stunthanger.com
Building Tips and technical articles. => Building techniques => Topic started by: Charles Hofacker on October 24, 2021, 01:43:39 PM
-
I have an electric Vector 40 on the way. So, I've been reading everything I can find about it. I came across a post that said put the push rod in the innermost hole in the bell crank and the outer most holes in the flap and elevator control horns. This is good to have info because the instructions just say, "attach the push rod..." no details on which control horn holes to use or ratios to achieve. This raised the question in my head, "why not a 1:1:1 bell crank/flap/elevator ratio"? It seems to me that the recommended inner hole to outer hole would give a non linear movement to control inputs: that is more sensitive around neutral and less sensitive at full deflection. Is that a desirable configuration? Perhaps due to airframe response at large deflection angles? I kind of expected a linear response at all deflection angles to be desired. Sensitivity of the 1:1:1 system could be controlled by the line spacing at the handle. Most of the posts I've read talk about flap to elevator but not flap to elevator to bell crank. perhaps y'all could direct me to a discussion of aircraft response to control inputs considering the entire system: handle/bell crank/flap horn/elevator horn.
Thanks to all!
-
First, we should all start naming our planes "The Compromise", because that's what a stunt plane is -- it's a giant compromise between competing requirements, and any time you completely solve one of them then (unless you're hard up against some limit imposed by the rules) you create another. That would make our naming perfectly honest -- except then you wouldn't be able to tell the planes apart. Ahh, compromises...
As an example that's pertinent to your question, if you go with a 1:1:1 crank, flap and elevator ratio then you get a nice mostly-linear ratio (but not quite, alas, because the bellcrank is working in a different plane than the flaps are). However, that means that you give up a whole lot of leverage on the bellcrank. That means that the lines don't have to move as much for a given control surface motion (nice!), which in turn means that the force on the lines for that given control surface motion is increased. Because your lines are curved in flight, not straight, they act like springs -- and by reducing the leverage that they have on the control surface, you increase the effect that their springiness has on the control system. This is an effect that I've experienced myself, on a Ringmaster which I built with the pushrod too far out on the bellcrank; when I made a super-long elevator horn for the thing, that dragged in the mud every time I landed on grass, the airplane performance perceptively improved.
I think that as pertains to your question, the things you want to achive are -- sort of in order of importance -- the following:
- Good workmanship -- make it so it doesn't wear out until over 1000 flights, passes pull test, etc.
- Make the controls free-moving and free of slop (already, two competing requirements)
- Use the biggest bellcrank that'll reasonably fit. For a Vector 40, that's probably a 4" bellcrank. "Reasonably" is a judgement call because the bigger the bellcrank, the more room you need for it and the more the leadouts will bend at the wingtip, which is bad, but a bigger bellcrank means that the lines have more leverage, which is good.
- You can go overboard on this -- at one point people were experimenting with pulleys instead of bellcranks (Ted Fancher either originated this or was the most famous guy doing it) it turned out to be more trouble than it was worth
- Get the elevator travel correct with respect to the bellcrank. +/- 20 degrees at full bellcrank travel is probably as much as you'll ever need, but I usually make plenty of room for the bellcrank to move and set the travel at +/- 30 degrees on the bench, before the fuselage is buttoned up. I'm probably giving up line leverage -- but my planes always have plenty of turn.
- A 1:1 flap:elevator ratio isn't always best -- and "best" depends on a number of factors, including pilot preference and how heavy the plane comes out. You won't find out until you start trimming. So, if you're serious, make the ratio adjustable. Most people do this with a slotted elevator horn and a carrier for the pivot pin that can be loosened, adjusted, and tightened.
- You won't know the best flap:elevator bias until you trim. So while you're making the elevator horn slotted, use a ball link on the end of the pushrod that can be screwed in and out (and note that this presupposes an access door unless you're using a take-apart plane)
- Other stuff I'm forgetting at the moment
- Oh, yes -- a perfect relationship between handle and elevator/flap motion. Note that I don't say "perfectly linear" because -- especially for heavier planes -- folks are going to logarithmic flap control, where the flaps tend to move a lot around center, but slow down after a while with respect to the elevator. The idea is that in level flight and in round maneuvers you want to vary the lift for smooth groovy level and round which is aided by lots of flap, but in square corners you want to turn the airplane -- having more elevator than flap at that point helps
Howard Rush has gone to the trouble of making an Excel spreadsheet that lets you enter in all sorts of parameters of your control system, then calculates the actual flap and elevator motion. This won't let you make that perfect control system, but it lets you figure out your own personal best compromise -- especially if you're fabricating your own bellcranks so that you can do things like offset the pivot point or the flap rod pick-off point. It's available for the asking.[/list]
-
I have been using a borrowed Vector 40 for the past 18 months as my sole source of PA, including contests, due to circumstances beyond my control. It is an excellent plane and flies better than most of us can fly it. I got it "as is" and had to keep it that way so I inherited a control setup that was not to my liking but after a few flights I adapted. It was ?:1:1 with 30 degrees total. It does beautiful rounds this way but it does not "snap" through corners like I prefer. If you have the luxury of equipping it with "logarithmic" - Do It.
Your search for balanced controls leads me to wonder why we spend an enormous amount of time trying to make our planes turn at the same rate inside and out with a wrist and fingers that are as asymmetric as you can get yet the new normal is to have the handle straight up and down. If I put my wrist in a "normal" relaxed position where my fingers are in line with my arm I can move my wrist 1" up and 3" down. I balance my control speed to give me 3" total leadout travel (1 1/2 each way). Why would I not want to bias my handle movement so I have 2 & 2 wrist and no elbow?
Ken
-
There are a bazillion solutions to this. On my Brodak Ventor 40 I went middle hole bellcrank to outer hole flap and middle hole flap to middle hole elevator. That's about (eyeball measurements) 1/2" to 7/8" and 1/2" to 1/2". The total unrestricted travel of the flaps reaches a bit over 30 degrees which in turn means the elevator reaches a bit over 30 degrees. My protractor wasn't anyplace easy to find although it was most likely within arms reach and I'm guessing a bit over 30 degrees could be close to 40. It doesn't matter a whole lot as I don't go to lock during flight. The same result or could be reached by going inside hole on the bellcrank to the middle hole on the flap lever and outside to outside flap and elevator lever but then there wouldn't be any option to reduce the elevator travel which I don't think would be necessary.
-
when I used the innermost hole on the bell crank (in other airplanes) the push rod would contact the bell crank bushing and mounting post. I had to move the push rod to the middle hole to get clearance to have full down. I've decided to purchase a premade control system probably from Tom Morris/Okie air. My previous ARF crashed when the lead out came off the bell crank in flight. Can some one give me the exact hole to hole distances for the bell crank to flap horn and flap horn to elevator horn? I don't have the electric ARF Vector 40 in my possession yet and probably couldn't make these measurements accurately enough if I did.
-
Uh, post #2 outlines the whole Tom Morris system for that size plane.
I think he has the Brodak model that doesn't have slotted flap and elevator. The setup I used on mine is not far from the TM. Bellcrank middle hole to outside flap hole. From middle flap hole to middle elevator hole. As I recall both flap and elevator horns are the same.
-
The 4" Brodak belcrank has 3 holes, the inner most is for racing, the middle is for stunt and the outer most hole is for combat.
I advise you can use the Brodak supplied control system with no problem if you set it up as Mark has suggested. The middle bellcrank hole to the tallest flap horn hole then use the middle hole for both ends of the elevator push rod for a 1:1 flap elevator ratio. This is a very standard set up for the Brodak ARF's and it works well.
Motorman 8)
This is where I will start. I still plan to assemble the ARF with hatches under the Stab/elevator to permit adjustment of neutral and equal upright/inverted control and if necessary change the elevator response ratio.
-
While we are on the subject of setting up the control system of a Brodak Vector 40 ARF, I have a question about "slop" in the elevator horn. ??? Back in the days of old my friends were flying the pattern and I was still trying to not crash... they intentionally built some "slop" into the elevator supposedly to stop "hunting" in level flight. Is this still common practice and does it really reduce or prevent hunting? If so, How much slop is desirable?
-
While we are on the subject of setting up the control system of a Brodak Vector 40 ARF, I have a question about "slop" in the elevator horn. ??? Back in the days of old my friends were flying the pattern and I was still trying to not crash... they intentionally built some "slop" into the elevator supposedly to stop "hunting" in level flight. Is this still common practice and does it really reduce or prevent hunting? If so, How much slop is desirable?
There are guys that do this. Unfortunately there are many causes of "hunting". Most of the "hunting" I experience are meat servo induced and slop might help this. For me though zero slop in the control system up front and making slop is a last ditch effort (will never happen) following other fixes like more handle time. Probably the leading cause of hunting not involving the meat servo is the wing wake impinging on the HT which is why we see vertical separation and the HT not being in line with the wing. Increasing the distance from wing to HT also works but impacts the turning radius. Having a too thin TE on the control surfaces can cause the surface to float back and forth causing oscillations which would be aggravated by having control system play. Rule of thumb is the TE needs to be around 3 times the boundary layer thickness thick. I dunno but 1/8" ish seems to work. A discussion on hunting will lead down a wide rabbit hole. The Vector 40 doesn't seem to have troubles with hunting and grooves nicely.
-
Let me follow up on Mark since the V40 has been virtually all I have flown in the nearly 2 years since our fire. It does not hunt on it's own, period and it turns fairly well nose heavy (electric) if you are so inclined (I am not). The one I flew had 1:1 with 30 degrees movement and no slop at all in the controls. It also recovers nicely from the drifting kind of hunting caused by wind with minimal effort on your part. I never once got the porpoise kind that slop in the elevator would "help". It had 1/4" slab flaps (which I hate - I build mine from 3/8" tapered to a 1/8" CF TE cap). IMHO if you are going to have slop it makes sense to have it between the BC and flap horn, not the elevator. Having the flaps move slightly w/o any elevator movement is going go create a movement opposite of the direction you thought you needed to go.
You mentioned that you will be electric. Very important to keep the controls free, especially around center. "Stickshion" (sp?) is real!
Ken
-
"Stickshion" (sp?) is real!
I've always seen it spelled stiction (stick + friction, or a contraction of "static friction"). In academic papers it's usually called "static friction", not to be confused with "coloumbic friction".
-
While we are on the subject of setting up the control system of a Brodak Vector 40 ARF, I have a question about "slop" in the elevator horn. ??? Back in the days of old my friends were flying the pattern and I was still trying to not crash... they intentionally built some "slop" into the elevator supposedly to stop "hunting" in level flight. Is this still common practice and does it really reduce or prevent hunting? If so, How much slop is desirable?
I think that probably got started when people rebuilt their controls from "loose"/worn out to "tight" and in the process created some binding - which is what causes hunting. I think as long as it is free to move, it has enough slop, you don't need more.
Brett
-
... How much slop is desirable?
None, as long as the controls are totally free. But controls that are oh so slightly sloppy are probably better than controls that are sticky, particularly if you're not running a giant methanol-powered vibrator on your plane to break the stiction.
-
Problems I see from poorly analyzed control systems are: 1) inadequate leverage over flap hinge moment and 2) asymmetry between up and down. Neither can be fixed with handle geometry.
-
There are guys that do this. Unfortunate like I there are many causes of "hunting". Most of the "hunting" I experience are meat servo induced and slop might help this. For me though zero slop in the control system up front and making slop is a last ditch effort (will never happen) following other fixes like more handle time. Probably the leading cause of hunting not involving the meat servo is the wing wake impinging on the HT which is why we see vertical separation and the HT not being in line with the wing. Increasing the distance from wing to HT also works but impacts the turning radius. Having a too thin TE on the control surfaces can cause the surface to float back and forth causing oscillations which would be aggravated by having control system play. Rule of thumb is the TE needs to be around 3 times the boundary layer thickness thick. I dunno but 1/8" ish seems to work. A discussion on hunting will lead down a wide rabbit hole. The Vector 40 doesn't seem to have troubles with hunting and grooves nicely.
I am trying to post using a new phone upon which I am not type (in two senses) rated, so I won’t separate issues. In the unlikely event somebody reads this, I apologize for it being hard to follow.
We saw vertical separation between the wing and stabilizer, but now we don’t so much, eg. the current world and US Nats champs. In-line airplanes are all the rage in my neighborhood.
I spent some effort on a 737 issue where hysteresis caused a limit cycle. You controls guys can explain to me how it fixes one.
My flap and elevator TEs were 1/16” thick. No oscillation. I now have a .3” elevator TE for a different reason and shall try .4” Saturday.
Igor Burger and Frank Williams have given on this forum credible explanations of hunting by causes other than friction.
-
I spent some effort on a 737 issue where hysteresis caused a limit cycle. You controls guys can explain to me how it fixes one.
While I was not the one advocating slop, you can turn a big one (caused by control binding) into a little one (caused by .005 worth of free play). Mark's rather strange description is right in at least one regard, your second-order pilot response causes the big one in the binding controls case, and free play removes that but causes an "automatic" little one.
My predecessors and mentors were the ultimate wizards of using intentional non-linearities to work around other, unavoidable, discontinuities and non-linearities.
Brett
-
My understanding of the action of elevator slop and a no-slop flap mechanism is that it provides the ultimate in logarithmic control, in that right around zero handle motion the elevator isn't engaged at all, only the flaps. So -- according to the theory -- gentle up or down corrections result in the airplane rising or falling without changing its direction of flight.
I'm not sure how true that is, or how much the actual behavior diverges from theory when you add in aerodynamic forces and a giant methanol-powered vibrator to the front of the plane.
Does anyone here know if the Igor Burger logarithmic flap thing could be made to affect this outcome? I'm pretty sure that it drives the flaps and not the elevator, so you really only get the option of desensitizing the flaps at the extremes, not desensitizing the elevator around neutral.
We could do this all with electronics and servos, if only some over-enthusiastic fellow hadn't allowed himself to be talked into filing a freaking rules proposal that rules that out.
-
My understanding of the action of elevator slop and a no-slop flap mechanism is that it provides the ultimate in logarithmic control, in that right around zero handle motion the elevator isn't engaged at all, only the flaps. So -- according to the theory -- gentle up or down corrections result in the airplane rising or falling without changing its direction of flight.
I'm not sure how true that is, or how much the actual behavior diverges from theory when you add in aerodynamic forces and a giant methanol-powered vibrator to the front of the plane.
Does anyone here know if the Igor Burger logarithmic flap thing could be made to affect this outcome? I'm pretty sure that it drives the flaps and not the elevator, so you really only get the option of desensitizing the flaps at the extremes, not desensitizing the elevator around neutral.
We could do this all with electronics and servos, if only some over-enthusiastic fellow hadn't allowed himself to be talked into filing a freaking rules proposal that rules that out.
I have in Howard's words, a RCH of slop in my logarythmic flaps and zero in the elevator. In the logarythmic the transition horn drives both the flaps and the elevator. The link to the elevator is basically ball links so zero slop. The flaps are driven by the bearing arm and that is where you have a tiny bit of slop. In my plane I have just under 1/8" total with the elevators locked down. So basically I have a zero slop setup from the handle since the elevator will start moving before the flaps.
It is different. When you have zero slop in the flaps and some in the elevator you get a negative response near center. For most of my competative years I flew with 1/8" or so slop in the elevator. My muscle memory knew that I should "turn into the drift" so to speak to keep the plane level. On my first flights with the logarythmic I nearly pancaked on my first inverted because the plane went in the direction I told it to. What I mean by that is that with normal controls, the plane will move slightly in the opposite direction from the control at the handle until the elevator engages. With logarythmic the elevator always moves first. That is why planes with too much slop tend to drift up and down. You are giving it a hint of control one way but the plane is getting the opposite from the flaps.
Ken