The headache: proof that there is no binding standardization in PEVs and it makes my head hurt

Ok, so im going to start trying to document my progress on the headache build but it’s been drawn out for like a year so far and is only just getting settled into a rideable config now but here is the plan. Just going to summarize my initial plan here then break out my ignorance in full glorious detail by system in the follow up replies. The general idea was to make something spec’d up beyond my intended use to bake in some safety factor and reliability so I don’t have to open it up on every ride to fix shit. Oh my sweet ignorant past me:

As much battery as i can fit, i need range. But I’m also not familiar with higher than 12s configurations and don’t want to introduce additional issues because of higher voltages. 12s8p P42a. I was thinking i would be near 25w/mile because everything else I’ve made has been hubs and thane… didn’t quite understand what pneumatics would do to this pie in the sky expectation lol. Was planning on building with .2mm nickel and cutting out my own from large wide nickel as I’ve done in the past but I used some of the duck battery pre-cut nickel and it has ruined me so pre cut it will be.

Good quality smart bms with more control over balancing and charging. Ennoid xlite v1. I do lots of long rides back to back and need to make sure im not getting out of balance because I don’t have sufficient downtime between rides to allow the pack to settle.

Metr pro UART because who doesn’t like telemetry?

Tomiboi deck in plain bamboo cause i like it. 41.4 MTB deck with double stack enclosure. I should have gone with a step down deck but I didn’t understand how hard it would be to get rkp trucks to work on here and in the end it was ok.

Boardnamics M1AT drive train at 5.2 gear ratio. I just wanted gear drives and these have been one of the best choices but not headache free as I’ll detail later. I can break anything but haven’t busted these even though i am learning as i go. I should have looked harder for a street setup - but this was a while ago and all the gear drive options that are popping up now just weren’t as readily available and I didn’t know what i was looking at.

I want lights, shiny functional lights, break lights running lights. Ground effect lights. All the dam lights! This was part of why i chose the Ubox v2 75v/100a as I has an integrated 12v lighting power and control scheme via the uni remote.

Led strips I grabbed a maxkgo controller module to push the strips ages ago and it was time to use this… never worked out but I upgraded to a ttl and it has been better but my setup is buggy - more details later

Wheels, initially i got some 6” solid tires from evolve, these didn’t work out for various reasons. I gave them to my brother and he printed up some adapters to fit the tire to different rims but it never worked out. My ground clearance with the double stack and rkp trucks with the m1at drives. I was afraid of flat tires, pleasantly surprised at how nice the 7” pneumies are working out though

So i had a plan, checked it twice, tore through the forums. Should be easy

IMG_7501

8 Likes

Tomiboi deck! OK! So i grabbed this deck and double stack at tomiboi and he layed it up just beautifully. got the bounce just right and it feels so good to jump on this thing and bounce around. Did a test fit for cells and whatnot and it looked so clean

if you just ignore reality the clearance looks good and my kluged together drive train/wheel combo seems ok…

reality sets in and its not gonna work. need bigger wheels and a wedge riser to get my turning radius and clearance somewhere acceptable but it will work, no major headache 10/10 would recommend

Got a big ass heatsink and trimmed it up, too many heat problems on my hills so i want MASS. lots of thermal mass and surface area. went way overkill here but i like how it turned out. went to Heatsinksusa.com and snatched two of these bad boys for 20ish freedum labor exchange tokens and chopped it to size. should have been more careful with the depth of cut when i was roughing the flange and shortening the fins but nothing a lot of Dremel, fileting, sanding couldn’t fix up.

it turned out pretty good imho so not too bad. made a stencil from the finest cardboard from the trash can and trimmed a slot in the last bay to just about completely replace the bottom with aluminum. was able to keep the fit nice an tight and secured it with 6 taped M3 bolts and some liberal application of clear silicone as a gasket.

With a click and a grunt i got it fitted up and poped some holes for the motor connections while i was messing around… this was a bad idea. the holes are exactly over the mounting screws for the enclosure… i just wasn’t paying attention riding the endorphin high of getting the fitment of the heat sink so nice. that’s what rtv silicone is for!

1 Like

Battery times: the plan was get a bunch of cells in there, make board go fas, fas good, drive long time. range good

ordered up the cells and made up the p-groups. Using the duck battery systems 8p tabs for a 12s8p p42a. if i could do it again i would have got non staggered cell tabs but i was able to make it work. the staggered cell extra height is just empty space in the build and i could have used the little bit of lateral space for cable management and small components. hindsight

everything was going great until I drilled the holes for the heatsink, my ennoid xlite took a shit when i dropped some shavings on it bridging a component connection. Booooo not repairable but i had a spare. Yayyyyyy

3 Likes

m1at and wheels. Did you know that Frenches yellow mustard makes a great gear mesh indicator paste? Smells funny but got my backlash dialed. Got some of that good moly grease in there and Gear drive was good but i was still dead set on running these evolve airless tires atm so i built the drive and got to work moding the shit out of some kind of adapter to make them work.

Fsr they have 9 spokes?!? Why would I do me this way. Such a pita to find a way to adapt the M1-AT well to a 9 spoke setup. didn’t know what I was talking about and ordered the adapter plate that works with 3, 5, or 10 spoke tires… worked kinda but had to abandon them anyway due to clearance issues.

so the M1-AT drives only showed up with 2 wheel bearing spacers… don’t know what the deal is there but whatever had a bunch of round stock chopped up by a guy locally before I realized the evolve wheels aren’t very dimensionally consistent. lots of shims and spacer kits to find the right setup for each wheel.

… and then I changed wheels. Because my wife is starting to get savi to my bull shittery I went with the 5 spoke evolve rims and 175mm wheels. had to mod the drive adapter plate because the bolt pattern was slightly narrower than the MBS hubs it was made for. now that it will actually engage with the drive i realized that the balance was off… like way off.

got some balancing tape and got them dialed nice with a little CA glue to keep the weights in place. nice, now to get the spacers done up again, more shims and many many trial and error attempts. got my bearing preload. thought i could just space the wheel out so i had about a 3mm gap between the rim and the adapter plate. runs great on the bench this way as you don’t need to be accurate with the spacing between the wheel assembly and the adapter plate - just let the rim bolts bridge the small game an Bobs your creepy uncle who likes to hug a bit too long.

This was not a good idea, the flex from the tires while carving quickly started wearing down the bolts and got me a few score marks on the drive covers. Had to get the spacers right and shorten the gap between the tire and adapter. shims… so many shims. took a few million attempts via trial and error to make a stack of spacers and shims to get the gap between the rim and drive adapter plate around .2mm. wear on the bolts dropped off and no more score marks! As a bonus it gives me a solid placebo effect - carves better than it did before! gtg

3 Likes

Ubox… the horror. So i went after the Ubox when it hadn’t gotten any flack yet for being finicky. found out the hard way that even though it has 5.3fw and 6.0fw… it really doesn’t. running the stock installed 5.3 fw was super frustrating, i was also having issues with the CANbus being complete dog shit. Motor detection was an exercise in patience as I would have to try and detect and adjust settings 10-20 times before anything would take, constant errors, time out and loss of communication. as my first experience working with a VESC based controller i was ready to give up many times. I could get it to work for one setting with a controller detected, then the second ubox would drop out after a mile of test ride and i would be walking home with new bruises. bluetooth was out of the question, the bluetooth module i purchased with the ubox was doa but had a blue light and i was way past return/refund and i didn’t know the metr pro was also a bluetooth connection… how did i miss this?

Tangent: Metr pro uart is the big winner here, fucking hell. it just keeps giving and giving. logo should link to the giving tree ffs. Except when I busted the external antenna off I never lose connection (now after FW 5.2) just a sold piece of kit and I only got it as a last minute no preplanning thing. Just saw it was in stock and made another poor financial decision. worked out fucking stellar though, i use the metr almost every ride because its just fucking cool.

getting long so here is some gore. most of my other wounds get flagged NSFW

back to the fw5.3 blues. was about to give up as i was always just blind with rage trying to get this thing to do what it was supposed to do, it wouldn’t accept the 6.0FW when i would try to update, and vesc tool was shitting so many “YoUr FiRmwaRE is oLD aNd misSMatCHeD sO i cAn’T dO NOFin” errors or similar unless i rebooted for every connection attempt and power cycled the board and prayed to RNGesus while sacrificing small mammals to Skaten. Took a long break. like a loooong break and just decided to work on other parts of the build and come back to this after.

Accidentally found one of the problems… my bms. so there is an edge case that Ubox just happened to fulfill where the way the ennoid xlite v2 CAN was terminated can destabilize the bus (fixed in later versions and they were super helpful guiding my ignorant ass to the issue), which was unfortunately only one of my compounding issues, at least after pulling it out of the circuit i could talk to both of the Ubox halves consistently. still having issues with the remote and outputs acting super sketch but at least i could blame that on user error as i fuck up the settings and also because i was using old FW that’s “mostly compatible”… so fucking sick of acknowledging that fucking error message every time i do anything FUCK

accidentally stumbled on the method of updating the FW while digging through every word published on the spintend website. you have to update the boot loader first before using their FW for it to take… FUCK that was hard to find, or I’m just so new i didn’t realize that was so fucking basic knowledge no one mentioned it before. GOLDEN, 6.0fw uploaded, local PNWesk8 build day coming up and someone who knows VESC tool close by to help me sort my shit out because I still cant figure this out. And they fucking did, thing was working like a champ by the end of the day and made a few test laps and it was solid

you know when someone asks a question on the forum or on reddit or somewhere… and you read it and think to yourself: the question they asked is so off base, so far from basic understanding, and kinda entitled that you say to yourself that it will take so much effort to get them up to speed and on the right track that you just let it sit… someone else will chime in probably and help this poor fuck. i just don’t have the two days its gonna take to teach them how to read first in order to fix their shit. Those are my questions

I don’t know why the fuck this started happining

like look at this shit, just started doing this randomly. thanks again to the Metr Pro. nothing like breathing on the controller and having it go 100% duty cycle for .5 seconds. kick you off real good and then when you wake up and try to figure out what the hell happened and limp home… one motor doesn’t work, the can bus is shit again, settings are all default and you busted your puck. LOL

it was acting so weird it had to be me, i must have broken something or got chips in the case or some shit. took it apart and it was so clean and fresh looking. not one thing wrong. went over the whole thing under magnification but anything beyond a visual inspection for complicated pcbs is outside of my wheelhouse.

took a fall and got a big sad. had to limp my as home with my broken toys. fuck this ubox and all its bull shit. fine! ill replace it and just take the L and move on. no more ubox. ordered a dual SKP - probably super fucking overkill for my needs but everyone testing that shit was all - it just fucking works - and i needed that energy or i was going to part some shit out.

couple days before the skp shows up I’m hooking up a Flipsky 50a dual just to see if i can make anything work but i’m warned about killing it on the 12s voltage and just put it back down. beginner question thread pulls through here. Thanks to @haven for the solid advise to roll back to 5.2 and @Common_good for spoon feeding how and the old vesc tool to me. rolled back the FW and everything just started working, I could talk to it on my pc EVERY FUCKING TIME! I could see both of them across the can and didn’t have to program them separately. Bluetooth connection was available! and working solid - even if i sometimes have to write twice to get it to stick it works more than it doesn’t. Controller works all the time, no more dropouts or weird outputs.

You have no idea how scared of riding my board I had become, it was going to kill me. I was about to just abandon the whole thing and that was making me so fucking sad I cant even describe. Like seriously if it was just not working it would be totally different than the Russian roulette I was playing. Board runs amazing for like 30 miles of testing then takes a shit on you right as you are starting to build confidence in it.

My boring build has started to be boring. press go, go. press break, breaks. not a big ask. glad that issue is solved… kept the skp as both a threat to hold over the ubox and because I really like building shit.

6 Likes

Reserve for lights and leds

5 Likes

Reserve for idk

1 Like

Hehe never goes as planned

4 Likes

What on earth are those settings my dude :skull_and_crossbones:

2 Likes

They aren’t settings, they are firmware issues and CANbus communication problems. I guess I need to fill in the context here lol, i got busy and it’s complicated.

800 mile update: 5.2fw for the win. everything has settled down nicely. set on 70a motor current. The boardnamics M1-AT gear drives got a full rebuild just to check on a few know issues with the helical gear causing lateral forces.

Just got to the end of life on the first set of evolve 175mm tires, swapped them out with the second set I had and rebalanced them. i like how soft the compound is and i feel like i stick to the road really well so i will probably just keep these guys, no need to fix something that isn’t broken. the rims are kinda bad though. had one of the nuts on the rim give me a hard time because it wasn’t captured well and was able to spin as i was disassembling the tire. pita but not too bad, swapped the rim with a spare and gtg

The pinion had indeed popped free on the right hand drive and just started to scratch the inside of the case. took everything apart and rebuilt it but got nervous after some noises on the next ride and opened it up again. just paranoia but i cooked off the retaining compound and did a super good job cleaning everything up and redid it. The left hand drive, getting the majority of its force in the opposite direction had murdered the wave spring washer under the circlip on the motor so i tore that one apart and redid everything also. On close inspection i think i might have gotten the 620 and 648 compound thread locker and retaining compound mixed up on the initial build as my bottle of 648 was sealed when i went to grab for it lol. don’t know how i missed that initially - the two bottles i have are identical except for the label and both compounds look almost identical.

all in all everything is good except the range estimates i get. seems like i start at 20-23miles of range on the davega, ride 10 miles, davega shows 16 miles left, run 10 miles, davega shows 6 miles left, run the last 10 home, davega shows 3 miles left. never ran it all the way to the cutoff. need to look into it but. metr is showing some crazy high wh/mile some times but also sometimes shows what seems to be very accurate wh/mile. i think i run somewhere in the 30-50 range depending on the route i take and how aggressive i am playing around.

1 Like