yaw damping) Promote spiraling in: roll moment due to yaw rate Cl_r (faster-moving tip has more lift) yaw moment due to yaw angle Cn_b (vertical tail's yaw stability ) The airplane is spirally stable if the first two forces are greater than the second two forces, all acting together. The symbols are standard aero jargon for the forces. So we have... Spirally stable if: Cl_b x Cn_rCl_r x Cn_b (*) After these four forces are estimated using the airplane's geometry, the spiral stability criterion (*) above turns out to be approximately the same as Blaine Rawdon's criterion after a bit of algebra: EDA x (tail_length/span)5 CL(**) So why doesn't the vertical tail area A_vert come into the picture? Because it affects both the yaw stability and the yaw damping equally: yaw damping Cn_r ~ A_vert x tail_length^2 yaw stability Cn_b ~ A_vert x tail_length These are opposing forces which appear on the left and right sides of criterion (*), so A_vert cannot influence the direction of the inequality. Only the extra tail_length factor in the yaw damping force remains, and can be seen in Blaine's form (**). The other quantities in Blaine's form come from estimation of the other forces, e.g. dihedral effect Cl_b ~ EDA x span^2 roll due to yaw rate Cl_r ~ CL x span^3 One can also influence the spiral stability criterion "artificially" by using a rate gyro. This can be done in one of two ways: 1) Yaw gyro driving the rudder.Increases Cn_r (yaw damping) 2) Yaw gyro driving the ailerons. Decreases Cl_r (roll due to yaw rate) In each case, the gyro feedback sign must be set correctly to move the Cn_r or Cl_r number in the right direction. Otherwise spiral stability will get worse, not better. Method 1) works because it increases yaw damping Cn_r, but DOES NOT increase yaw stability Cn_b. Just adding vertical tail area increases both, which then cancel each other out. Blaine Rawdon and Bill Watson have demonstrated the effectiveness of method 2) in artificially obtaining spiral stability on a full-house Open Class glider: http://members.home.net/evdesign/pages/technical_articles.html RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
Re: [RCSE] Spiral stability
Tony Estep writes: This sez that if your tail moment is 40" and span is 120" and you fly at CL = 1, you need 15 deg of dihedral to be spirally stable. This is indeed what Blaine's Crossbow design has, but lots of RES designs including the Allegro have quite a bit less. How much spiral stability (in terms of that constant) is necessary in a practical RES design? Your CL = 1 assumption is excessive. Typical thermalling for most airfoils will be at CL=0.8 or less, since the inside tip sees a higher local cl. The overall average CL must be significantly reduced below the straight-line stall CL. The Allegro-Lite has a 30" vertical tail arm, a 78.4" span, and thinner than usual airfoils with a max circling CL = 0.75 or so. Therefore its minimum EDA needed for spiral stability is EDA_min = 5 * 0.75 * 78.4" / 30" = 9.8 degrees. The actual EDA is 12.1 degrees, which is a comfy 23% margin. - Mark RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
Re: [RCSE] Victory C-RES rudder
Does an increase in rudder area have any effect on the cg? I was thinking it may have an effect in a bank... I assume you mean whether it affects the aftmost-allowable cg. It does not. - Mark RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
Re: [RCSE] airfoil question
These look different than anything I've ever seen on any r/c sailplane - or for that matter, anything else that has wings! What's with that sharp edge of the front? And what's the deal with that flat surface? The flat part is on the top of one airfoil and it's on the bottom of the other - not that there's anything wrong with that! The sharp LE acts as a turbulator. At min sink trim, the flow ideally stagnates just under the LE, and curls around the sharp edge as it goes up and over the top. At high speed in the launch, the flow ideally stagnates right on the sharp edge which "turns off" the turbulating action. The geometry has to be just right for both flight conditions to work well. The approach taken by most serious IHLG fliers is systematic trial and error via prolific building. These airfoils are extremely difficult to design the "modern" way via computation, since the turbulating action is impossible to predict reliably, and it depends on lot on the construction details. Note that their Reynolds numbers are much lower than on RCHLG's, so don't get any funny ideas! :-) The flat facet on the bottom surface serves no purpose other than provide some "LE upsweep" which we know is essential for low drag at launch. I think he'd be a bit better off if the facet was rounded off and blended smoothly with the flat bottom surface. Ron Wittman's record-holding IHLG "Supersweep" had such a rounded upsweep. FWIW, he got 90 seconds with a conventional throw. 22 in span, 67 in^2 area, about 24g (1.8 oz/ft^2). - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
RE: [RCSE] Glider turning dynamics ideas
Jim Cubbage wrote: I am curious, most of the methods mentioned seem to me, and I am really no expert, to create a left momentum. What I mean by that is that it seems to me that the tendency for the planes will be to roll and continue to roll to the left, probably until the plane is in a knife edge position. That means, you will have to have some means counter the left tendency with a right tendency, otherwise your plane would continue to fall to the left. The tendency for a plane to steepen its bank simply means it is spirally unstable. All aileron gliders I'm aware of are spirally unstable to some degree, and hence must be "flown" constantly to keep them in a circle. The most common way to eliminate spiral instability is to use sufficient dihedral -- at least 8-10 degrees or so is usually needed (the exact threshold number depends on the tail length, rudder area, and a few other things). Many aileron gliders have some dihedral to make them LESS spirally unstable. Their bank will still steepen, but not as fast. All free-flight models and most properly set up RC poly ships are spirally stable. They will hold their airspeed and turn radius without pilot input. One way to eliminate spiral instability _without_ dihedral is to drive the ailerons though a rate gyro which senses yaw and roll rate. Blaine Rawdon and Bill Watson did some nice work on this: http://www.rc-soar.com/tech/spiral.htm Blaine's Plane Geometry spreadsheet and document also delves into spiral stability and dihedral a lot more. - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
RE: [RCSE] Aramid/CF material
Matt Gewain writes: Aramid fiber has a much lower density than carbon or glass fiber so when you replace some of the carbon with Aramid the weight of the model can be reduced. That's a good point. I was being perhaps overly harsh on the hybrid, since I was only considering cases where the material is pushed close to failure. In many model applications stiffness is a more relevant consideration, or there is a minimum-gauge constraint. The lower density makes the hybrid much more attractive in these cases. -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
Fw: [RCSE] Aramid/CF material (slightly OT)
However within a year of getting there new Kevlar contraptions some people where noticing a certain amount of sag or flex with their boats sitting on trestles or trailers or when lots of power was put into the riggers, the reason was that after 6 months in UV Aramid/Kevlar cloths loose 50% of their rigidity and all the load was being taken by the carbon which was now only half the amount as usual. This years boat models are all carbon including the seats, riggers and oars! No flex, higher efficiency. Regards Daniel Armstrong Chester UK Kevlar and carbon have similar tension properties. However, Kevlar is relatively crummy in compression. When a 50/50 mixed material is put under high compression, the Kevlar in it may fail to mush, even though the carbon still holds it in one piece. Maybe that's what was observed in the boats. One could oversize the part so that the Kevlar in it doesn't fail, but then the carbon is stressed to only a small fraction of its compression load capability. The bottom line is that Kevlar and carbon by themselves are great when used appropriately, but mixing them together in a weave doesn't seem to make much sense. In a pure tension situation their strain-to-failure is comparable, so they can share the load efficiently. America's Cup sails contain a carbon/kevlar mix. This works well since a sail is always under tension. Carbon and Kevlar can be effectively used together in a compression structure if their fiber orientations are different, e.g. carbon spar caps for bending with +/-45 Kevlar cloth skin for torsion. The carbon cap will fail in compression first, since the bias Kevlar can have a larger bending deformation without failure. This would not be true for a 0/90 Kevlar skin. -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] spar structures questions
Jon Stone writes (Re: Allegro spar at CRRC):
What drives the value of sigma (design cap stress)? The reason I ask, is
that this is the only "variable" driving the equation for tip-deflection.
And I do not know what parameters define the value of sigma.
sigma is simply the stress the carbon fiber can withstand.
The appropriate value depends on the quality of the layup
and whether the cap can buckle or not. The numbers I used
were large for wet layup carbon, but still somewhat conservative
for prepreg carbon which cannot buckle. It is difficult to
predict the allowable sigma for structures which MIGHT buckle.
Specimen testing to failure is usually required.
Furthermore, what parameters define your values for sheer stress (tau) and
core stress (sig_C). I would like to use heavier balsa (say 8 lb) for the
shear web. Is tau the same as Shear Modulus?
The web's shear modulus (usual symbol is G) is not an issue in most
spars.
What matters is the max shear stress tau ("parallelogramming" stress)
and the compressive stress sig_C (vertical stress). These two
stresses depend on how the spar is set up. The max tau and max sigma
of balsa scale pretty linearly with density.
In the Allegro spar, the shear is taken up entirely by the half of the
skin fibers which are under tension, and the associated vertical
compression
is taken up almost entirely by the endgrain balsa core.
So the core stresses are as given in the article:
tau_C = 0
sig_C = tau As / h w
If there is no skin, then
tau_C = S / h w
sig_C = 0
This is more difficult, since the max tau of balsa is only
about 1/5 of its max sigma, and the spar may also be subject
to peel failure at the core/cap bond. This is why I used a spar skin.
But I have used this second system in my follow-up Allegro-Lite
glider to appear at CRRC soon. The core takes all the shear load.
Peel is prevented by a Kevlar tow wrap. Not as weight-efficient,
but it's very compatible with built-up balsa/Monokote structure.
One gent I've talked to used .086" x 1/2" wide carbon spar caps for his
3-meter glider. The thickness tapered to .040. He did not use any shear
web at all. He stated that if the caps were strong (thick) enough, that no
appreciable shear stress was transmitted to the foam core.
Not true. Whatever is between the caps takes the same total shear load
S.
What the caps look like is immaterial. The only way to isolate
the core from shear stress is to use wing skins with huge shear
stiffness
all around. Picture the Allegro spar skin blown up chordwise
into an airfoil shape. But the shear has to go all the way around the
airfoil instead of directly though the thickness, so it's not an
efficient use of shear material, though.
The size of the caps does affect their buckling tendency, which
is a separate issue -- you can have either shear failure of
the core or spar shear skin, or buckling failure of the caps.
The Allegro spar is aimed at dealing with both of these efficiently.
If you don't use a separate web, then you need
1) sufficiently thick sparcaps so they don't buckle with only
the soft foam restraining them. This favors narrow and thick sparcaps
(for a given cap X-sectional area).
2) sufficient other "shear stuff" between the caps to take the total S.
Normally the bit of foam between the caps cannot take the
required tau_C value as computed above. I think blue foam
has a max tau of 20 psi or so. However, the total S load
can be spread over a wider chordwise extent by using +/-45 skins
which are sufficiently stiff in shear (G * skin_thickness is adequate).
Sort of a compromise between the core-only and skin-only shear transfer.
This may require Kevlar or carbon skin, I don't know for sure.
And if the skin is required to transfer the shear from the
cap to the core, it too can buckle with 45 deg wrinkles.
It gets complicated.
Tom Kiesling's private Mantis planes use massive 1" x 0.125"
caps in the center, with 45 deg Kevlar skins and no web
(that's what I gathered from chatting --- corrections welcome).
So it can be made to work, but it's very tough to predict
the max loads without testing, and peel failure from defects
is always a concern. One reason I went to the Allegro
spar system is that it's sorta idiot-proof. Predictable
by analysis, tolerant of defects, and it cannot peel.
--
- Mark Drela
RCSE-List facilities provided by Model Airplane News. Send "subscribe" and
"unsubscribe" requests to [EMAIL PROTECTED]
RE: [RCSE] people with molded experience (construction) contact me
Are the spar-sizing calculations there the final, corrected ones? Yes. The actual web page is below. Just modify the input numbers as appropriate for your glider and recalculate as indicated. http://www.charlesriverrc.org/articles/allegro2m/spar_sizing.txt -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Fuselage Layups
I'm pretty interested in making a couple of HLG's over the winter, and I had a couple of questions about the Apogee in specific: 1) What is a good source for stiff tailbooms? (CST, ACP?) 32.5" long Avia "Skinny Ultralight" 7.6g before trimming www.intothewind.com catalog #4500 $7.70 You might find a local source to avoid the shipping charge. The ACP and CST tailbooms are grossly oversized. 2) Pushrods? 0.012" piano wire. www.smallparts.com sells it straightened, and also sells small-diameter thinwall Teflon tubing for housings. You can also use 0.015" KS wire, but that is gross overkill and it will have more friction which is the major battery juice sink. 3) What layup schedule and materials would suit the fuselage? I've never worked with Kevlar before, but am interested in trying it out. Three layers of 1.8oz Kevlar in the front, two in the back. It must be bagged into the mold since it is too stiff to stay put. Another way to do it is to make a silicone cast in each half and use that to press the Kevlar into the mold. It's a pain to mold but worth it IMO. -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
Re: [RCSE] How should an HLG launch?
Dick Barker writes: The most efficient climb angle is almost straight up. Seventy five or so degrees is about right as anything greater will drift behind you on the way up and make it hard to get a good transition at the top. The steeper it goes, the less drag is induced by the coef of lift. This might be misinterpreted. I think you meant to say that the _launch_ should be almost level, and the _climbout_ after that very steep. This will extract maximum additional energy from the wind. A near-vertical launch does not benefit from the wind at all. -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] V tails
This is probably old hat, but what special considerations are there when deciding the angle between the surfaces of a v-tail? The first thing you have to do is to figure out the appropriate horizontal and vertical tail areas, just like on any other airplane. This is done by picking appropriate tail volumes. Once this conventional tail is designed, the equivalent V-tail geometry then follows. See the following article for the conversion formulas. http://www.charlesriverrc.org/articles/design/markdrela_vtailsizing.htm The DJ Aerotech site also discusses this stuff. Or you can fire up Blaine Rawdon's Plane Geometry spreadsheet. -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
Re: [RCSE] Tail sections vs. handling
Derek Boyer writes: Thanks John. I stumbled across the CRCC website yesterday and found Drela's new HLG. I'm always looking for cool HLG airfoils and will give them a look. Never did think about a turbulator as you suggested. That might even be easier than sanding kevlar, carbon, glass and foam :-) Turbulating HLG tails is almost futile. The Reynolds number is too low to permit significant turbulent flow, so if there's laminar separation it is very hard to quash. A much better approach is to rely on *attached* fully laminar flow, so that tripping is a non-issue. Attached laminar flow can be obtained with a thin airfoil with the max thickness point far forward, as you discovered. Such airfoils always give well behaved control response no matter how low the Reynolds number is. The HT05 airfoil on the CRRC site takes this one step further. It has attached laminar flow even with a modest flap deflection, so that crisp control is maintained when holding fixed up elevator in a thermal circle. The trick that does this is the slight dimple in the HT05 surface at the hinge line. This is "not your father's simple slab tail section", but it's worth the extra effort to build IMO. - Mark -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Proper HLG throwing technique
Use a Javelin throw, and enough "up" preset on a spring-loaded toggle switch to rotate HARD from the near-level or slightly down angle you release the plane at, to vertical as instantly as possible. The plane should "park" nearly over your head, NOT out in front (much). You want to pull up fast, but not "as instantly as possible". Two reasons: 1) The induced drag energy loss in the pullup goes up with the g's pulled, so there is an optimum turn radius. For a typical HLG a 20-25 ft pullup radius is optimum. 2) Energy is extracted from the wind when the path curves upward in the presence of headwind. If there is a wind gradient (and there usually is), a larger radius pullup will see more path curvature higher up where the headwind is stronger. If the glider enters the larger wind while already going vertical, the additional wind doesn't do any good. I fly at Davis Field in Sudbury MA which often has a noticable wind shear at about 30 ft up due to surrounding dense trees. The best trajectory seems to be a rather sedate pullup so that the glider punches through the shear layer at about a 45 deg angle rather than vertically. -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Transitional Airfoils (again)
what I need is info on how to modify (by increasing camber and washout) a particular airfoil across the span such that tip stalling is minimized with minimal impact on performance, and what airfoils do or do not lend themselves to such manipulation. Specifically, I'd like to do this with the RG-15. (In fact I have already done it on two wings, both of which seem to fly well, but I was more or less guessing at the amount to increase the camber and would like to know if there is a more "scientific" method?) A very safe way to do this is to interpolate two baseline airfoils in various proportions. The polar of the interpolated airfoil will be quite close to what you get by interpolating the two baseline polars themselves. This seems to work with any two sailplane-type airfoils. At least that's what calculations indicate. Note: A foam cutter with two different templates at the ends does this interpolation for you. On the other hand, simply scaling the camber and/or thickness of an existing airfoil is more hazardous. Scaling the camber from 2% to 4% may increase CLmax a little or a lot, depending onthe % thickness and Reynolds number. It may even decrease CLmax in some cases! One would have to run tests or calculations on the modified airfoil to be certain how it performs. There's no guaranteed simple rule to predict it. -- - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Decalage (was: Stab size)
It seems to me that decalage is greatly oversold. Changing decalage simply biases the elevator position, and can be entirely compensated via the elevator trim on the TX as long as the elevator deflection remains modest. There should be no effect on handling. This is obvious on an all-flying tail... Increase the wing incidence 1 degree, and add 1 degree down elevator trim on the TX. The wing and tail are at the same relative orientation as before, and the glider should fly as before. In a conventional hinged elevator, you'll have to add about 2 degrees of down trim for 1 degree of wing incidence, but again, the glider should fly as before, even though 1 degree of "decalage" has been added. If changing the decalage on the glider DOES produce a noticable change in handling, then one can conclude that the elevator response is nonlinear, which indicates something bad and draggy is happening. Two possible causes for tail nonlinearity are: 1) The elevator deflection in trimmed flight is excessive ( 5 degrees, say) before or after the decalage change. Changing decalage to reduce this deflection will surely reduce the tail's profile drag. 2) The horizontal tail has a crummy airfoil which suffers from hysteresis or other flow pathologies. Such airfoils are usually draggy. A proper airfoil will eliminate this behavior and will reduce the profile drag as a bonus. If the tail response is linear and the handling doesn't change, then a minor effect of adding the 1 degree of wing incidence in the example above is that it lowers the angle of the fuselage by 1 degree. This may increase or decrease the fuselage drag. Another minor effect af adding 1 degree of wing incidence is to raise the tail relative to the wing's wake. This is not an issue on most gliders, which have the tail flying well above the wing's wake to begin with. - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Re: MH32 VS SD7037
For the wing loading you gave, Xfoil sez that the 7037 has about 10% higher usable CLmax than the MH32. But the MH32 has a lower Cd at higher speeds. But as Don Stackhouse pointed out, the two airfoils want different wing loadings (different chords) for a meaningful comparison. If you assume the MH32 wing has the same span but a 10% bigger chord than the 7037 wing, then Xfoil says the two wings are almost indistinguishible in both thermalling speeds and penetration. The appropriately scaled polars are right on top of each other. The larger MH32 wing area almost exactly cancels its lower Clmax and lower Cdmin. The larger MH32 wing has more volume and skin area normally means it may be heavier, but on the other hand the larger thickness means the skin can be thinner which may compensate. Still may be a toss-up. In reality the quality of construction will obviously have a significant impact on the relative performance. This is tough to quantify. - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Re: Composite spar sizing
Tom Seitz writes: I understand that for applications where the wing will be really stressed, it is better to put the spar at the 1/4 chord point, the reason being that this is where all the bending force is applied. If the spar is offset from this point, the wing will tend to warp. Is this correct? Or is the max thickness point close enough to the 1/4 chord on many airfoils not to be significant? The center of lift location as a fraction of chord isx/c = 0.25 - CM/CL. In a typical winch launch of a typical glider CM = -0.05, CL = 1.0, and so the center of lift is closer to 30% chord. This is the place to put the spar if you want to minimize twisting. If the wing is very stiff torsionally this may not matter. - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Camber change versus span
Chris Adams writes: I am trying to understand the reasons for changing camber as one progresses from the root to the tip on small wings. Can anyone tell me more about the lower camber at the tip and whether this induces effective washout on the wing? There are numerous airfoil/planform/twist conflicts between the following requirements: 1) Good penetration L/D or good handlaunch height. This wants all spanwise locations to go to zero Cl at the same time as the aircraft's AoA is reduced, and also for each location to remain within its airfoil's drag bucket. 2) Tip stall resistance in tight circling maneuvers. This wants smaller Cl towards the tip, preferably with more stall-resistant tip airfoils. This is complicated by the lower tip Reynolds numbers due to taper. 3) Minimum induced drag. Assuming the span is fixed, this ideally wants the c*Cl distribution to be elliptical at slow thermal speeds. Two extreme possibilities are : i) a constant chord c, with elliptical Cl via washout --- great for 2), awful for 1) ii) an elliptical planform c, constant Cl via flat wing and constant airfoil --- OK for 1), bad for 2) The simplest safe baseline compromise solution is: a) A constant airfoil, zero twist, and a planform with a considerably wider tip than elliptical. This is nearly ideal for 1), OK for 2), and least favorable for 3). The following fine-tuning mods can be done: b) The tip airfoils are thinned, while maintaining their camber and keeping the zero twist. This benefits 2) the most, since it compensates for the lower tip Re and usually gives a larger local Clmax. But this thinning narrows the tip airfoil's bucket somewhat, which may penalize 1). The c*Cl stays the same, and so 3) is unaffected by this mod. Note: The smaller thickness makes the tip airfoils appear more undercambered, even though their camber has not really been changed. c) The tip chords are narrowed slightly from the "simple" wider-tip solution, and some washout is added. This mod can make the loading nearly elliptical, and benefits 3) the most. On the other hand, 2) is more or less unaffected, but 1) will suffer if the washout is done to excess. d) The tip chords are narrowed slightly as in c), but the tip chords are decambered the correct amount in lieu of washout. This benefits 1) at some cost to 2). The benefit to 3) is same as with c). My HLGs use a blend of mods b) and d). My current 2-meter RES project uses a blend of b),c),d) in suitable proportions. I don't know what's best for F3B. The best combination of mods b), c), d) depends on which performance consideration is most important. If 1) is most important, like in a windy-day HLG, the simple flat wing solution a) may be best. If you want a calm-day floater for small and weak thermals, then making mods b) and c) is most appropriate. Mod d) is useful in lieu of c) to keep the wing flat for easier construction perhaps. - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
[RCSE] Composite spar sizing
I recently built a composite spar sample and tested it to destruction. The following is a writeup of the test, followed by some calculations for sizing an actual glider spar to take a given load. It should be useful for do-it-yourselvers. Note: A fixed-width font is required to view the high tech ASCII graphics :-) Spars of this type have incredible strength/weight ratios. They make much better use of carbon fiber properties than piling carbon layers into the skin, since the compact spar cap is supported very well against buckling. With the spar taking all bending loads, only a light bias glass skin is required to take the remaining torsion loads. My 180lb-winchload 2-meter wing has a 3.5 oz spar, and will end up under 12 oz total. A 120in Open Class spar should come in at about 8 oz. Tough to beat that with any other type of construction, although it does take more work than a simple thick-skin wing. The CST website www.cstsales.com shows how to build a spar of this type. They use sparcaps tapered in thickness rather than in width, and use hard Rohacell rather than endgrain balsa for the core, but these are details. They also use a heavy woven carbon sock as the shear skin, which appears to be major overkill according to my test. Light glass cloth is adequate. - Mark Drela Standard 4-point bending test on 11 inch long composite spar sample. --- Mark Drela 8 Feb 00 Test description The spar sample was designed to produce material stresses in the spar caps and shear skin comparable to those in an RC glider on a hard winch launch. = CF spar cap | | | | endgrain balsa core | | | | = CF spar cap Glass skin on the bias is wrapped twice all the way around, with the seam along the top of one spar cap. 4.5 lb endgrain balsa core: 0.35 in wide x 0.48 in high Precured prepreg CF caps : 0.35 in wide x 0.021 in thick (3 plies) 1.5 oz glass skin (+/-45) : 2 layers, 0.005 in thick total Small endgrain basswood inserts replace balsa core at load points. Sample weight = 11.2 g Caps were first bonded to balsa core with slow epoxy. Glass skin was wrapped completely around spar with West 105/206 epoxy, under 15 in-Hg vacuum (intentionally reduced vacuum to avoid crushing the soft balsa core). Sample was cured for several days before test. Four load points L = 3.7" apart: F/2F/2 | | v v ==X=== ^^ || F/2 F/2 Skin failed in shear at F = 180 lb at location X, with diagonal buckle lines and tears. Failure appears to have started along the edge or the spar cap, where the skin goes over the 90 degree bend. Rounding off the cap edge to reduce this kink curvature is likely a good idea if the skin is marginal for the shear load. Stress calculations Cap thickness : Tc = 0.021 in Spar width : w = 0.35 in Net spar height: h = 0.50 in 1 spar cap area: Ac = 0.00735 in^2 = w Tc Bending inertia: I = 0.00092 in^4 = h^2 Ac / 2 Shear skin thk : Ts = 0.005 in Shear skin area: As = 0.005 in^2 = 2 h Ts Max bending moment: M = L F/2 = 330 lb-in Spar cap load : P = M/h = +/-660 lb Cap axial stress : sigma = P/Ac = +/-9 psi (no sign of failure) Max skin shear load: S = F/2 = 90 lb Skin shear stress : tau = S/As = 18000 psi (at failure) Scaling test results to winchproof RC glider spar - Typical two-meter glider example... Sparcap thickness : Tc = 0.042 in (6 plies, commonly sold as 0.045 in) Design cap stress : sigma = 10 psi (somewhat conservative I think) Design shear stress:tau = 12000 psi (50% safety margin) Note: The bottom spar cap is in tension, and can surely tolerate at least 15 psi or even 20 psi stress. The bottom cap can therefore be only 2/3 or even 1/2 as thick as the top. Conditions at wing root: wing span : b = 78 in wing load : F = 160 lb(winch line force) spar height : h = 0.9 in(9% airfoil with 10 in chord) bending mom.: M = b F/4 = 3120 lb-in cap load: P = M/h = 3467 lb shear load : S = F/2 = 80 lb required cap area : Ac = P / sigma = 0.035 in^2 required cap width : w = Ac / Tc = 0.83 in required shear area : As = S / tau = 0.0067 in^2 required skin thick.: Ts = As / 2h = 0.0037 in (2 layers of 1.5 oz glass) Ideally one wants to taper the spar dimensions so as to maintain constant sigma and tau. This will give minimum weight for a given strength. Assuming the cap thickness Tc and spar height h are roughly constant, the relative spar width and shear skin thickness should taper as follows: center .. midspan .. tip --- w ~ 1.00.560.250.060.0 Ts ~ 1.00.750.500.25
[RCSE] Re: SD 6063 for Handlaunch?
Martin Brungard wrote: I've done some fiddling with airfoils using Hepperle's code to come up with something that better suits the HLG flight regime than the S6063. I don't think that the S6063 was designed for very low Re anyhow. I would sure love to see Selig or Drela create a better HLG airfoil since they have more knowledge and tools than me. Well, since you asked, I've attached the AG04 section which I described earlier. I've had very good success with it on my 36" and 40" HLG's with 4.5" avg chord. It has 6.4% thickness and 1.75% camber. At the tips I used a slightly thinned version --- 5.8% thick, about the same camber. This section will NOT benefit from reflex on launch, since its drag bucket goes down to zero lift, provided the bottom surface is built accurately, especially at the front. According to XFOIL, the CLmax is significantly higher than the S6063's. It seems to float reasonably well, in any case. At the low Re numbers it also benefits very little from downward flap deflection (i.e. flaps or flaperons just screw it up in every way). So it's best suited for poly ships. Mark, I concur with your recommendation to increase chord for the S6063 to get the loading down and the Re up. I was very surprised when DJ Aerotech came out with the higher aspect ratio Spectra recently. The higher aspect ratio does pay off provided the weight is reduced correspondingly. DJ Aerotech certainly succeeded doing that. The "optimum" AR is higher if you keep thinning the airfoil as the Re dwindles, rather than keep the airfoil fixed. The best airfoil for the Spectre will be thinner than the best airfoil for the Monarch. I don't know if it actually is thinner. Structural issues push the thickness the other way. High aspect ratio wings are inherently more pitch sensitive and difficult to fly. I don't think that's the case. If the tail is sized correctly, bad pitch behavior may be due to low Re effects on the wing's Cm or the tail's Cl. Either problem can be avoided if proper airfoils are used. - Mark Drela AG 04 0.990.000670 0.9948700.001167 0.9849630.002107 0.9732880.003174 0.9609100.004278 0.9483320.005403 0.9357080.006542 0.9230680.007688 0.9104230.008838 0.8977820.009989 0.8851480.011139 0.8725220.012283 0.8599010.013420 0.8472830.014552 0.8346690.015674 0.8220560.016786 0.8094430.017889 0.7968310.018980 0.7842170.020061 0.7716010.021132 0.7589850.022192 0.7463660.023242 0.7337430.024281 0.7211200.025311 0.7084960.026332 0.6958720.027342 0.6832490.028342 0.6706290.029330 0.6580100.030306 0.6453930.031267 0.6327790.032216 0.6201670.033147 0.6075570.034064 0.5949500.034963 0.5823460.035844 0.5697450.036705 0.5571470.037545 0.5445510.038364 0.5319590.039158 0.5193700.039929 0.5067830.040673 0.4942000.041391 0.4816200.042082 0.4690420.042741 0.4564670.043370 0.4438950.043967 0.4313250.044532 0.4187580.045062 0.4061950.045558 0.3936360.046016 0.3810820.046436 0.3685330.046816 0.3559900.047155 0.3434560.047449 0.3309290.047699 0.3184130.047899 0.3059080.048048 0.2934160.048143 0.2809380.048180 0.2684770.048156 0.2560360.048067 0.2436150.047906 0.2312190.047672 0.2188520.047357 0.2065160.046956 0.1942190.046460 0.1819650.045864 0.1697620.045159 0.1576180.044335 0.1455440.043381 0.1335520.042286 0.1216590.041034 0.1098830.039614 0.0982520.038005 0.0867960.036191 0.0755610.034151 0.0646020.031864 0.0540030.029317 0.0438990.026506 0.0345160.023466 0.0261870.020297 0.0192500.017178 0.0138390.014300 0.0097980.011763 0.0068150.009556 0.0046040.007625 0.0029580.005908 0.0017390.004355 0.0008610.002932 0.0002850.001613 0.180.000382 0.85 -0.000805 0.000554 -0.001985 0.001456 -0.003123 0.002756 -0.004215 0.004455 -0.005308 0.006633 -0.006445 0.009447 -0.007655 0.013152 -0.008964 0.018112 -0.010393 0.024689 -0.011906 0.032994 -0.013392 0.042729 -0.014711 0.053413 -0.015788 0.064674 -0.016627 0.076299 -0.017257 0.088178 -0.017709 0.100238 -0.018017 0.112430 -0.018202 0.124717 -0.018284 0.137072 -0.018277
[RCSE] subject author date 49103 Re: Soaring V1 #741 Robin Badenoch Mon Jan 17, 2000 49104 Re: Carbon tailboom source Tom Broeski Mon Jan 17, 2000 49105 METRIC 2METRE Garry Whitfield Mon Jan 17, 2000 49106 Re: Dynamic Soaring in Kentucky Michael Neverdosky Mon Jan 17, 2000 49107 DS a Building Rudy Siegel Re: Carbon tailboom source
Tom Broeski writes: http://www.intothewind.com Their Avia wrapped tapered carbon tubes are perfect for HLG tailbooms. At about $8 these aren't cheap, but they are almost half the weight of pultruded tubes for a given bending stiffness. They are also much better in torsion. They have three weight - stiffness grades: Skinny Ultralight 7.7g , stiffness =1.8 Skinny Super UL11.7g , stiffness =3.2 G-force Ultralight 13.0g , stiffness=4.9 The weights are for the shorter 32.5" lengths. The stiffness numbers are relative. The two Skinnys have the same ID's but different wall thicknesses. The G-force has a larger ID and hence a much better stiffness/weight ratio. The big end of the Skinny Ultralight is perfect for 3-4oz , 30"-40" gliders. For a 60" glider the small end of the G-force is probably just right. - Mark Drela RCSE-List facilities provided by Model Airplane News. Send "subscribe" and "unsubscribe" requests to [EMAIL PROTECTED]
