Rotating Chair Project Progress

[Karl Sainz Martinez]

As Mike suggested in my previous thread, I took some progress pictures of the chair project. I thought I'd put them into a new thread, and explain the project a bit more while I'm at it.

I am a PhD student at Reading Uni, formerly at the now defunct School Systems Engineering. For my research I have found myself needing to pop people on a chair that can rotate by itself using an electronic motor. Since the university doesn't seem to have such a chair, and since it saw fit to dissolve SSE and demolish the entire engineering building, workshop and all, I've found myself having to improvise.

First things first. This poor little motor was at some point separated from the mobility scooter it used to be part of, so I was only too happy to adopt him for this project:

Not too shabby!

Next, the chair. I was originally planning to use an Ikea swivel chair I had lying around. However, as I was finalising my plans and preparing to bring it over to be welded/drilled, I realised that welding onto a pressurised gas-operated lifter chair was probably not the smartest thing to do. So instead, I went out and picked up one of these:

The chair bit is probably not the best, but the base was somewhat solid, hollow and most importantly not explosive. Technically not a swivel chair, but it seemed easy enough to convert. The plan was to weld a bike sprocket to the top shaft, then drive that from the motor using a bike chain. For that, the motor would have to be lifted up so that the attached cassette was level with the end of the bottom shaft. I figured the easiest way to do that would be to construct some wooden blocks to prop it up. Arthur very kindly offered to show me around the table saw, and with his help I put together a little platform for the motor:

That big aluminium plate came with the motor and looks quite complicated to remove so I figured I'd make the best of it. The motor itself now rests on a solid block of wood built from layered offcuts, and the plate is bolted to two lighter plywood blocks for stability. I also made two little wooden sleeves for the chair base to hold it steady and maintain a constant distance between the chair shaft and the motor.

My plan of welding the sprocket to the shaft fell apart when Arthur pointed out that the sprocket was stainless steel, but the shaft was mild steel. So instead I hacked this together using a spare sprocket and the hex key that came with the chair:

Quick and dirty, but it works!

So, where are we now? Everything seems to turn fine, but the chair is a little wobbly. At the moment, the points of contact are the screw heads at the bottom of the sprocket assembly, which rest on the plastic at the top of the lower shaft. Again, this works but isn't ideal: the top shaft wobbles, and under load the screws dig into the plastic (which isn't good for something meant to have a person's weight on top). My current plan is to fit some kind of flat metal disc to the bottom of the sprocket assembly, that the screws fit into nicely and that will offer a smooth point of contact with the top of the bottom shaft. That should reduce the wobble and remove the need for the smaller sprocket in the assembly (as the hex key would be actuated by the metal ring. I have an STL file for this disc, and was hoping to build it in the downstairs metal machining... machine. One of the thinner squares of aluminium would be enough.

Once I've fixed that, onto the electronics!

 

[Karl Sainz Martinez]

Thanks to Alex, I was able to switch out the chain support plate (previously a sprocket wrapped in tape) for a laser-cut acrylic disk. Still prevents the chain from derailing from the chair sprocket, but now it actually looks decent and holds the transmission lever in place nicely:

Also on the laser cutting front, following Rupert's advice and thanks to the ever-helpful Arthur's assistance with the laser cutter, I fitted a series of rings to the upper chair shaft, which made it fit nice and snugly into the lower shaft and greatly reduced the wobble:

Everything seems pretty good on the mechanical front now. A thrust bearing was suggested which I'm looking into, but I'm not entirely sure how I'd fit that. It would be helpful though, as at the moment the whole weight of the person is resting on the sprocket assembly which is hardly ideal.

[mikethebee]

Looks great. Thank you for posting the pics, really good to hear how suggestions are helping improve the design, - Mike

[Karl Sainz Martinez]

More progress (pics to come soon)! I have installed an arduino, motor driver and power supply. The chair turns both ways at a good range of speeds, but under load the motor pulled itself up by the chain instead of turning the chair, so I relocated the shaft and placed it on top of the motor plate so the person's weight stopped that from happening. So far so good, but putting the chair shaft on the plate has meant not being able to use the original lower shaft for the chair. I built a replacement out of wood, adding a bearing to the bottom, but (surprise surprise) it wobbled! In fact it wobbled so much that the chain would very quickly derail under load. Thanks to the helpful suggestions of Ian and David, I decided to scrap the wooden bit I made and put in something a little more robust. Ian instructed me in the use of the lathe (I swear, by the end of this I will have used every single machine in the downstairs area) so I took a big aluminium cylinder and bored out some holes for the shaft and 2 bearings. Also on the lathe, I made a little sleeve out of some scrap steel to go between the shaft and the upper bearing for a nice and tight fit.

TL,DR: Finally, I have a robust, non-wobbly, bearing-equipped aluminium holder for the chair shaft! Now I just need to attach it to the motor plate. Both of them being aluminium, I'm told TIG welding can do the job. Anybody able to teach me how to do that?

[dz]

glad it worked ;) 

[Karl Sainz Martinez]

And here are the pics:

Lathe-worked, bearing-equipped cylinder to hold the chair steady

The electronics box, containing an Arduino Uno, a motor driver and a power supply

The control unit, with an emergency stop button (top), a rotary encoder for participants to report their relative position (mid) and a 3-way switch to turn left and right (bottom). I should really find some covers for the controls.

The new gear assembly for the motor. Much more levelled than the previous iteration, plus you can now adjust the height!

Not pictured is a photo-interrupting encoder + wheel that I have yet to fit to the chair shaft to provide closed-loop position and speed control. Will post up a picture of the complete chair as soon as the welding is finished!