Semi-Recumbent Recycled-Parts Cargo eBike: "CrazyBike2"

[Alan B]

With all the weight of your setup, and the daily use and reliability that you need, a moped tire would be good. They are also inexpensive. They do require clearance and to fit the rims.

[amberwolf]

I've been looking for a set of them locally for cheap enough to afford, preferably whole moped wheels with tires on them, used, that I can recycle into the bikes. So far a few sets have come up via people I know or meet, etc., but all of them either flaked out, sold or gave them to others before I got there (even though they said they were held for me), raised the price beyond new cost, or decided not to sell.

I have two sets of dirtbike wheels with tires, but A) they're knobbies and B) they're really big and heavy. I'm planning on using one of them anyway, on the newer bike being built for heavy-duty use, but on CrazyBike2 it's a bit much. I have considered it several times now, anyway, recently. But I'd need to either go front wheel drive again, or setup a middrive, and if I'm going that far I need to spend that time working on the new bike instead.



In today's news, that battery pack appears to be working normally. I left it on a headlight until the pack dropped to 10.8V (3.6V/cell), which took more than two hours, close to three I think. Unfortunately I forgot what time I started the test at, so I don't know exactly, and I also forgot to write down the wattmeter readings so I don't know what all the readings were. Too many distractions today with dogs and yardwork and housework. I almost forgot I was doing the test at all, and managed to remember only at the last minute.

Oh, and the final wattmeter results from the charging session just prior to that showed that it took in 22.3Ah, from 9.8V up to 12.42V (4.14V/cell).

I haven't recharged it yet, as the SLA used to run the Venom are still recharging on the Sorenson.


A couple of days ago, a coworker gave me one of his old MC helmets, a half-helmet by AFX. It's from 2007, FX-3 II, with the POW-MIA "you are not forgotten" logo on each side. It fits at least as well as my bicycle helmet, maybe better, and so far it doesn't seem to be too hot to wear. It should provide much better protection than a bicycle helmet, even the type like my swirly-colored Nutcase, should I ever need head protection in a crash or fall. I would prefer it not be black both for visibility and to stay cooler, but hey, I can't argue with it for the price. I'll probably add reflective tape to front and rear to increase visibility at least at night.

DSC06358.JPG (48.11 KiB) Viewed 451 times

DSC06359.JPG (53.13 KiB) Viewed 451 times

DSC06360.JPG (24.76 KiB) Viewed 451 times

DSC06361.JPG (46.4 KiB) Viewed 451 times

[texaspyro]

I'll probably add reflective tape to front and rear to increase visibility at least at night.

Don't forget the obligitory duct tape and day-glo paint...

[The fingers]

Be sure to write your blood type and emergency phone number on the inside if you do not carry a Red Cross donor card in your wallet. (got the idea from a MC rider @ work) Free stuff is great!

[amberwolf]

On those valve stem leaks, have you ever tried shrink tube to cover the whole stem, then air it up ?? Might be you are getting some kind of flexing and that fatigues the old rubber as the hot dry climate messes with the rubber ?

I actually never thought of heatshrink. That will be my next trick. I'll do that on the one I patched the stem of, and try it out, see if it keeps the patch in place. I'm pretty sure I have some of the right diameter.

This is now under test here:
viewtopic.php?p=576368#p576368

though it hasn't seen a road test yet.


This morning I left to go help a friend put a harbor freight trailer together for his motorcycle, and about half a mile from the house the headlight on CrazyBike2 fell off.

DSC06401.JPG (50.9 KiB) Viewed 428 times

Metal fatigue from vibration on the road, with that heavy cantilevered load out on the old bike headlight mount, which was never meant to take this weight at that angle with the bending it had already taken to aim the light.

DSC06402.JPG (29.37 KiB) Viewed 428 times

I went back home and used a retail shelf bracket that was being tossed out during the remodels as a quick fix. Have to actually custom-make a new bracket that will let me pivot the light to aim it without just bending the bracket to get it right.

DSC06405.JPG (32.74 KiB) Viewed 428 times

Only a bit of filing/grinding at the top of that slot was needed to let the pivot bolt fit thru it, but it can't actually pivot because the bracket sits right against the weld for the square tube to the headlight frame.

DSC06406.JPG (45.42 KiB) Viewed 428 times

Another wierd problem is that somewhere along the way, a couple of miles or so, the disc brakes just stopped grabbing. They work mechanically, and the pads "grip" the rotor fine. Adjusting the pre-tension on the cable changes nothing; I can already get full braking movement well before the brake handle reaches the handlebar.

While I was at the friend's place, I even swapped the pads on the caliper for the ones on the complete spare caliper I have been carrying with me since I installed this, just in case the cheap-looking flimsy-feeling caliper came apart (which hasn't happened). No change. Same with using the whole spare caliper.

So that leaves a problem wwith the disc itself? I see and feel nothing wrong with it, no oiliness, no dust, color changes, etc. But there is ZERO friction on it--I can hand-spin the wheel offground, and squeeze the disc brake lever, and no change in spin rate happens, even though the pads appear to be making full contact with the disc.

Fortunatley I still have regen braking and the rim brakes, so I used those for the ride home. About 90% of the way home, maybe 3/4 mile from the house, the very light sprinkling rain that had just started turned into a heckuva windstorm (enough to almost blow me off the road, and nearly overturning a tall but small van when it passed out of the wind-shadow of a building), and really started raining cats and dogs. Fur everywhere, it was terrible.

I had to slam on my brakes for a couple of car drivers that suddenly forgot how to drive (this happens every time it rains here--the water melts their brains or something), and guess what? The disc brakes worked--screaming horribly, and nowhere near as well as they should, but they worked.

So....contamination on the disc itself is all I can think of. I got home and used ammonia to clean the disc, in case it is oily it should help remove it. It is still raining a little out there, though the wind has just about died to a breeze, so I am not yet going to go test it. When the rain stops I will see if the brakes are working again as they should. They do stop the wheel completely now in the off-ground test, and I can't hand-spin it against them.


The contamination theory bothers me, though, because there is NOTHING I could have ridden thru that would do that, especially in the very short time between leaving the house and the failure of the brakes. Nothing on the wheel/hub/etc that could have done it either. It is just hte only logical remaingin answer.


As a side note, I have for a while been checking the motro and controller temperatures by "manual" feel test, and they're a lot hotter than they used to be when in front. Teh cotnroller is still only warm, but it is quite noticeably so, where it wasn't really when in front frame. The motor is too hot to touch the windings right after stopping on any portion of the ride. I can just barely touch them before I must remove my fingers from them. The covers are pretty warm near the coils, but not super-hot. Between spoke flanges is also not hot, just warm. Axle and whatnot is all cool.

So it is only at the windings, meaning that being in teh rear between the cargo pods and behind/under the seat, it is getting nowhere near the airflow it used to when in the fornt fork. Also, the change in covers with different cooling hole types is not doing what I had hoped (though it probably is cooling it significantly--I need to test this with a thermometer and sensor, with cooling as-is and with the holes all sealed with tape).

[amberwolf]

The disc brake in front is still acting wierd, and I haven't figured it out yet. It doesn't really have any braking power at all, no matter how hard I squeeze it. Kinda like wet rubber rim brakes on a shiny steel rim. And they make a horrendous noise when just beginning to engage; just about like car disc brakes that are worn down a lot do--not a squeal, but a metallic grinding that is beyond loud. I need to take them all apart and clean the rotor again, then check and clean the whole caliper, pads, etc., and see what happens.



The rear wheel flat problem has returned as well. It is a fast-slow leak, in that while I ride and am at work (6-8 hours or so) it does not seem to lose any pressure at 55psi, but overnight (12-14 hours) it loses all of it's pressure. I haven't taken the wheel and tire off to check the tube yet, but I don't see a problem with the valve stem so far, so I'm not sure where the leak might be.



The temporary headlight bracket hasn't failed yet, which is good cuz I havent' had time to dig out the stuff to make the new one from.



Other than that, things seem to be working ok.


It's getting hot enough (102F outside when I left for work today about 3pm, and 90F in the house, 92F inside and out when I got home about 9pm), that I need to get cracking on the swamp cooler fixes for this year. Stuff I have meant to do all year last year and this winter and never took time to get to. Definitely gotta do it now. Will probably delay several other projects.

[JohnC]

I need to get cracking on the swamp cooler fixes for this year.

AW, Can you post your swamp cooler design on your youtube page? Thanks.

[amberwolf]

The existing design, or my planned fixes?

The existing one is a Mastercool from probably 15-20 years ago, at a guess, and the casing is all rusted out on the bottom. A temporary fix a few years ago was to use some clogged-up sprayfoam cans and smear their contents around the outside of the tray, with remnants of old clothes stuffed into the holes left when I poked at some of the more rusted sections. It worked ok, but still leaked, and birds have pecked and pulled quite a bit of the foam in the leaky wet areas, making it mostly pointless after a few months or so.

At some point the old rusty pump died, and I connected the water inlet directly to the drip pipe over the cardboard on the air inlet, and set the valves for just a trickle, which mostly wet the cardboard enough for cooling the air as it passed thru, but it wastes a fair bit of water. Technically it's not wasted, as it waters the mulberry tree that partly shades that end of the house and the cooler itself, but it is a lot of water--probably a hundred gallons a day including the water evaporated into the cooling air, if I run it 24/7. So it only gets run long enough to cool the house down to livability, then shut off both water and motor.

I now have an old aquarium powerhead that should work as a pump, but the theory isn't tested yet.

Later I was given a disintegrating misting "rope", which I reduced to the non-broken sections and sort of wrapped around the intake end of the cooler. The misting wastes very little water, as almost all of it gets sucked into the air intake, and it cools more efficiently than the drip-pipe-soaked cardboard the cooler was designed around. But it makes it much more humid in the house, whcih makes it feel a lot hotter once I shut the cooler off. :/


So my current plans revolve around welding up a new water pan for the cooler, and making up whatever I need to get the aquarium powerhead to work in place of the old pump. Most likely the pan will be made from the sheet metal of the bodies of the several salvaged clothes washers and dryers I have here, as that's the only larger sheets I have available, other than old retail shelving, which isn't any thicker metal and is harder to get apart.


The main motor of the system is pretty old, and I have to lube the bearings several times a season now, or the motor fails to startup and just hums. I'm considering an oil-filled hubmotor to replace it with. I fleetingly considered Auraslip's broken-axle 9C, but I'd have to re-seal the covers to do the oil-filling. I have to scrape the deposits off the motor nameplate and see what it's rated at, to figure out if the hubmotor could survive continuous usage at those power levels. I don't have a power meter that can read the wattage usage of the existing AC motor, but I might see about borrowing a Killawatt meter from SRP (they let you do that thru the libraries here, IIRC). Not sure if the KAW can handle the startup current, though.

[amberwolf]

Guess he wasn't really interested in the swamp cooler, with no reply after a couple of weeks.


Anyway, speaking of swamp coolers, since I still haven't finished fixing mine, a friend brought me the "portable" one he uses in his shop/patio/etc., to borrow for the summer (or until I fix mine, at least):

DSC06768.JPG (40.56 KiB) Viewed 325 times

I will have to make a duct to run from it's output to a window so I can properly use it to push air thru the house (rather than just keep adding humidity to existing internal air), but it works very well. I have not yet measured the temperature difference it can create at its output vs the intake air, but it feels like 20F or more. Should be enough.


FWIW, I haven't been riding CrazyBike2 (except for shopping trips) for the last week and a half or so; instead I have been riding the "new" Fusin Test Bike here:
viewtopic.php?f=4&t=39877
I definitely prefer CB2, mostly due to seating and cargo capacity, but it is nice to have full-suspension on the Fusin Test bike, making me want to finish CB2's ultimate replacement even more.

I had been doing some testing/troubleshooting on that Fusin test bike, when I was interrupted by the postman delivering a box of goodies from Methods. He had said he was sending it, but no full list of what was in there. Naturally I had to stop and see what was in it...

Batteries. Lots of 18650s, some individual little prismatics, and some RC LiPo -looking cells.

DSC06760.JPG (54.36 KiB) Viewed 325 times

And a single A123 cell (I've never had one in my hands before, and somehow expected them to be smaller, though bigger than the 18650s).

DSC06763.JPG (29.58 KiB) Viewed 325 times

No time to even really look thru them, so for now they all go in a box in the grill outside, just like my known-dud RC LiPo cells. (just in case, you know? cuz the USPS beat the box up pretty bad)


Lots of connectors, including some alligator clips I actually was wishing I had more of, just a couple days ago...now I do! Lots of JST cables, which I can use to make Y-cables or extensions or whatever. Lots of little red high-current plugs that I forget what they're called, polarized. Some other assorted bits and bobs I didn't have time to really look thru yet.

DSC06764.JPG (70.26 KiB) Viewed 325 times

At least part of an Eagletree setup; I haven't even read the labels on everything yet.

DSC06765.JPG (83.55 KiB) Viewed 325 times

A throttle...

DSC06762.JPG (69.24 KiB) Viewed 325 times

...presumably for this controller, which appears to be a monster, a prototype experiment, I think.

DSC06761.JPG (36.51 KiB) Viewed 325 times

I hooked it up to CrazyBike2's 9C real quick, and it powers on and argues with the motor about which way it's trying to go, but I didn't have time to find the right phase/hall combo yet (I tried a couple of quick phase combo changes, one is forward but grindy and one is backward but grindy, and another is nohting at all but incredibly high current even just touching the throttle).


After I get done with the Fusin testing I need to finish (probably tomorrow night, I hope), I want to try hooking this thing up right and see if I can burn out CrazyBike2's motor (not really....but maybe see if I can wheelie it a little ).

There's also some prototype bits of the LVC/HVC system he's working on:

DSC06766.JPG (73.53 KiB) Viewed 325 times

DSC06767.JPG (72.71 KiB) Viewed 325 times

These will also eventually go on CB2, on the RC LiPo packs I am eventually going to get tested and repaired and built into a bigger /longer range pack or add-on pack to the existing experimental one. For now, they will get wired into the ammocan pack that's on the Fusin bike, as soon as I get a chance to do it.

Then those friends came by with that swamp cooler I pictured above, we went to the store and stuff, and time got away from me again. Then when I got back to the house (later than planned), another friend was already waiting for me, to pick me up so we could go get this:

DSC06773.JPG (44.59 KiB) Viewed 325 times

It's an older powerchair, brushed motors just like what I used on CrazyBike2 the first time around, the little 300W 2-pole type I think.

DSC06781.JPG (53.36 KiB) Viewed 325 times

Uses 2 35Ah SLA,

DSC06772.JPG (67.46 KiB) Viewed 325 times

which were "recently replaced" by the last owner (probably several months or more ago, at a guess); who then replaced the whole chair when that didnt' fix his problems with it. It sat unused, until now when he's having his old place cleaned out, and rather than toss this like so much other stuff has had to be, my friend thought (correctly!) that I could use it. For ease of lifting in transport, we took the seat off, batteries out, etc., so we could actually lift the things into and out of his truck.

DSC06771.JPG (39.62 KiB) Viewed 325 times

It's recharging right now; was at 23.5V resting according to my WU meter (whcih already has the right Anderson SB50 on it to just plug right in). It's now charging using it's built-in charger, monitored with the WU, at about 2.8A (charger's max is 3A).

DSC06776.JPG (35.3 KiB) Viewed 325 times

It does seem to work, though I only tested it as the chassis, with the joystick unit taken off the chair.

DSC06785.JPG (44.57 KiB) Viewed 325 times

I didn't test all teh controls as the cover/labels has come off of them, and I have no idea which button does what (except I found the power button by accident, first try, by lucky guess, using a bit of metal to short the switch pads).


I'm not quite sure yet what I will turn it into....



Another tidbit that was going to be tossed is this old Sony Stereocorder, which is almsot identical to the old Sony Tapecorder I first experimented with recording my compositions on from my junky keyboard of the time. (I still have the tapes, but not the Tapecorder). I only had time to just plug it in and see if it would play, and it doesnt, but the capstan doesn't pop up and hold the tape against the rollers/etc like it should, so it will need disassembly and cleaning or repair (it was very dusty with sticky dust on the outside, could be inside, too). Gonna be a long while bfore I have time to do anything with it.

DSC06786.JPG (52.83 KiB) Viewed 325 times

Now if I could just find some spare time (while still having an income) like I find these tidbits to work on, I could actually *do* something with them!

[auraslip]

What a haul!

[amberwolf]

Aaaaand, I blew up the controller.

Took me something like 3 hours of dicking around to find the right phase/hall combo, because the connector on my motor end of the halls apparently had gone intermittent on a pin somewhere.

First, I thought I could be clever, and use the lights on the ebike motor/controller tester unit (like the one Lyen sells) that Mdd0127 gave me, to figure out what the combo was without actually hooking up the motor/controller together and potentially blowing up FETs or something (I did that once with a bad combo, on an 18FET). Theoretically, it should have worked. Really, it probably did, but due to the intermittent connector problem at the motor end, I woudlnt' have known it--it would just end up as one of the "growly" bad combos.

So after wasting way more time than I should have allowed for that, I figured the theory was flawed and just tried the John in CR method, of trying the phases till I got a good combo (backward in this case), then I swapped two of them, and then searched for the right hall sequence, but I could never get the right one--they were *ALL* bad, which is impossible, because I got a good REVERSE.

So I went back to the reverse that worked, and now THAT didnt' work. I got hte tester out and everything worked on that...so halls were good and phases were good on both motor and controller.

WTF.

I was about ready to give up, cuz I was really tired and I was also really hot, as I had not yet noticed that that evap cooler pictured above (which I was trying out) had run out of water, but had already succeeded in making it more than 50% humidity in the front half of the house, and was now just making the air warmer as the big motor runnign the squirrel cage fan added it's heat to the air, without any cooling going on. I can't even take off my shirt or I'll get eaten alive by the little mosquitos that seem to warp in from nowhere in the bajillions everytime a room gets darker than full sunlight (which in my house is all the time, as it's usually more like twilight in here at best, trying to save energy/money).


But I persevered, and just cut the connectors off both controller and motor, as I had no more JSTs to replace the motor connector with, and just started alligator-clipping them together in the combos. On the fourth one, I got it, finally. WHen I checked, it turned out to be the one that it SHOULD have been based on my ebike-tester experiment, although now I can't remember how I arrived at the experiment OR the results, of course.


Now, all excited, I looped all the wires on there (not tying anything down), soldered and taped the hall wire connections, moved the CA back to this bike off the FUsin test bike, and then took CrazyBike2 outside to test it on the road.


First thing that happened was....nothing. Huh? It had just worked...so I checked all the connectiosn, power cycled it, and it worked for an instant, then nothing. Great. Played with the throttle connector, because as I bumped that with my knee it seemed to work again. Nothing. Red herring. PUlled the controller out of the tool-box side (left black box) where I'd kinda just dropped it in there for the test, and realized it probably had clicked the little Hi-Low voltage pack button, putting it into 100V pack mode. My pack is only 58.3 max, so.... Clicked the switch, retried it, and whammo! Powah!


Ok, so now I FINALLY get it on the road, and start going....160A on the CA for realtime battery current as I took off....wow. Guess I shouldn't hard-throttle it on THIS controller like I usually do.

MUCH gentler on the throttle, I let it build up speed, till I hit 20MPH and coasted to a stop...because I forgot to hook up the ebrake, so no regen braking, and the disc brake still hasn't been fixed, so it does almost nothing, and the rim brakes aren't adjsuted tight enough cuz the rim is not true enough and they just rub, so they only help to slow and stop at low speeds, after regen has done most of the work, on the regualr 12FET. Good thing it was after midnight with no one on the road.


Turned around, and took off slower this time, but it was still pretty high--124A at startup! Didn't tail off much till I got past 12MPH, where it was still at 60A, but it sure did accelerate fast compared to normal. I hit the throttle hard, and saw 196A on there as the bike practically shot forward, feeling the G's for a second, and let it go, again coasting aroudn 17MPH down to my front yard...then I touched the throttle to push it up over the sloping curb and onto my front walkway, to the house, and halfway up that it died. No response.

I figured I must've hit the hi/lo voltage button with the bump getting over the sidewalk, but no, that wasnt' it. Noticed the phase wires at the motor side were actually kinda hot, not melty-hot but definitely not happy. They're not beefy or anything, just standard 9C 2807 phase wires, so I didn't expect them to take this abuse for long (which is why this was just a short 1/8 mile trip at most).

My first worry was that I'd melted the motor, but it wasnt' hot, not even as warm on the windings as it would be if I had a full commute to work. Lifted wheel off ground by tilting bike up on a cargo pod edge, and spun wheel by hand--could definitely feel resistance, though not "cogging". Unplugged phase wires from controller, and resistance went away, so the problem is in the controller.


I haven't opened it up yet; I just don't have time to fix it right now even if it's easy; I already wasted way way too much time tonight as it was. (and 20 minutes typing this all up, too, so I don't forget what happened/etc).

So, Methods, sorry it didnt' last very long....I didnt' even get a wheelie out of it. I'll have to see what I blew up in it later, fix it, and maybe undo some of the shunt mod so it isn't at quite so much risk of frying next time.

[methods]

Just throw it in the gutter and move on. That's my advice. Use it for parts.
That entire run of prototypes were failure prone due (at least in part) to the crooked drilling of the heat sink. We saw a lot of blown fets on those. In hind sight I would have trimmed the fet tabs more aggressively or looped insulator around each one.

Dont post any pics of the inside of that controller or I will be forced to fly to your house and dismantle all your bikes.

I have a run of 24 fet controllers starting to trickle in. Once they get here maybe I can send you one of my fussy 12fets. I have a couple with sensitive throttle response. 100V 70A.



-methods

[methods]

Just throw it in the gutter and move on. That's my advice. Use it for parts.
That entire run of prototypes were failure prone due (at least in part) to the crooked drilling of the heat sink. We saw a lot of blown fets on those. In hind sight I would have trimmed the fet tabs more aggressively or looped insulator around each one.

Dont post any pics of the inside of that controller or I will be forced to fly to your house and dismantle all your bikes.

I have a run of 24 fet controllers starting to trickle in. Once they get here maybe I can send you one of my fussy 12fets. I have a couple with sensitive throttle response. 100V 70A.



-methods

[amberwolf]

Ok, no surgery pics.

OTOH, I'd get to meet you shortly before we killed each other.

Still probably gonna fix it someday when I get back to the powerchair motor project.

Crazybike2 still has it's working 12FET on there, so no problems about that. I just figured it'd be fun to try this one on it, just to see how differently it might perform.

[methods]

That will teach you to try and have fun with free stuff you found in a box.
If you want to troubleshoot something go after that Eagle Tree. I think I accidentally plugged 100V into it and let some smoke out. Never even bothered to look - just threw it in the bin. There is an RPM sensor, air speed sensor, and some other goodies. I instrumented my RC car before it exploded and got some good data. Airspeed rocks.

-methods

Crazybike2 still has it's working 12FET on there, so no problems about that. I just figured it'd be fun to try this one on it, just to see how differently it might perform.

[amberwolf]

Not that I started out to do it, but I fixed that controller. At least, it runs the wheel off-ground.

I actually set out to weld up the evap cooler tray this morning while it was still relatively cool, but I somehow popped the main breaker to the house, and resetting it didnt' fix the problem. I troubleshot it and tried to find my spare breakers when suddenly everything came back on, and I realized that it probably wasn't anything to do with whatever I did here--there must've been a power surge that popped the breaker and then killed power to the block or whatever.

I can't imagine anything at my end that would cause that without also causing a fire or being pretty obvious....

Being daytime, I hadn't noticed that anyone else's power was out, especially since the laptop's wifi picked up a signal good enough to peek at ES a couple times.

So there was around 3-4 hours wasted; I don't know how long it was for sure, but by the time the power was back on, it was past midday and past 100F and I'd given up on any outside projects.


Anyhow, I was too overheated to think well, and I tried to take a nap after failing to remember what else I had planned for today, and decided today will just be a "day off", since it was already an "off day". I helped a friend in the evening, replacing his ipod touch's messed up touchscreen--fortunatley that went much better than I anticipated and I didn't break anything else on it. While doing this I was thinkng about the thngs I probably ought to accomplish tonight, and kinda made a mental list of them.

He donated a few lengths of PVC pipe to my solar water heater project, though as I was reaching into the bed of his truck to pick them up when he dropped me off back at my house, and I had just grabbed one end of the pieces, he just drove off, running over my left foot's toes.

Fortunatley the shoes I was wearing are steel-toed, and were stronger than the small amount of the vehicle weight on them for the short time as he drove over them, and nothing bad happened. but it was terribly surprising, and I can't imagine why he drove off at that moment rather than waiting till I was clear of the truck.


Anyhow, somewhat dazed and confused and unable to remember what I had been pondering doing earlier, I went in to greet and feed the dogs and let them out for potty time and playtime and such, and passed CrazyBike2 many times and it just bugged me that I'd broken that controller so quickly.


So after doing the various household chores I decided must get done, like clothes washing and prepping a few food ingredients for stuff to make later in the week, and then popping in here on ES a few times, I wanted to open up the controller to see what kind of carnage I'd done.

First, though, I took your advice and poked around at the Eagletree, but right now I can't see well enough or hold my hands steady enough to work on the tiny little thing.

So after a little while of that, I did open up the controller:

No visible damage, though I certainly see what you mean about the crooked-drilling of the FET mounting holes. They must've been drunk that day, or doing it blindfolded with both hands tied behind their back. but there were sufficient gaps that there shouldn't be any shorting--at least 1mm, sometimes 2-3, between FEt groups.

The controller doesn't really look that messy to me, though I am not exactly the best judge of that sort of thing.

Ohming out the FETs I found the rightmost (from the top, FETs on top) group had one side of the bridge (leftmost, cant' remember if thats' positive or negative side of the bridge) shorted completely out, gate to source to drain less than 1ohm!

I used that soldering gun you sent me ages back in an earlier "care package" to remove the FETs one at a time, and of course it was the last of the three that was dead shorted--the others tested fine. They might be damaged, but I still reused them, at least to see if the controller would work at all. I managed to peel off the gate trace to the leftmost FET, on the top of the PCB, when I removed it. So I had to scrape some coating off the gate trace to the others, and use the FET lead to solder to the top of the trace instead. I replaced the shorted FET with one of the ones I'd bought from you way back when (I still have most of those new in their rail, fortunatley).

Was about to reassemble and test it, when I thought that I had better maybe kinda put the shunts back to something close to what they started out as. So I again used taht soldering gun to melt the solder, and sucked it off off there with the little thumb-cocked springloaded soldersucker I'd inherited from my dad, and then used needlenose to pull off the (presumably) desoldering braid wrapped around the shunts, while continuing to heat the shunts with that gun.

It was starting to smell funny from the PCB heating up by the time I was done. Thankfully I was running the evap's high-volume fan (no water) to move the air thru the house at this point, with it finally down below 80F outside, at somewhere after 1am I think.

I also enabled regen BK to GND with a solder bridge, though I dont' remember now if I checked that the brake wires are actually hooked up to an external connector or not.... I guess I'll find out later when I go to set up a switch on there.


Anyway, then I carefully but quickly stuffed it all back into the casing, after re-verifying no shorts on anything major, and no escaped solder globs. Reconnected to the bike (except the hall wires, still soldered to the motor halls) stuff and powered on. No smoke yet. Tilted bike to get rear wheel offground and carefully pulled throttle...nothing. Remembered pack voltage high/low switch on end of controller, clicked it, and retried--now it worked!

Then I realized what time it is, and that I'd unintentionally spent at least two hours on this.... Well, at least this time it was productive!


So if I have to, I can now just hop on CrazyBIke2 and ride it...though I need to do some finishing work on the new battery box, which is a larger 50-cal ammocan, already painted white a long time ago--I used to use it as a toolbox, and long ago forgot I even had it till I was looking for something else a week or two ago, I think it was. I realized when I saw it that it should be just big enough for the experimental pack, and it turns out be exactly right to hold that pack and the Methods LVC/HVC system, and *maybe* the HF cutoff switch JEB sent me in a care package a long time back.

Sorry no pics on anything tonight, as I just am too wiped to get back up and go find wher eI left the camera.

[methods]

Good job bro
I suppose you noticed the "insulator" that I jammed in between the PCB and extruded housing. Make sure you jam it back in there else the board might touch the housing.

Lowering that current limit was a wise move - I think you have yourself a good controller now. Soft start might even come in handy.

If that big soldering iron ever gives you trouble I remember having to "push" on it some times to get it to kick in. Not sure why - probably something stupid. You can feel it buzzing in your hand when it is working - but the light on the front always seemed to have a mind of its own.

Fixing things is satisfying.
A true value added activity that makes the world a better place.

-methods

[amberwolf]

I did see the insulator card, and ensured it went back where it is supposed to when I put it back together, just in case.

I would bet that evertythign mounts via "bus bar" screwed together in that soldering gun, and that the mounting screws are loose. I didn't even get the lights on at all at anytime, but I didn't need them anyway then. Have to open it up and check later, if it gives me problems on the soldering part (didn't so far).


I figured if I didn't put the shunt mostly back the way it was, I'd blow up the controller again, because even with the controller in it's highly-modded condition, on THIS bike with that knobby Maxxis tire from Mdd0127, I can't get wheelspin evne on loose dirt/gravel in my front yard, with this thing on the rear wheel (whereas the front version spun even on less than 1KW even on asphalt if it wasn't sticky). So it would just ahve locked rotor current levels at startup every time, because it takes a LOT longer to get going than on a regular bike that's probably 70-80lbs less.


Anyway, I set up the ebrake and verified it works to stop the motor, but ot doesnt' do regen. I dunno if that's a voltage-level thing I can fix with the "R12 mod" or some other mod, or if it would have to be reprogrammed. (or if it even *can* be). Without regen braking, this bike does not have sufficient braking power to be safe to ride at higher speeds, above maybe 12-15MPH, in traffic conditions.

The front disc brake I spent about an hour and a half fiddling with, and basically I guess the rotor is hosed or something, because it feels like glass on the pad contact areas, and no matter how hard the lever is squeeze or how far pads are adjusted (even full contact with disc before braking!) you can easily spin the wheel by hand. Same with both calipers I have handy, probalby the same with all the others wherever I put them.

The front rim brake is ok, but I need to move it to a real lever, not the crappy ebrake lever, so I can actualy get the right leverage so it doesn't hit the bars before I get full braking power (or have the calipers so close to the wheel taht they rub a lot). Wheel is still not perfectly true either, plus I had to offset to the right to fix the disc rotor/ fork leg interference issue, so regardless of adjustments to brake arms the right arm still never really springs away from the right rim side properly. So I can't adjust them down as closely as I would like.


That means that I really need that rear regen braking, to do most of the slowdown work from higher speeds. So if I run into more experimental time (and energy) I will poke around to see about how to figure out which mod, if any, will let regen actually work. (it already has BK soldered to GND pad next to it).


Anyway, the main reason for hte post was to note the test ride data. I used TWM1, not the CA (which is on the Fusin test bike), so I don't have speed/distance data. Ride was about half a mile, kind of just around the block. At a guess I might've gotten close to 20MPh, probably not even that, becuause of the braking issue.

Wh/mile might be more than 70, but I was deliberately riding it hard to see if I coudl blow it up again. (rather do it now than later).

0.669Ah
36.1Wh
84.89Amax
4440.7Wmax

57.6Vstart
56.92Vrest
51.79Vmin

Motor was definitely very warm, as the stator coils were hot enough to not leave my finger on them very long. Nothing else in the motor had noticeably warmed up yet, as the ride was very short. I probably would not want to ride this motor as hard as this normally, without working out some additional cooling, especially for when I am stopped.


Acceleration from a complete stop is way better than before (should be, with what might be up to twice the power!), so that part is a success. Plus that strength of acceleration continues for longer than with the ecrazyman 12FET.

[amberwolf]

Since this controller can take up to 100V for sure (whereas I haven't verified beyond 60V on the 12FET Ecrazyman), I decided to add 4s to the 14s experimental pack to see what difference that makes to acceleration (since I dont' need a higher top speed, even if it would be a lot of fun).

Since I don't want to mess with the existing charger (as I also use it for my 14s2p RC LiPo Ammocan pack currently on the Fusin test bike, which I'm commuting with every workday to review it), I dug out the other charger I got from Dirtdad/Comcycle's super clearance sale a while back. It's an HP8204L3, marked as a LiFePO4 charger, which while it has a different charging curve than this pack needs, will still get it to the final termination voltage correctly. Or should--I'm testing the theory right now.

Using Kingfish's instructions in this thread:
viewtopic.php?f=14&t=16800
I found that it only goes up to 67.0V--not high enough for 18s (74.7V for 4.15V/cell), so the closest I can get is 16s (66.4V).

I'm hoping the other charger can be made to go to that 74.7V, but I'm not experimenting with it until I am sure this one works fine, so I have a backup.

This one is only a "3A" charger, and the other is 4A. FWIW, this charger's initial current was about 2.5A, which increased to over 5A during a period of 15-20 seconds, then it went back down to less than 1A, then back up to 2.5A, and swinging back and forth like that for a while. Eventually it settled down to around 1.8A and slowly dropped over the first 0.8Ah of charge to about 0.9A it's at right now, still dropping as the pack voltage increases.


Ideally I'd like to use the big Sorenson lab PSU for a charger, but it only goes up to just under 56V (rated at 55V 55A). So to use it for high-current charging at even 14s, I'd have to find one of those server supplies and mod it for series operation wtih the Sorenson. If they're 12V output, I'd actually need two, as I need another almost 20V to get up to 18s charging voltage. Eventually I may want to go to 22s, as I have enough cells to do that (but they're currently in use as the lighting pack).


OH WAIT: I totally just remembered about those DC-DC units I found in that networking rack unit's PSU!
viewtopic.php?p=555639#p555639
If the outputs can be isolated from each other, I can take an intact PSU and series 3 of the DC-DCs on there, to get 12V (15A) + 5V (25A) + 5V (25A). Assumign that the input side is isolated from the output; I can't remmeber if I determined that or not yet. I have at least one more of these PSUs (several more I think), so I can experiment with this...soon, I hope. Probably not, though.

Then I would have 22V @ 15A out of that, plus the 55V out of the Sorenson, for up to 77V @ 15A for charging. Of course I can also parallel more DC-DC units if I could use more than that charging current, but I have trouble imagining a need for it just at the moment. I *will* eventually want higher charging *voltage*, and for that I can simply series another of these modules on there....


I think I'm really glad I didn't just desolder all the DC-DCs off those PSUs yet.

[amberwolf]

Pack charged perfectly to correct terminal voltage, and the charger shut off like the other one does.

Will try to dig out one of those netwokring PSUs and see if I can play wiht putting the DC-DCs on it in series.

[amberwolf]

Care package from Ohzee came in. I can even use the packing material, which is insulation styrofoam sheets, as insulation in my windows for this summer.

DSC06835.JPG (49.25 KiB) Viewed 477 times

DSC06836.JPG (38.27 KiB) Viewed 477 times

Very useful will be a couple of Infinity Armadillo-armored kevlar lining tires, almost road-slick tread just like i prefer, round profiles. These will go on CrazyBike2 as soon as I get the chance to do it.

DSC06837.JPG (34.67 KiB) Viewed 477 times

A mirror (Electra?) that will go on Nishik-E when I get the chance to, as that really needs a mirror and hasn't had one yet.

DSC06839.JPG (51.19 KiB) Viewed 477 times

Strangely, a couple days ago when adding the cells to the experimental pack, I pulled 3 of those Epbuddy balance-connector grips off 3 packs he'd sent me before,
viewtopic.php?f=2&t=12500&p=559270#p559270
to help protect the balance taps on that pack...and here in this care package were three more grips.


Also a rear rack, whcih I used on Nishik-E to replace the one I'd moved from it to the Fusin test bike. It's not structurally as good a rack, but I don't need it to be on Nishik-E, as it won't see the kinds of loads the other bike will, as Nishik-E is still not E yet. But it fits better than the other rack as it's more adjustable.




The main thing was really an MXUS motor from AmpedBikes, which had been repaired but with some sort of undetermined stiction/friction issue inside.
viewtopic.php?p=575454#p575454
I played iwth it for around half an hour or so and couldnt find the problem, but I suspect one of two things:

1) side cover lips are not quite centered on bearing hole, causing it to rub the then off-center magnets on the stator. This shoudn't be it cuz he marked the covers.

or

2) one of the wire sets is rubbing on the cover right at the outer edge, like maybe the halls on the lip.

I don't know which it is, if either, but it's definitely going to take some time and experimentation to figure out.

DSC06840.JPG (67.33 KiB) Viewed 477 times

DSC06841.JPG (57.62 KiB) Viewed 477 times

DSC06842.JPG (44.64 KiB) Viewed 477 times

DSC06843.JPG (67.5 KiB) Viewed 477 times

DSC06846.JPG (47.55 KiB) Viewed 477 times

DSC06847.JPG (50.13 KiB) Viewed 477 times

DSC06850.JPG (47.18 KiB) Viewed 477 times

DSC06853.JPG (49.77 KiB) Viewed 477 times

Some other bits and bobs from two other (local) friends:

A couple of those "shake to recharge" flashlights, but both seem to have leaky caps and so they dont' really charge much and don't hold it long, never get very bright.

flashlights and busbarsDSC06787.JPG (44.37 KiB) Viewed 477 times

Some very thick copper busbar from netwroking rackmount stuff.

busbarsDSC06789.JPG (47.19 KiB) Viewed 477 times

And some depressed dogs melting in the heat.

hachi and nana DSC06791.JPG (58.75 KiB) Viewed 477 times

I finally got a chance to remember to recharge those powerchair batteries, and they took 12.363Ah to reach full charge.

DSC06794.JPG (35.01 KiB) Viewed 477 times

A couple of racks made of 3/4" square tubing and 1/4" wire rack, saved from a tossed-out display stand at work, hauled home with the Fusin test bike and the Bell flatbed trailer. Might be using some of the tubing to make a better battery cage for CB2.

racks from CM display DSC06857.JPG (36.84 KiB) Viewed 477 times

[ohzee]

glad you found use for everything.. that foam is from JRH when he shipped my 5404.

I worked that mxux motor pretty good when I had it , but I don't think i weakened the magnets or anything.
Just made it smell funny with the insulating varnish. I had thought of maybe using small washers between the
cover and the magnets for a bit of additional space so the hall wires did not rub. Otherwise I was out of ideas.

Really glad you like the tires.. they were pricey , but once I got into fat tires and the ride they provide I could
not go back.

Wish I had a spare air conditioner to send you. they have some decent 120$ units here - I hate being sweaty
near my pc so I have one in my office myself.

[amberwolf]

Oh, I have window AC units and a main house AC unit...it just costs a lot of electricity to run any of them enough to make any difference, so I only run them when there isnt' any other option, or if I and the dogs will only be in one room for a long time (like sometimes when it's bedtime), and I can close up that room to keep the cool air inside there so it doesn't have to run very much on it's highest-temperature setting.

Usually I wait until it's cooler outside than inside, and run the cheaper main evap cooler (minus water) to blow cooler air thru the house in bursts, longer each time until I just keep it on till dawn after about 2-3am or so, to help cool the house. If it's *really* hot, I'll turn on the water too, but usually that just ends up making the house feel hotter the next day as the humidity comes back out of everything.

[texaspyro]

A couple of those "shake to recharge" flashlights, but both seem to have leaky caps and so they dont' really charge much and don't hold it long, never get very bright.

Ahhh, yes... wankerlights...

[amberwolf]

I thought I would test out the tires from Ohzee here:
viewtopic.php?p=596613#p596613
on the Fusin test bike, as that'd be easier to change than CB2, but I was wrong, and wsted about 5-6 hours diealing with a bunch of things, leaving me with an unrideable mess for now.

In a few hours I gotta go pick up 2 35lb bags of dog food i bought last night but couldn't take home on the Fusin test bike I had with me, at least not with the trailer I didn't ahve with me, and with the fusin bike down, i'll be using CB2 for it. First "real" road test with it; I really wish i had regen braking working on the controller, because I can't safely go 20MPH really without it. The mechanical brakes just aren't good enough, and I don't have time to fix the disc brake issue, which probably requires a new rotor, which means filing out holes on another of these nonstandard 180mm rotors.

Im too tired to do it now, and it's already dawn, so i eithe rtake a nap which i desperately need, and then work on a rotor which could take me hours to fix up, the way I feel now, which would put me out at riding this thing in the midday/early afternoon hot sun. Then if/when something goes wrong roadside, i'll be baking with no shade. Would rather just go slower and have my nap and start out earlier when i'st stll reasonably cool (less than 100F) with shade still around.

i'd be sleeping now but i am still too wound up and frustrated wth myself to doze off yet. am hopeing "venting" here will help me nod off.

I wonder if any of the above will make any sense to me when I wake up.