WillChair Three - Let's Build it!

[Williamclark77]

Lube up your modem.

I learned a lot building the first two. Number three will have a LOT of minor design improvements and a few major. It'll be slightly more compact, stronger, 30% more battery capacity, REAL four wheel independent suspension, gear drive, tilt, leg extension, 6.5" of ground clearance, yadda yadda.

It will have a touchscreen with speedometer, odometer, and trip reset if I can jive my Arduino programming well enough.

I already get 38-42 miles and charge about once per week on the W2. Range varies by a huge amount because I use mine like an ATV or utility vehicle. This one has 31ah more capacity (I haven't capacity tested the cells yet though)

The custom brushless motors on W2 are nearly perfect. These should be as smooth as brushed down low and nearly silent. They have encoders and will use sinusoidal commutation. The trapezoidal growl on my first two builds can get annoying in a quiet place. Fyi - Custom motors in low quantities ain't cheap. $3,865 and took about six weeks to get. Being a gimp is expensive!

This build is not a rush. So, updates may be few and far between.

Zero wheelchair parts on the whole thing except for the cushion. Most every part started as raw billets or sheets.

Caster barrels. From 2" Schedule 80 seamless pipe to





Yes, those are 3/4" id tapered roller bearings as on my W2. They'll last forever.



Rear wheel hubs.











Back in the lathe to be trued up perfect







Final reaming to size. And no, you shouldn't use a straight flute reamer with a keyway. Feel free to buy me one for my next build.



Stainless studs welded in place to ease assembly and a coat of pretty color rust preventative These don't look like much, but they're as accurate and concentric as the wheels on your car, if not moreso.



Snazzy. I also made tight fitting aluminum washers that are on it.





Tilt pivots and armrest pivots. Lots of work there. Oilite bushings press into them.







Swingarm and suspension mounts. Already ground the weld off here. You can't tell it was two pieces.



Mount for the tilt joystick. Neat little piece.



The rod ends for the rear suspension mount on these



Pivot point for the rear suspension. That slot is just to locate the plate the motor goes on.



Motor mount plate. This is how most of the flat parts were made. Rough cut oversize by saw then bolted to a board to be CNC machined out. Very hard on tools and not easy to figure out how to hold.







That slot in the middle is to locate the motor faceplate and allow a full penetration weld with less heat.



Like so. It was made the same way.



I'm not pro welder but she ain't going anywhere.



Fits perfect. Same process.







Gussets added.



The pins they pivot on. These pressed in then welded on both sides.



Ignore the apprentice marks



Front swingarm mounts

Big expensive chunk of aluminum.







Put in the lathe to drill a center hole to home the CNC by.





Back side finished.



Flipped. I make that tool also. It goes through aluminum like butter.







A lot of labor represented here.



Front axles. That's precision ground and hardened 1055 shaft. I just cut it to length then drilled/tapped the ends.



Spacers









Coping the swingarms. I don't own a proper sized notcher and can't physically operate a tubing notcher if I did have one. But, I do have CNC and know how to program 3d shapes. A notcher would not have located as accurately anyway. These came out perfect.

Short video of cutting the copes.

https://photos.smugmug.com/photos/i-7FzkP7k/0/2743ec20/1280/i-7FzkP7k-1280.mp4











After the bearing races were pressed in. I sprayed the inside first with primer to prevent rust. I don't ever plan on removing them. They can be taped for painting.



Part of the tilt mechanism



She got hot cutting. Endmill getting dull too.



Parts of the caster forks. It takes some time to figure out how to arrange them all, hold them, add support tabs in the right places, and not crash your tool into anything.



It can get sketchy



Short video cutting some similar. I slowly ramp down. It takes several loops to get all the way through. You can see the tool rise to leave tabs in spots. Without them the part will fly out as it cuts through - that's if it doesn't vibrate enough to break the endmill first. You can break the tabs with a crescent wrench and vise.

https://photos.smugmug.com/photos/i-TdmvPq3/0/0905bbcd/1280/i-TdmvPq3-1280.mp4

Jig for the caster forks. They came out great.



Short video power tapping it on the lathe

https://photos.smugmug.com/photos/i-SxM8qhw/0/367e984a/1280/i-SxM8qhw-1280.mp4

Video of how I bored and countersunk the bolt holes.

https://photos.smugmug.com/photos/i-V4LMXb9/0/b79ab6ee/1280/i-V4LMXb9-1280.mp4

I only have 9x18" of travel. Making the 10x22" side plates took some thinking ahead and four setups. They came out closer than can be measured with calipers.



Shock mounts. They'll be welded on.



Short video machining the shock mounts. This is cool to see. The endmill is getting dull plus that steel was very hard is why it was glowing orange.

https://photos.smugmug.com/photos/i-2HQ7HTf/0/a9a934f5/1280/i-2HQ7HTf-1280.mp4

This is how you tap 20 holes in three minutes. And no, it doesn't need to retract that far and can go faster. I need time to stop it quick if the tap breaks.

https://photos.smugmug.com/photos/i-58pxFGb/0/a5b50cb2/1280/i-58pxFGb-1280.mp4

There's a buttload more that I didn't take pictures of and lost a lot due to a failed memory card. I should have the battery box buttoned up into a rolling assembly this weekend if I can get a hand. I have other projects too so it may not happen.

Feel free to ask any questions.

[Burgerman]

Wow. Very industrious. Were you tempted to do double the parts? 2 at once isnt much harder than 1.

[Williamclark77]

It's definitely much cheaper and not too much more effort to build two. If the motors, gearboxes, and lithium cells weren't so expensive I would. Lots is prototype and may need changing along the way.

My W2 came out nice with a fit and finish better than any factory chair I have seen. It still looks pretty good and performs great after several years of HARD use. This one will too.

[flagman1776]

Stunning workmanship. And I'm sure that the planning is several times the work.

[Burgerman]

Since you built the first ones, lithium has moved on a little. Its now possible to fit 16 prismatics in the space of the olde grp 24 batts. Giving a 5x greater range, from a real 210Ah. Or 105Ah at 48V.

[rustyjames]

Wow, once again speechless! When do you find the time to do all this? It's a big task producing all these parts, yet alone documenting the process. Thanks for sharing, Will.

[shirley_hkg]

The custom brushless motors on W2 are nearly perfect. These should be as smooth as brushed down low and nearly silent. They have encoders and will use sinusoidal commutation. The trapezoidal growl on my first two builds can get annoying in a quiet place. Fyi - Custom motors in low quantities ain't cheap. $3,865 and took about six weeks to get.

Feel free to ask any questions.

Have you tried the sinusoidal sensor mode yet, Will ?

Hope to know how it performs .

[Williamclark77]

Flagman, thanks. You're also right. The designing and 3d CAD modeling took longer than it'll probably take to build. Laying out the parts to fit the material, figuring out how to hold them, and writing the toolpaths may take several hours and only 15 minutes to machine.

BM, this one is 14S with 91ah cells. Lots of watt hours. It's also using quite efficient 90 degree planetary gearboxs. Should do good with brushless.

Rusty, thanks. My day job slowed down to the point of layoff two weeks ago, which has given me some time to actually do this. I've gotten so used to going at it from 4am until 10pm or later over the past 12 years that it's weird not to be pushed all day. Unfortunately, I'm pretty much at the point now where not much gets done on it without a hand to lift/position stuff.

Shirley, untested. The encoders I spec'd are what Roboteq calls for. It will be quite a while before wiring.

[Scooterman]

Wow!

Does swarf have any scrap value?

[rustyjames]

Will, I can relate to your situation with limited hand function. If I had just one hand with good pinch and grip my life would be quite different. The stuff you're getting done is impressive!

[greybeard]

Will, you're a magician.

[Williamclark77]

Scooterman, none different than the material it was cut from as far as I know. It's pretty light weight for its volume.

Rusty, club hands are fun, huh? Everything I do is an interesting process I had to figure out. I won't do most things if people are around because the way I have to is embarrassing. I get it done though.... Eventually.

Greybeard, I don't know about magician. More like stubborn.

[flagman1776]

Virtually every task I do has to broken down into tiny steps that work around my defisits. People with no impediment have no clue. Your defisits are different than mine, so your work-arounds will be different. The incredible quality of your build... the tollerances, the care are an inspiration to us all.

[expresso]

thats excellent work - puts them to shame - very good !!!!

besides having tools - you have the skills and knowledge - great work

[rustyjames]

Rusty, club hands are fun, huh? Everything I do is an interesting process I had to figure out. I won't do most things if people are around because the way I have to is embarrassing. I get it done though.... Eventually.

I hear you, I spend a big portion of my time everyday picking up stuff that I drop. And I'm the exact same, if someone stops by while I'm working, I stop. Fortunately, I have a few friends that have mechanical talent so I ask them, but just for a easy, quick hit on something that will save me a lot of time. It's never more than a few minute easy task, for functions hands.

I think about the handiest things I keep a good stock of are small blocks of wood, of all different sizes.

[Williamclark77]

Flagman and Expresso, thanks!

Rusty, my better 7/8 is instrumental in me completing these projects. When she has time she helps with the assembly and provides working hands. In turn I do all of the artsy woman BS that I think is pointless without complaining. At least not out loud. A good high quality telescoping magnet is the most useful tool I own. On is stuck under the armrest of my chair at all times.

Shock mounts





Rear motor mounts or whatever you want to call them completed. The bushings are pressed in also.



Caster forks completed



That aluminum piece has two O-ring grooves. The caster barrel housings are water tight. They'll be put on at final assembly. These are about the same as on my W2.



Those are real mountain bike shocks. Pretty good quality. I'm pretty sure I'll need softer springs though. Haven't had time to source any.



The pivot points for the swingarms are above the caster axle centerline. Not that it'll make much difference at these speeds but they should work better at straight-on impacts.



Can't get much more compact than that. The phone camera distortion makes the rear tire look out of proportion.







Anyone who knows suspension designs (or basic trigonometry) will realize there's a bind point in the arc of travel with the single axis of rotation at the front pivot point. On paper yes. I'm well aware and knew it when designing. Yadda. Yadda. Factor in the radius of the links' arc and the side flex needed is miniscule. Without the spring on you can feel it moving the rear suspension. In use it will not be noticeable.



Nothing is adjusted properly yet. 27.75" wide. It was suppossed to be 27" even. They sent me four 2.5" wheel halfs instead of two 2" ones like I ordered. I didn't notice until starting to put the wheels together. If that extra 3/4" causes any issues I'll swap them and discard the tubeless insert to gain even more clearance.



Ground clearance with the shocks setup at nearly the lowest point. It can be adjusted to add another 1/2" or so. Seat pan height is still only 17.2" (tire pressure and suspension sag will alter it a bit)



Don't know when I'll make more progress. I have other projects I need to get to first now that I can roll this one out of the way.

[Burgerman]

I still think that you would get a much better ride, and more predictable and controlled suspention if thosefront arms pivoted from the bottom of that box, inline with the rear suspension mounting point. As it is if it fits a bump its forces go straight into the pivot rather than in lifting the wheel against the spring load.

Rear, not symetrical, would reduce arc swing angles by 2. But wont look as pretty. Work better.
In orange...

Side...
As it stands, when the wheel hits a bump, BLACK line = ground, the forces are transmited almost directly to the pivot point. So initially the shock loads the chairs chassis. And you will feel that. instead of letting the spring take the load. If a bump was large enough it it could actually cause the suspension to extend, and want to lift the chair UP.

Moving the pivot to the bottom, and making it longer (further from the wheel) means the load is mostly on the spring and the caster wants to move up instead of back. My rubbish drawing attempts to show forces. And where shock would go if it was me building.

[Burgerman]

Please dont shoot me!

[Williamclark77]

In a perfect world they'd pivot from in front of the tire! If I hit something big enough to matter the rearward impact direction will not be my biggest problem

I'm out of real estate for much different. Two actuators still have to be fitted. One is big. The bolt holes between the shocks are for it. The bottom of the chair is above the caster axle centerline anyway. No way to completely eliminate it without affecting ground clearance.

Oh, my W2 chair weighed 264 American pounds. This one should be around 20 pounds lighter.

[Burgerman]

Just sayin. It would have a much smoother ride. Theres nothing wrong with leading link suspension. As long as it pivots from a low position. So that the bumps load the spring. But if you dont have the room, no choice anyway.

[Williamclark77]

Please dont shoot me!

Nope. Not at all. I completely agree and VERY much appreciate your input. I've designed enough to know, mostly three and four link drag car chassis. Bring on the criticism. If it can be changed, now is the time.

The rear - completely aesthetics. I drew it with longer links and just didn't like it. If it feels like it will need improvements when i test it I'll make a new back plate and links to try. Not very difficult. I originally designed it with a slide that attached near the gearbox that would take the place of the links. I figured it would pack with mud and stick.

[ex-Gooserider]

Wonderful job - you are doing fantastic work....

I did see a link a while back to an article that suggested a possibly better workholding solution than your wooden board and bolt approach - I haven't tried it, but plan to.....

FWIW the link was on Practical Machinist, and the article was from one of the major mill brands 'tutorial' section - Tormach's IIRC... So it is a reputable source...

Basic idea was to make a flat sacrificial plate of aluminum (fly-cut to very flat) and similarly flatten the bottom side of the workpiece. Then cover each with one layer of blue 'painters tape' (if multiple strips needed don't overlap). Then bond the two parts together with industrial grade cyano-acrylic (super-glue) and clamp tight until cured...

Machine part using similar approach to a vacuum table - leave a thin layer of material at the bottom until the last pass, possibly even treat it as something to be removed in deburring... If careful setting depth of cut, finish pass should just get the tape, and not touch the sacrificial plate - if it does, flycut again for cleanup...

Soak finished part / plate in acetone to dissolve glue and release the part....

The video showed them doing some pretty aggressive high speed machining so it seems worth a try....

ex-Gooserider

[Williamclark77]

Thank you Goose. I'm very familiar with the super glue method. It actually works really well. It's about the best way I know to hold really thin aluminum.

I've never tried it slotting steel like this though. The super glue can't take much more than warm-to-the-touch before it fails. Slotting this steel gets hot. Smoking hot. Flood coolant is not an option with an open machine like mine. I ASSumed glue would fail and didn't even try it. I could be very wrong!

One upside to the wood is, if you look closely, you can see I cut well below the bottom of some of the parts into the wood. I did that to get to the good unused carbide instead of trashing the endmill when the bottom got dull. Same with tapping. I drilled well into the wood to make certain that the tap didn't bottom out. My machine can't rigid tap. I use floating holders. Where the tap stops can vary quite a bit from hole to hole.

[terry2]

That is totally awesome

Get your ass to the UK and build some

What is the size of your rear tires?

[Burgerman]

Same as the ones we use. 15 x 6.00 - 6.

[wheelinghome]

ABSOLUTELY BRILLIANT!

in the first post, second photo- how did you make/assemble those tapered roller bearings? is there a benefit in terms of function/longevity etc in having them tapered?

[Burgerman]

He may not be back fopr a while. But yes theres a benefit. They are intended to take both vertical (weight) loads, and sideways loads too, as the caster itself creates. As well as bumps. Although I might have used stainless ones. But if greased initially well, and very occasionally say yearly they should not corrode anyway.

Bycicles use round ball bearings. Motorcycles use sets of taper rollers. They are stronger as theres a much bigger contact area.

[LROBBINS]

With Will's o-ring sealed caster barrels, I doubt that corrosion of those bearings will ever be a problem - even with him trucking through Mississippi swamps.

[Williamclark77]

Get your ass to the UK and build some

$$$ that I don't have!

Tires are what BM said. Mine aren't the kevlar reinforced ones. Hard to get those in the US. They're the standard 4 ply. I've pulled welding rods and finger sized chunks of metal out of them on my W2. No flats yet in several years.

ABSOLUTELY BRILLIANT!

in the first post, second photo- how did you make/assemble those tapered roller bearings? is there a benefit in terms of function/longevity etc in having them tapered?

Let the page load longer. That pic was about number 25, not 2

Those bearings fit into a hardened steel race that's pressed into the caster barrels. They are MUCH more durable when subjected to axial loads (up/down pressure with the bearings sitting flat). The downside is they are more expensive, larger, must be preloaded with a little pressure, and have more turning resistance (a plus in this application).

To add to what BM said, most caster forks have a lot of offset. This reduces the axial/up/down loads while increasing the radial/side loads. Cheap standard roller ball bearings are designed for radial loads. This helps them survive. My forks don't have as much offset to make the chair more compact with my larger casters, which adds a lot of axial load. Roller ball bearings won't last a week. I tried. These will and have in my W1 and W2.

My caster barrels are water tight, negating the need for stainless. A bit of PROPERLY applied grease (yes, there's a right way to get grease inside them) will do just fine. I've had my W1 in the ocean five or so years ago. Literally. That almost ended badly. Bearings never got wet.

This is my W1. The camera is mounted right at the motor making it sound very loud. Those casters take a beating. You can see the suspension working too.

https://photos.smugmug.com/photos/i-MNZFRpG/0/30b74b2a/1280/i-MNZFRpG-1280.mp4

[Scooterman]

You’re going really fast You is very brave

The front suspension works really well, I can see why you need it

How fast, 10mph+?