DougM's latest build - Sand Runner

[DougM]

Bearings are here!!

This is how they interface to the driven gear (prototype here)

I’m 3D printing the bearing housings right now

And the rims are here:

Aside from the part about, you know, designing them and printing them. I’m definitely looking forward to cracking a beer tomorrow night and designing rims

Over the weekend my goal is to make the inner and outer plates and actually put a full gearbox together. Probably I’ll make the plates out of a sheet of PVC rather than commit to aluminum.

Still outstanding are the truck mounts. They are completely designed in my head but I hope to have them sitting on the table by Sunday night.

Also, bad news, I ordered the wrong thickness of aluminum for the plates. I meant to order 0.200 but I accidentally ordered 0.250, which is too thick (and surprisingly heavy). I know how it happened, I went back to the website to check my order and I noticed the price, per sheet for 8" by 7" for the .250 is $11, and for .200 is $26, so my little brain chose the right price but the wrong size Now I have to decide if I’m just going to use them or if I’m going to try to mill them down, which is not really a strong suit of my particular machine. But given enough time it could do it; so still noodling. I could go with 0.125 which I have lots of, but it seems like it would be too flimsy.

[DougM]

The plastic wheels arrived

I actually quite like them, they are decent looking and very tough - I literally can’t flex them. So I’m going to continue down the path of using these wheels before attempting to use the rubber tires. Weirdly I ordered 4 and only received 2. I’m hoping the other 2 are out there somewhere and will arrive next week.

I started this morning with parts and a dream and ended out the day with a working (still with the prototype driven gear) transmission. I’m kinda pleased.

Here are probably more assembly piccies than you want to see.

Step 1, cut the inner plate out of PVC before you commit to aluminum. This worked out really well because I made 3 mistakes, 2 of which you can see here. The center hole for the motor mount should be oblong similar to the screw slots, and the screw slots are too wide for M4’s. I made them for 10-32’s but obviously that’s not what the motor takes.

The third mistake I didn’t catch until after I’d cut the plate in aluminum, and that is that the motor has a snap ring on the outside of the driveshaft, and even oblong that center cutout wasn’t big enough to clear the snap ring. So I had to do some manual intervention. Fortunately nobody will ever see it

The next question I know you’ve been meaning to ask is, Doug, how are you going to keep this transmission from exploding if you’ve 3D printed the surround? Won’t that delaminate under stress? Well yes it will, but I have a plan.

If you line each hole with aluminum rod that’s been drilled and tapped, then screw into that you put the housing in compression rather than in tension.

These things are stupidly labor intensive to make, but they work great. I’ve used them before on electronics housings for my other boards.

Now press the big bearings into their carriers (PLA, 3D printed) and add some heat-set threaded inserts (10-32)

Stuff some parts in

add a motor and a drive gear

Bolt it all together

and spin it up!!!
It’s really loud. Hopefully with the production gear and some lube in there it’ll quiet down a bit, but I won’t be sneaking up on a lot of seagulls with this thing.

So now that I’ve proven that the thing works, doesn’t explode and is more or less correct, go ahead and commit to aluminum plates. In the end I decided to go with 0.125 since it really doesn’t span a long distance, so is plenty rigid.

And here is the (almost) final product!

The really eagle eyed amongst you will notice that the 3D printed housing has 8 screw positions, but the plates only have 6. That’s because, um, I screwed up. But I’ll fix it for the other side and may go back and fix it for this side if I get the need.

I also got some work done on the truck mount - here I’m 3D printing it for initial fit testing

[HarryForks]

Loving this build. A smooth wet-sand beach must be such a glorious place to ride!

[DougM]

@HarryForks it is as close as you can get to snowboarding - if you do it right you can slide the back out somewhat predictably and if you fall it doesn’t hurt as much as concrete

[DougM]

It’s hard to get much work done during the week but I managed to get a bit of redesign done on the housings. I was going to go back and just drill the holes for the screw bosses that I forgot on the gearbox, but it turns out that the screw would have interfered with the motor covers, so I couldn’t do that. So for that gearbox I opted to ignore it in the hopes that it won’t be a problem

However, for the second gearbox I went ahead and updated the 3D printed surround to move the bosses to the outside so they wouldn’t interfere. Here they are side by side

The one on the left is the new one.

As you can see the second motor came in today as well as the grease.

Next up is the motor mounts. Probably won’t have a chance to make them until the weekend. Hopefully that giant block of Acetal will make it here before the weekend or I won’t be able to make the driven gears.

[DougM]

If you’re ever wondering what a $54 block of Acetal looks like

Hopefully by the end of the weekend this will be two driven gears.

[DougM]

Ok, so that block of Acetal is too big to fit on my lathe, too big to fit through the band saw, and much too dangerous to try to cut on the table saw, so I had to go old school

But I mananged to get it cut into 2 halves and loaded into the CNC machine

You can’t really tell from the picture, but I cut a 6.15" hole in a block of 3/4" ply and the acetal is inset down into the plywood hold. Then holding it down is easy because I don’t need to worry about it sliding around.

Outcome is that I got the first set of holes into the things

However, as you know if you’ve ever worked with rotating components as soon as you remove something from the fixture you completely lose registration, and therefore you lose concentricity. The neverending battle is to get as close as possible to concentric - if you don’t the gears won’t be quite round and your interface with your drive gear will have a weak spot where the teeth are not fully engaged.

In my case the center hole is not quite round, so I can’t just stick the wheels on the lathe and start cutting the sides down, I need to build a fixture that will make use of the small threaded holes. So that’s tonight’s project - a highly accurate lathe mount plate. Designing it. After that it’s pizza and beer all the way down.

I also got the motor mounts cut and attached to the gearbox. It took me a long time to figure out exactly what the profile of the MTB Matrix truck is - it’s really weird, so I kept modelling it, then 3D printing a little sample, checking fit, modelling again, printing, checking, probably 10 iterations before I finally got it good.

so pretty soon I’m going to need to start thinking about the frame. Here is a pointer to my mountain board build where I talk about the frame a little bit. I’ll be using the same extruded rail design for this one.

[sesat]

thank you so much for sharing your progress in this incredible project

just curious why did you choose to cut out the holes in the acetal before hobbing the gear teeth, rather than the other way 'round?

[DougM]

@sesat good question, all of the initial steps don’t have to be that accurate. My CNC machine isn’t accurate enough to trust with final dimensions of rotating components, I need to do that work on the lathe.

So the four relationships that are critical are the concentricty between the threaded holes and the inner bearing, the outer bearing, and the teeth.

The next step will be to build the precision plate that will allow me to mount this thing on the lathe, then I can turn down the sides. As you can see in these pictures the gear isn’t just flat, it’s got an outer rim and an inner rim that slide through the bearings. All that needs to be turned out of that big block of Acetal.

So the short answer is the things that need to be super accurate go last because I have the best chance of not screwing them up then

[Venom121212]

Damn dude nice work. I have the profile of the truck I could have given you

If you need the baseplate for any reason, lmk

[DougM]

@Venom121212 if you’ve got a .STEP or SLDPRT of the entire truck that would be fantastic!

[DougM]

Ok, so here’s my question for all y’all:

These motors

Maytech Brushless 6374 170/190/220KV Sealed Cover Sensored 8mm Shaft Motor...

are advertised as IP54, which is “limited dust ingress”. Do you think they will resist beach sand intrusion? That would save me a lot of time.

[Venom121212]

I made the baseplate, @creavenger provided me the entire hanger which was nice so I’ll have to make sure it’s cool with him before sharing but I’ll pm you the baseplate in a bit.

[Movation]

You are creating a thing of beauty. But my trashbin called and it wants it’s wheels back…

[michondr]

great job man, beautiful craftmanship

lookig forward for the final result

[Creavenger]

@DougM Ill upload the files to thingiverse today or tomorrow

[DougM]

Ok, well it’s been a really long day, but I got one of the transmissions completed. I had no idea it was going to take so long to make the driven gear.

Caveat I am not a machinist. If you are a machinist you probably already know this

First, as I was talking about last night, I made a mount plate based on the bolt pattern on one side of the blank, and machined the outside flat so that if there were any wobble it would be in the blank and not the plate.

I have one of those little Taig CNC micro mills which is accurate enough to make this part - the key to the success of the whole thing is that the bolt pattern has to be true to the rotational axis or I’ll never get the transmissions attached to the wheels. The downside to these mills is they are quite small so I couldn’t just do the whole wheel on it.

Then, for no particular reason I tested the wheel (I guess just to see how far out my CNC router is) and it was out by about 13 thousands, so that’s not bad across 6 inches, but not good enough obviously.

Then I machined the outside flat…

Now I have 2 reference surfaces, the inside and the outside. I then machined the inner circle true

Now I have 3 reference surfaces that I need to do the whole thing, so I just went to town on it.

Here is one side machined and with the bearing and bearing housing in place

and here it is on the gearcutter getting its teeth cut.

Then I put the whole thing together, ran it for a while to break it in, took the outer cover off, cleaned out all the dust, greased the crap out of it and put it back together.

Unfortunately I won’t be able to get to the other transmission until next weekend. So I didn’t get as far as I wanted to, but I will admit that I took the eMountain board out late in the afternoon for a couple of hours of crashing around the trails.

[HarryForks]

I was wondering about this, eager to hear more about he the reliability of this rating.

[sk8l8r]

was watching some trucks on discovery channel and could not take my eyes of those tires… if only we could get some baby versions

they reminded me of this thread

[Venom121212]

I have those link saved for a rainy day if you want to try.

Nvm they changed the damn link on me. Great.