F3F radio set ups (OpenTX)

[Konrad]

Well, it looks like I'm now moving up in my progression as an F3F pilot. At the last race I learned a lot about expo in the set up of a plane's response. I normally only use expo values of 18 to 25 on my elevator and ailerons. I learned that many of the top fliers are using expo in the 35 to 55 range. I tried this and can report much smoother laps.

I also learned of the concept of asymmetrical expo. That is while one might have a large Expo value for up elevator. Many have a curve for the down input (forward) stick much closer to a linear response. This is done to allow a fast response when bunting a turn (going slightly inverted to avoid a cut).

I see with OpenTX that this can be done with the custom curve. I'm using a 9 point curve where I try to emulate the shape of a high expo number in the up direction. On the down side (forward stick) the curve looks close to a linear line (maybe half the expo of the up). I have to to ask is there a way to get these curves using the same units we use in the Expo screen? (Right now I'm manually manipulating the points on the curve and using the smooth function).

Well, it looks like I need to bring out my Stregas' to get a feel for this next level of F3F programing.

FYI: I'm using Mike Shellim's F3F templets with a X9E using OpenTX 2.2.4.

 

[Wayne]

As I recall a V-tail does not give a balanced response like a flat tail will for elevator inputs. As I recall you get more response in the up elevator than you will in the down elevator. Basically when the elevators go up, they are creating a higher pressure responses due to the "V"

Sorry, not well worded..

 

[Konrad]

Well, I find that the V tail actually is more powerful in the down as the tail gets out of the wing's downwash as it goes up.
Now for a few reasons we add rudder differential to add some down when using the V-tail rudder.

What I'm looking to learn is how to change the exponential curve between the up command and the down command. This has nothing to do with the tail geometry.

To get nice smooth inputs most guys are using a very large expo value on the elevator. Now in a bunt turn the need for a smooth inverted push is long gone. What is needed is a fast response to keep the nose up while inverted.

Here I'm using OpenTX and assigning a custom curve to the elevator curve. In OpenTX you define this curve by the number of points (7 here) then manually move them to the shape you want. You can see that on the left is the high expo like curve (Up elevator). On the right is a fast linear curve (Down elevator).

I just tried this out on my Schwing Corsa and actually like the feel. very smooth when making the passes. but if needed I can set up the model very quickly to fly to the pylon inverted should I have cut a base.

Now what I want to learn is there a way to define these curves like we do normally with a command Expo XX? Or am I forced to manipulate the points by hand to get an approximation on the expo curve.

 

[RC-Soar]

I also learned of the concept of asymmetrical expo. That is while one might have a large Expo value for up elevator. Many have a curve for the down input (forward) stick much closer to a linear response. This is done to allow a fast response when bunting a turn (going slightly inverted to avoid a cut). [...] I'm using a 9 point curve where I try to emulate the shape of a high expo number in the up direction.

FYI: I'm using Mike Shellim's F3F templets with a X9E using OpenTX 2.2.4.

Here's a cleaner, more flexible method without the need for custom curves. It gives you independent travel up/down ('diff'), as well as independent up/down expo.

The secret is to use two Input lines, one for up and one for down. The active line is determined by the 'side' parameter, '<0' is stick back '>0' is stick forward.

 

[Konrad]

THANK YOU !

This alone is worth the time spent on the forum.

All the best,
Konrad

 

[RC-Soar]

You’re welcome, Konrad

 

[Konrad]

That was slick! If you knew how poorly I and programs get along that’s saying a lot!

I’ve added this functionality to all my racers. Now all I need to do is actually fly the models to find the values appropriate to each model.

 

[Wayne]

Interesting solution. had not thought of that one, I love that there are still things popping up that I have not seen before.

 

[Konrad]

I flew this dual expo curve in the heat of competition and found that it actually works well. Getting the nose up fast while inverted added a bit of confidence to my flying the cut.

Now another thing has popped up, that is the dreaded V-tail wiggle. As a result of a crash I was forced to fly a Cruciform tail racer right after a V-tail racer. I think I might be seeing a negitive interaction between the flaps mixed as ailerons disturbing the airflow over the V-Tail. At speed I think I'm not going to full aileron deflection, as a result the flaps don't move as much. And much to my surprise the V-tail racer looks to be coming out of the turn a lot smoother ( no wiggle). Now if I've lost speed or am rushed and use close to full aileron input the V-tail racers wiggle their tails.

I'd like to ask if you guys that use Aileron to Flap mix with V-Tails, have you seen this speed sensitive behavior. With Cruciform, and maybe even the T-tail, the disruption of the airflow from the flaps only effects pitch. And as we are pulling the elevator we don't notice any other interaction. But with the V-tail there is close to a 50% yaw component, hence the wiggle.

With Cruciform tails I use the flaps to get a better span wise load distribution for the roll input. I've thought that the less I have to deflect the aileron the less over all drag for any given control response. But with V-tails I think I'm adding a lot of unwanted yaw from the air disturbed by the flaps in the opposite directions.

Unfortunately I'm all out of V-tail racer to test this right now. But I'm thinking that with the V-Tail I'll leave the flaps fixed when making an aileron input. I may take a slight drag hit in roll but I think the over all response will be a huge drop in drag as the fuselage won't be skidding all over the place!

I was talking with Darrell Zaballos and he too is noticing this and is experimenting with this as he now has the Aileron to Flap mix on a switch.

 

[2MUCH2DO]

V-Tail wiggle form aileron to flaps usually has to do with differential, or lack of. Your flaps create more drag than your rudders and you get yaw. I had one glider (an Erwin) that needed less differential, and another that needed loads more. I usually start with 30% and adjust from there, that Erwin was interesting, its had horrible tail wag, after this tuning its nicely coordinated across a wide speed range. I tend to do a series of abrupt aileron rolls and see how bad the effects are, then change and repeat. If its coming from your snap-flaps, than its possible you are mixing in some aileron thumb with your elevator?

BTW, I was just told over that weekend that the danger to differential is that its effect reverses in negative G's, so it may be better to combat it with aileron->rudder to keep you smoothly coordinated.

anyhow...

 

[Konrad]

Thank you.

My understanding is a bit different from yours.

Aileron differential is to address the change in the lift vector. The upward aileron tilts this forward and the downward aileron tilts this aft. This looks like a drag.

All yaw correction are reversed when inverted. Unlike FAI F3A, F3F does not have a large inverted flight regimen. So we can trim the nose to go we’re we want with some inputs from the rudder.

I think you might be on to something with the idea of setting the aileron diff with the flaps kept in neutral. Then add the flaps to aid the roll rate. Adjust the interaction with the V-tail using differential in the flaps only. Don’t change the aileron setting.

It will be interesting to see if this flap and V-tail interaction is speed dependent.