The SB Cruiser : Amberwolf's 2WD Heavy Cargo Trike & Dog Carrier

:lol:

I actually had a few cars/trucks slow down as they passed me on 31st Ave to yell that out their windows. :)

Three different cyclists turned around and followed me at a distance for a bit, during different parts of the ride back, though none approached.

Some pedestrians here and there (mostly people getting out of their cars as they got home, or coming out of their houses for mail, etc., in the residential areas I was passing thru to avoid the main roads) turned their phones toward me, presumably taking pics or video. (too bad I can never find any of it on the web to link to these posts).


I had two (but only two) idiots that nearly scraped my paint as they held their horns on loud, gunned their engines, and acted like they were trying to force me off the road (even though I was completely within the bike lane), one on 31st Ave on the way there (unloaded), and one on Sweetwater on the way back (with the piano on the trailer).

Still, that's quite a lot more than I usually have, since on the trike most people go quite far around it even if I'm in a different lane (like maybe it's contagious :lol:)


Was actually a fun trip; if it weren't for the heat (over 112F in this area when I was out there, and hotter on the roads with traffic) it would've been even more fun. Took something like an hour to get there and back (almost all of that after picking it up), and at least an hour to load it up and tie it down.
 
Wow that trailer is huge! Very nice, I like the efficient loading ramp concept and the dualies look very cool. No dogs are getting left behind anymore :) If you could put solar panels on the roof of the trike and on the trailer you might not ever have to plug it in to charge it!
 
I've considred the solar roof, but unless I got the panels and conversion electronics for free, it's not worth it to me--plug-in electricity is cheap.

The trike isn't efficient enough to be fully recharged by solar without a lot of panel, since it's indoors (to protect it from idiots) while I'm at work, and only gets sunlight on it at home in the morning, and then while I'm riding to work for a few minutes near midday. If I'm out on other trips, it'll be parked outside, so it'd get some charge that way, possibly enough to keep it full, depending on conditions, since I tend to ride either in middle of the day, or well after dark.

Takes at least a couple hundred Wh for work commutes; if I had a hundred watts of solar (at the battery charge input point) it could take at least a couple hours to recharge it each commute. I don't know what efficiencies I'd really get, but I expect I'd need 200-300w of panel to get 100w at the battery, on average depending on conditions and time of day.


Anyway, If I already had the panels and electronics, I'd probably do it just cuz. :)


The trailer's usable surface deck is not any bigger than it was as the MkIV, not yet. I still have to widen it to turn it into the two-dog enclosed cooled MkV. So far the only work done on it was just the stuff absolutely necessary to handle the piano moving. So still only fits one dog at a time; the other still has to go on the trike.

The work to convert it to the full MkV is quite a bit of stuff, and will have to be done in stages to leave it functional in case I need it. Details on that will be over in the MkV thread:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=94215
as I get it done, or work out the ideas.
 
Power usage has been fairly steady at around 45-48wh/mile according to the CAv3.1. I still don't know if it's reading low vs the CAv2.x, using the same shunt and setting, or if the the 2.x was reading high, but was more like almost 60wh/mile. I still need to do the experiment of hooking them both up to the same shunt at the same time to monitor usage, and see what the difference between them is, and figure out why.


Earlier this week, on my way to work, passing thru the parking lot for what used to be Tim Finnegan's (has been shuttered for a couple years, so that part of the parking lot is always empty, pretty much just like the satellite image linked below),

https://www.google.com/maps/search/tim+finnegans+metrocenter/@33.5697162,-112.1211944,181m/data=!3m1!1e3



a gold pickup truck going the opposite direction from me suddenly swerved directly towards me as it was passing me, like he was trying to hit me. I just gunned it a little bit to scoot past his front end, and the guy just finished pulling into the parking space I'd been passing at that moment, and sat there, engine running.

Dunno why he did that, or what he was thinking, but he had to have tried to hit me on purpose.

There's no chance he couldn't see me--he was looking directly at me; our eyes met just before he did that.

There's also zero reason for him to just randomly happen to be pulling into that space; there's no businesses there anymore, havent' been for ages now. (there are some several store fronts further down, but the entire parking lot was empty except for a handful of cars near the Subway at the very far end of the complex).

Just glad he didn't chase me to keep trying to hit me; there are a couple places I could've gone he couldn't have got the truck thru, but rather not have to do that.

I've never had anyone do anything like that before, in a parking lot.

I *have* had a few of instances on the road, one not all that long ago, maybe last year, where he ended up driving into construction signs and stuff because he was too busy trying to cross lanes to get to me to run me off or over, and one around a decade ago on DayGlo Avenger where some guy swerved across lanes to try to run me down but instead got chased by the police for it. Other similar stuff over the decades before that, where they often got themselves in trouble or actually crashed into things because they were concentrating on trying to run me over/off the road (or sidewalk or separated bike path) instead of watching where they were going.
 
Second rainstorm of the day ended more than an hour ago, was still sprinkling on my way home from work, and had dumped close to an inch in my area, all the way from work to home.

This left lakes in the streets for me to ride thru on the way home, one of which soaked the underside pretty well, causing intermittent partial-throttle from there all the way home, on that side, so I had to keep using the ebrake (which triggers both sides) in pulses now and then to reduce speed sometimes.

When I got home I tipped the trike up on it's side and found the water had gotten thru my minimal waterresistanting on the controller wiring--specifically on the unused CA connector of the Grinfineon powering the rightside wheel. I flicked the water off the connector a few times, and it fixed the issue for a moment, but I'll need to fix it more permanently before I leave tomorrow, as there's some chance of rain pretty much the rest of the week.
 
For now I've simply cut the Throttle and SpeedSensor wires off the CA connector, and insulated them separately from the rest of the wires. (since even if I ever use the CA connector on there, I wouldn't be using it for either of those, specifically because of the risk of shorting battery voltage into the throttle or motor hall sensors or the controller MCU)

This prevents the problem, as long as this separate insulation remains intact and the wires don't come near the rest of the wires on that connector.

Longterm I should remove teh wires from the controller baord entirely.
 
Worked so far, puddles and rain both defeated.

Still need to waterproof teh handlebar controls themselves; no problems with them yet but eventually it'll happen, with a big enough drenching for long enough.


Once it arrives, I'll be beta-testing this controller for a little while:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680&start=25#p1393153
file.php


on the righthand wheel of the trike. (and after that, if there's time, on the lefthand wheel, since they're different windings).

It's much higher power than I've had on here before, should be interesting to see how the trike performs with it.

Hopefully I wont' break anything with it (axles, frame). :)
 
Hey, I hope you didn't use to go shopping at the Safeway on 35th and Northern.

I did my first ebiking in the rain on Wednesday evening - same night that the Safeway died. My waterproofing was shopping bags and electrical tape. :^) That worked out OK. But then I avoided major showers and puddles.
 
I stopped shopping there some time back, after the housefire, because of their change in attitudes to various things (including bikes and people that ride them), but they were my primary grocery store for many years before that.

I've bagged the bar controls and CA a number of times for various rainstorms, on various bikes--it definitely helps, however the style of bars with more vertical grips I use on CrazyBike2 and SB Cruiser (whcih make a much more comfy ride than the straight or MTB / etc type bars) tend to let water run down inside the bag ends and into the controls. So it's not as effective as it could be, unless I use a much larger (transparent) bag that encloses teh whole handlebars.

It's less of a problem in vertical rains on SBC, because of the canopy that extends out over the bars, but wehn I'm moving rain does still get under it and on the bars.


I keep myself dry with motorcycle rain gear, works fine; the canopy keeps my head dry except in driving rain with gusty strong winds.



I can't always avoid the "puddles" ;) some of them are more like rivers or lakes filling a whole side of the road, or intersection.

I actually like riding in the rain, wind, etc.; it's easier with an ebike than a pedal-only bike, as long as I can keep the water out of the electrics. :)
 
FWIW, I have had many drivers make eye contact, yet obviously not see me. Sounds to me like he got a big surprise, and then pulled over to calm down.

One time when I was about 20, I actually failed to see a cement truck. I looked right at it, then pulled out in front of it. its unbelievable what you can look at and not see it.

That was the day I decided to cultivate an ability to see, vs looking, which ended up making me a fortune on the job compared to the lookers.

Hard to belive somebody could not see that trike, but if i could not see a cement truck bearing down on me, anything is possible. He was driving thinking about other shit, as most do.


and if the rain is a pain, send it here, we'll take it. :mrgreen:
 
dogman dan said:
Hard to belive somebody could not see that trike, but if i could not see a cement truck bearing down on me, anything is possible. He was driving thinking about other shit, as most do.


and if the rain is a pain, send it here, we'll take it. :mrgreen:

There is a famous psychological experiment where you are asked to count the number of bounce passes that occur between a group of basketball players during the playing of the video. So like 90+ percent of people I dutifully did that. What a surprise to look at the video later and see that a man in a gorilla suit had walked into and through the scene. Sure, my eyes must have seen him. But my brain didn't.

So yes, it is very possible to look at something yet not see it.

I don't mind summer rains at all. And even when I find winter rains uncomfortable, I have no problem putting up with it. We live in a desert after all and periodic rains are a good thing.
 
I wish there was more of that rain; what we did get is making the yard stuff grow like crazy compared to the water I can afford to put into it. :)




The trike is now ready for whenever the SFOC5 controller arrives for field/beta testing (from here:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680&start=25#p1393153
)

So that I can easily swap back and forth between the existing Grinfineon controller on the righthand wheel, and the SFOC5 on that wheel. Only the phase and throttle will be switched; power, halls, brake, reverse, etc are all still direct solder connections on the Grinfineon, and the power to the SFOC5 will be a separately-pluggable from the spare charging port under the trike.

I've put PP75s in the phase wires of the existing Grinfineon (leaving the hall wires directly soldered to the motor).

I put a connector into the ground and throttle signal, too. I didnt'have much in teh way of even remotely water-resistant connectors (since its' the rainy season, such as we get), and the only one I had three of (one for the Grinfineon end, one for the throttle end, and one for the SFOC5 end) was the "turnigy" bullets off some old dead LiPo packs. I'll put the last one onto the SFOC5's open throttle wires once it arrives.

Since the SFOC5 already has PP45s on the power connector, then at first I'll just use the existing spare PP45 charging port to power it (it's right by the space the controller will be mounted to under teh cargo deck). If that turns out to be problematic due to the relatively small wires on that charging port, I'll change the PP45s on the SFOC to an SB50 and mate that to the regular SB50 charging port that has pairs of much thicker wires.

I dug out some old computer extensions (VGA, serial, keyboard, etc) to use for the "dashboard" LEDs from teh handlebars to the controller.


BTW, I wasn't sure based on the manual if it's possible to stream the log data live out to a PC. If it is, I can carry the laptop in either the seatbox or the cargo area, and log stuff directly to it.

I'd like to also log the CA3.1 data to teh laptop, which would make a comparison for some things. It's shunt reads both controllers, but I have separate throttles and use them in specific situations so I know generally which data should be for which controller (or both).

I should be able to use two separate USB-serial devices on different USB ports, and two separate instances of terminal/logging programs, on one of the laptops. I have some pretty old ones, Win98, maybe WinXP, with actual serial ports, if I have to use those with just plain serial cables. I also have a win10 laptop but it's my main computer so don't really wanna carry that around much. :)

I also have an old android tablet, with an external USB port but I have no idea if I can read USB-serial stuff with that.
 
The SFOC5 controller arrived,
https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680&p=1398175#p1397464
pics and a specific thread for testing/etc of it to come. In the meantime:

Crosspost from the Hall Sensors'B'gone thread:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680&p=1398175#p1398175


Couldn't sleep, so now that it's (just) under 100F outside (after midnight), I wired up the connector ends that came with it to the connector ends that mate with what I put on the trike, based on the pinouts in the manual.

Throttle only connects signal and ground; the 5v from SFOC5 is not used; my throttle still gets it's 5v from the Grinfineon it would normally be powered from. (this may be the source of a problem noted later; have to troubleshoot).

I cut an old VGA extension cable and used that to extend the "dashboard" LED/switch PCB so it will be able to reach up to the handlebars and mount there.

The motor temperature sensor is not connected yet, as the manual shows a second 3-pin connector for it, but there isnt' one on the controller itself. It does have a multiwire cable coming out of it, but there isn't a connector on it, so I'll need to determine the wiring of that cable; most likely it'll be easy enough if I open up the controller to see which color goes where. (assuming the fourth cable is actually for the temperature sensor; all the other cables are used--throttle (verified), data/programming port, and LED/switch dashboard panel. )


Didn't mount anything physically to the trike yet, but setup the trike with the righthand wheel offground, and did a basic power-on test.

As noted, I'll have to troubleshoot these issues and find out what I've setup or wired incorrectly.

On power-up, it sequentially lights the 1, 2, 4, and 8 LEDS. Then it blinks the center two, then the outside two also light (so all four are lit), then the two outside LEDS go out, leaving only the center two lit. I was being distracted by a couple of very scared St. Bernards due to someone setting off mortar-style fireworks at that moment, so I lost count of the number of times it did this. When it stopped it left the center two LEDs lit steady. It did repeat this at least once while I was testing, probably more than once, as I was not always looking at the display.

If I'm reading the codes correctly, the two center steady LEDs would be that the temperature sensor is not connected (makes sense), and the all four on would be either "internal error" if it's presumed this is a blinking code, or battery voltage gauge 4/4 (voltage less than 40% full) if it's a steady code.

I haven't calculated what the voltage is on the battery vs the settings I'd put into the spreadsheet, but that might be correct since I gave myself a bit of wiggle room on the LVC.


During this, the motor made some hissy-sounding tones; presumably the controller doing some checks.

After it remained steady, the motor continued to make a steady hissy-sounding noise. It's pretty quiet, but I suspect it's not supposed to keep making that sound all the time.

At no time during this did the motor move at all (the wheel is off-ground, just in case it did so).

Since there did not appear to be any actual error condition (except perhaps for the hissy sound from the coils of teh motor), I tried out a very gentle throttle test, and it did begin to spin up the wheel, though rotation is backwards (have to swap a pair of phase wires to fix that).

As I gently and slowly gave it more throttle, it reached a point where it sounded like a slam or a very loud fast grinding, just for an instant, and continued to spin. I backed off throttle at that point, and did not push it further.

At that point I was hot enough I decided to let the dogs guide me inside (cuz they were still scared from the fireworks boom), and write this up.

Sorry I still havent' posted the pics from the previous post, gotta fix teh card reader cable (I think Kirin tangled up in the cable and pulled it's connection apart...her and wires.... :roll: )
 
Another crosspost from the SFOC5 / hall sensors b gone thread:



MOUNTING:

Haven't got the live feed / logging going yet, but did mount up the controller and LED status board (inside a mini tic-tac box) and tie the wiring down. Everything still behaved as before, so I didn't break anything. (yet) :)

Then I swapped phase wires to get the motor running forward.

STATUS LED CODES:

I fully charged the traction battery, and now the LED status only reads a steady six, except every 20 seconds or so it goes to a 13 for less than a second, then back to the six.

13 is undefined (and I think 11 would be it's inverse in case I'm reading these backwards, but it's also undefined). With the switch on the lower left, and the LEDs across the top, I think they are 8 4 2 1, correct?


LOADED TESTING:

I'll get the trip log tomorrow, too tired to follow the steps to do it right now.


First, I did a low-throttle on-ground power test in the yard. There was no feeling of response at all; upping the throttle eventually got (at about half throttle) the motor to begin "rocking" but didn't spin forward on it's own at all.

I used the generic sensorless controller on the left wheel to get started, while still holding throttle at mid on the right wheel's SFOC controller, and at 4-5MPH I began to feel the right side pull; then I ran out of yard and had to stop.

Next was a test up and down the street at the front of my house, as that's the longest straight run right here. This late at night (about 1am at the time) there's no one on the street so is safe enough to do testing where I have to look at the displays on the handlebars a lot and mentally note stuff. :)

I had to startup each time using the generic sensorless on the left wheel, and sometimes a partial pedal stroke. Even if I then let off the throttle on the left wheel completely, and no pedalling, and even full throttle on the right wheel (SFOC5) then unless I was already somewhere around 4-5MPH I couldn't feel any acceleration/pull from the right wheel.

Once at that point, I could feel gentle smooth acceleration, like it was barely pulling me along. I saw around 28A on the CA at this point, fluctuating up and down, but I was paying more attention to the speed and feel so don't recall the full set of readings.

There were a few random "jerks" indicating most likley that the motor parameters are not fully tuned (connections are fine).

As noted above, the hissing noise is audible/feelable even at this stage.


While very quiet, even silent (ignoring the hiss), it doesn't have anything like the acceleration power I'd hoped it would. Might need to up the amps settings, if motor parameters don't fix it.


NOTES

NOISE:
The "hissing" is really loud. Like a large tire with a very bad leak.

Loud enough to hear while riding (even just coasting), over the wind and road noise, up to something like 12-13MPH. After that point I have to strain to hear it, but I can still pick it out in quieter moments.

Worse, I can *feel* it thru the frame and suspended-mesh seat, over the road vibrations, as a low random-frequency-feeling vibration. As speeds increase so does amplitude of road vibrations, so at some point around 10-12MPH it becomes difficult to tell which is which.

It doesn't change amplitude under any circumstances so far, though it does change frequency (upward?) during the 13 LED code.


The actual motor operation, while being driven by the controller, is silent as far as I can tell--I can't hear it over the hiss or the road/wind noise, at any rate.

I suspect different motors might sound different, and be quieter or louder depending on their design and size.

I can imagine many people being quite disturbed by the noise and hence not using this controller. If the point of having this controller on a bike was silence, I wouldn't want to ride it on bike paths or other quiet places without sufficient traffic/etc to drown out the hiss.

(the noise of a typical/generic non-sinewave controller is much louder, but it also only occurs while actively powering the motor to move the wheel. Since SFOC5's hiss is continuous anytime it is on, it's actually much more annoying.

EXTRA DRAG?
Along with the noise, the controller "holds" the wheel in position a bit more than lightly, whenever no throttle is applied, and appears to slightly attempt to fight manually turning the wheel. At a guess, this probably means wasted power, and extra heat generation inside the motor, and extra drag on the wheel when not actively powering the motor with the controller.

When it's off ground, I can see the wheel jittering back and forth just a teensy tiny bit sometimes.



STARTUP FROM STOP

I knew it wouldn't be able to take off snappily from a stop like a sensored controller, but I did expect it to do quite a lot better than it does right now (a crappy generic 15FET beats it all up). There may be settings to change to fix this (motor parameters, etc), which I'll explore tomorrow.

The generic controller stutters and occasionally "grinds" for a second at startup, but in most cases it is at least capable of really quickly getting the trike going from a stop. In some cases it won't move and I have to just very slightly push the trike forward either flatfooted fred-flintstone style, or pedalling just a partial rotation. Neither one of those is good, because it hurts my knees/ankle to do it either way, but as it's only a second or less I can usually handle it. (normally the sensored Grinfineon on the right, used along with teh sensorless generic on the left, gets the trike started with no help from me).

Even with just the generic sensorless, it doesnt' take long to reach 20MPH, and acceleration from a stop is very quick (as long as it starts forward smoothly or I help it).

However, at least with the settings it presently has, the SFOC5 is not able to get the trike started quickly. The worst part is that it accelerates slowly, feeling like it is either a linear acceleration curve or that it is only building speed faster as speed increases. The opposite of the way the other controllers do it, and the opposite of the way I need the trike to accelerate in traffic.



TEMPERATURE SENSOR:

The temperature sensor wiring requires +5v on one side, but all the (hub) motor sensors I know of (including on mine) that come built in are wired with ground on one side, so you might want to redesign that part of your circuit so the sensor wires can be plugged into existing motor temperature sensors (if they're compatible types, but it's a common sensor type). Or have it switchable, if there's a design option that lets you do that.

Before I can use the temperature sensor with the present design, I will have to open up the motor and rewire the sensor to put it's ground leg on the hall 5V line. That part isn't difficult, but having to take the wheel off the trike to do it is.

(otherwise I'd have to cut the hall cable entirely off the Grinfineon-to-motor connection to disconnect the ground wire from anything, so I could use it as the 5v line from the SFOC5. I really don't want to do that, since that would leave me unable to just move the phase and throttle wires back over to the Grinfineon in just a moment, should I have to do that roadside for any reason. I could put a connector in there, but I don't have any waterproof ones that have enough wires, which is why I soldered them in the first place).

More later; finally getting tired enough to doze off soon, I think.
 
I forgot to keep crossposting results from the SFOC thread, but most of the issues are resolved now. link to most recent results over there

https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680&p=1400680#p1400680

The hissing noise/vibration appears to be caused by a multiplication factor being too high in a control loop. Reducing it to 10% of what it was fixed it without introducing other issues AFAICT so far.

Programming/etc fixed partly in firmware (updated now) and partly with an extension cable on the serial end. Not as elegant as I'd prefer, but the other stuff I tried didn't work as expected, and a firmware fix could affect realtime operation of the controller.

Performance of the system improved partly with motor parameter refinements, and partly with a firmware change, and partly with lowering the phase current max. Still refining settings.


ATM, the biggest issue is low-speed acceleration. It doesn't do that very well, but it does GREAT once it reaches the speed I have to stop accelerating at, and pretty good a bit below that.

It doesn't startup from a complete stop, but I knew it wouldn't do that to start with. It does now at least begin to startup with only a little pedal stroke to get rolling a bit. However it does not yet have any "oomph" from there, until it gets to at least several MPH, and starts to really get going near 15MPH, and then whooosh like I'm on a catapult from there. ;)

Just need to get the catapult stuff at low RPM.



Started a thread for dealing with the realtime logging output from the serial port. First I guess that will need a program to parse an exsting logged file, and convert it to a table of human readable data. AAfter that works, then use stuff learned from that to take in the data live and convert it then and there (and log it, too), and display it on screen as it changes.

https://endless-sphere.com/forums/viewtopic.php?f=30&t=95633


Still need to work out something for motor temperature monitoring.
 
Broke the inboard axle of the rightside motor now. :(

https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680&p=1402067#p1402067

PIcs once I get it off and start working on the fix, probably tomorrow.
EDIT: pics added, 081218




Should be able to fix it the same way I fixed the leftside inboard axle, by grinding a notch out of the axle shoulder area for the axle stub to sit down inside of, then weld that stub in there good and solid.

Am thinking about a torqe plate, where I flatten the outside of the thick axle shoulder area as either a hexagonal or a square shape, then use the matching shaped hole in a plate that can be clamped or bolted to the trike frame.

Alternately with a hex shape, I could use the ring end of a wrench around it, and bolt that to the frame.

Either of these is probably capable of handling the torque much better than the little flatted axle.
 
EDIT: added pics 081218

Fixed the axle...
View attachment 19
dsc07673.jpg
dsc07674.jpg
dsc07675.jpg
View attachment 15
dsc07677.jpg
dsc07678.jpg


but screwed up the fix a bit: I was SURE I had made the slot in the shoulder area for the axle stub parallel to the other side's axle, because I had followed the lines of the broken bit of stub. But I failed somewhere along the way, as the welded-back-on stub is some 20+ degrees off from the unbroken side. Still don't see how I could've managed it, but I obviously did.

Anyway, can't fix that without cutting it off and regrinding/filing the slot, etc., and it took at least a couple of hours the first time. So...I pass on that.

Since I pretty much *have* to have the fixed (inboard) side vertical, as that's where the massive clamping dropouts are, I had to widen the slot in the outboard dropout so the now-angled outboard axle will still go up into it. If I fix the axle later, I can always readd and flatten material to fix that too.


While I had the motor off the trike anyway, I went ahead and opened it up to do a couple of things.

The first was to add the thermistor supplied with the SFOC5 controller, gluing it under the windings in a similar place to the built-in KTY83-110 sensor. The SFOC could be setup for a different sensor, but I don't know the procedure yet, so since I was in there it was simpler to just put it in there.
dsc07680.jpg
dsc07682.jpg




I did make a mistake, though: SFOC uses a +5v reference for the thermistor, whereas the MXUS and all the other ones I've seen with temperature sensors in them use a ground reference. I had meant to wire the thermistor from the red hall 5V to the white wire temperature outpout, but instead I wired it the same as the original sensor (disconnecting the original sensor's output from the white wire), not realizing this until after I'd closed up the motor and got it mounted to the trike. REALLY don't wanna take it off and open it up again, so Im just going to do the same thing I did with the KTY sensor wiring to SFOC, and leave ground to the halls (on the spare grinfineon left wired up to them) open, so the SFOC can have exclusive access to the sensor with no common wires to anything else in the system.



Next up was something better than the axle flats for resisting torque.

The plan was to file the larger diameter axle shoulder (which is about thumb-sized, vs the pinky-sized axle) into a hex shape that would fit a 14mm or 15mm socket (whichever turned out to fit best). Then drill teh socket's square end out in the press so the axle would fit thru it. Then weld an arm to the socket that I could bolt to the trike frame, and grind the socket down in height as needed to fit available space between dropouts and hub.

But it took at least three hours to file just two of the flats parallel enough, using a wrench to test fit them in between each tiny bit of filing. If I'd been smart, I would've gone inside to do this, but instead I was at the trike, outside, under the trees but still in the heat. So I was hot, exhausted, and a bit frustrated by the time I got that far. I just didn't have it in me to do another at least six hours for this side of the motor and then another nine hours on the other side.

So for now I gave up on the idea of the hex flats, and just went with the pair. And I cheated a bit on the other side, and used the grinder to get started, then filed the rest of it a bit, checked with the wrench, filed, checked, etc. Still took almost a couple of hours.
dsc07684.jpg
View attachment 9
View attachment 7



Eventually I got to a point I could test fit the whole thing on the trike, and find out where the anchor points of the wrenches would end up (so I could bolt thru the closed end of each to the frame). Before getting to the bolting on, I tested it by tightening down all my normal hardware, tehn hose-clamping the wrench ends to the frame.
View attachment 8
View attachment 6


The hose clamps sucked, and couldn't hold anything in place even against just a little startup torque. WOrse, the inboard wrench came completley off when the motor stuttered and banged from the controller doing whateve rit does to cause that, as I eased on the throttle.

So I got the metal plate to weld to teh outboard frame for that wrench to bolt to, and drilled the hole in the inboard frame for that wrench to bolt to. I used a 14mm nut on the bolt just below the head and washer, for the outboard wrench, to help fix the wrench to the bolt/support with no movement possible. The inboard wrench I was able to use the 15mm head to clamp down onto and partly into the wrench, to keep it from being able to move around. (bolt too short to do it the same way as the outboard one, didn't have another bolt with the otherwise-correct characteristics).
dsc07690.jpg
dsc07700.jpg
dsc07691.jpg
View attachment 3
dsc07698.jpg
dsc07699.jpg

Went to do some test rides, with their results in teh SFOC thread linked in some above posts.

Seems to work well enough so far, haven't felt or heard any wheel wiggle.



PIcs coming as soon as I can convince myself to go back outside and grab the camera to upload the card contents...but ATM my head is still hurting from the overheatedness; sipping water and cooling down and resting till then.
 
Test ride of 5 miles ok so far. Might be my imagination but it "feels" stiffer than it used to, just a little; perhaps it's always been rocking a little in there and that's what eventually broke it (probably).

Anyway, I'm testing the SFOC here:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=30680&p=1402444#p1402444
on a 10-mile ride today, am halfway thru it. First half tested at only 150A phase (had been 250); acceleration definitely suffers but it's easier on the motor/etc. Am going to up that to 200A and see if I can find a happy medium between stress and acceleration times.

wifi bandwidth too low to upload pics so later when I get home.
 
EDIT: added pics 081218

On the way home, the SFOC "BANG"ed again, which hasn't happened with this firmware and settings before. When it happens, the noise is quite loud, there's an instantaneous braking jerk as if it suddenly completely reversed rotation at full torque (or more) for an instant. It jerks the trike to the right very hard, I'd almost say violently.

Anyway, right after this, the wheel felt different, and then I began to hear rubbing of the tire on the frame...the axle broke again on the inboard side. :roll:
dsc07702.jpg
dsc07704.jpg
View attachment 2
dsc07713.jpg





The wrenches I put on the new flats on the shoulders may have prevented it from breaking both ends; the one on the inboard side did prevent the wheel from tilting enough in the frame to jam, and actually enabled me to ride home with it that way, several miles, and then go out for a grocery trip for another several miles (I just had to put a thick washer on the inner side of the outboard axle dropout to push the motor inboard enough so the tire wouldn't rub on the outboard frame).


Side note: The wrenches had had to be tapped onto the flats originally; now they are loose on them, and wiggle. So the BANG or other issue has also either bent the wrenches, or deformed the flats. I can check which by testing the wrenches on the nuts used to hold their bolts in place--they previously had to be wiggled onto the flats of the nuts, so if they don't, the wrenches are bent open. If so, unsurprising and why I wanted to go with a closed-ended solution to start with.



Unlike everything else I've welded up, this did break at my weld, becuase I didn't weld it as deep as I should have, trying to avoid damage to the wires in the axle (as I didn't want to take it all apart and pull the wire thru, etc; instead I kept ice and water right there, and wrapped the wire bundle out of the axle in a wet rag, and managed to avoid wire damage (it did damage the outer insulation jacket). The weld only reached a small part of the outer surface; never got into the "trench". Plus the rapid cooling I had to do to keep the wires intact probably made the welds that *were* there brittle. :/


Fixing it is going to require taking the wires out completely, so I can get good enough penetration, and let it cool down naturally.


HOwever, if I'm going to fix it again, I might as well do it "right".

I wanted to put thicker phase wires on it anyway, just to see if they make any difference to performance in the SFOC5 controller. but there's no way to do that with the small "standard" sized axle this old v2 MXUS 4503 has. To do it, I'd need a bigger sidecover bearing with much larger ID.

Well, to fix this right, given the loads and torques I'm after, I'd need a larger bearing anyway, to fit a fatter axle.

I have some old cut off sections of 3/4" and 1" bolts, from railroad rails I think. If I cut the axle back to it's widest shoulder inside the motor, and thoroughly weld these onto that, then as long as there is sufficient penetration it'll be a lot stronger than the maybe 15-16mm or so thick axle it has now (between the threaded sections, which are only 13 or 14mm tall on the threaded area, and only 10mm between the flats).

If the bearing's ID is large enough to also accomodate a nut sized for those bolt sections, I can also weld that nut to the bolt, and use it's wide flat hex sides as a gripping point for a torque arm, and still be able to get the cover on/off. No "axle flats" will be needed, so the axle should be thick enough and strong enough to take the regular weight loads of the trike, inclidng potholes and the like, that have also broken at least one axle (on the X5304).


Unfortunatley I don't have any single piece of metal that's big enough and long enough to just replace the whole axle all the way thru, so welding new ends to what's there is about the only "easy" option I have.


Then I just have to make a "filler ring" or spacer that fills the gap between the OD of the bolt, and the ID of the bearing. The phase wires can go thru this, and so they can be as large a diameter as the gap between the two. If necesary I can make the ring in segments, and epoxy them in place on the axle, lined up with the bearing ID.


My limitation on bearing OD is that I'll have to carve out the disc brake rotor mounting area from the inside to put it in the cover.

I don't have that option on the freewheel side (outboard on this side), so I cna't use a bigger bearing without either a new cover, or making a sort of "tophat" adapter to hold the bearing, that bolts to the cover. So since the outboard side is holding up ok so far, I'll leave that as-is until I can work out a cover solution.


To get the larger bearing, I have some powerchair brushed mtoors with gearboxes, and the axle output has a pretyt large bearing with large ID too. I'm soaking one in a citrus and soap solution to get the rest of the grease out of the gearbox, so I can see in there to undo the stuff securing the bearings to the box housing, and the axles, and whatnot, to get it apart more easily. Once it's apart I can see what I've got, and compare it to the cover's disc brake rotor mount, and plan around all that.

Not gonna be all that quick a "fix", cuz I don't wanna do it twice, ;) so I'm hoping the present mounting stuff holds the inboard axle down well enough for my commutes until then. Could be at least a week or two.

If not, I may have to swap back to the old HSR3548 with the bent rim, until I fix this axle, since that's only just about a couple hour's work to pull the 4503 wheel off, take its' tire and tube off and put them on the HSR3548 wheel, then mount that wheel back on the trike, and move the PP75 connectors over from the 4503 so I can plug it into the Grinfineon rather than the SFOC5 (as I don't want to break the HSR with any SFOC5 misbehavior; it's my backup wheel).


I'm going to have another week or so off sometime before the middle of september, but don't know when yet, so gonna be workign on this fix with any spare time that weather/etc permits.
 
Front braking squealed in every turn today, both to and from work, regardless of how they were adjusted. Warranted investigation, so after resting and dinner and a bit of forum stuff, I went out and checked, and found the locknuts inboard of the dropouts to be loose, which should be impossible.

Undid the QR skewer, and found the hollow axle is actually broken at both ends, just inboard of the cone nuts. :/ (pics tomorrow)

I've never seen that before...I guess it's from teh sideloading the front wheel sees in turns, being a non-tilting trike that doesn't have to go slow around corners.

Several of the bearings have flat spots on them, probably from grinding against bits of the broken axle.

I have other QR wheels I could take the axle out of, but I thought first I'd try to put a slighly larger axle in there; unforutnately the bearing cup is just a hair too small to allow any of the bearings I have to fit between the bearing surfaces of the cone nut and the cup.

Since I already have an old solid front-wheel axle (slightly smaller than the QR axle) in a drawer with nuts, bearings, cones, washers, etc., I just used that for now; if I have problems I'll replace it with something else.

So that's fixed.




Now it's bedtime, accordng to my body (though my brain doesn't want to sleep; i'll crash anyway soon enough), so other fixes will have to wait till tomorrow.


Something else that happened in the last couple days is that the BB shell has broken away from the trike frame. I'll have to take the plastic side panels off (the right one is riveted on), and move the wiring out of the area so I can reweld it. I also have to take the TDCM torque sensor out.


So those things plus still needing to fix the axle on the 4503....


I am also hoping to get to weld up some mounting points for the under-deck battery box, which will require moving teh controllers out of the way, and rewiring them (because all their wires are just exactly long enough to be where they are). Once the box is made, I can move the main battery down there, and have the whole cargo box for...cargo. :)
 
Test rode the front axle today on a 10-mile ride; worked fine. Also eliminated the "regular" brake squeal I had been having, whenever applying brakes lightly rather than fully. I guess that was caused by the broken axle, too, though I dont' understand how (I have theories, but they're probably wildly wrong).


After the ride, I pulled apart the front triangle, and found the entire BB shell (which I'd cut off an old 10-speed made from SS lugs brazed to chromoly tubes) had simply broken away from the frame at the welds...because I had only hard-tacked it in place, and forgotten to go back and weld it good and solid. :roll:

I remember way back when I changed to this drivetrain, I was experimenting with BB positions for setting up the pedalling position, and I guess the tack welds were enough to hold it until now. Guess that says something for the welds, but not for my memory and planning. :/


It took about 5 hours to fix, because I kept screwing things up (kinda normal). Only took about 20 minutes to take off the panels and get the wires out of the way, probably a record. Then another most-of-an-hour to get the TDCM BB sensor out of the shell, because the tools I used to install it seem to have been mislaid, and I had to improvise.

The welding up of the BB shell was quick and easy; about another 20 minutes.

But in order to prevent the threads inside the BB from getting weld-splatter stuck in them, I threaded in the endcaps from the TDCM.

That was a mistake. I should've used "sacrificial" ones off an old BB that had the "nut" style ends on them, to make it possible to wrench them out after welding, because of course the BB warped a little bit from the heat.

So I ended up damaging the threads on the TDCM endcaps, having to then file and scrape them to allow them to be reinstalled once I got them out. :/

The left cap was too damaged, and I couldn't get it back in far enough to be used. I ended up using one off another old sealed-bearing BB set; the threads in the BB shell itself are also damaged so even this one was tough to get in there in a way I could still get it back out later.

After more than 3 hours of messing with it, though, it's done and works. And the TDCM still works; at least the CA seems to read it correctly within Setup; didn't yet test it in operation.


After all that, in the heat and humidity, I had to come in and take a shower and a long break and dinner....and now I'm pretty wiped out.

If I can, I'm gonna try to at least get the motor apart, stator out of rotor, and wires pulled out of axle, so I can attempt to weld on the new axle tomorrow. I'll still have to get the larger OD bearing out of the powerchair gearbox and see if it'll fit where I need it to, and if it does, then hollow out the disc-mount area for it.

But first, it's naptime. :/
 
Was so exhausted last night that once I got sleeping I stayed that way until about noon today; even the dogs didnt' wake me for breakfast. :shock:

Without having done the prepwork of taking it all apart last night, there wasn't enough time to be sure of being able to do that today, especially as hot as it was (all the work has to be done outside; despite the shade it's still too much to stay out there too long at a time, especially when there is not even a slight breeze).

So maybe this coming weekend. Or if there's enough time after work day after tomorrow (won't be tomorrow), I can move the tube and tire over from the 4503 wheel onto the old HSR3548 wheel with the bent-up rim, and put that on the trike in place of the broken-axle 4503 until I fix it up.




Side note:

I need to finish up making the "trike shed" insulated enough to make using a window A/C unit effective in there, then I can do a lot of work inside there, once I move the unrelated stuff out of it. I had put an old window unit in the uninsulated shed under the biggest mulberry, but it's simply insufficient to make but a very small difference, evne running constantly.

The trike shed is half-under a denser "eucalyptus" tree that shades it from the west, at least, and the roof not under the tree is covered in a couple of layers of styrofoam cooler lids, which are covered by some long thick heavy rubber mats, and a large area rug salvaged from the alley bulk trash some years back. That is significant insulation, and means it doesn't heat up as quickly in there as it would otherwise, and it doesnt' get as hot as it would (20F+ more than the outside air even with the door open), but it's still as hot as the outside air. Parts of the walls are insulated with a single layer of cooler lids (about an inch thick), but the job isnt' done yet, and the many air leaks of the typical tin-shed aren't covered yet either.

So I ahve to do those up first, then open up a hole in the back (under the tree) for the window A/C unit, so it's in the shade and operates more efficiently than otherwise.

But...today I had to redo all the roof insulation, because a storm a couple weeks ago or so had blown it off, along with the old aquarium (which it destroyed) I used for solar cooking/tea/etc. Then I had to secure it all down, which I hadn't done previously, so that the next high winds wouldn't do that again. (hopefully). (and come up with a new solar-aquarium-thing).



If I didn't have so many things that have to get done in order to do all the other thigns that have to get done, so that I can then do the other things I'd like to get done, and so little time to do them, whenever the weather is good enough, or there's enough time in a row to do them, or I have the energy and ability, etc....I'd feel like I accomplished a lot more than I do.

I do actually get a lot of stuff done...it's just that so often it is repairs to otehr things already done, or prepwork for something else (which is sometimes wasted because something prevents me from doing the something else, until it's been long enough I have to redo the prepwork, etc). Or just some other thing that comes up that has to get done "now" instead of whatever it was I'd actually planned. (like this broken axle thing...it's already taken a couple of days of time away from other things, and it's going to take at least one or two more, maybe a week if stuff doesn't go as planned).

Sometimes I wish I just had money I could throw at all these things instead. :lol: Sure would save time!
 
Tomorrow I'm going to have to swap out the righthand wheel with the MXUS 4503 in it, to the old bent-rimmed one with the HSR3548 in it.

Today on the way to work, I had a sudden stop caused by the inboard axle completely failing past the shoulder, so nothing keeping it from tilting outboard at teh top and essentially turning the tire into a giant brake pad against the trike fender frame. :/

Was stuck at the side of the road for 20-30 minutes while I worked out a temporary fix, since I was too far from home to go back and nowhere near close enough to work to get there with the problem.

Was actually easy to fix, but had to test it a bit at low speeds in turns on the back street before I headed back to work at normal speed in traffic.

What I did was to pry the rivets off the rightside fender cover, and unscrew the wooden top of the fender, leaving just the steel frame around the wheel.


Then I used the longest wrench I had with me, 1" open/closed box type, and place it's open end over the remaining axle shoulder, against the inboard dropout face. The body of the wrench points straight up from there, and the closed ring end of the wrench was tapped (with anohter wrench) to fit snugly under the top inboard fender frame tube.

Right at that point there is a hook that was still welded to the tube, from it's previous life as some part of some retail signage; I left it in case I might ever use it to hold wires or whatever in place. In this case, I used it to run an X of zip ties thru the ring and around the tube, on either side of the hook, so the round ring can't slide back and forth against the bottom of the tube under vibration and bumps.

I'd've used a hose clamp but didn't have any with me.

It worked well enough to get me to work and back, which is all it had to do.



Another problem to deal with tomorrow is the front wheel / axle again:

Yesterday, I had to tighten the axle's cone and locknuts, becuase I could feel the wheel wiggling and vibrating on the way to work. Today, before I left work for home, I had to do it again, and I could feel it almost as bad as originally before even halfway home.

So something is wrong with the hub itself; possibly the cups are pushed inward, detached from the hub body. :/

I'll have to swap the wheel out until I resolve it's problem. Thankfully I have a number of usable front wheels; don't think any of them have as nice or as wide a rim as this one though. Some are as wide, but they don't have machined braking surfaces (still aluminum though) Guess I'll experiment with what works best for braking (most important) after I swap the rear wheel out.



I *might* also just try swapping out the front fork for the fatbike fork, and put the ex-Stromer Ultramotor wheel in it, after running wires out from the motor to use it with an external controller. Kind of a lot of work so probably won't happen just yet, since I have a fair bit of work to do swapping the righthand wheel out:

--remove the 4503
--remove tire and tube from 4503
--install tire and tube on HSR3548
--add PP75s (or SB50; same contacts/etc) to HSR3548 phases to plug into Grinfineon
--move SFOC5 to left side, to hook up to 4504
--take 4504 off so I can add flats to shoulders for wrenches to act as torque arms to ensure it's axle won't be destroyed like the 4503's was, should the BANG! occur over on that side, too (it shouldn't with the updated firmware, but...).
--Add mounts to bolt those wrenches to the trike frame
--add PP75 (or SB50) connectors to the 4504 so ti can connect to the SFOC5, and add them also to the generic controller in case I ahve to swap over to it from the SFOC5 for testing or emergency or whatever reason.
--mount the 4504 back on the trike, including the wrench torque arms on the axle shoulders

The reason for the shoulder-mounted torque arms is because it's obviously not enough to have my clamping dropout on the axle flat itself, when the motor torque is high enough and/or direction change is fast enough (which also equates to high torque) it'll snap at the dropout face (it's happened on the 4503 twice, and the 4504 once, for different reasons).

Also, these arms are going to be vertical, like the big wrench I'm using temporarily above, so that they can also hold the wheel in position verticaly even if the flatted part of the axle does somehow break off, and prevent it from rubbing on the frame.

If I had the tools/time to do it, I'd just add stuff to the existing dropouts to act on these shoulders.

If I can, I'll use "spare" wrenches of the sizes that will work best on these shoulders, and actually weld them into place on the dropout faces, so the axle and shoulders will simply slide up into these as part of the dropouts.

But doing this requires that I get everything perfectly parallel on both axle shoulders, , and I don't think I can do that with what I have, in the time I have. So the best shot I presently have at doing this in the most secure way possible is to have the wrenches bolt on separately.
 
(Pics once I can get out from under the dogs on my legs and go get teh camera card)

Today at least I got these steps done:
--remove the 4503
--remove tire and tube from 4503
--install tire and tube on HSR3548
--add PP75s (or SB50; same contacts/etc) to HSR3548 phases to plug into Grinfineon
--mount HSR3548 on trike
--add a wrench torque arm on the outboard axle, just outside the motor cover, on the top side of the axle so it also provides some vertical support should the outer end of the axle fail. Since the outboard sides haven't failed yet, that's just a precaution, and probably actually wasted time. But better safe than sorry.

I tried to setup that big wrench as a vertical support for the axle shoulder like on the 4503's emergency repair, but it won't reach it / rest on it correctly; it only worked on the 4503 because the axle was already broken.

With the wire exit at the shoulder, there's not quite enough shoulder / space to let me make vertical flats there for a different wrench to go on the inboard end like I did for the outboard one. If I change the length of the spacer between outboard dropout and outboard shoulder, I can fix that by shifting the wheel more outboard and giving a bit more axle space inboard, and putting the wrench on the axle itself, but I'd already spent almost all day getting the rest of this done, and needed to fix the front wheel, so I can go get some groceries and other stuff tomorrow morning, without worrying about potential failures.

PUtting the wrench on the axle rather than the shoulder wouldn't really accomplish anything the dropout doesn't already do anyway.





Didn't have PP75s left, so used two SB50s, the second with only one contact in it, since they'll mate with PP75s. Went ahead and wired it up to the Grinfineon for now; hall wires are soldered (since the SFOC5 doesn't use them no connector is needed to swap between the two controllers for testing).



Tested this a little bit around the yard, then out on the street around the block, and it was so loud that at first I thought something was wrong with the phase/hall combo (even though I"d verified it off-ground both forward and reverse), but then realized I was just used to the silence from the SFOC5! :lol:

It worked under hard acceleration, braking, etc., though it has so little torque compared to the MXUS / SFOC5 combo that it's difficult to accept. :/




Next up was fixing the front wheel's axle issue.

Took it off and brougth it inside, since it was getting dark outside and I needed light to see, so as not to lose bearings, etc.

ONce I got the axle out, I found the grease on the left side was silver with ground up metal particles (it started out translucent red). There were 9 bearings on each side to start with...but there were only 7 on the left now. Rigth still had 9.

There's no way at all for any of them to have "fallen out", so that means the metal particles are all that's left of two whole bearings, and all of the others on that side were damaged, with flat spots and missing chunks. I'm guessing some bit of debris somehow got into the grease as I reassembled the hub/axle/bearings when I replaced the axle...but I'll never really know.

Races do not appear to ahve any pits or scratches, which is kind of amazing. So at least I can keep using the hub if I replace the bearings.

Cleaned it all out, squeaky clean, and took apart the hub from the BMX wheel I'd used to get the spokes for Raine's trike for the left rear wheel with the Ezee v1 hub. That gave me 10 bearings for each side, same size as the previous ones both original and the last set that disintegrated on that side. Wasn't sure they'd all fit, but they do.

The axle is thicker, so I can't use it; it won't let the bearings run on the races of both the hub cups and the axle cones.

Greased and adjusted, installed, and works fine now. Tested around the block with hard turns and braking, etc., no problems.




Didnt' get any of the rest of the list done yet, except
--move SFOC5 to left side, to hook up to 4504
I moved it to the center instead, between the other two controllers, so it can reach either side's motor. That way for now I can test it with the HSR3548 until I can get the 4504 setup for safe testing.

This stuff is still to come.
--take 4504 off so I can add flats to shoulders for wrenches to act as torque arms to ensure it's axle won't be destroyed like the 4503's was, should the BANG! occur over on that side, too (it shouldn't with the updated firmware, but...).
--Add mounts to bolt those wrenches to the trike frame
--add PP75 (or SB50) connectors to the 4504 so ti can connect to the SFOC5, and add them also to the generic controller in case I ahve to swap over to it from the SFOC5 for testing or emergency or whatever reason.
--mount the 4504 back on the trike, including the wrench torque arms on the axle shoulders
 
The SFOC5 is back at Incememed's for repair; somehow the 5v line it uses for thermal sensor was shorted inside the motor or axle of the HSR3548 to the axle or supports, which meant shorted to the trike frame. I verified nothing else was shorted to the frame, so nothing should have happened despite this, but it still killed the SFOC5. Incememed said it was that the frame acted as an antenna and essentially funneled radiated energy from the phases/etc into the 5v line. So some changes are made in the SFOC5 now to prevent this.


In the meantime, I got a new set of tires (same kind, Shinko SR714), since the old ones have just a hair's thickness of tread left in the middle, and even little 1/4" long thorns are getting thru to the tube.

I also got a new set of tubes, since the ones I'd gotten before (Sierra?) have had several problems with unpatchable holes in the inner circumference, where it rests against the rim. The actual cause of the holes appears to be my "rim tape" stretching and failing. allowing nicking or pinching of the tube, except for the most recent one, which appears to be that I unavoidably ran over a broken-off branch (from a storm), and it was probably mesquite or similar, with thorns. One or more of those made it thru the tire and tube, and though the slime I had in the tube (just in case, because of a problem noted next) let me reinflate it, the thorns had made it all the way thru the inner circumference too, so it wouldn't stay inflated.

But since the patches/adhesive won't bond to the tube surface sufficiently (regardless of source of the patch/adhesive), the tubes aren't trustworthy since I can't be sure I can fix them if there is a problem on the road.

I sort of patched the inner-circumference hole enough to let me get home, but had to change out the tube to make sure it'd last. Was the first flat I'd really had to worry about in a long while. But it was the trigger for getting the new tubes and tires.


The new tubes are mostly natural rubber, and much thicker (and heavier) than the others. Larger diameter too, so they don't have to stretch nearly as much as the others, and stay thicker while inflated vs the others. Same cost. I'll edit in the brand once I am not weighed down by a quarter ton of dogs surrounding me on the bed. :lol:

I also got a roll of wide gorilla tape to add as rim tape in a double layer. This ought to be tough enough, over what's already there, to not let the tube be damaged by the nipple holes in the rim.

For several days I only had one new tire and tube on, just on the right side, until I had time to take the left wheel off and do a bunch of things.

The first was to find out what was causing the squeak/creak every time I accelerated/braked with that motor wheel (the 4504). I assumed it was the axle I'd previously repaired, broken again, but it wasn't. Neither was it the frame, or dropouts. In fact, I couldn't find anything that could explain it, and after reassembly, the noise was gone...so I don't know what caused it.

Next was upgrading the phase wires, and reinstalling hall and temperature sensor wires (which I'd removed when I'd repaired the phase wires back when installing it). I at least tripled, probably quadrupled, the crosssection of the conductors, barely fit thru the axle hole with the extremely thin hall/sensor wires, and a layer of heatshrink around them.

The temperature sensor wire came off in my hand from the built-in sensor, whihc is on the other side of the stator from the cover I opened up, so rather than take that cover off too, I just installed an old 10k NTC I had laying around from a computer fan/etc monitoring system I no longer have; this will work with the SFOC5 when it comes back without modification. I used independent wires for it from the hall sensors; it doesnt share any ground or power with them.

I also changed from direct soldered connections to PP75s on the phases (so I can just unplug it from this to the SFOC5 when it comes back from repair). The halls aren't connected to anything at all rightnow, just bare wires, since the present controller is sensorless.

I haven't done any of the axle flats/etc upgrades I wanted to do; there wasnt' time that day. Don't know when I'll get back to it to do that.


'The new tires handle much better than the old worn ones, even though they're the same kind. Slightly less grip, since they're rounder profile than the worn-flat ones, so less contact patch.
 
Back
Top