Showroom Full-Frame 2 DOF, 80/20 Strut, JRKs, Wind, Flying/Driving Rig

It's time to make the build thread now that it’s almost done. Still need to mount the Warthog throttle quadrant but it’s basically finished. It’s been a process and there’s been lots of revisions along the way. Tried lots of things and most worked and some didn’t work so well. Didn’t anticipate a number of things and had to redesign my way out. Learned a lot from all the people here and browsing all the build threads.

It’s been a lot of work but the design/build process has been a blast. The way I built the rig turned out to not be inexpensive going with the 80/20 strut, but the cost was spread over about 9 months so it wasn’t too bad overall but I didn’t keep track of the costs, both because I didn’t want to know but also because once I started, I wasn’t going to stop. I wanted to both fly and drive and the changeout to be quick and easy. It just takes a couple of minutes and really is easy.

Features: Full frame, 2 dof, wind, seat belt tensioning (needs more power but the design works), 4 Buttkicker Minis under the seat with two plate amps for right/left vibrations, reclineable seat is on rails for adjustability and balance, bar graph ammeters for setting balance, steering wheel adjusts forward and back for driving and to clear the joystick, joystick base slides out of the way when not in use, rudder pedals and driving pedals swap out, JRKs mount on fan-cooled heat sinks, chokes help block back-emf from motors, total non-contact Hall sensors, 45 mm C-C output arms pinned and clamped.

There are a number of compromises but that’s how it goes. It’s not perfect but a whole lot of fun. Still learning how to properly set up SimTools but flying and driving in VR is a blast.

Will post links to other threads like the bar graph ammeters and stl files for the printed parts, but this is my baby. The only thing that gets even warm are the plate amps for the Buttkickers. It seems sturdy but the proof will be in how it holds up long term. Will post more info, pictures, and a video later.

The bar graph ammeters have turned out to be surprisingly useful. Between those and my seat being on adjustable rails, it doesn’t matter what size person is trying it or what seat back tilt, all I do is slide the seat forward or back to null out the motor current and minimize the bars to nothing. A sim doesn’t even have to be running. SimTools doesn’t need to be running. I just soft start the rig where it’s at center and adjust to no current at rest. It’s then balanced.

Program to soft start JRK motor controllers:
https://www.xsimulator.net/community/threads/soft-starting-motion-platforms.11051/

Info on the bar graph ammeters:
https://www.xsimulator.net/communit...o-do-motor-current-sensing-for-balance.10510/

More info on the JRK 12v12 and the boxes:
https://www.xsimulator.net/community/threads/jrk-12v12-overheating.10327/#post-133655
https://www.xsimulator.net/community/threads/jrk-power-turning-off.10477/#post-135419

Swapped out the SeaFlo wind fans for TerraBloom fans and love it! They are quiet, put out as much air, and should be much more reliable due to them being dual ball bearing as opposed to sleeve. No Monster Moto or power supply needed. It was a big change in sim quality with the much reduced fan noise and I love the setup. Just need a level shifter. The TerraBloom fans accept a PWM signal so the Arduino code doesn’t even need to change:
https://www.xsimulator.net/community/threads/wind-simulator-fans.14124/

Added some of the .stl 3D print files - the box for the bar graph ammeter circuit board, the mount for the bar graph display, the boxes for the choke coils in the motor power circuits, the Mini DIN clamp to hold the connector on my driving pedals, the ducted and un-ducted Pololu 12v12 mounting boxes, the power entry box, the clamps for the USB hub mounted on the rig, and the parts for the narrow-stance pedal modification for the MFG Crosswind pedals.

Links:
Heat sink fan assemblies for Pololu 12v12: http://a.co/i5btPWy
600 Watt power supplies: https://www.mouser.com/ProductDetail/MEAN-WELL/SE-600-12
The outlet strip from Amazon: http://a.co/fcRAoxD
POWERTEC 4-Inch Elbow: http://a.co/eLEZLjz
14 port USB 3 hub: http://a.co/i5pi3ck
Subwoofer amps: https://www.parts-express.com/dayton-audio-sa100-100w-subwoofer-plate-amplifier--300-802
Power chokes: http://www.coilws.com/images/HF5712-700M-60AH Rev.A.pdf
Hall sensors: https://www.digikey.com/short/q8qj5v

@Jerry Atrick - the bottom frame itself is 24" x 48". The power strip at the back extends another 2" and the wind system and pedal mount overhang adds another 6" at the front. Weight is just a guess but it feels like about 40 lbs on each of the back wheels withe about 25-30 lbs on each of the front wheels. Somewhere around 140 lbs total?



There is some stuff that sticks out a little on the sides though. I was trying to keep everything inside the frame but needed to add some brackets on the outside of the frame to strengthen the motor mountings. Then I had to add some snap-on dress plates to hide the sharp edges and some fake press-fit acorns to dress the nylock nuts that held onto the motor gearbox top mounts. You can also see where I clearanced the heck out of the Hall sensor mounts so that there was no way the heim/rose joints would hit them when the power was off and the system sagged down in the front. I used some spare 3D printed feet from Avenga76's wind generator to be rests for the far front ends of the top frame side rails.



These last three photos show the joystick base slide and the USB hub that’s mounted under the seat but ahead of the cross beam that supports the seat U-joint.

And while the rig has worked out well, the benefit of these threads is as much about what didn’t work, or what was unexpectedly valuable.

What didn’t work was I thought I could get out of machining output arms by printing shells and filling them with composite and all thread for strength. It didn’t work. With the offset of the heim/rose joints, the arms just twisted. Not sure if they would have worked with clevis ends or something that would minimize the twisting moment, but I had to just bite the bullet and just get arms machined out of aluminum.



I also tried to use flexible tubing to route the wind onto the wheel mount table but the corrugated tubing really cut into wind velocity due to turbulence in the tubing. I found 4" drain elbows for cheap and in black to go with the rest of the system. Big difference and now nothing is tied to the wheel platform. I 3D printed thin adapter rings to go on the inside of the elbows and friction fit both the blower IDs and the flow straighteners IDs then secured them with a couple of wraps of black vinyl electrical tape.



The other stuff that worked out really well was the JRK mounting on CPU heat sink fan assemblies with thermally-conductive foam. They run very cool. No overheating at all and the components don’t even get warm.

Something else that turned out way better than I thought it would were the bar graph ammeters. @noorbeast mentions balance a lot and how important it is to have a rig balanced. I think he is absolutely right that it is important. It’s immediately obvious when my rig isn’t balanced and with the seat on rails it was trivial to find proper balance. Just a few notches of seat adjustment out makes a big difference in the forward and reverse currents and I have to believe makes it easier to hit current limits on the side more heavily loaded. https://www.xsimulator.net/communit...o-do-motor-current-sensing-for-balance.10510/

Hi Zed, looks great and those wind tunnels you have mounted must be a blast!

Great job!

Can you tell me the footprint size (floor) and the approximate weight please.

Jerry.

Fantastic job Zed!

Thanks, guys! Was just telling Random Coder how I missed a lot of stuff early on and had to redesign quite a bit as I went. It’s why I had to narrow up the rudder pedals to fit in the frame, why the curved joystick, why the wheel had to go on slides to push forward out of the way, etc. I even started out thinking I could put the motors up front but the top frame design nixed that. Luckily the strut made it easy to rearrange everything. Much to consider in these things especially when trying to keep things compact.

@Jerry Atrick - will get the dimensions but weight will have to be an estimate. It’s not too bad there’s a lot of metal. Lots of detail yet to post.

Zed,

I have no words...only big hugs

Well done, my friend.
Mirko

Wow Zed, great job, I'll follow the thread

Hi Zed, is this your invention
https://roadtovr.us11.list-manage.c...747d896bcd22323cd7&id=aba3ce072f&e=98bb026ed4
Regards Jerry.

Nope. But I do like their name.

@Bastiuscha asked for details on how I mounted my Hall sensors on my pgsaw gearmotors so am adding to my build thread if it might help others.

The Hall sensors I used aren’t like others. They are Honeywell units that are hardened for automotive use. (Link in the first post.) That’s not why I used them, though. I used them because there are no alignment issues, no shafts, no flex couplings, and no shaft galling like so many get with the shafted kind. There is no contact at all between the magnet and the sensor body. You just get them in position to within tolerance and you’re done. They do cost more though and you have to buy the connector and contacts separate and build those but it’s easy.

They are just blocks with mounting ears on them for the actual sensor and then the magnet is what mounts to the pgsaw output.



There you can see the sensors with one mounted on a mount. I’ve thinned up the mount since these photos to give more clearance to the helm/rose joints on my output arms, but you get the idea. The three ears of the mount line up with the three bolt holes on the pgsaw gearbox. These are all 3D printed and I put pockets in the mounts for nuts for the Hall sensor mounting screws. There are also 3D printed mounting studs and nuts that hold everything together at the proper distances/register. It can all be plastic because being non-contact, there is no stress on the parts. They just hold sensors and magnets in the proper locations in relation to each other.

The magnets use some non-magnetic setscrews to attach to mounts I made that slip into the pgsaw output shafts. In these photos you can see the mounting plates are way overbuilt. There are zip files of the new design .stl files in the first post of my build thread and the new design is better. This was one of the first things I designed and printed so isn’t well done but it works. In the last photo you can see the relationship between the magnet body and Hall mount. That plastic output arm was a failure and the standoff’s aren’t in that photo. That’s why the magnet sticks through the clearance hole. In actual application, the face of that magnet is just a millimeter below the mount face.



These show the newer version mounting plate design with the cuts to clear the helm/rose joint. You can also see the standoff’s used to properly space the pieces. Doesn’t show the whole thing but shows the relationships and general idea.

Hey @Zed - Did you machine the motor levers or are they off the shelf? If off the shelf, do you have a link to s supplier? Thank you!

Hey @Archie, a friend of mine machined them for me. I drew it up in Fusion 360. I'm attaching the .stl file so you'll have the dimensions. If you need more info I can open the drawing up in Fusion and get information from there. As popular as these motors are, someone could make some money machining them and making them available. It's a 45 mm C-C on this arm in the .stl file.

They were the only things I couldn't manage to do myself. I originally had designed in the thinned end of the arm to make it fit clevises from McMaster Carr (https://www.mcmaster.com/#Standard-Clevis-Rod-Ends) but then I realized that I needed a ball end joint at both ends. Turns out that with washers, I still needed the ends of the arms cut to that thickness to fit the stud on the ball joints and have the nuts clear the bosses on the worm gear drive housings.

The drawing/stl shows the clamp open but my friend did them fancy so that tightening the clamp would increase clamping pressure without bending the clamp arms. These were CNC machined with hand cleanup but the flats on either side of the output shaft for the roll pins would help with drilling the holes for the pins. I can't say the pins are absolutely necessary but I'm using them and nothing has come loose.

They've been absolutely perfect. They haven't tried to loosen on the output shafts. There is no wobble or play (other than what is inherent in the worm gear itself), and no failure or issues with quite a few hours on the rig so far. He did a good job and I'm really thankful he could do that for me. I found some online machine shops that would to the outline on a standard thickness plate for about $40 each but no thinning at the end of the arm and no roll pin hole which is why I added the flats since I figured I could always just buy a drill press and do those myself.

All of the hardware on those output arms and the linkages to the moving frame are from McMaster Carr (https://www.mcmaster.com). I'd guess you have similar companies in your neck of the woods.

Thanks @Zed - I'm ok with Fusion360 so I'll take a look at the STL in there.
I appreciate you sharing the design!

I have a machine shop near me (laser cutter) so I might see what can be done.

I agree with someone making these, as even the guys that sell the motors can never recommend fitments.
It's odd though as the rods are made to interface with something, surely!

I'm thinking of going back to a foot mount/chassis shaker again as I've improved my build skills over time and any new one I build would be a bit more "finished"

Hey @Archie - any luck? I was looking in my documents folder and found this drawing. It’s rough but has the information. It doesn’t show the flats or the thinner arm out where the ball joint goes, though.

Hey mate - Yes - I sent for a quote at a local laser cutter place and they sent me a quote back that was way too expensive.
AU$418 for both .

I understand it's a one-off job so there is CNC setup etc, but damn... expensive!

Very impressive! You've done some very innovative engineering. The ideas on this forum continue to amaze me! Keep up the good work!

Damn. That’s a lot!

There are online shops where you download their software and draw up the part. They quote online and you just pay and submit to get them cut. As a uniform thickness where all they do is cut an outline and the holes for the motor shaft and ball joint stud and you drill the clamp bolt hole and hole for the shaft pin, it was pretty cheap here in the States. About $50 or so IIRC. AU has got to have shops like that too I’d bet.

And thanks, Frank!

We had guests over last night and I put the guy in the the rig for some PCars2. He's a motorhead with a beautifully-restored El Camino (he bought it new and has owned it the whole time) and he didn’t want to get out. It was pretty funny. They just live here in the neighborhood and his wife went home without him.

And I finally fired up my AMC1280 USB yesterday to start checking it out. Very cool! I guess it’s time to start building a 6 now. First stop is just a mini demo rig using R/C servos to sort out all the settings. Got to decide if I go with worm gear motors or linear actuators for the real thing. Platform or space frame. Damn you, @Thanos

So, as I threatened, I'm looking harder and harder at a 6 dof. Haven't decided if linear actuators or gear motors, yet. But looking at the controls and setup just to keep greasing the downhill slide into a new project and the concomitant honeydos that come with that...

In order to explore this madness, I'm printing out a tiny platform base now for some Futaba S3003 servos. Here's the Fusion 360 drawings. Once I get it all going, I'll post the .stl files.

Edit - here are the stl files: https://www.thingiverse.com/thing:2916371