PDA

View Full Version : trim



hercules
01-11-2009, 05:08 PM
Hello all,
first i wanna say sorry for my poor english...
We build a trim machanism for my control columns and i want show you the first results.
On the right site of the picture you see the motor. It has a gear (145:1) and drives the plate. In the middle of the plate is a hole. Through this hole comes a pipe that is mounted to the columns. There are 2 dumpers mounted on flexible arms. When the motor moves the plate, the dumpers pressing to the pipe and moves the control columns. This is the trim function.
You can move the columns manuel at ervery time. In this case you work againts the force of the dumpers. The plate wont move, beacause the motor is not turning. So you have manuel control at any time.
Iam not sure yet how i should connect the trim to the sim, beacause there is one poti thats work fo both movements..the trim and manuel.
Any ideas are welcome:-)
Best Ragards
Thomas

Michael Carter
01-11-2009, 05:10 PM
Nice work Thomas!

Mike.Powell
01-11-2009, 05:31 PM
Yes, I agree, very nice work.

These sorts of mechanisms are not discussed nearly frequently enough. The complexity stops many people from working on them.

I think more pictures, and information about the type of dampers you chose, why you chose them, and the motor configuration would be welcomed by many. I, certainly, would find it interesting.

hercules
01-11-2009, 06:00 PM
yes thats right, this kind of sim parts are very rare to find on sim-builder-sites!

Since my english knowledge is not the best in town, its difficult to describe such technical things. I will try it...so dont hit me if it sounds like banana:-)

The problem with trim is to move the control colums with a trim motor without loosing the normal function of the column. The motor will allways blogg the column if you dont find a solution to decouple the trim from normal capability to move the yoke manuel.

So i had the idea to take a plate and link it to a gear motor via a threaded rod. Now you could move the column with the motor in a very fine way, but the old problem is still exist....the colum ist fixed and blogged by the plate or better the hole it the plate.
When the part of the plate thats make the touch to the column isnīt fixed you can move it with your hands. For this "touchzone" i use the dampers (200N). When you move the column by hand, the middle pipe gives this movement to the little arms and from here it goes to the the dampers.

Thomas

hercules
01-12-2009, 02:35 PM
Hello again,

can somebody help me to get something clear in my head?! The controls are connected via 2 potentiometers (1 for Ailerons, 1 for Elevator).

When i move the trim motor the elevator poti gives the changes to the flightsim. But i think that the trim need a own poti. I imagine that the trim position works with a needed- and existing position thats registered with a this poti. The 2 values are compared and when the existing position differs from the needed position the motor moves the column to the correct position.

When i use the same poti for trim and normal movements it is not the correct way.

Anybody have some ideas about this??

Regards

Thomas

Michael Carter
01-12-2009, 02:45 PM
I don't know how to solve your problem, but this might help understand the system a bit.

When you use the trim switch you are trimming the entire horizontal stabilizer. When you are using the yoke to fly it, the elevator tabs on the horizontal stab are moving.

Two separate but related systems.

hercules
01-12-2009, 04:04 PM
Michael,

yes i know the way trim works in real. I guess the problem is easy to solve.

I try to explain my problem on another way (to LA:-))

Without trim capabilities, the yoke construction needs one poti for each ruder (elevator and Ailerons). It means that the elevator poti catch the movements of the column (back-/forward). The additional trim-motor does just the same...it moves the column for- and backward, so the elevator poti catch this movements also! But this poti is not linked via FSUIPC to the TRIM offset, and will not have effects to the trim-offset inside FS9, but to the elevator offset!
I think its important to use the trim offset additional (takeoff warning and so on). But when i add an additional poti to the construction, i cant prevent that both potis catch the column movements and so the sim wont work correct. It gets nose down /nose up values from boh potis at the same time.

For that problem i need a solution.

Hope my crasy english is good enough to explain it:-)

Regards

Thomas

Michael Carter
01-12-2009, 04:13 PM
Youe english is much better than my Deutsch. :)

dc6driver
01-12-2009, 05:30 PM
Hi Thomas,

I believe you need 2 pots, 1 for the primary flight control (Elevator) and 1 for the secondary flight control (Elevator trim). Attach 1 pot to the column movement for the elevator and the second pot to the moving trim plate for the trim function. The aileron will need the same setup.

Fritz

Trevor Hale
01-12-2009, 05:36 PM
Your engineering is incredible.

I agree, attach a pot to the trim plate, and attach the elevator pot to the column.

Great stuff.

Roland
01-12-2009, 10:27 PM
I agree, Very nice mechanics.

For elevator trim in FS9 or FSX, it is hard to make it behave like the real world trim, as FS build the trim function for mechanical joysticks, which have their zero force setting always in center position. So when adjusting trim, FS actually adjusts a bit of elevator offset, and the stick can be brought back to the center while the plane will be flying level with the extra elevator ofset. Not exactly like the real world trim.
See also http://www.fs-force.com/help2/trim_system.htm. To make it more realy world, Dirk modified the trim by intercepting the pilot trim from reaching FS, and only use it to shift the zero centerforce position of the yoke forcefeedback system.

I have also experimented a bit with force feedback trim, and you need a fair bit of experimenting to see what works best for your setup.

Roland

AndyT
01-12-2009, 11:40 PM
I would place the other pot on the plate and connect it to the motor screw. Then when the motor adjusts, so does the pot. You can use the signal in a feedback loop to control the position of the plate that way.
Turn the trim wheel and the plate moves.

Wonderful design.

hercules
01-13-2009, 01:20 PM
Thanks a lot for all replys. Nice to see the big interests in this meachanics:-) My friend marcus has all the needed machines on his job and he support me in a great way:-)

Yes, i believe that a second pot is what i need, but there is a big problem. The second pot would be linked to the trim-offset of the fs! The first pot is linkend off course to the elevator-offset.

As far as good, but the first pot (elevator) does register all movements of the columns....also that movements that was initiate from trim motor. So two commands will send to FS at same time...one from each pot.
In other words....when the trim moves the column...the elevator and the trim would move on the stab, because both pots turns and change both offsets.

I hope you unterstand what i mean....

Regards

Thomas

Roland
01-14-2009, 10:42 AM
I know it is a bit confusing. The result will depend a bit on the sensitivity setting of the FS-trim value to amount of motor drive and resulting plate shift. As I mentioned, FS does not correctly deal with elevator trim. I'd say, try it out and do some experimenting with sensitivity. You could also do it as Dirk described, and make the the up/down trim switch act on the motor only, not to FS, and only link the yoke rotation potmeter to elevator control in FS.

hercules
01-14-2009, 12:04 PM
Using only the elevator could be a problem, because i couldnīt setup the correct Takeoff-Trim due to the missing trim indication inside FS.

I think about using both potis, one linked to elevator and the other to the trim offset, but this with dissabled trim-effects in the aircraft.cfg.

So i would have both: Climb/ Descent with elevator via the elevator pot and trim indication with the trim pot within Project Magenta without effects to the aircraft.

Could this work?

Thomas

dc6driver
01-14-2009, 05:11 PM
Hi Thomas,

Very nice work. I was wondering if you could post a few pictures detailing the threaded rod portion specifically where the rod goes through the bearing and the connection to the electric motor.

Fritz

Roland
01-15-2009, 08:35 AM
Disabling trim in the aircraft config? Haven't tried that one. Interesting. Not sure how autopilot is going to behave then, since A/P ALT is controlled via trim, and the joystick input is disabled in A/P mode.

hercules
01-17-2009, 06:06 PM
Hi,

I found a solution for the problem! Its so easy that i can`t believe that i ditnīt saw it sooner.

The elevator pot is a rotating pot, that is connected to the column via a flexible arm. But the pot itself was mounted to the main wood-plate far away from the middle section of the coulmns. Now i changed its position direct on the moving trim-plate and the flexible-arm is mounted to the middle (shorter) column arm. You know this arm from my pictures!

When i now push or pull the column, the elevator-pot does turning like before! But when i now move the trim plate to the correct trim-position, the elevator pot goes automaticaly back to its center-position, since its base plate is the trim-plate. In other words, the trim plate neutralize the pior column movement, because the trim movement is always the opposite direction of the prior column movement. I know...its difficult to understand, since my english is strange, but will upload pictures as soon it is ready for a photo-shooting:-)

From now i can move the trim mechanics complete independently. The one an only "Bug" is, that the column position of a real aircraft does not change during trim action! But my construction does it....

Is this a reason to cry??

Nice sunday to you

Thomas

rsandor
01-21-2009, 02:52 PM
Hi Thomas,

I have an idea how to overcome the problem of simultaneous movement of pitch by both potmeters. What if You mount the pitch potmeter to the previous position (where it was originally), and replace the trim potmeter with a stereo one. Now if you connect the second Chanel of the stereo potmeter in parallel with the pitch one (maybe with some current reducing resistors on both ends), to result pitch down when the other Chanel on the same potmeter commands pitch up trim. I`m sure that some experimenting with resistor values is a must (too small values will result in over steering, and vice-versa), but... Is this understandable, or I have tho draw a quick schematic?

Robert

hercules
01-22-2009, 08:27 AM
Hello all,

thanks for the tip with the stereo pot, but as i wrote before, i found a solution that is much easier as you can see on the attached pictures!

i made some more detailed pictures now.

On picture e.jpg you can see the arm that is connected to the elevator-pot an the middle column pipe. When i push/ pull the column this arm moves the pot forward/ backward. The force on on coulmns i allways neutralized when the position of short part of the arm its 100% vertical. When i push the column the angle of the arn moves some degrees. Now i have to move the trim-plate as long till the arm i vertical again. Just now i can leave the hands from the yoke without lost controls. Its a great feeling to feel the force change during trimming..:-) You can also see the slider pot for the trim.
Next step ist the trim wheel. I use a multiturn pot for it and i let comepare the values of the trim-/ Wheelpoti with sioc! When the trim pot value i different than the wheel pot, the trim motor moves until the values are not different.

Thomas

Thomas

hercules
02-01-2009, 07:06 PM
Hi all,

i just wanna let you know that the trim mechanics works now exatly how it should. You canīt imagine the great feeling when you "trim away" the forces coming from the system when the a/c is not trimed correctly.

Addional i realized the A/P function! Since it was a big problem to fly in A/P mode and deactivate the A/P, cause the trim-poti is not in sync with the A/P.
It a horrible situation on final approach when you dissengage the A/P and the A/C goes of control, cause the trim is far awy from the A/P trim.

It just the same like missing A/T lever servo. When you dissengage the A/T the lever wont be in in correct position, so you allways running in a trouble situation.

I wrote a SIOC code, that moves the columns and hold them in Sync with the A/P trim. I does not move endless. Only when difference between the A/P trim value and the trim-poti is more than 0,5, the motor bring it back in Snyc.

Thomas