Design > Engineering board

New log flap drive crank idea

<< < (5/6) > >>

Istvan Travnik:
Dear Mark,
Wouldn't it be nice, if you "drill" another hole for the pushrod junction in the luffing arm, as close, as possible to the fixed pin (or in other words: to the end of the slot)?
That diagram could show more dramatic change of the input-output ratio...

Mark wood:

--- Quote from: Istvan Travnik on November 12, 2021, 02:07:48 PM ---Dear Mark,
Wouldn't it be nice, if you "drill" another hole for the pushrod junction in the luffing arm, as close, as possible to the fixed pin (or in other words: to the end of the slot)?
That diagram could show more dramatic change of the input-output ratio...

--- End quote ---

Oh yeah. I did a number of iterations starting from what I could determine from the drawing. From that I played around with the pivot and the push rod dimensions. Too much curvature can lead to poor response at low angle and and excessive at higher angles. Intuitively, I think we want the two curve to intersect around 20-25 degrees bellcrank angle. That's just a bit before a typical square corner input is. My intent is to use this in combination with the log flap drive which does some "flattening" of it's own around center.

The struggle with this setup is getting the two conditions to converge, flattening the slope around neutral and crossing the "linear" curve at 20-25 degrees. Here are few condition analysis'.

First is an even ratio pivot length to bellcrank length 1:1 and PR located 1/2 between. This is how I interpret your drawing.

Next is when the pivot is very long and the PR actually goes backwards initially and never does converge with 45 degree bellcrank travel.. This could actually be a math thing and not physically capable of being made.

Finally a really good exponential curve but the trouble is that it doesn't catch up to the linear curve until 35 degrees or so which is beyond what we would be doing during flight.


Ultimately it isn't an easily modifiable system and is extremely sensitive on the ratios. Just a very minor change creates a big difference in the output. So, I settled on an iteration I felt was a good middle ground meeting the highest requirement and crossing the linear curve at 25 ish degrees for the one I would create a working prototype.

Istvan Travnik:
Wonderful curves!
Thanks a lot!
I think, almost nobody will try to make in real life this "extremist" and too hidden arrangement, but the other one, which one built into my 2012 model, can be built more simple, and freely variable.
...and some pilots, who feel himself a little bit "woodenhanded", like me, can find it useful, helpful, to fly better.
Since my 2012 model is broken now, I could make some photos, showing input and output junctions.
And, finally, I can design a more simple, conversion mechanism, which is additionally mountable, just for the elevator. (When I began, my very first step was that. )

Mark wood:

--- Quote from: Istvan Travnik on November 12, 2021, 04:02:11 PM ---Wonderful curves!
Thanks a lot!
I think, almost nobody will try to make in real life this "extremist" and too hidden arrangement, but the other one, which one built into my 2012 model, can be built more simple, and freely variable.
...and some pilots, who feel himself a little bit "woodenhanded", like me, can find it useful, helpful, to fly better.
Since my 2012 model is broken now, I could make some photos, showing input and output junctions.
And, finally, I can design a more simple, conversion mechanism, which is additionally mountable, just for the elevator. (When I began, my very first step was that. )

--- End quote ---

Thanks for the photos Istvan. I had already made some assumptions on how your system was connected and I may have not understood the connectivity. The solution I evaluated looks more like a logarithmic function. I will return to that calculation and reevaluate it.


Howard Rush:
Here's a discussion of similar mechanisms:  https://stunthanger.com/smf/stunt-design/interesting-bellcrank/

Navigation

[0] Message Index

[#] Next page

[*] Previous page

Go to full version