Hi all, designing myself a set of pendular rudders to replace my 25 year old TM RCS that are stiff, clunky, noisy and not that accurate; they also lack toebrakes but that's not the end of the world.
I'm limited with space and want to retain my T3PA at the same time so the TM TPR layout with the central tower is no good. Similarly I'd like to clear the floor area so I have a proper heel rest for driving so a repeat of the floor mounted parallelogram design wouldn't be a step forwards. I want the centring spring mechanism to be easily adjustable and capable of being backed right off so there is no centre detent and the pedals will stay in one position for realistic helicopter usage. I also want the pedals to be damped to an extent and run on proper bearings where possible.
Finally the must attach to the vertical uprights on my desk.
So with my design brief in order I fired up solidworks and started putting ideas down. The design in principle is based upon a combination of the TPR and MFG crosswinds but wholly within my capacity to manufacture the majority of it by hand.
The initial assembly model before refining. Essentially just fleshing out the design to check the mechanisms will work before detailing further.
Added gas struts for the toe brakes, this will give both a spring to push against and a damped return so they don't bang against a stop.
Toebrakes off.
Toebrakes on.
Then added an additional pair of dampers to the main arms/cross beam to act in opposition to the rose joint links. They are both set to be at midstroke in the centre position and because gas struts are hydraulically damped when extending it should act as an effective damper system.
Updated the spring pull rod to incorporate a clevis that I found under my work desk (had been there for about 10 years!). It both simplifies the assembly and manufacture and works nicely.
As the pedals are deflected fore and aft the rose joint links swing the cross beam accordingly. The cam follower roller pushes the cam plate backwards which in turn pulls the pull rod which compresses the spring. The collar on the end of the spring pull rod is threaded so both retains the spring and adjusts the tension.
I'm using hall sensors for positional input into the USB controller board, modelled up the sensor with the flying leads and some 30x6x2 neodymium magnets I had laying about and positioned then to get maximum travel across the magnet to improve the resolution available.
A similar setup is used on the top of the cross beam to give the position of the rudders themselves.
Whilst finishing off the detailed engineering drawings so that I could manufacture it all I realised that I had inadvertently built in a massive hard centre on the cam and it would be horrible to use.
Decided to go for a separate cam profile part machined from delrin, that way if I want to change the profile to adjust the centre detent I can just get a new cam machined.
I also came up with a method of clamping the main 16mm spindle to the horizontal chassis box.
Because it's a dual direction cam with a single pivot follower the two halves of the cam profile need to be different; it's not quite perfectly balanced but at full deflection there is only a 0.3mm difference in spring preload so I thing it will be fine.
Finally for now I've started cutting, marking and drilling the aluminium flat. The kinked sections needed to be done in a pressbrake, the cracks welded up (because I couldn't be bothered to anneal the aluminium) and dressed. Cost me 4 packs of biscuits for my mate to do them!
I've dropped the drawings for the machined delrin components on the CNC operator at work and the laser cut stainless toebrakes should be done next week.
Once I've got all the mechanical bits assembled I've then got to wire up the Leo Bodnar USB controller and get it all mounted to the desk and calibrated. Hopefully it all works as intended
I'm limited with space and want to retain my T3PA at the same time so the TM TPR layout with the central tower is no good. Similarly I'd like to clear the floor area so I have a proper heel rest for driving so a repeat of the floor mounted parallelogram design wouldn't be a step forwards. I want the centring spring mechanism to be easily adjustable and capable of being backed right off so there is no centre detent and the pedals will stay in one position for realistic helicopter usage. I also want the pedals to be damped to an extent and run on proper bearings where possible.
Finally the must attach to the vertical uprights on my desk.
So with my design brief in order I fired up solidworks and started putting ideas down. The design in principle is based upon a combination of the TPR and MFG crosswinds but wholly within my capacity to manufacture the majority of it by hand.
The initial assembly model before refining. Essentially just fleshing out the design to check the mechanisms will work before detailing further.
Added gas struts for the toe brakes, this will give both a spring to push against and a damped return so they don't bang against a stop.
Toebrakes off.
Toebrakes on.
Then added an additional pair of dampers to the main arms/cross beam to act in opposition to the rose joint links. They are both set to be at midstroke in the centre position and because gas struts are hydraulically damped when extending it should act as an effective damper system.
Updated the spring pull rod to incorporate a clevis that I found under my work desk (had been there for about 10 years!). It both simplifies the assembly and manufacture and works nicely.
As the pedals are deflected fore and aft the rose joint links swing the cross beam accordingly. The cam follower roller pushes the cam plate backwards which in turn pulls the pull rod which compresses the spring. The collar on the end of the spring pull rod is threaded so both retains the spring and adjusts the tension.
I'm using hall sensors for positional input into the USB controller board, modelled up the sensor with the flying leads and some 30x6x2 neodymium magnets I had laying about and positioned then to get maximum travel across the magnet to improve the resolution available.
A similar setup is used on the top of the cross beam to give the position of the rudders themselves.
Whilst finishing off the detailed engineering drawings so that I could manufacture it all I realised that I had inadvertently built in a massive hard centre on the cam and it would be horrible to use.
Decided to go for a separate cam profile part machined from delrin, that way if I want to change the profile to adjust the centre detent I can just get a new cam machined.
I also came up with a method of clamping the main 16mm spindle to the horizontal chassis box.
Because it's a dual direction cam with a single pivot follower the two halves of the cam profile need to be different; it's not quite perfectly balanced but at full deflection there is only a 0.3mm difference in spring preload so I thing it will be fine.
Finally for now I've started cutting, marking and drilling the aluminium flat. The kinked sections needed to be done in a pressbrake, the cracks welded up (because I couldn't be bothered to anneal the aluminium) and dressed. Cost me 4 packs of biscuits for my mate to do them!
I've dropped the drawings for the machined delrin components on the CNC operator at work and the laser cut stainless toebrakes should be done next week.
Once I've got all the mechanical bits assembled I've then got to wire up the Leo Bodnar USB controller and get it all mounted to the desk and calibrated. Hopefully it all works as intended
Last edited:
