I finished putting the headlight and signals/taillight on today.
I used the Honda's ex-wiring harness, stripped down to just the wires needed for the lights and horn.
Kind of a mess at the stage above, which was about 3 hours to untangle, sort, and splice as needed to fit the harness to the bike (which is much longer than the Honda Spree).
When completed, it is fairly neat, though I still have to pull some of the corrugated wire-protection tubes from the Ford LTD engine compartment to cover the rest of the harness.
All I had for now were the plastic partial covers from the Honda's harness, so I used them mostly in places the wires might get damaged, or where they would look better.
I also did some more DayGlo paint, such as the rear rack, the wooden parts of the seat, the chains and sprockets, and the new headlight's casing. For now I left the rest of it the powder blue and purple, to contrast with the DayGlo orange and red-orange, painting only the parts that aren't likely to change much, if at all.
For the battery to run the lights, I needed a secure holder to mount in the open frame. I tackwelded the base of a dead UPS (a different one from the one the battery holder on my upright bike came from) into the Magna's rear triangle, currently unused by anything else.
It would be nice to put the 12V 12Ah batteries there for the quickie road motor tests (they'll all be very short due to the battery condition), but I'm not sure I can fit more than one in that space. The one used here is a PowerSonic 12V 7Ah, just about 2/3 the width of the 12Ah, and about the same height and length. It doesn't have it's full capacity anymore, as I believe it came from one of the dead UPSs, but it is better than nothing, and will do fine for the LED lighting I will be replacing the scooter's automotive-type bulbs with. (The scooter uses 1537 dual-filament bulb for tail and brake, and 1536 single-filament for turn signals, and they suck down several amps of current during operation, without generating even as much light as the *much* lower power LEDs on my upright bike!).
As I will eventually want a keyed switch for the motor power, I went ahead and temporarily installed this Briggs & Stratton keyswitch with foam strips into the top of the steering stem extension.
Turned flat like in the image, power is off. Turned parallel to the bike frame, power is on. Rather like the traffic signals for the valley's new light rail--horizontal = stop, vertical = go. Right now it's just stuffed in there, but I will eventually secure it to a plate that will then be welded to the top of the tube. It also for now directly passes all the power for the lighting, but it will hook to a relay to switch all power once I put the motor and controller in there, because the lighting voltage is different from the motor voltage, and will also come from a different battery (or set of them).
The new headlight is just a Hewlett-Packard Scanjet XPA transparency/slide/film/etc adapter, mounted using a bolt thru it's casing (with a large washer inside to spread the load against the plastic more evenly) onto a standard front reflector stem-mount.
I've had an identical one as a very bright reading lamp for several years, and it's CCFL inverter is powered by about 12V (an old scanner power adapter, which I think runs a volt or two high with so little load on it) at about 800mA. It fills the whole bed area with enough light to easily and comfortably read by, and is too bright to directly look at when laying in bed and first waking up. ;-) Since it could be possible to destroy it by going too high a voltage, I put a 12.1V zener (and resistor) on the input just inside the casing, to prevent the battery charging voltage from burning it out (or any possible feedback from the motor circuits, should I end up having to run the motor and lights on the same battery set sometime).
I long ago did the same thing on the upright bike's CCFL, as well as a diode in series with it, after destroying one inverter board by hooking the charger up backwards to the battery with the charger's clips (this was before I used a 1/4" audio jack for the charger tip, and just used the charger's alligator clips instead). Battery and charger were unharmed due to the charger's protection circuitry, but since I had power on to the lights at the time, it destroyed the CCFL inverter in a smoky/poppy fashion the very instant I clipped the second lead on.
Unfortunately, at some point during the test ride the headlight became quite dim, for reasons that will probably require disassembly to find out.
I also have a front reflector on there, but I didn't have it installed when I was taking pics, and forgot to go back and do it before making this post. The reflector is just mounted using a flat bit of metal going up from the back of that protruding bolt (secured with another nut, washer, and lockwasher), then a screw thru this second metal bracket/tab into the back of the reflector, just as it was when it was mounted to this original bracket.
The taillight is bright enough to be visible even in sunlight with the camera's automatic dimming feature, and is perhaps three times as bright when braking. The signals are even brighter, since they are amber rather than red, and more of the incandescent bulbs' light is in that part of the spectrum.
This was still during the stage of sorting out the wiring harness, so the bike is quite a mess in the pic above.
A few hours later, nearing sunset, I finished the wiring harness, rechecked everything, then just in case I pulled the 12V 7Ah battery off the upright bike light system in case this one died before I could get back, since I was planning as long a test ride as I felt comfortable with based on ride results. I also figured I might get tired and have to stop somewhere for a while, since I have no motor on this one yet, and all testing is being done by human-power-only for now.
I found during a little yard-based ride to shakedown the harness that the right-hand turns still really needed a limiter, so I started to get out more of that angle bracket, until I saw that the way the handlebar-stem mounting-plate of the steering tie rod moves against the stem's headstock tube where it is welded to the seatpost of the Magna frame. I realized that all it needed was a small bump of metal to get in the way of the mounting plate, and keep it from pivoting as far right. A quick job with the welder (which I did not remember a pic of, but will edit in here later if I remember tomorrow) and now it steers only about 40-45° to either side, and doesn't have the problem of the wheel edge catching on pavement or rough terrain during a really hard turn.
While fixing this, I also decided that the amount of handlebar turn vs wheel turn was going to be a problem, because there are moments when I keep turning it much farther much faster than I need to, and it is going to cause a crash if I don't adapt faster or fix it. Since adapting will take me a while (I already did all the *quick* adapting I'll be able to do), reducing the ratio is better. I welded shut the bolt holes in the handlebar-stem mounting plate, then redrilled the hole as close as I could possibly get to the inside edge of it and still clear the plate with the actual steering tie rod eye. That's still a significant difference (about 1.5" from pivot center to tie-rod swivel center at the front end, the fork, and about 1.75" or so from pivot center to tie-rod swivel center at the back end, the handlebars. Maybe a 1:1.17 ratio. Much easier to steer, though--it had been perhaps 1:1.5 or more. Then I removed as much of the plate's outer end (now not needed) as I could, to eliminate it's catching on pant-legs/knees during top of pedal stroke.
I also put Slime into the tire tubes, as I don't yet have any good tires or tubes for the 24" size, and will have to use these for a long while, most likely. I still have the old Slime protector strips from my upright bike over a year ago, one of which failed and let thru a nail (they're guaranteed against that, and Slime replaced them under warranty, but didn't want the old ones back, so I kept them around for emergency use). I'm considering putting them into these tires, since they would be better than nothing, even if they're not as good as the newer versions (like the replacements they sent, which have kept my upright Columbia bike flat-free for over a year now, despite everything it's been thru!)
After packing the foot pump, the toolkit, my aluminum sketch folder (formerly a hospital chartbook from many years ago), and a handful of repair materials like zip ties and duct tape into a backpack, I clipped the backpack into the rear rack using it's lift-up spring retainer, and then for good measure bungeed around the pack's straps so they could not possibly tangle in the wheel spokes. :-)
First I rode around the local blocks again, as I did for it's first road test ever (see previous posts), and this time the handling was much better and more comfortable, as I had the real seat installed now (and repaired from the breaks documented previously). Overall I REALLY like this bike, but it does have a few issues.
I noticed a very slight wiggle in the steering, and checked everything, finding no breaks or bends but instead that the bolts used to hold the steering tie-rod ends to the mounting points at each end do not perfectly fit the holes in the bearing units. It's very small, but that translates into a fairly large pivot of the steering by wobbling. Enough to make it hard to keep a straight line, instead sort of weaving quickly a very small sideways distance at each pedal stroke or tiny bump in the road, then overcorrecting for that and weaving the opposite direction, and so on. It's something I can live with until I make some shims to surround those bolts with, but it tires the arms.
A more serious issue is that apparently something is wrong with the outer (pedal) freewheel on the rear BB stack. Sometimes the palls aren't catching after they've freewheeled a moment, and then pedalling does NOTHING but freely spin. I have to stop, reach down and back, and turn and wiggle the motor's receiver sprocket (for the 1/2" chain) and pedal, then it will cause the palls to latch and start moving again. It does not happen on any pattern I can yet predict. I had seen this happen once when i was playing with the freewheel before assembly, but only the one time. I oiled the freewheel since a dry freewheel could theoreticaly have enough dirt/etc to catch on the palls and stop them from springing back into line and forcing the sprocket back into action--no change. It happened several times during the test trip, both locally and on major streets, forcing me to pull over and get out of traffic.
A bit more experimentation simply shows the freewheel itself is probably worn-out. I have others, but I will have to make a new mounting plate for it as I did for this one, cut off the broken freewheel, then reweld another one on there. For now, since I don't have the motor on there and don't need this sprocket to freewheel, I'm probably just going to tackweld it in place so it will always just transfer power. Once I put the motor on there, I'll need to change it out, though.
I was forced to test the horn when an idiot in a motor vehicle (some sort of minivan-ish thing) attempted to run me down by turning right directly in front of me, on 29th Avenue, when they had no business doing so--if they had waited literally ONE second, I would have been past the street they wanted to turn into, and would not have had to gun their engine to squeal around me on the left--they were behind me at normal speeds just before that. I had to slam on the brakes and stand up out of the bike to stop it in time, and still only just managed to keep it from sliding underneath the back of their vehicle as they turned in front of me. I used the horn as soon as I could see that they were about to turn in front of me illegally, but they ignored it completely and continued their dangerous (potentially fatal) maneuver, then sped on down the side street away from the accident they caused. Fortunately for me and them, no one was hurt, and the bike just got some scratches on it--if there had been more traffic on the road, I could have been run over by any vehicles behind us, if they were too close to stop in time.
It's not like they couldn't see me--it was still very much light outside, and you can clearly see the dayglo and light-colored paintjob, and I was the only vehicle on the road in front of them for quite some distance, including parked cars. They most likely simply didn't care, since I'm "just" a bicycle, and figured I couldn't possibly be going fast enough to be in danger if they did what they did--but I was going 25MPH at that point, as fast as *they* were going before they gunned their engine and raced around me!
It's a danger all cyclists (even motorcyclists!) face, whether they ride in traffic by the rules of the road as I do, or not, because some motor vehicle drivers either do not understand that cycles are part of road traffic and must be treated as such (if the cyclists are doing their part) or they don't comprehend how fast a cyclist can go, and that they can under good conditions keep up with much of regular traffic, or give a fair attempt at doing so.
Rant aside, the rest of the ride went fairly well, except for the several times the pedal-side freewheel failed to engage again, but none of those was at a dangerous time, just inconvenient. Because I could not trust that I would have motive power when I needed it, from a complete stop, I did not attempt any street crossings while on the bike unless there was absolutely no traffic for some distance in any direction, and walked it across instead. Sometimes there was little enough traffic that I was able to do my normal street crossings, lane changes, left turns, etc, but more of the time walking was safer--if I did not get pedal power when I needed to start, I might end up a bit too late crossing and cause a hazardous condition, or else I might piss off whoever was waiting behind me at a turn lane, etc.
I get a fair amount of people, usually once a day or so, that comment on the Columbia upright bike's lights or paintjob or cargo pods, either when I'm stopped at a light or as I pass a pedestrian on a less-trafficked route, or in a parking lot. But on this trip, total of around 6 miles, I must've had several DOZEN people stop and ask about it, or comment on it (some are making fun of it, I'm sure, but at least I know they noticed it!). Lots of them were kids, maybe 8-12 years old, some teens, but most were adults of varying ages. One said that she had a lot of old bikes and parts piled up in the garage, and had been thinking of getting at least one of them working for a while, but now was considering building something like this. I explained what a challenge it's been, and gave her this blog address, so hopefully she'll get enough pointers from this and other places to finish her own! :-) Several kids asked their parents if they could have one, to which there were many kinds of replies.
After stopping by work to pickup the schedule for next week (and giving a preview of the bike to those who were there at the time), I had gone about 3 miles and was getting hungry, as I hadn't eaten since breakfast, and things had run longer than planned getting the wiring harness and such ready before doing this test run--I'd intended to start it about 3pm, but was at least 2 hours later before I could do so. I don't have enough money to eat fast food very often, but decided that if I didn't eat some real food and sit down for a while, I might have trouble making it home (my hands and body start shaking if I lack food for long enough, and I felt close to that point). I ended up going to Taco Bell, as it's the closest inexpensive fast food place to where I was--$4.75 will buy a complete meal, including tax.
The part I don't like about this place is that there is major "construction" in the middle lanes of Peoria Avenue right there, and I can't ride safely for some distance east and west of there--people seem to drive faster and more hazardously in construction zones, and are particularly hateful of cyclists and pedestrians in these areas, no matter how careful the cyclists and pedestrians might be. Even if speeds are greatly reduced by signage in the area, such as down to 15-20MPH, and I pedal that fast, drivers often are lined up behind me gunning their engines and honking, even though they cannot go any faster than that anyway. Just because I am a bicycle in front of them, and not a car, they seem to perceive that they are forced to go slower, even though they are not. (Unless perhaps they intend to illegally speed thru the construction zone, where fines for doing so can be at minimum doubled, and often there are police officers visibly present!).
Anyhow, around 8pm I safely got there, and locked up the bike in their rack, which I like because it is viewable from the dining area, from most of the tables. Many places don't *have* racks, and those that do tend to place them out of sight, perhaps to keep other customers from seeing these strange human-powered vehicles. ;-) There were a few people around, but none seemed interested in the bike, beyond looking at it (or me) strangely, which isn't uncommon even on the upright bike. I went in, ordered, and sat down to eat and think.
I have been pondering how to build essentially this bike, but with a lighter and more elegant frame, and have made a few "napkin sketches" of these ideas. But this time I had an inspiration, based on an idea I'd thought of a few days before, and drew it up.
It's complete enough I decided to scan it in and clean up the pencil a little, running it thru a contrast-enhancement in GIMP. Basically it's all curves wherever I could do that. Really hard to actually *make* because of that, but a lot nicer-looking than the CrazyBike2.0 is. :-)
Another difference is that this one has a rear shock, in a way I never thought to try before this. The swivel point is still the same type I designed for the very first recumbent I was going to make, where the bottom bracket has a "U" bolt around it holding the rear to the front, in such a way that the chainline never changes length, and the pivot point is also the center of rotation of the forward sprockets of the wheel's chainline. (I still don't understand why no one does this yet on bikes built in factories and such--I doubt very much that no one thought of it before *I* did!).
But the shock absorber is very different. It uses a leaf-spring, or rather, one leaf. My first thought is that I could probably use one from the Ford LTD's rear suspension, assuming I can get it apart and out of there, which isn't certain. It is used differently than in a car, as here the center of the arch is pointed upward rather than downward, and is not attached to anything at all. Only the ends are attached, one to just above the rear dropouts, and one to a few inches forward of the bottom bracket (far enough for all the hardware to clear chains, sprockets, derailers, etc., regardless of position of swivel).
The chainstays are the only part of the rear frame that is attached to the wheel, as there are no seatstays in the normal sense. They are also curved, rather than straight. To keep side-sway and warping down, a cross-X is made near the front of them, with the same diameter tubing. These chainstays are very long for the wheel size, so that I can keep the front chainline shorter by making this one a bit longer. It also allows for a larger rear wheel should I desire it--24" or 26" would work here. It would require different brake pivot stud points to be welded on for whichever wheel size was used, but that can even be worked out by putting one set on top of the chainstays, and the other set on the bottom. Something I'm willing to do to make it easier to experiment with.
The rear rack is made as an extension of the seat mounting base, which is a twin tube splitting off from the toptube/downtube, as you can see more clearly in the ortho view than in the side view. This curves up and around back to the top of the seat as a stiffener there, too. There are diagonal braces zigzagging across between the twin tubes to stiffen them and to make a rack to set things on/tie them to.
The seat itself on this one would be tubing with open-weave stretched between them (probably laced on), which is lighter and cooler to ride on than my wooden seat. Probably just as comfortable, too, or even moreso.
Steering presents a challenge, since I don't want a straight tube anywhere if I can avoid it, just for sake of consistency in the design. Cable-steering is the only easy option, but it might be possible to use a number of what amount to long U-joints in series down the curve of the tube from the handlebars to the top of the headstock. I'd have to make each U-joint from scratch, as I have none (except the ones in the Ford LTD, which I think are all too large and heavy for this). Basically making a "driveshaft". It would also be possible to use a steering tie rod like I'm using now, but it would not look very good against the backdrop of the rest of the bike.
More to come, but all I have time for today.
Saturday, February 28, 2009
I finished putting the headlight and signals/taillight on today.