2WD Semi-Recumbent Recycled-Parts Cargo eBike: "CrazyBike2"

[amberwolf]

Just about the same readings for everything the last couple of commutes, except that the peak amps is dropping on the ride home by several amps, and voltage is dipping more, by about 2 volts, probably because of the cold back room the bike is in while I'm working (it's probably no more than 50-55F back there most of the day, and quickly gets to the cold end of that, or below, by the time we close and I leave, when I'm working closing shift).

It has been very cold (for Phoenix) the last few nights. Every night has been below freezing (at my house, dunno about "officially") for at least the last 6 days, with last night the coldest at 24F. (actually the backyard thermometer said 18F but since I did not reference check it with another one like I did in front, I'm just going with the 24F).

My electronic one (from a PC) I monitor from my bedside only goes down to 32F, I have to turn it off after that because it just alarms constantly. It was designed to monitor a PC case/etc., so if it got cold enough to freeze in there it means any liquid cooling might be in danger of failure, and also condensation is pretty likely to form and possibly cause damage. So it doesn't go any lower than that, I guess they figured once it got that far it didn't matter anymore.

Right now in the front yard (monitored by the above unit) it is 41F, and dropping about 1 degree every 15-30 minutes. The backyard will be colder by around 5F right now because the street heats up in the daytime and that keeps the front yard warmer than the back, for a few hours after sunset. Usually by midnight, and for sure before dawn, they're the same temperature during this season.

I now have to wear a thin sweater inside at work under my uniform shirt, and a pair of sweatpants under my dress pants, or else I get way too cold at the cash register where I am stationed most of most days/nights I'm working. On the way to and from work I have to add another thicker wool sweater over that to keep warm, with the windchill of riding at ~20MPH, and I am glad of that motorcycle helmet from Goodwill, to keep warm enough even in daytime (55-65F in the warmest part of the day).

I'm about to dig out the liners for my snowboarding gloves that I wear to keep my hands warm, because on the ride home at night they're not warm enough without them.

It'll be even colder on Thursday morning when I have to be there at 8am, not sure if I'll just slow down a lot to keep windchill down or if I'll bundle up even more (I already feel like I'm wearing a fat spacesuit or something, and this weather is nothing compared to what lots of ESers have to go thru!).

The real problem with the cold, though, is what it does to my joints--I feel like there's sand in them all, and glue, and even just the bones all ache. I almost can't pedal at all, and have a hard time moving my fingers to manipulate controls. Glad of the thumb throttle as I couldn't squeeze hard enough to keep a good grip on a regular twist throttle; at least with the thumb I can lever against the bars and push with my hand if I have to.

I love the cold for every reason other than that. Makes staying in bed under the warm blankets with dogs piled around me a lot more fun than in the summer, though.

[amberwolf]

Day off, tired, being lazy, so only thing I did today so far was take a trip to the grocery store closest to me to check out their clearance and sales, ended up hauling about 53lbs of groceries home. Had bolted the rightside cargo pod on beforehand, and glad I did as I filled them both just to capacity.

Data from the trip is unfortunatley only as a whole trip, rather than outbound and inbound legs, as I forgot to note them down before heading home from the store. Two people, one right after the other, approached me to talk about the bike and distracted me. Told the second about Endless Sphere as he was interested in motorizing his bike, and has been considering a gas kit but didn't really feel it was environmentally friendly and they're noisy.

As I was finishing the discussion with the second man, one of the factory-built ebikes drove past, although the rider didn't appear to have any interest in stopping to talk at all, and ignored us both as we tried to at least say hi. I don't know what kind it was, but it was a metallic red and I think copper, pretty nice looking, rear hubmotor, tiny cargo panniers to the top sides of the rear wheel, battery pack on the downtube of a step-thru style frame.


Data from the trip:
56.6V start (hot off charger)
53.2V rest
40.3Vmin
1.043Ah
50Wh
24Wh/m
57.07Apeak
2.013miles
21.4mph max
11.8mph avg

Lower Wh/mile than my work commute even though I have the cargo pods both on there plus the heavier load for half the trip. Probably mostly due to the lower average speed, and far less stop-and-starts (only two each way).

[amberwolf]

The only other exciting thing to happen today (ebike-related) is that Bigmoose's package arrived.

DSC03849.JPG (55.92 KiB) Viewed 5173 times

Primarily sent to get me a spare (more reliable, too) PDA for VeloAce, the bike computer/speedo app I've used for a long time, it also had as "packing material" a number of other interesting things.

DSC03850.JPG (79.5 KiB) Viewed 5174 times

The PDA is an old Palm Vx, which unlike most of the previous ones uses an internal Li-Ion rechargeable battery. My current DGA unit is a Palm m100, which was designed around 2 AAA non-rechargeables, and lasts maybe 2-3 days on AAA NiMHs due to their lower nominal voltage. I'd attached a celphone battery to the back of it to solve that problem, which lets it last for up to weeks, but it's unwieldy and the charge indicator in the PDA is useless, as it only indicates a low battery when it is actually really right about to die (like in seconds). So this Vx should be a great help, and as it is a tad smaller it will also be easier to mount into the DGA's "dashboard" if I get a chance to get back to that.

DSC03854.JPG (59.53 KiB) Viewed 5174 times

It has the wall charger and serial sync cable, but also has the serial sync cable and PCB from a dock, which makes it fairly easy to adapt to a mount inside a dashboard, using the serial input for the reed switch speedo sensor.

Also included as part of the packing material are some 1000uF 100V caps; only 85C rated but given that I have tended to leave the ends off the cases should be sufficient for the controllers I've got, if/when I need to replace more caps, or if I ever get around to building one of the controllers others have designed or finish with the ones I started. Never heard of the brand, but they sound affectionate: HUG.

DSC03851.JPG (95.25 KiB) Viewed 5174 times

There are also some others that can be used as external caps, by Mallory, 750uF 65V, screw-terminal type (and thus probably a lot higher current capable than the usual type like those above). Not sure if the 12-8-1971 is a date code, but if so that makes them only 3 years younger than I am. I'll have to test them out on the scope for what ripple they allow filtering under a load, vs the caps above them, for curiosity's sake.

DSC03852.JPG (75.66 KiB) Viewed 5173 times

No separate pic, but the big brother of the little cheap meter I have by Centech is also included. It turns on but I haven't tested it's functions yet. Can do the usual Amps Ohms Volts, plus a little transistor tester socket (like the HF meters have), and capacitance (don't yet know what limit that has, have to look up the manual online--probably 200uF or so, seems a typical limit for these). Also can test duty cycle/frequency. Very similar to (but larger than) my Extech meter I bought at Fry's several years back to test small capacitors with (but which is very annoying to actually use for that; it's very slow and I have to series most caps with other known good values to be able to test them at all, as it has such a low capacitance limit).


A couple of linear power supplies that look as if they should be adjustable to a small degree, should be useful for charging things and just regular bench supplies. Marked at 48VDC 1.5A, not sure what adjustment range they'd have yet.

DSC03853.JPG (73.15 KiB) Viewed 5174 times

Two little breakout boxes may prove useful for anything with an IDC header. Haven't opened them to check but they don't appear to have any electronics. They might be handy to break out the Vpower pack's cell balancing wires for much easier readings of them.

DSC03855.JPG (65.97 KiB) Viewed 5174 times

Two things that really touch my inner geek are these two panel meters.

DSC03856.JPG (89.65 KiB) Viewed 5173 times

Unfortunately the larger one, calibrated for 0-15V AC or DC, by GE, Signal Corps type IS-122, Type AW-41, model 8AW-41, didnt' quite survive shipping intact, despite the double-layer of bubble wrap it was in by itself.

DSC03857.JPG (60.51 KiB) Viewed 5173 times

The bakelite casing received some sharp impact along the line that fractured it, so I will need to glue the casing back together before I can use it.

DSC03858.JPG (54.82 KiB) Viewed 5173 times

DSC03859.JPG (60.52 KiB) Viewed 5174 times

I don't see any faults with the meter movement itself, so it should still work.

DSC03860.JPG (56.84 KiB) Viewed 5174 times

The smaller one, which is intact in it's separate bubble wrap, by Simpson (MR15SIDIDCAV) is a 1mA-full-scale DC100mV unit, type WF, Model 182. This one could easily be used with a modified version of the program by Jeremy Harris for the "simple ebike fuel gauge". It would just need a little LED edgelight installed either inside the meter or on the face pointing down at it. It would be quite appropriate here on CB2.

Also included is an interesting belt buckle, which is nearly identical to a painting my mom used to have on her bedroom wall until she passed away, and one of my sisters has it now.

DSC03861.JPG (102.38 KiB) Viewed 5173 times

DSC03862.JPG (91.31 KiB) Viewed 5174 times

[bigmoose]

An eclectic mix of "stuff" for a very creative guy! I am sorry the panel meter imploded... Perhaps I shouldn't have put so much in the box. Not sure on adhesives. Perhaps epoxy? There are an awful lot of pieces to the case of the meter. I really wanted to get the palm Vx to you, and the rest was just sort of "filler"... thought filling it with utilitarian stuff was better than styrofoam peanuts!

[amberwolf]

Definitely much better packing material. Only way I can see to improve it is to sift resistors, little transistors, and other tiny components down into all the other gaps, and then use PCBs bolted together to make the box itself. Then it would be a fairly solid mass, not allowing for moving around at any velocity sufficient to break something (hopefully). Method only to be used with flat rate shipping, though, because the weight would cause killer shipping costs otherwise.

I'm sure I'll come up with something to glue the bakelite back together, once I figure out where I'll use the meter. If it turns out to be more useful stationary, I can just use superglue as vibration/etc. won't reallly matter at that point.

I spent a bit of time searching for the Centech P35017 DMM manual online, but unfortunately haven't found it.

The major block to this is the hundreds (maybe thousands) of "keyword spam" ad sites that clog the various search engines. Those search engines really need to have a way for users to just check a box next to each link to such sites, then click a "report all checked as spam" button at the top or bottom of the search page. Even an individual button would be better than nothing. Then just tally up such reports so that if it is reported more than a time or two, take it out of the search results. If nothing else, at least have a little button next to each one that does something like "never show me results from this site again", and remember that for your personal searches.

[docnjoj]

I'm rooting for you to get the trike working. The high CG might br reduced with 20" rears but if you look at Lightfoot Cycles trikes one has similar 26" rears and they warn you that it could become "unstable" on some turns at speed. My wifes USX looks like it has similar size and head tube angles but her's has 20" wheels and the rears canted out (negative camber) at the bottom. I think they had to do that due to problems similar to yours. Her's is quite stable and she has learned to recover from "2 wheel" turns. She hasn't gone over yet! It would be easier to increase the rear track instead of changing the rear camber but then you take up more room on the street. Tilting trikes really make sense! Just rambling and enjoying reading of your adventures (not the dog bites, however)!
otherDoc

[amberwolf]

CB2 itself is too high in it's mass and design to work with the trike even with 20" wheels, but I have been ponderously pondering a version that is more like my old abandoned Recycle (not pictured anywhere on these boards I don't think), with the seat a lot lower, down between the trike section and the front triangle/wheel.

Another possibility that keeps the seat higher is a sort of tilting trike, where a roll pivot would be placed at the connection point between the trike section and the rest of the vehicle. Then I could still lean into turns for the majority of the bike's mass, and if using 20" wheels on the trike kit, most of that mass would be kept low, too.

I also want to put the trike kit on a suspension, as diagrammed/pictured in the Delta Tripper thread.
http://www.endless-sphere.com/forums/vi ... 20&start=0
So probably that roll pivot would also be a pitch pivot for the suspension, or at least they would be close to each other if not at the same point.

I'm not totally sure how I will do it yet, and have been browsing images of various mechanisms on the web to see if any inspire me.


I cannot increase the rear track, or it will not fit in the front door, and that's a requirement. Plus I'd have to seriously modify the trike kit to do that, and I'd rather leave it just as it is for the most part, since any serious modifications to it would mean I might as well just build my own trike section the way I really want to, and tha tmeans I might as well do a tadpole instead of a delta.

[docnjoj]

Interesting. I've been eyeing her trike as with a bit of welded extension and some pivot bearings it might be made into a tilt trike. It would need a centering mechanism (either gravity/pendulum or springs/shocks but the engineering appears somewhat worked out for delta trikes. She won't let me touch it since she likes it the way it is! Oh well. Dream, dream dream. Good luck on your quest.
otherDoc

[amberwolf]

I figured on putting the pivot as high up as I could, so most of the bike weight would be below it, so when I lean it'll tilt with me but return to vertical otherwise. I'm guessing I'll have to experiment a lot with that to get it right.

I'm not sure what I'll use for a bearing yet, but my first choice is a headset and steerer tube off a cheap bike, as I have a few to play with in the 1" size. Since longer is probably better for more leverage against the frame, and spreading the load, I may use one off one of the three large frames I have, one of which is seen on the trike in the Delta Tripper thread.

Ideally I also want the rear to be suspended, so that complicates things a little, adding in the swingarm behind the pivot point. I still don't have a clear picture in my head of what I want yet, but I can feel it forming back there somewhere in all the mush.

[docnjoj]

There will automatically be a "solid axle" type suspension at the pivot for lateral bumps. The front to back suspension will definitely be the difficult engineering challenge. See the USX Sun website for the central suspension on my wife's bike. It seems to work better than most folks give credit for absorbing impact along the frame. A suspension front fork may be the simple solution. Trikes really are the answer for we older "balanced challenged" folks. You are too young to know that yet!
otherDoc

[amberwolf]

I wish I was--I've always been "balance challenged", and it is one reason I crashed a lot on upright bikes (but so far the only one on CB2 was at the Undead Race, not counting a mechanical failure or two). I am not as bad as I could be, I'm sure, and it gets worse sometimes.

I'll take a look as the USX stuff once I have enough time to do some pondering with the info (this holiday season is far busier for me than last, as I actually am scheduled about six to eight times the amount of hours this year as last. I only got one day off last week, same for this week, unknown about the next couple).


My work commute today was interrupted about 0.31 miles in, by losing all power suddenly with no warning or prior symptoms. CA went dark too, so I knew it wasn't a controller death (or probably not, anyway). I checked the obvious, fuses, breaker, andersons, etc, with no visible problems. No time to go further than that as I would barely have time to pedal CB2 to work at that point, so I just gritted my teeth and shifted as low as I could get and still balance it, and barely made it to work with literally a minute to spare.

I spent most of my lunch rechecking everything, using the CA as a voltmeter as I had no DMM with me. I isolated it to the forward NiMH pack (Pack A), and began examining the wiring/etc more closely. Eventually I got back to the fuse, and found it to be the problem.

The thing is, several weeks (months? since just before Undead Race) ago I put these packs on there and something I was doing in "bench testing" blew the 20A fuse on PackA. At the time I couldn't find the many other 30A and 40A blade fuses I have out of old UPSs, so I just soldered a thin wire (20gauge-ish) across it, with a couple inches of loop, and tucked it all inside the cap of the fuseholder, where I promptly forgot about it, never remembering to go back and replace the wired fuse with a real one.

Apparently that wire (about 10-ish strands of thinner wire inside typical insulation, salvaged off of something long ago) fractured inside the insulation, either from something that happened to it prior to salvage, or from vibration or whatever while on CB2. Strands probably broke, leaving less strands to carry the 40A+ surge currents, and eventually more strands broke and/or burned thru, then at some point it was too much for the remaining strand(s), and poof--no more connection, and it arced across and burned away both wire and insulation enough to give over 1/4" gap inside the insulation.


To let myself be able to ride home isntead of pedalling, I just cut away that part of the wire, and twisted the ends together, shortening it to less than 1" total of wire across the fuse. I guess my next diving operation into my boxes of salvaged parts needs to be a few of these fuses, so I can keep some spares with me.

I'd meant to do that several times before, but sometimes forgot and sometimes didn't have time....

[amberwolf]

Ugh. Not my week for fuses/fuseholders, I guess. Today, on the way to work no problems, but on the way home in the cold (~50F) and dark, I kept getting this very short sharp noise and jerk as if something was jamming in a wheel, but when I stopped to check nothing would be wrong.

After about the dozenth time or so, I finally happened to be glancing at the CA to check my speed when it occured, and saw the voltage dip wildly just at that instant, telling me that what I was probably hearing and feeling was the motor losing power completely for an instant. That led me to stop under a streetlight next to a business sign in a parking lot (not even a mile from work, about 1/3 of the way home), and check the "fuse" from the previous post, only to find no problems at all.

But checking the still-intact 30A fuse in PackB, I found that it was loose in it's holder. Wierd, because it was VERY tight during the last check in the above post (and I did not at that time actually remove it from the holder for that reason, since I could see it was intact thru it's translucent housing). I can think of nothing that would cause it to suddenly become loose, either, unless someone was messing with it when i was in the grocery store today before work, or when it was in the back room while I was at work today. Or just Murphy having a laugh.

Anyway, I squished the fuseholder contacts a bit and forced the fuse in tight, and had no more problems on the way home, but now I am peeved enough that I am going to SOLDER a fuse across that point, and leave the fuseholder itself present but empty. That way there will BE no connector to have problems, unless a fuse blows and I have to use the fuseholder to plug the replacement into (until I can get home to solder a new one in place). That will stop these particular problems, at least.

Once I have the Vpower/CammyCC pack in there, I'm sure I'll find a whole host of new problems to entertain me.

[amberwolf]

Ok, I am seriously considering one of these:
http://www.allelectronics.com/make-a-st ... SED/1.html
as a taillight for either CrazyBike2 or it's descendant (it would in all likelihood be tested on CB2 first).

In case you don't get the scale, it'd look kinda like this on CB2:

DSC03670.JPG (65.14 KiB) Viewed 4779 times

cuz it's almost a foot across (about 10.5" diameter).

Do you think anyone would have a problem seeing that?

Since it is wired up for 120VAC operation, I'd be looking at how it's wired inside (guessing it's got a DC supply inside that bump on the back, current-limiting the LEDs as a few parallel strings). Probably I can use half the strings for taillight, and half for brake light. There are about 150 LEDs in these things.

I can also wire up the taillight as two halves itself, so that the left and right sides can each be flashed along with the existing amber turn signals.

Alternately, if I had the extra $33 or so for a pair of them, I could just use these for the turn signals:
http://www.allelectronics.com/make-a-st ... SED/1.html
THey are a little bigger, at about 12" across, and have about 100 LEDs in each one, already lined up in the arrow shape.

I wish they had the amber version of these, too, because I'd like those even more than the red ones.

I also wish I had a couple hundred bucks to spend on these, as I could buy a few just for all the LEDs in them.

[texaspyro]

How about http://cgi.ebay.com/12V-10W-White-LED-L ... 3422wt_782 or http://cgi.ebay.com/20W-White-LED-Flood ... 3138wt_922 if you need 20 watt. or http://cgi.ebay.com/50W-LED-FloodLight- ... 2953wt_922 for 50 watt.


You can get them in white for the headlight. The 12V models are specified for 12-24V (usually) Something tells me visibility won't be a problem (at least your visibility, their visibility might be slightly impaired).

[amberwolf]

For a headlight I've been figuring eventually going with something like what's been talked about in that Super Light thread, once I can find some that are used (and cheap enough). Those floods look nice, and would be very bright even if they're only half the spec they give, but are out of my price range, with the cheapest at $50. (I might be able to justify the $23 for the red traffic light, but would prefer to figure a way to buy several at once to get more out of the $7 shipping charge ).

At the moment I have no issues with other people being able to see me from the front; that CFL headlight is pretty bright and fairly wide-angle (though not as wide-angle as the CCFL scanner adapters I'd been using before, which are not nearly as bright). Me seeing far enough down the road to make good decisions at 20MPH in partially-lit conditions is a problem though (especially after my HID lens fell out and shattered), and one I could rectify with a large enough parabolic and Fresnel, but currently haven't found a used parabolic of the right size/shape.

Taillights and rear signals/brake lights/etc are more important than headlights for me, since in the city I can generally see where I"m going with streetlighting and lights from other traffic and such, on most streets where going faster is a requirement. But I can't guarantee anyone else is going to see me and I can't watch behind me all the time; even with the rearview mirror I can only see so much.

So more rear lighting, with more surface area (rather than just brighter but smaller) would be a better solution to grab attention when it's needed.

The CFL in the old scooter tail/brake light housing is really bright, but it is small, and like the headlight requires at least 50V or so to really start well, and needs even more to stay bright, especially in the cold, so I also have that 19V laptop adapter in series with the traction pack to run it, and it's one more thing to break. Also, if my traction pack fails for any reason, as happened essentially twice this week, I don't have any taillight (just brake and turn since those are on a separate 12V lighting pack).

Rigging these LED traffic signals for 12V should be easy enough, just by cutting traces and changing the series/parallel configurations of the strings, depending on how they are already wired inside.

[texaspyro]

You can get those 10W red LEDs for fairly cheap ($10 shipped) They put out around 300 lumens. The 20W units are around 20 bucks and put out 1000 lumens.

Something tells me that rewiring that array is not going to be easy... some of those are sealed and potted.

[amberwolf]

I'm sure it is sealed, and it might be potted, but I can most likely defeat that. The challenge of doing that kinda makes them more attractive.

I don't want to use a single essentially point-source LED for a light, because it is not as effective as a large surface area in getting attention without causing visual problems. When i am in traffic, especially at night in poor and changing lighting conditions, small bright lights on vehicles are certainly visible, but not nearly as much so as larger surface area ones that have about the same equivalent brightness. Plus the larger surface areas are more visible more clearly from a longer distance. This might not be true of everyone, but I'm pretty sure it cant' be unique to me.

Another issue is that with just an LED, I'd have to put it in a housing, reflector, etc. Sure, I could use the existing taillight housing that has the CFL in it, but I'd kinda like to keep that one intact to use as-is on this bike, when I move whatever new taillight/signal system I end up with onto the next bike. Or vice-versa. These traffic signals are already in housings, which if I am careful enough in opening up, will remain intact and can be resealed and used essentially as-is.

It's not quite such a big factor, but the fact that these signals are used and would be repurposed/recycled makes them preferable to using a new LED or other light array, too. If used LEDs like that came available, they'd be more likely to be on my "radar".

One other small factor is cost of shipping vs cost of product--wherever possible I prefer to use places that do not charge so much for shipping, or have a flat rate, like All Electronics.

Yet one more small factor is that when possible I want to put what money I have to spend back into the North American economy, as locally to me as possible. Preferably into companies doing the sort of thing that I'd like more of them to do, and not doing things I would rather have less of them do, though this is really hard to achieve.

[amberwolf]

I've been continuing to have that intermittent connection problem while riding when the bike is vibrating from bumps/etc. It gets REALLY bad sometimes, and other times does not happen at all. I haven't eliminated the fuseholders yet, but was planning to tonite when I got home from work.

However, one of the dogs (probably Loki or Hachi) had different plans.

DSC03874.JPG (85.55 KiB) Viewed 4712 times

Those used to be the little Andersons I had on the two 24V NiMH chargers I'd been using for the packs on CB2. So I'm probably not charging it up tonite, as I don't have more of them to put on there (except ones already in use on other things). I'll have to dig out some different connector hardware to install on the packs for charging until I get more (fortunately some are on the way from Veloman's sale thread, where I spent a bit of money I probably shouldn't have, but too good a deal to pass up on some parts I'd been wishing I had).


I think though that I will just put the mini-XLR connectors on there that were originally on the chargers, if I can find the female ends around here (I can't even remember if I ever had them). Then I can plug the chargers in without disconnecting the series pack connections; I can just leave the main breaker disconnected to keep it off the controller during charge. Since I have not been monitoring the charging of PackA and B for a few weeks now, excepting once or twice for curiosity, as they appeared to be just about balanced in usage/charge now, I don't really have to have the andersons to plug the TWM's into for charge monitoring and pack comparison (though I'd rather have them, it's moot for now).


Maybe if I'm lucky I'll find the connection problem while I'm wiring in the charger connectors. It is possible that the problem is not in the series pack connections or fuses at all, and that it is at the controller or the CA shunt connections to the pack or controller. I don't think it is, since I already closely examined the anderson springs and contacts on both, and I ziptied across the connectors for those points to ensure they could not vibrate loose. Would have done it on teh series pack connections too, except I (used to) have to unplug those each night for charging. (not now, I guess).

It is also possible that a tab weld or solder joint within a pack itself is failing; I'd have to unwrap the packs to verify that. I'd rather not even take them off the bike if I can avoid it, but I think I'd best check that.

I hope it's just the fuseholders; that's the easiest fix.

Problem is that I cannot easily test the fix; I just have to ride it and see if it has the problem again, which can be gone for an entire ride or two and then come back.

[amberwolf]

I came home tonite to find the connector fairy had left a little baggie with 6 andersons and housings and a 30A blade fuse. Since I am off work tomorrow I will probably spend time tonite installing those, so I can recharge the NiMH packs (now used for two trips, about 10 miles, without recharging). Three are black and three red, so I'll have to mark one of the red and one of the black with something so I know those are thermistor connectors and not power.

Today on my way to work the cutout problem became really bad, to the point where I basically couldn't even accelerate from a stop without juddering, so I took a look and found the tape had come off that big capacitor I'd wired to the board in the controller (to replace it's exploded ones), so it was possible that it's leads could be touching something. I didnt' have any tape with me so I just flipped the cap around so the leads and wires pionted towards the open end of the case and reinserted it, so it couldn't touch anything. That seemed to solve the problem but it began again just as I entered the parking lot at work, proving it is not.

I rechecked a bunch of things at lunch and found no issues, and had no problems for almost all of the trip home, except for one very bumpy/rough part of the road, where I got a little cutout a couple of times. It's just so unpredictable.

Gonna have to try wiring around the fuseholders with the soldered-in fuses to see if that helps, and if not I think I'm gonna have to open up the packs to make sure there's no inter-cell connection problems.

[amberwolf]

There's really nothing visibly different about anything after the work below from previous posted pics, so I didn't take any pics. I can if anyone wants to see something in particular.

Wiring around the fuseholders did nothing: I did it without fuses first, to make sure it would even make a difference before wasting time soldering to fuses. After that test I decided to fix the chargers and examine the whole harness.

I replaced the dog-chewed andersons on the NiMH chargers and began charging them up; they're at about 4Ah back in right now, just after finishing the other work below.

I dug thru some project boxes and finally found my better crimper, Molex branded, which though it's not designed for Andersons does a MUCH better job than any other crimper I have, or my questionable job with pliers. So I went back and recrimped every Anderson that is in the power path. I also partly soldered just the contact-end of the crimp, using my lower-wattage fine-tip Velleman station on max, so that even if a crimp is electrically faulty it will still have a connection via the solder, but the solder is not wicked out of the crimp into the wire that must remain flexible.

At the same time, I examined all the contacts for pitting/etc, and none show any sign of it except at the sacrificial tips of the pairs at the battery connection ends and the controller-battery ends, which is normal. I was hoping I'd see pitting or discoloration at the contact surface somewhere, because that would indicate poor connection there, and would probably have been the source of the problem. And something I could easily fix.


As the NiMH is still charging, I have not yet tested the harness/anderson fixes, but I hope that's what the problem was.


While I was at it, I separated out the controller's thin power wire for a keyswitch, and hooked it up to that B&S keyswitch in my stem, which has been disconnected since taking the Curtis and brushed powerchair motor off for the Undead Race. This will allow me to "lock" the bike in a safe mode so that even if someone flips the breaker/etc. on they still won't be able to run the motor unless they start messing with wiring, too.

Eventually I'd like to seal in the wiring harnesses into a metal housing with the controllers and batteries, but that'll be a while, given that I planned to do that over a year ago and never did have the ideas, time, and materials at the same time. Too many other experiments to run to worry about that so much, since not many seem interested in messing with this junky bike.

[amberwolf]

Charging finished up; had a little hitch though in that one of the dogs must've stepped on PackA's charger and switched it off, pulling it away from the pack at the same time and partially unplugging one of the TWM1's leads from the pack. So I don't know the full charge that went into PackA, because the TWM was reset by that.

pack a:
5.839Ah
165Wh
26.08Vm (restart voltage)

Pack b:
7.425Ah
208Wh
24.94Vm (start voltage)

Guess I need to rearrange the way I charge the stuff so A) the dogs can't get to these things to unplug them/etc., and B) keep all my charger stuff well clear of areas they have access to when I'm not here.


A friend brought some small pieces of Kydex by for me to check out, it's about 1/4" thick. One piece about the size of my palm and one about teh size of my foot, with another piece of ABS of similar size. The ABS is much more flexible and distortable than the Kydex, at teh same thickness.

A possible plan for making a protective shell for the two halves of the Vpower pack is to make two U shapes of Kydex, and then use some strapping with a tensioner on it to clamp them around. Should be stiff enough not to flex significantly under vibration, unlike the duct tape/cardstock packing they come wrapped in. Might help keep spotwelds from breaking.

Vpower Kydex Battery Box Idea.PNG (26.3 KiB) Viewed 5358 times

[docnjoj]

I've been using polycarbonate for my battery holder (split ping 36v20ah) for over a year with no problems. I use 1/4" and double the floor to 1/2" and have had no problems. It is bolted together and industrial cement (quite toxic) used to hold the pieces at edge bond. Mine keeps the battery weight low on the trike under my seat. It ain't pretty but it works and holds either the 54 volt fatpacks or the 36 volt ping solidly in place.

100_4541.JPG (109.9 KiB) Viewed 5452 times

otherDoc

[amberwolf]

I'd been planning on perhaps using aluminum salvaged from various equipment I have, or coroplast, etc., but when the Kydex was offered and I saw what it would be like, I figured I'd be much better off using that.

Polycarbonate would probably be about as good, maybe even better since I could see thru it to see if there were problems inside without opening it up, but I don't have access to that stuff.

Most likely I will have some small gap between box walls and cells, and I will see if I can find my stacks of old mouse pads to use them for their closed-cell foam (many are only 1/8" thick or so, and are almost big enough to cover one side of the pack by themselves, thus taking only two or three mousepads per pack-half to pad them within the boxes).

[amberwolf]

It rained today, easing from a mist into a light drizzle just in time for my ride to work. It's pretty much nearly always dry here, so I don't have to worry about keeping CrazyBike2 waterproofed, and thus I don't usually try. DayGlo Avenger is fairly waterproofed because most things are under that headlight fairing/housing module now, but a good soaking rain would probably not be good for it either.

Just in case, I decided before leaving to duct tape CB2 around all the battery connections, fuses, phase and hall connections, and down at the axle opening too. I also put a piece across the back of the large-screen CA, where it has the extra holes for user-installed cabling and screwmount for camera mounts, all of which are the only ways water could get into it. Since mine does not face upwards, but rather rearwards, those holes are in a position to gather water and allow it inside on the electronics.

About the only thing that would still really be vulnerable is the hall throttle itself, but that is partly shielded by the metal control set off the scooter against rain running down into it, and since it is about 90 degrees off the normal axis (since my bars point down not out), it wouldn't be able to pool inside on anything electronic.

I also taped up the fronts of the two controllers (the spare Methods controller and the in-use 12FET), which normally are open for ventilation to keep them cooler. Would've just put the metal covers on but I can't find them, of course.

It appears to have worked fine to tape things up, as I had no problems on the way there or on the way back. Unfortunately I'll have to untape the battery connections to charge up tonight, but I can hope the rain goes away by the time I need to go to work tomorrow afternoon, and stay away or at least stop long enough for me to go home a few hours later. I'll be carrying the roll of tape with me just in case (as I did today in case I had to redo anything).

[amberwolf]

The intermittent problem has not returned since the last work done, so hopefully it fixed it. I rode 14.74 miles today on a trip to a picnic, including some very rough sections of road, and had no cutouts, and had none on the last work trips either. Hopefully it was something I fixed permanently rather than something just hidden for a while.

After the trip I went to Goodwill's half-off sale at a couple of their stores, only finding something useful at one of them:

DSC03884.JPG (43.87 KiB) Viewed 5419 times

One of these will replace that drinking-cup arrangement I have the CFL headlight in right now. Since I have no lens for it anyway, it won't really affect the light distribution, I think, and it will look a LOT better:

DSC03885.JPG (34.99 KiB) Viewed 5419 times

Data from the trip is for a route:
1) from my house to the entrance to the center path of Arizona canal at 29th Ave just north of Dunlap; this is all by motor at usual speeds of 19-20MPH, and 5 complete stops/starts. I think this is about a mile.

2) On the canal until it crosses 12th street near Glendale Ave; this is all at 13-15MPH, while constantly pedalling very lightly (so I don't upset anyone on the path by being a "motorcycle" since it already looks so much like one to some people). There were 6 complete stops/starts that I recall, maybe 7, having to cross various streets. This was about 6 miles. Also, I didn't intend to go that far, was only supposed to go about 2 miles, and turn south at 15th Ave, but I screwed up daydreaming and didn't notice where I was until the canal began going southward and I started trying to figure out why and where I was.

3) At 12th street, I continued south to Rose Ln and then east to Central, South to Bethany Home Rd and then south on 15th Ave to Encanto Park where the picnic was being held. That's about 4 miles back to 15th Ave, and about 5 more miles south to the park. This was all motor, no pedalling, and at anything from 15MPH for the Rose Ln part (lots of kids playing and other slow traffic), to 17-18MPH for the rest of the way (trying to save a bit of power so I could still get home later, but that turned out irrelevant).

(all these are very round mileage numbers and actually don't add up right for the actual mileage).

Total Wh/mile ended up as 16.5, whcih is very low for CB2; typically it's about 23-27Wh/mile for my work commutes, up to 30wh/mi for ones where it's heavily loaded or there's a lot of headwinds.

14.74 miles
20.0 mph max
13.7 mph avg (Seems really low for an average of the above)
1h 4m 27s trip time

242.78Wh
16.5Wh/mile

5.040Ah
48.8A peak

56.6Vstart
50.0Vrest
41.4Vmin

I'm quite surprised that I only used 5Ah, because with all the faster riding in there of about half the trip (say 7 miles), and all but maybe 6 miles of the trip (say 9 miles) being motor-only, I'd expected to be way past half-used on the 13Ah NiMH pack. Although of course I don't expect I'd actually get 13Ah out of the pack anymore, even if I only get 10 that's still only half used.

As said before, once I had wasted power correcting my daydream overshoot of several miles, I rode slower to use a few less watts against air resistance, especially since it seemed to be fairly windy (against me both east and south, of course, the majority of the trip distance).

I didn't end up having to worry about it, though, because a couple of friends at the picnic had some stuff to bring me and had brought their truck to carry it and also CB2 back to my house. Partly since I wasnt' totally certain I'd have enough power to get back home without haivng to really pedal hard enough to hurt my knees, I decided to take them up on it.

This is some of the stuff, including an abandoned project one of their friends gave them, whcih I might just keep as-is and try to finish (though I'd put functioning pedals on it somehow so that it would be legal to ride here; without them it's not).

DSC03877.JPG (68.04 KiB) Viewed 5419 times

The Trek frame is similar to the one I have that's intended for my cargo bike, but not identical. Is probably older, given the seatstay attachment style:

DSC03883.JPG (54.83 KiB) Viewed 5419 times

This is probably the shock fork off of that FS frame I noted previously:

DSC03878.JPG (67.88 KiB) Viewed 5424 times

It's not a great fork, or much suspension, but I guess it's better than none. The dropout ends are actually separate, and appear to be pressed into the bottom of the shock tube, not bonded into it or welded or bolted in place.

DSC03881.JPG (48.46 KiB) Viewed 5424 times

So no hubmotor on these. Plus they're alloy of some sort.

The scooter wheels are tubeless but not airless, maybe 8" diameter I think. The original small-chain sprocket has a regular bike sprocket bolted to it; presumably he'd planned to use that to drive it from whatever motor he'd planned to mount on it (don't know what that might've been). It also has a drum brake that appears to have a rubbing problem inside. I'd expected a band brake, but it seems to be drum.

DSC03879.JPG (62.85 KiB) Viewed 5419 times

They also donated a couple of different kinds of 3/4" video tape to my media collection (I collect various kinds of information storage media, anything from core memory to punch cards to magnetic tape, optical media, solid-state memory devices, etc., although I'm not sure why I started I still keep it up).

DSC03886.JPG (67.72 KiB) Viewed 5424 times

Then there is a box of assorted stuff, a few celphone chargers/etc., and some bike parts:

DSC03887.JPG (76.18 KiB) Viewed 5419 times

There are some Biopace rings (28-38-48) and the spider/crank for them (though not all of the bolts for the set), part of an ISIS type crankset with just one ring (I think the other ring is the one on the scooter wheel but I'm not sure; they're both 4-bolt types), a couple of freewheel cassettes (one missing most of the bearings inside from the sound of it, and the other just about exactly like the 7-speed Chinese cassettes I see advertised on various rear ebike hub kits.), a couple of Shimano trigger shifters that are separate (not part of brakes), perhaps early models?, most of some rear derailers and all of some front ones, assorted screws, etc. And a front hub with no axle or bearings or cones (these parts probably being what was used as the footpegs on the scooter-to-be).

These are the Biopace rings directly from above, next to a granny ring from something else:

DSC03888.JPG (66.27 KiB) Viewed 5419 times

One thing I really really didn't expect to come home to, though, was that now I need a new back bedroom door:

DSC03875.JPG (41.44 KiB) Viewed 5419 times

Nana decided she didn't like being cooped up in there while I was gone for some reason, though she hasn't had a problem with it before.

DSC03876.JPG (67.14 KiB) Viewed 5424 times

It's patched with 3/4" thick plywood right now (she should have a hard time digging thru that if she should try again), but I still have to replace the door at some point. (I don't think I can repair it in a way that will look like it is original, with anythign I have around here at the moment).