In the interest of having a static display Deathblade on Thursday, I’ve been pushing up the timeline on parts work which was originally going to wait for summer. The frame pieces have been cut out, but I will only assemble 1 for now. The motor cores will not be wound for the presentation. Now that I finally received my delayed (at no fault to them!) McMaster shipment, I can make the motors’ center shafts.

That’s about it, actually. I also need to think of a wittier and less fear-inducing name than Deathblade. Let’s play the Name Charles’ Latest Stupid Vehicle Project game! I’m thinking it should stick to the $SINGLE_SYLLABLE + “Blades” theme.

Failblades? Motorblades? E-blades? Z-blades? Cheeseblades? Another suggestion I received was “EDM”, for both electrical discharge machining and “Electric Death Machine” which, while it appeals to my affinity for puns, still has the word death in it which I would like to…err, avoid.

Anyways.

We have motor magnets! I ordered these from my all-time favorite source of super magnets, Mr. Supermagnetgeorge back in early March some time when I decided to go ahead with the Deathblades project more seriously. They are custom-ground arc segments, 14 to a circle.  I elected to go with the strongest grade available in order to maximize torque. I might find it useful later on to switch to a high temperature grade for better magnetic field retention at healthy motor temperatures of 40-50 degrees Celsius.

Into the cans they go. I used some generic 2 part, 24-hour laminating epoxy that we happened to have a gallon of. The installation process was easy due to the fact that the magnets complete a circle.

I found that the magnet-making process is essentially dead on, dimension-wise. I keep overboring my can internal diameters because I possess fears that magnet #14 will barely not fit (lastmagnetdoesntfitophobia), but it seems like their manufacturing tolerances are tighter than what I can hold on the Old Mercedes.

After shoving in the magnets on both motors, I left the cans to cure under a desk lamp. The elevated temperatures allow the epoxy to set quicker and stronger.

Now that I know how far the magnets actually stick up, I could make the removable endcaps. They’re structurally the same as the permanent endcaps, just a few thousandths of an inch smaller in diameter.

After making the endcaps, I put the radial screw holes into both components.

Astute motor builders will probably wonder why I drilled and machined the steel endcap after I installed the permanent magnets. I did that because I’m an idiot and it was late at night.

Yeah. Make sure your motor cans are done before gluing the magnets, unless you like picking every last tiny chip of steel out of the can internals.

Yeah, um… about that “late at night” business.

It looks like I over-indexed by one peghole on my small adorable dividing head and didn’t notice. Actually, I don’t even know what on earth happened here. That’s two different directions of “slightly off”.

Oh well, at least the rest seem to be okay.

Moving on now to the skate wheel frame. Remember the rendering? Well, here it is, in 1/8″ aluminum form. The waterjet still stands as my most favorite machine tool ever, right next to bacon bit dryers.

Also in the picture is my weapon of choice for putting together the wheel frames. Now that I have had practice on Cold Arbor, zinc-aluminum brazing is one of my preferred methods of putting together thin(ish) aluminum structures when t-nuts are not optimal.

To zab something, you need a

• giant-ass torch
• zinc-aluminum solder (available on McMaster as part number 7664A4, or through places like Durafix or Alumiweld
• and a good stainless steel brush. Preferably a bunch, since they seem to vaporize if you accidentally wave them across the torch path.. I do it on the cheap and make them out of a piece of tubing and a few strands of SS aircraft cable.

Also, a set of heat-resistant gloves are nice. No, I couldn’t dodge the torch exhaust when building Arbor, no matter how hard I tried.

First, I hit all the parts with the belt sander in order to remove the waterjet’s characteristic features such as edge draft (minimal in 1/8″) and underside burrs. I also reduce the size of each tab slightly so it can fit into its appointed slot with only a small amount of beating and forcing.

I put center punch dots on a few slot-tab interfaces to keep the assembly together while it undergoes the joining operation.

Next is the globbing phase. This is where I heat each joint area to the point where it keeps a pool of solder alloy molten, then brush it thoroughly into the metal. All over the place. Vigorously. Enough to actually start scraping up and eroding the aluminum alloy underneath. I’m fairly certain this is actually desirable, since it mixes the solder into the parent metal and the whole things then solidifies as one piece.

As its name entails, the process leaves huge alloy blobs everywhere, which have to be cleaned up on the belt sander.

The results after a while of 120 grit assault.

I would have much preferred 60 or even coarser grades, since they remove the globs quickly, but we didn’t have a single coarse belt left. Time to place yet another McMaster order.

These results are great for me only having the patience to do the outer edges (that and the brush I made this time wasn’t long enough!). For extra strength, I should really fillet the “bucket” inside edges.

Here’s the eventual wheelmotor frame next to the original skate wheel frame. I kept the wheelbase identical in order to keep the handling characteristics the same.

…

Alright, here it comes…

IT’S DONE!!!!

… no, not really. I parked the original skate boot on top just to check if I got the mounting dimensions correct. And correct they are!

Here’s the dummy shot, c.f. the CAD mockup.

Clearly, the motors have no center shafts. This will be taken care of soon, hopefully today.  After that’s done, this *blade can stand on its own.

Well, not really. I’ll make a snazzy display stand for it or something.

Look! It’s a Deathblade!

… or not. But this most excellent screenshot gives an idea of what the final product should look like – hopefully a little more 3-dimensional.

I’ve decided to condense the research and development that I have completed so far about small hub motors, refine it, and present it at this year’s de Florez Award Competition, hosted by the MechE department. The theme will be personal electric vehicle (PEV) propulsion and DIY conversion. Thus, my sudden attention to the Deathblades. While it’s unlikely I’ll have a whole power unit running by next week (and this is certainly better for my health), I at least want one on static display.

Anyway, you can’t have a Deathblade without hub motors. I decided to use the slightly scaled RazEr motor design because I already “knew” how to make one. The whole “can change a tire without taking apart the whole motor” feature is also desirable. Before starting, however, I went and simplified the design even further by changing the way the wire exits the motor.

The previous shaft involved drilling holes at an angle to pass the wires out. While this would be a better arrangement from a strain relief perspective, I didn’t want to have to put up with yet another machine setup. So the design was reduced to a slot meeting an axially drilled hole.

This is actually a design that I wanted to try on RazEr when I have to replace the existing motor, but it seems the Deathblades will get it first.

I ordered the material for these motors about a month ago, and they have been sitting on the stuff pile since then. Likeall my motors, I start with a giant steel pipe and cut it down to bite-sized chunks.

The pipe had to have an OD greater than the largest motor OD, which was the fixed wheel retaining flange, and also a bore smaller than the magnet ring design diameter. I chose 2.75″ OD x 2″ ID tubing – that’s a solid 3/8″ wall. Clearly not all of it will be used…

Leap of faith!

Here’s the protoform magnet ring/outer case. Protoform because it does not yet have holes drilled for the removable side plate. There are plenty of pictures of me building wheelmotors on the site already, so I’ll skip the cut-by-cut commentary for now.

I elected to beat the steel into submission using the meatiest lathe in my general vicinity at the time, which was the auto shop’s machine. I believe it’s a Rivett 1030. Either way, huge meaty tools means huge meaty cuts, and this let me pop off a can in about an hour.

A modern production CNC lathe could probably knock one out in under 30 seconds.

Moving onto the threaded ring of wheel retaining (+1). I decided to be smart for once and ordered aluminum tube so I didn’t have to cut a big dish out from solid, then thread while risking crashing the tool into the bottom.

I went back to MITERS to do the threading part because the Old Mercedes, while a bit shaky for heavy metal removal, was more responsive and gave me better feedback for the sensitive threading operation.

I single-point machined the 2.5″-24 GYF threads using my cheesy internal threading tool. I’m getting pretty quick at it after discovering that it’s faster to reverse the machine to back the threading index up, than to wait for it to be driven all the way around again…

Leap of faith, again!

Well, two. The permanent side plates have already been installed in this picture. I’ve also fit the bearings in. They are the same 6902 type 15mm bore bearings I use on RazEr. I’ve moved away from the 6802s because of their stunning lack of curbhop durability.

I dropped a stator in for fun to see how things were fitting. At least the stator fits inside – this tells me I didn’t do too badly, right?

The removable side plate will wait until I install magnets in each rotor. Usually, some aspect of manufacturing means that the magnets sit a little higher or are a little thicker than I had anticipated – RazEr’s motor needed two trips back to the shop to actually seat properly inside the can rim.

While the Loctite was hardening, I cored out a 98mm Razor scooter wheel. The Deathblades will run 98mm wheels, which are about the largest found on common road skates. Motor torque scales with stator volume, so smaller wheels would mean even even even less torque.

(See? 3 even’s! It’s cubic! So is volume! Get it?!!)

After the coring…

…and with the threaded ring of wheel retaining installed. I have yet to drill the spanner wrench holes, and might just get lazy and Loctite the whole thing again.

But, that would be unprofessional.

Steps to go for 1 static display Deathblade:

• Get the frame waterjet cut from 1/8″ aluminum, and braze it together
• Epoxy the magnets
• Make the other side plate to fit, times two
• Make two shafts
• Maybe wind the motor, but because of its nature, I might skip this step for the presentation.