DougM's winter project 2022-23

[DougM]

So anyone who’s followed any of my builds knows I’m a bit unique in the construction of my boards, but I believe I have stumbled upon the perfect carving machine (for me). Yes it’s heavy and slow but it seriously rides like a dream and hasn’t tried to kill me in at least a year.

In fact, I like it so much I’m going to make another one, exactly the same, only different.

This constitutes the beginning of my build log.

For reference here’s the original one:

It’s 10S5P 2170’s with a FocBox Unity, Torqueboard 6355’s on Trampa 12Fifties with 6" pneumies. Headlights, taillights and a e-ink display up front.

The new one’s not going to be quite so fancy. it’ll be 12S5P with a Stormcore 60D, same motors, same trucks, but Hoyt St. 7" Pneumies, but no lights (at least not at the start) and no fancy display.

So slightly larger tires, slightly more battery, slightly newer electronics. And I’m actually going out on a limb and putting an onboard BMS on it.

Here’s all the goodies

Batteries
This board is using BAK 2170’s with a 5,000mAh capacity. The original board was running Lishen 2170’s at 4500mAh. So not only 12S vs the old 10S but higher capacity.

A few minor improvements to the battery packs - They are the same design as the old ones, but I’m taking it a couple of steps further this time towards my goal of completely solder/weld free battery packs. The idea is that if you have a weak bank or a weak cell you can simply crack the thing open, R&R some of the batteries in the pack and be back on the road.

And it gives me the ability to swap out batteries for higher-capacity at a later time with no welding required.

And the old weaker cells that come out go into other projects, so don’t have the spot weld marks on them.

Anyway, it worked pretty well last time, so I’m doing it again with the following improvements.

I changed out the contacts for ones with a larger contact patch. I also swapped out the original aluminum connectors for nylon. A little less strength, but non-conductive.

Here are all the connectors, cut to length drilled and tapped

Here are some of the the frames fresh out of the CNC ,machine and PCB’s.

and here’s a couple of completed banks

As you can see they have screw-terminal connectors so I no longer have to solder wires directly to the PCB’s. I can now solder the wires into little harnesses.

Drivetrain

Ok Vendors, please consider the following - Standards! Yes, I know there’s so many to chose from but wouldn’t it be cool if we had a standard method of connecting the pulley to the wheel? Here’s my problem

It won’t be that hard to cnc some little inserts out of Delrin that slip into the slots on the Hoyt wheel and attach them to an aluminum plate that then attaches to the pulley (an MBS 47072).

Frame

The frame is recycled from one of my very early builds - the first machine I built that had 18650’s and a not-home made ESC and unarguably the best motor every made, the Tacon Bigfoot. In fact, if I can get HFI working correctly on the Stormcore I might actually try to find a second Bigfoot and use those instead of the Torqueboard motors.

This board, as you see in the picture, chain drive and single motor and thanes, had more range than any board I’ve built since. If I were to load it up with some modern batteries and I swear I could get 60 miles out of it. I would totally win a single charge iron foot competition.

That’s it for tonight.

[Plurf]

He’s back ! He’s back !!

[That_Jamie_S_Guy]

I know this thread isnt about the other build but the e ink display makes me irrationally happy. That and the use of ally extrusions gives proper mechatronics engineering vibes

[rafaelinmissouri]

So dude.
Very cool.
How long is the standing platform and how long is the wheel base?

[DougM]

@rafaelinmissouri

How long is the standing platform and how long is the wheel base?

standing platform is 23 1/2 - I think, but can’t guarantee that I based it on my snowboard stance.

I checked tonight, though, and it’s going to have to be longer for the new board since I can’t fit 6 battery banks in the existing space. So probably closer to 28 or 30". Another change I’m making on this board is to mount the electronics housing on top of the board rather than underneath, so I might be able to keep the same 40" wheelbase despite the longer platform.

Wheelbase is exactly 40"

[DougM]

Tonight I finished all the battery banks and soldered up the inter-bank connectors.

I’ve got some edgelit stuff I need to do for a lights festival so probably won’t be back to this in earnest until after Dec 10…

[DougM]

@That_Jamie_S_Guy

I know this thread isnt about the other build but the e ink display makes me irrationally happy. That and the use of ally extrusions gives proper mechatronics engineering vibes

Amazingly and by sheer happenstance the almuninum extrusion is just about the right amount of flex.

The best thing about the e-ink is the sheer size of the thing - I can see at a quick glance down what is going on (mostly how much battery I have left). The stats are kind of fun too. I think the speed is about right but the mileage measurement is wildly inaccurate. Probably because my wheel diameter isn’t exactly 6 inches. Or possibly because I’m bad at math.

I do have another display and most of the hardware to put it together, so at some point in the future the new board will likely have the same display. Of course all the code is based on the FocBox (and liberated from @janpom’s davega project). I’ll have to see how hard it is to port to StormCore.

[rafaelinmissouri]

@DougM

So probably closer to 28 or 30

I like longer platforms so I would like this. I hope it does not ruin the shorter one for you…

[Savage1]

Neat project dude. Following…

[DougM]

I lengthened the board to accommodate the 12S packs, here’s a test fit.

The new standing platform length is 24", so only a half inch longer than the old bed. Weird. But I guess it’s because I used to have the electronics box encroaching into the space where now the box will be on the top of the tail.

There is one little gotcha, the PCB’s are of copper fill so if they were to migrate and start rubbing on the aluminum risers I could get a short. But because I have a loop key this actually isn’t a problem. Either the board would short-on (bypassing the loop key) or the short would happen at the other end, in which case when I put the loop key in I’d get a big fat spark.

But nonetheless I’ll put some adhesive thick rubber in to make sure nothing untoward happens. It’s never been an issue because the packs are hard bolted to the rail, but one can never be too careful when it comes to batteries.

Next time I do a run of PCB’s I’m going to make them thicker and also mask off this section so the copper fill stops short.

Anyway, wrt rims here’s a nice 3D print test fit:

(got it right the first time!)

and mill up the tabs

and the plate

I just did peck drilling because I wasn’t sure of the final diameter - my plan is to tap this plate and I want everything to be 10-32 since I have every conceivable length of 10-32 in stock and very few metrics, but I think the rim won’t take them.

[DougM]

Got the hubs mounted to the plates, but as I suspected it’s going to take some M4’s to get them attached to the wheels.

and I’m going to need to hog out a bit of the plastic to get the screws in. I might throw some nylocks on there too just to make sure, since the rim only has 3 mount bolts. Which means I’d need to hog out a bit more but not that much. Actually now that I think about it I should press inserts in there - otherwise I’d have to get a wrench on the nuts, which would be a hog too far.

[DougM]

Incremental progress - got the threaded inserts and pressed them in. Not my best pressing job but I got better by the time I got to the 6th one…

CNC’d my motor mounts

And, of course, once I have the threaded inserts I don’t need nuts, so no need to hog out anything, I just trimmed the screws flush. I made the mistake of setting the length when I had the plate on but not the insets, so all the screws turned out slightly too short and I had to do it all over again, but it’s only 6 screws.

Then I found these Giant raccoon prints in the snow in my back yard. I knew I had a family back there, but I had no idea how big they are getting…

Here’s the plate with the insets just before final assembly:

Then I put the wheels on the trucks and realized that the Hoyt wheels also do not have the correct size bearings for the trucks. They are close, but definitely not right, so first I had to remove them, which required making a bearing removal tool:

Now I’ll have to go hunting for the correct size bearing facing, once again, the eternal dilemma of finding a bearing where the ID is US standard (3/8") and the OD is metric (26mm). This is exactly the same problem I have with some MBS wheels I got, where I need the same 3/8" ID but the OD is 28mm. But I think MBS stocks those.

So not being able to completely assemble the trucks I turned my attention to attaching them to the board and made these cool new brackets that I think are going to dramatically decrease frame twist. I plan on making more of them to retrofit my old board.

Lastly CNC’d the second half of the motor mount assembly

And that’s all I can get into this weekend - time to get back to more pressing but less fun stuff.

[DougM]

In another thread we had a bit of a discussion about bearing sizes (like how the truck axle is 3/8" and the wheel ID’s were 26mm and 28mm - I pitched a fit about this in an earlier post here. So MBS had the 3/8 ID 28 OD (for retrofitting the original board) but I had no luck finding the 26"

So I made some sleeves.

Starting with 9.5 x 22 bearings and some Delrin blank

Here’s the obligatory lathe shot

Followed by the final product, completely ignoring the several attempts I made and screwed up along the way

I made the sleeves quite a bit taller than the bearings because the non-threaded part of the axle was longer than the depth of the wheel so I would have had to put a spacer in anyway - this is more convenient and gives me a touch more track width.

The drivetrain is coming together

Aaaaaaand the thing is starting to look like a skateboard

I also spent some time retrofitting the old board (seen in the background in the above shot) with the new version of the truck mount plates, which meant I had to re-design and re 3D print the taillight parts.

(Yes, I’ll have to do something about the giant rain collector slot)

I can tell already that the frame is significantly stiffened up. On the old board I had a lot of squeaking and groaning when I was carving - hopefully that’ll get cleaned up. I’m probably going to add two more cross-braces between the risers

to give it additional stability. Probably will retrofit the old board the same, though that will require moving some wiring and re-designing more 3D printed parts.

Fortunately spring is a ways away still.

[glyphiks]

@DougM

happenstance

I love this word.

Great to see you at work again!

[DougM]

Worked on the electronics box - pretty tricky fitting everything in since it’s so skinny, but fortunately I was able to go long. It’s taller than I had hoped.

Here are some glam shots of the design in CAD.

The thing on the top (inside the housing) is the BMS, the thing on the bottom is the StormCore

Bottom view of the Bud fittings - all wires from the motors will go through the 2 big ones, and the wires from the battery pack will go through the small ones. Though I just now realized I have no way of getting the balance leads out. Crap.

And another view that shows the charge port and the hole for the power button

Also ordered the belts. Length = 420

And I installed the cross-braces but I didn’t get any pictures. It doesn’t help as much as I was hoping - the end assemblies are super rigid now, but still there are only 2 long skinny extruded rails connecting the front to the back so it’s pretty twisty. Installing the foot board will help a lot, but even then it’s not great because it’s wood. Some sort of rigid lightweight honeycomb thing would be cool but maybe I’m overthinking this - I want the flexy but not the twisty. Riiiiiight.

[DougM]

Was able to get the drive system completed, here it is half done

But let’s face it that box is just too tall.

So I decided to separate the two functions out - I’m going to have a low-profile (with luck 1.25") box on both ends. the Stormcore will be in the rear one and the BMS in the front - so they will kind of match. This unfortunately makes wiring more complicated as I’ve got to run big cables from front to rear (If I want the BMS cutoff).

Hopefully the box will be about this high:

But then I decided to replace the tires on my daily driver board with newer ones which, as it turns out, are a different diameter, which explains a lot about why my display isn’t accurate. I mistakenly assumed that if a tire says it’s 6" it is, but turns out these were 6.25". A big difference given how many revolutions they do in a mile.

Anyway, the new ones are 6" and have a significantly skinnier contact patch which I hope will translate to more range.

Here you can see the 3D printed pulleys

and this is literally where the whole project fell apart. I had exactly 1 piece of Delrin large enough to cut 72T pulleys and I mistakenly assumed that the inside curves were at or larger than 0.125 and blindly went in with my 125 cutter and…

I’m fucked.

I thought about fixing it but if the cutter is smaller than 125 it’s not longer than 3/8" so even if I could get the part back on the machine accurately there is no cutter that will do the complete depth.

So now I’m looking at caving and just buying some off the shelf pulleys but I’m balking at the $30 each they are charging for them, or I could come up with a clever way to piece together a complete pulley from smaller pieces of Delrin. So that’s what I tried to do.

This is not actually a new idea, the original Delrin pulleys on the daily driver were 2-piece because the cutter wasn’t deep enough to cut them out all at once.

But these are a bridge too far, so to speak. They are 10 interlocking pieces.

TRY 1:

As a proof of concept this actually works pretty well.

Until you do a bit more testing and realize that your interlock allows the part to rotate out which would catch the belt and fling you to an imminent death.

So that one was out. But re-designing the interlock wasn’t actually that difficult so I did that and I cut a whole bunch of new parts only to realize that somehow my alignment of the teeth to the mount hole wasn’t correct, so while one level goes together perfectly, when you alter the direciton of the other row the teeth don’t line up. I didn’t get a picture of this because I was so pissed off that I threw the whole thing in the trash.

TRY 2:

But it takes too much energy to be pissed about stupid mistakes, so I went ahead and re-designed the interlock and re-aligned the holes to the teeth and I think this one is going to work:

The only other problem is a bit of interference between the pulley bolts and the motor mount bolts but that’s a pretty easy fix. Oh, and different pulley = smaller diameter = belts are too long, so new belts on order.

[marketthrowaway]

Nice machine work. I’ve been away from a well equipped shop for too long.

[DougM]

Vacay between Christmas and New Year affords me some time to keep going on this board.

The strategy of separating the BMS from the driver and putting them in separate housings is going well - I’m able to get both housings down to about an inch high.

Here is the driver box test fit

Uploading: PXL_20221226_184102523.jpg…

And some wires down (up I guess) through the Bud fittings

And some more wires

Until the thing is fully wired

Then I CNC’d out a shiny new deck

Beginning to wire the BMS box

Balance Leads installed

A bit about BMS’s though. This is my first foray into them and now I understand them better than I did before, so be gentle while I explain how I screwed this up.

I bought an “80A” BMS assuming that that was the cutoff current and that the power would be routed through the BMS, but when I received it it only had 2 connections: C- (charge -) and B- (battery -) and they both had 10AWG wires. So what I bought, apparently, is an 80A charge rate BMS, which is a bit drastic for a 12s5P pack. But next time I’ll know better.

In any event, the board passed the first smoke test. Now I’m going to download the BMS app and see if I can watch it charge, then fire up the driver and see if I can get the wheels to spin.

This is also my first foray into a Driver other than the original VESC 4.1x and FocBox Unity series, so I’ve no idea what’s in store for me. I did read that FW 5.2 is the sweet spot for StormCores - hopefully that’s what it has on it.

Next up will be 3D printing and installing the rib cage, the battery cover, the UHMW scrape strips, the (blindingly expensive) carbon fiber covers for the two electrical boxes and then, hopefully, a test ride!

Though if it’s dry enough I might be able to get a short test ride in without any of those things.

[haven]

Loving all of the custom stuff! Looking like a BOTY candidate forsure.

[DougM]

We are almost down to details. Got the Battery cover on - this is a surprisingly persnickety operation

So the way the ribcage works is I have a bunch of 3D printed ribs in the middle and end caps on the fore and aft, each one of which has heat sets pressed into it

And you have to locate the heat sets underneath the plastic sheet. I cut this HDPE strip

and drilled holes in it so I could mount it to the rail. Then I cut slots in the orange sheet plastic to get past the screws and compression fit it between the HDPE and the rail.

Then I roll it over and hold it down with heavy things:

and, using a strip of UHMW I’ve already cut and drilled I can locate the heat sets pretty accurately. Unfortunately I didn’t get pictures of the process so it’s pretty obtuse using just words.

But at the end of the whole thing you can give it a light blow with a heat gun and the plastic will soften up a bit and the curve will take permanently.

The result looks like this:

I was looking at my earlier rev of this design and I noticed that the curved part had hit something and taken some scrape marks and punched through in one spot, so I should have spaced the UHMW strips closer to the edge - right before the curve starts. But anything that I hit square on will slide fine.

I got one of the 2 blindingly expensive decorative Carbon Fiber lids cut, drilled and installed temporarily

Left to do is:

• install the wood end blocks on the riser
• take delivery of, cut and drill the second Carbon Fiber plate
• take delivery of, design a mount for and install the charge port
• install and wire the volt meter

So I’m coming in to the finish line, but I have to go back to work tomorrow, so it might be a month. Which is fine since it’s going to be raining the whole time anyway.