YAXC (Yet Another Xtracycle Conversion)

[thickhead]

Greetings ES denizens!

I'm a relatively new ES member who's been mostly just lurking for months. This thread is intended to be my build diary for an Xtracycle FreeRadical conversion that I started at the end of August. I expect there won't be anything particularly new here for the experts, or exciting for the speed demons, but, in addition to placing a marker for myself, the thread may be of interest to other noobs with similar goals. As everyone always says, I am extremely grateful to the ES community. I learned a ton by just reading this site for weeks before actually buying any components. I hope that in some small way I can return the favor. That, and get help when I (inevitably) run into problems!

How did I get here?
This summer I found out about Xtracycles and e-bikes at about the same time, and they just immediately seemed like two great tastes that would go great together. Funny thing: I stumbled on the e-bike movement while watching old on-demand episodes of Weeds. There is a series of episodes that takes place in Denmark, during which one of the characters becomes infatuated with the Copenhagen Wheel. I was very intrigued. A little web browsing showed me that this was more-or-less highfalutin vaporware of the MIT made-for-TV variety, but it did lead me to many interesting sites with real stuff--blogs, vendors, YouTube videos--and, eventually, ES. I was hooked. I became a blathering e-bike wannabee, overcome by the urge to join the cult. My wallet started hemorrhaging shortly thereafter.

While I'm an enthusiastic fanwhen it comes to e-bike hot-rodding exploits, my current personal interests are not so much in that direction. Though there are many great builds to learn from on ES, my interest in e-bikes as green and groovy transportation make teklektik's gorgeous 2WD masterpiece my favorite to date--it sets a standard of technical artistry to which I can only aspire. I have also been inspired by Justin's vision of a transformative transportation technology. (Thanks, Ebike Nerdcast for the long interview, and thanks, Justin, for great videos, as well as for ES itself.) I'd like to contribute to that vision in my local community. Well, that--but mostly I just want to have fun tinkering with, and cruising around on, my e-bike!

What I wanted
My goal was to build a reliable commuter that I can also use for recreational touring on our local Appalachian mountain roads. In order of importance, my bike is going to be used for: commuting while lugging my computer, books, notes; shopping, running errands, and light hauling; and touring on local mountain roads. I want to build my own because, well, honestly I'm just into this sort of thing... but, for less money than it would cost for a ready-made I will get better performance (that's the theory, anyway )--and I'll also be better able to maintain the bike and tinker with it. I am looking forward to learning about electric motor controllers and energy storage systems.

My commute is about 8.5 miles each way. I'd like to be able to get to work in about 20 minutes, so I figured 30 mph would be a good top speed if I could get it, but I expect to generally cruise in the high teens to low 20's. I generally wanted to err on the side of relatively low cost, simplicity, and reliability, so decided to go with a single rear BMC hub motor--I wanted torque and relatively easy peddling if I run out of juice. To get the right speed range, I decided to go with a 48V system. At roughly 25-30 Wh/mile, that meant I'd want at least 10-13 Ah of storage. Given that it's my first build, I decided to go mostly with a single trusted vendor who could help me with integration. In the end, Ebikes SF won out by a nose over Grin because I thought it would be easier to deal with a vendor in the US. When I called Ebikes SF and got through immediately to Ilia, I found that he was very friendly and generous with his time, knowledgeable, but a no bullshit kind of guy. So that sealed the deal. In the end, I'm happy because I still am getting substantial stuff from Grin, as well.

What I settled on
The intent was to use my 1997 Jamis Eureka hardtail as the base, which has a 7005 aluminum frame. I was concerned about the age of the frame (given the material), but I had used it mostly for just getting around and occasional joy-riding when I lived in town. Now that I've moved out into the woods it's just been a spider house in the garage. Here's what I ordered from Ilia:

• BMC rear V2T on a Velocity Cliffhanger, with matching unpowered front wheel
• DNP Epoch freewheel 11-34T 7 spd
• Lyen MK2 12 FET controller
• Half Twist Throttle with power switch
• Lumenator 1000 lumen headlight w/ variable brightness
• Grin 8 LED rear light 0.6 watt rear light
• Crystalyte Cruise control unit
• Tektro E-Comp hydraulic e-brakes with 203mm rotors, front and rear
• Direct Plug-in Cycle Analyst (didn't want to wait for v2.3!)
• AllCell 48V 13Ah LiNiMnCO battery pack & charger
• Xtracycle FreeRadical Classic
• USB-TTL programming adapter cable for the Lyen

I got the FreeRadical on closeout for $390 (saved some bucks, but the older version without waterproof deck and FreeLoaders...). Total cost for everything was $2,646.60. More than I wanted to spend (now that's a real shocker, ain't it??!), but a pretty competitive deal overall. Based on actual quotes I received, comparable pre-built cargo e-bikes would have cost me about $3500--and that was for 36V systems!

So what I bought was close to what I initially wanted--with the exception that I opted for the torque of the V2T at the expense of a little top end speed (according to the ebikes.ca simulator). For a hauling bike, that seemed like a good idea. I went with the AllCell batteries because in principle the phase change protection should make them more reliable (they come with a two year warranty, which is a rare thing for batteries). More importantly (to me), they weigh less: the NMC chemistry has a bit under twice the energy density of LiFePo4 (the 13 Ah AllCell pack weighs 8.1 lbs, compared to 16.5 lbs for a 15 Ah Ping pack).

But of course, this was all just guesswork (educated by y'all) since I have no actual experience with any of it. We'll have to see how it all goes... I'll be posting build milestones to bring the diary up to date as I can finish them.

Cheers!

 

[thickhead]

Initial Build

The feedstock for the project was my 1997 Jamis Eureka hardtail 7005 alloy frame.

I replaced the old Rock Shox, which did not have disk brake mounts and were kind of rickety, with a Big Dummy front fork. I took advantage of the Big Dummy's long steering tube to raise the headset relative to the seat, and lowered the seat on top of a compact Thudbuster.

The initial assembly went pretty quickly. Here's the first successful power system test, using my military-grade test stand: the back of the FreeRadical was just balanced on two boards across boxes full of empty jugs--was glad it didn't drop off the boards during the WOT tests!

My impression of the Xtracycle was and continues to be more positive than the general tone found here. The "looseness" is not a problem: I just greased the bolts and hand-tightened and everything pulled nicely into tension. But I, too, was disturbed by the small area of contact between the tongue and the front attachment plate (FAP) bolt head. I wanted to put a smooth washer and maybe a lock nut help distribute the load across the tongue. I asked Xtracycle about this and Ross Evans promptly replied:

Works great. Been doing it for years. Aesthetically nicer too. And one less thing to deal with. Functionally, the space in the front tongue is very close to the bolt size so the head of the bolt seats well on the lip. Pull out is not a concern because the material is so thick. The only issue would be if the tongue sides were to spread which they don't do. That said, there are a few washers in the kit and one could be used there.

That calmed me down and I just went with it as instructed. After some very initial creaking, it's really been rock solid. I am still considering adding some small Nord-Locks to the FAP bolt and dropout bolts... On the other hand, introducing extra rigidity is not necessarily a good idea, fatigue-wise. I'm proceeding using a "trust but verify" policy. Also, I have to say I still would not want to be running this platform above 30 mph. At least not based on my limited experience to date. If I wanted to build a 30-40 mph long-tail (and I already do!), I'd definitely go with a purpose-built frame. I believe the Xtracycle is very solid for quasi-static loading. For highly dynamic loading... my confidence isn't there (yet).

One anomaly came out during initial tests: with the motor off and the Lumenator running on power off of the CA, the CA showed negative watts. Unfortunately, I didn't write down the number, but I think it was something like -130 W. The Lumnator is actually about +10 W at max level. I have since modified the connector wiring between the CA and Lyen, as explained in a future post, but not yet resolved the problem completely.

Shortly after this static test I rode the beast for the first time with no rear derailleur, chain or rear brake. I cruised up the steep hill of my woodland neighborhood on a beautiful day WOT with no pedaling and suddenly realized I had a big, dumb, e-bike grin ! I also loved my first night cruise down the dark state game land road, with nothing but the wind and forest insects providing the soundtrack.

Here's the completed beta version of the bike:

For now I'm glad I didn't spend the extra money on the KickBack kickstand, since the standard one is much more stable than I expected it to be. I decided to get rid of the front derailleur. I picked up a left-handed SRAM 7-speed twist shifter from Ilia. The original rear 7-speed derailleur is pretty worn out, but it works well enough since I noticed almost immediately that I mostly ride stuck on the the smallest cog/largest chain ring combo, anyway. But I still need the lower speeds for pedaling without power and maybe for longer, steeper hills when I get in the mountains.

To get things going, I stashed the battery in the left inner pouch, and the controller in the right (it's best to keep the heavier load on the kickstand side). No apparent heat issues yet, but there could be under heavy load, and they do definitely get jostled and bounced around occasionally, which is not good for wiring and connector reliability. So I realize this is not an acceptable long-term solution. But it got me on the road fast! I do try to minimize big bumps and rough terrain because of it, though. I am pleased with the very stealthy look, since I will be riding on bike paths with "no motor vehicles" posted. Since so much can be hidden in or behind the bags, even at this crude stage it's pretty clean (as long as you don't look too closely). Down the road I want to secure the battery and controller on the frame and/or under the rear deck, freeing up the loader space, and tidy up the zip-tied wiring harness further, perhaps with some sort of cable organizer like Techflex.

Here's the initial handlebars:

Left to right, inside the bar ends: grip, 7-speed twist shifter, e-brake, Lumenator, CA, cruise control, e-brake, throttle with on/off switch, grip. Minimal.

 

[thickhead]

Basic performance
After initial test rides, I rode into town (about 20 miles round trip). Because of the negative watts reading from the Lumenator, which I was using as a running light on the more dangerous, shoulderless country roads, I lost track of my Ah, and ended up running out of power 10 miles from home. Much to my surprise and delight I was able to pedal my bus home in about 40 minutes. My bike was about 60 pounds, I'm 195. With fresh batteries, the bike runs very smoothly and hits an honest 25 mph on the flats, as predicted by Ilia and a bit better than the ebikes.ca simulator. I can easily cruise up the steep hill to my house (a 10+% grade) at about 13-15 mph, and I can start on the hill and get up to speed without pedaling (though I generally don't want to tax the motor by doing that). To date neither the controller nor motor has been more than barely warm to the touch, but I have yet to do any serious mountain riding.

Flat tires
Fellow noobs: you know all the stuff you read on ES about needing tough tires? It's true. On my first 3 rides into town I went 3 for 3 with flats! I was trying to run on the original Kenda MTB 26X1.95 road/trail tires (about 5 years old but barely used) and thin standard tubes. The first was a pinch flat that didn't go flat until overnight. The other two were punctures (metal and glass). All three flats were on the rear, the last actually involved multiple punctures, apparently from glass shards moving around between the tube and tire from the initial puncture site. I rarely had flats before: the extra weight and speed of the e-bike definitely makes you more vulnerable. Lesson learned. The good news is I got very good at the "dutch style" of tube repair (a must for e-bikes), and can highly recommend Rema patch kits. The patches worked flawlessly, and I could have ridden on them quite a while.

That said, given that I need reliability in my commute, I immediately made sure I had a CO2 inflator (and a Topeak portable pump with gauge). I picked up an Alien II (hat tip to this thread) and a cheapo small multimeter (I have a Fluke at home). Most importantly, I immediately ordered 26X2.35 Big Apples (the new ones with RaceGuard protection), 2.3-2.5 thorn tubes, and Mr. Tuffy brown liners. I was impressed by the quality of the Schwalbes: they locked onto the rims much more solidly and uniformly with less effort than the old Kendas. I did not get slimed tubes, which I sort of regret. I have presta valves, and it just seemed like too much hassle. I could not find pre-slimed 2.3+ thorn tubes.

Banshee brake from hell
The most annoying components are the Tektro e-brakes. I originally wanted BB7s, since I trusted mechanical brakes more, reliability-wise, but was convinced that hydraulics would require less maintenance because they are "self adjusting". So far this seems completely bogus to me. The 203mm rotors and brakes give me plenty of stopping power, and the e-brake switches works as advertised. But the rear brake squeals constantly and the front brake gets out of alignment frequently. I don't see any "self adjusting" going on. Now, these are my first disk brakes, so I freely admit that this all may be the result of my inexperience... but I don't think it all is.

The recommended installation procedure is to loosely mount the calipers, squeeze the lever, and tighten the adapter bolts. This is supposed to center the calipers. It does not: I always ended up with the calipers pulling the rotor laterally when the brakes were applied, even after trying the procedure with a tiny dab of Phil Wood grease applied between the adapter and caliper mount to let it slide laterally while tightening. So I ended up having to use a flashlight pointing up from the floor to carefully center the rotor in the calipers by eye as I slowly tightened the bolts. This works, at first, but the calipers seem to shift laterally after only a few dozen miles, and I get rubbing. Not only is it annoying to hear, I don't want those parasitic losses!

A big problem is the initial alignment. The front adapter does not clear the 203mm rotor (even though it has "203" printed on it), so I had to use washers to get extra spacing:

I got new bolts with slightly longer threads to compensate for the extra distance of the spacers. I believe the drifting might be caused by the generic washers being used, and so I'm going to try a couple unthreaded precision spacers from McMaster-Carr in their place (one each for each stack of three washers).

At least the front brakes have now worn in and are relatively quiet (except for the rubbing). The rear brakes squeal like a banshee. This is because of a different alignment problem: for some reason, whether it's in the FreeRadical mount, or the adapter, or in the axle/droput fit, I'm having a hard time making the rotor parallel to the brake pad surfaces. I have somewhat ameliorated the problem by placing DIY shims made from aluminum foil (!!) on the outside of the adapter mounting bolts (between the caliper mount and the adapter). Again, I'm going to get some super thin feeler gauges from M-C (you can get them down to 0.0005", I think) and make a more durable shim. I'm hoping I won't then still have to reface the rotor--it's easy to do, but the fricking wheel would have to come off again, and I'm just starting to have fun!

 

[dogman dan]

Well, for a thickhead you've done really well. Good work hooking up with the right company. I keep hearing good things about him. Longtails rock, I got what I wanted (full suspension longtail) with a welder and some good ideas seen here. The bikes rediculously heavy, and has that heavy feel, but something about the long wheelbase feels so right.

I'm not sure about your brakes, but my cheapo mechanical disks have an adjuster for the alignment. Maybe yours are similar, and not automatic. Or maybe you just need a shim.

 

[thickhead]

CA/Lyen metering issue

First, let me say this is not a problem with the CA itself, nor with the Lyen Mk II. It's about the connection between the two. During the initial build, I noticed that with the motor not running but the Lumenator on high I was big getting negative power readings. The Lumenator is taking its power directly from the CA and is supposed to take about 10 W on its max setting. I didn't record it, unfortunately, but the reading I got was in the hundreds at least--enough to really spin down the Ah number, and could even make it go negative as if there were regen, which there definitely is not. This is unfortunate, since I want to use the super-bright light as a running light on dangerous roads and keep track of my Wh consumed (without having to do extra math). BTW, my CA is version 2.25.

I contacted Grin about this. Justin himself replied with:

That means your CA-DP connector wiring inside the Lyen controller isn't quite correct. You should have the black and white wires on the (+, mosfet) side of the shunt, and the blue wire on the (-, battery) side of the shunt, and then ground current from the CA will show up as positive current on the CA.

Sounds simply enough, but the Lyen does not use the same color coding, and the connections are not what Justin seems to be expecting, anyway. Here's a shot of my shunt resistor:

The S+ and S- labels are mine: I am not 100% sure what's going on, since I think there may also be a "shunt mod" under the board, so maybe the shunt isn't where I think it is?? Anyway, I told Justin about the observed difference as I best understood it, and he replied:

...there should actually be 3 wires there, not just 2 [the yellow at S+ and big black negative main power lead at S- (my edit)]. I have no idea what colour scheme Lyen is using, so you should use a multimeter and check which wires go to pins 2, 3, and 4 of the 6-pin CA-DP plug.

Pins 2 and 4 (Gns and S+) connect where the yellow wire is shown, while the wire from pin 3 should go to the ground side of the shunt near the black battery cable.

So I checked as instructed, and determined that controller-side of the Lyen CA connector had pinouts as follows (with the important bits highlighted in blue):

Lyen CA connector pinout

1. Vbatt
2. Gnd
3. Gnd = S- (by jumper with 2)
4. S+
5. Speed
6. Throttle

Note that the CA manual seems to say that S- is close to ground, but not exactly tied to it (it just needs to be within 0.4 V of ground). The Lyen connector is set up to force S- to ground. Not saying that's bad or good or indifferent--just sayin'. What Justin was expecting was this:

Grin CA connector pinout

1. Vbatt
2. Gnd (expected to be tied to 4 inside controller at S+)
3. S- (directly with no jumper to 2)
4. S+
5. Speed
6. Throttle

Examining the two, it's apparent that if one is willing to keep S- tied to Gnd, the polarity can be changed if one rearranges the Lyen connector as follows using the existing wires:

Naive modded Lyen CA connector pinout

1. Vbatt, red
2. S+ (jumpered to 4 w/ old jumper)
3. Gnd = S- (using original controller wire to pin 2)
4. S+ (soldered jumper from 2 onto old pin)
5. Speed
6. Throttle

So, long story longer, I did the above change, and it works... sorta. I now get positive power with the motor off and the Lumenator on high, but the reading is about 240 W (it's supposed to be around 10 according to Grin's website). Oh well. I went and bought a Watts-Up meter (what the hell, I like measurement toys anyway) and it shows the light using 12 W, which is much closer to the 10 W figure. So there's a discrepancy by a factor of 20. Furthermore, when I just run the bike with no lights, the power and energy usage of shown by the CA and Watts-Up are pretty close (the CA is slightly higher, by about 3%). So it doesn't appear to be simply a matter of, say, a wrong Rshunt value entered in the CA. I suspect my error is in finding the right way to separately connect to Gnd and S- inside the Mk II.

I can see I'm going to have to dig further into this when I have time to poke around with the DVM and do some rewiring on the controller board. In the mean time, I'm going with the connector wiring that gives 20X too high Lumenator power. At least it's positive :?.

A sincere thanks to Ilia, Edward and Justin for being so willing to help. All three replied very promptly to email info requests. I'm going to keep everyone posted if and when I figure this out. If any of y'all know the answer, don't be shy about sharing!!

 

[thickhead]

Disaster strikes! Enter the X-Troll-cycle!

OK, so after these various, mostly mild problems, I had a completed beta+ version, mostly like the picture in my second post, but with new Big Apples and tweaked connections. The Big Apples are wonderful. I'm running the rear at 45 and the front at 40 psi. Very comfy, and after couple additional commutes and grocery trips I feel pretty secure about the Big Apple + tire liner + thorn tube set up. Finally, I was ready to log some miles. My plan was to just use the bike this fall without making too many more changes, gain experience, and think about how I wanted to more cleanly mount the battery and controller over winter.

So, one day I road into town to pick up a new bicycle mirror (the first one I had never stayed put, and finally fell out without me realizing). I'd heard good things about Mirracycle mirrors, so I got one of those at the LBS. The guys came out to check out "El Burro" and then one of them says: "Hey, do you know you have a big crack there?" Uh, no... where? Well, here's where:

The freaking head tube. It wasn't even a single simple crack. The damn thing had already forked into a nice Y shape, so I couldn't just put a nice little stress-relieving hole at one tip. Big bummer. I suspect the old 7005 frame was already fatigue cracked, but I also fear I may have caused/exacerbated the problem when swapping out the front fork (let's just say I did it without refined tools, or much experience--hammers were used!)

In any case, that frame was toast. I spent a couple days trying to convince myself to just hose clamp it, or get it TIG welded... but the latter is not very cheap and not something I could do myself. And I really don't want have the thing disintegrate at 25 mph on a narrow two lane road with lots of traffic. That would involve a nasty face plant, or worse. I then thought about getting another used frame, or a cheapo "Wal Mart" bike. I knew I wanted a CroMo frame this time around, not aluminum, but the used steel frames I could find on short order were either as old as my Jamis (like Trek 800s, which I like). And, wierdly enough, the new bikes, even the cheap ones, are favoring 7005 or 6061. Unless you get really cheap high tensile steel, which I did not want. Steel has become the domain of fancy high-end custom builds, with a few exceptions. Which brought me back to Surly... in the end I decided to spend the bucks on a new Surly Troll frame, which I was able to get for $390. The bad news is: it was 390 bucks. The good news is: it's a well made, tough, reinforced CroMo frame with bosses galore for mounting all kinds of stuff.

Oh the irony... I've ended up with a Big Dummy stand-in! I figure I'm still at least a bit below below where I would have been starting with a Dummy frameset, which costs about $950 (I paid $770 for my Troll + FreeRadical combo). If I had known, though, I probably would have just started with the Big Dummy.

On the bright side, my X-Troll-cycle is pretty sweet:

The frame is a bit longer now, with a bit more flex, both of which I like (more directional stability, better passive suspension). Here's the current handlebars:

Not much different, except I've added the battery-powered headlamp to use as running lights, and if I'm stuck at night without power. I also like the fact that if I should decide to build a cargo bike based on a Dummy or Mundo, I will be able to use the Troll frame for a sweet high-speed commuter! Yes, the addiction has set in!

 

[amberwolf]

FWIW, the old butted-tube Treks are pretty strong, and very light. I'm using the front of one for this project:
http://endless-sphere.com/forums/viewtopic.php?f=28&t=31255
whcih is still only half off the drawing board, but I don't foresee any issues with the Trek frame part, despite the condition of it when I got it used.

FWIW, there is a reason parts sometimes fall off--as a warning for us to check the rest of the bike. I doubt the crack had anything to do with the mirror falling off, but I have had times when something fell off becuase what it was mounted to was broken, but I couldn't see that until the other part was gone. So I tend to take any kind of failure as an "opportunity" to examine the entire bike in great detail...as if it were an aircraft overhaul.

 

[thickhead]

Initial impressions
So that brings the build up to the present. I've been riding the X-Troll-cycle back in forth into town for the past week for work and errands. On a fast commute, I can get into work in about 25 minutes, which is pretty decent. I think when I get the new route optimized, I'll be able to average pretty close to my original 20 min goal. The tire/frame/thudbuster combo has been a dream to ride, and the scary two-lane road has been manageable with my mirror and by riding with the general attitude that everybody is trying to kill me!

So far I've ridden 188 miles using 124 Ah in 14 cycles. So that gives a bulk rate of 31 Wh/mi. This is about right because these have mostly been fast rides with average speeds around 20+ mph and WOT much of the time. On top of that, it's overstating the energy, since it doesn't take voltage sag into account, which is substantial. On the CA I'm getting more like 27 Wh/mi for the same conditions. I know for a fact that I can get this down to around 20 Wh/mi if I cruise at, say, 18 mph. I just haven't had time for long rides to test my maximum range. But I'm pretty confident I'll be able to hit a 30 mile range if I keep the speed in the high teens. I hope to be collecting a lot of data on performance over the next few months.

On top of that, my current chainring only has 42 teeth, so it's hard for me to put meaningful power into the bike much above 20 mph. I've ordered a 50T since that's the largest the Troll can handle, and will run only a single chainring with a different old crankset. With the new chainring I'll be able to to put in more power, and so can expect better range in "hybrid mode". When I get more mountain experience, I'll see if I want to add a "grandpaw" chainring (like 40T) to hand-switch if I need it.

Near-term plans
The main things on the agenda for the bike this fall (other than just riding):

• Weatherization
  • Make boots out of old inner tubes to keep water out of FreeRadical joints. JP Weigle Frame Saver?
  • Spray waterproof the FreeLoader bags (didn't get the new waterproof ones...) and make water-proof screens between rear wheel and bags using Kingfish's preferred fairing material (HDPE). If I get the feel for the material, I wouldn't mind trying my hand at a fairing.
  • Weatherize the wiring and controls. Already applied dialectric grease to connectors, but will bundle controller wires and wrap with... silicone tape?
  • Some big fenders.
• Figure out erroneous power reading (20X too high) from the DC power output on the CA (powering the Lumenator).
• Miscellaneous mechanical stuff
  • 50T chainring and a less worn out rear derailleur
  • torque arm and Nord-Locks on the rear axle for added safety.

Over winter I'll be figuring out how I want to mount my controller and battery. The controller wiring is a rat's nest as is, and the standard JST connectors are not confidence-inspiring. I dunno, maybe after they're hot glued they're fine... I'd like to make a proper connector panel with screwable, weatherized connectors. WWTD? (What would teklektik do? )

Before next summer I might be getting another 13 Ah battery for longer trips into the forest, budget permitting. One disadvantage of e-bikes I hadn't thought of before I had one was that they are not very transportable. I can't put my 60-70 lb bike on my old strap-on bike carrier. I need a motorcycle trailer! Barring that, I have to drive to mountain roads I want to go on under my own power, which means at least 30 mile round trips with big elevation changes (sure, not Rockies big, but still Appalachian big).

Comments and suggestions are welcome and appreciated!

 

[thickhead]

Well, for a thickhead you've done really well. Good work hooking up with the right company. I keep hearing good things about him. Longtails rock, I got what I wanted (full suspension longtail) with a welder and some good ideas seen here. The bikes rediculously heavy, and has that heavy feel, but something about the long wheelbase feels so right.

I'm not sure about your brakes, but my cheapo mechanical disks have an adjuster for the alignment. Maybe yours are similar, and not automatic. Or maybe you just need a shim.

Thanks for the supportive words, dogman. I've learned so much from so many of your posts. Yes, Ilia has definitely lived up to his positive reputation. He was very willing to share his knowledge responding to my numerous questions. Very happy to about having gone with him this time around.

I'm a huge fan of the Frankenbike and similar hacked longtails. Now that I know what I'm looking for a bit better, and I'm not in a hurry to go from zero to reliable commuting in a month, I'll be looking around for sturdy old frames to use as fodder for future projects. I am fond of old Trek frames, and I've been looking at some of the Atomic Zombie plans, too, for ideas. I'll have to get my welding skills back... haven't done that for decades.

Regarding the brakes: I've got an inquiry in with Tektro, but I do think shimming will work.

 

[thickhead]

FWIW, the old butted-tube Treks are pretty strong, and very light. I'm using the front of one for this project:
http://endless-sphere.com/forums/viewtopic.php?f=28&t=31255
whcih is still only half off the drawing board, but I don't foresee any issues with the Trek frame part, despite the condition of it when I got it used.

FWIW, there is a reason parts sometimes fall off--as a warning for us to check the rest of the bike. I doubt the crack had anything to do with the mirror falling off, but I have had times when something fell off becuase what it was mounted to was broken, but I couldn't see that until the other part was gone. So I tend to take any kind of failure as an "opportunity" to examine the entire bike in great detail...as if it were an aircraft overhaul.

Hi amberwolf. Like I told dogman, I've learned a ton from reading many of your posts. Thanks. Also--great interview on Ebike Nerdcast! Really enjoyed hearing about your personal creative path in e-biking.

I, too, am a fan of the old steel Treks. I would have used one to replace the cracked Jamis if I had found a decent one that was cheap enough. I found an 800, but the headtube was not big enough for my fork. The other stuff I could find on short notice nearby was not that cheap. In the end, I'm happy having a shiny new frame... this time, anyway.

I'l be following the NuVinci mid-drive with great interest because it's something I wondered about pretty early on in my explorations. It will be interesting to see how it handles the power.

Interesting observation about failure signatures. Insipient failures can have little cascading effects that result in other things breaking first--usually from changes in the vibration signature of the structure. I'm pretty paranoid about the bike breaking. I rode a moped when I was a teenager for a year, and this e-bike, and that's it for my powered two-wheel experience. Maybe that's why I'm so cautious. That and not wanting my 55 year old body to hit the pavement hard!

 

[amberwolf]

Hi amberwolf. Like I told dogman, I've learned a ton from reading many of your posts. Thanks. Also--great interview on Ebike Nerdcast! Really enjoyed hearing about your personal creative path in e-biking.

it's all twisty and turny and jumps across chasms without bridges, so I'm not sure you could really call it a "path", but hey.

I, too, am a fan of the old steel Treks. I would have used one to replace the cracked Jamis if I had found a decent one that was cheap enough. I found an 800, but the headtube was not big enough from my fork. The other stuff I could find on short notice nearby was not that cheap. In the end, I'm happy having a shiny new frame... this time, anyway.

The 800s I've seen have the old smaller headtube, but the 930s all seem to ahve the larger 1-1/8". There is also a larger size than that, but I don't think any of the old Treks have it. The 930 is what I have as the front of the in-progress bike that still doesnt' have a name.

I'l be following the NuVinci mid-drive with great interest because it's something I wondered about pretty early on in my explorations. It will be interesting to see how it handles the power.

A few people have done NV mid-drives, and only one I know of has ended up as a failure--ddk's latest trike. But it is pretty certain the thing came from Staton defective, and not failed because of the power thru it.

My Loooooooongbike will probably start with a hubmotor for initial ride testing and geometry fixes, and then move to a middrive; probably the powerchair-NVjackshaft-wheel I had planned for the in-progress bike, if Loooooooongbike works out like I hope it will.

Interesting observation about failure signatures. Insipient failures can have little cascading effects that result in other things breaking first--usually from changes in the vibration signature of the structure. I'm pretty paranoid about the bike breaking. I rode a moped when I was a teenager for a year, and this e-bike, and that's it for my powered two-wheel experience. Maybe that's why I'm so cautious. That and not wanting my 55 year old body to hit the pavement hard!

Yeah, it's not much fun. I've done that more than a few times. Lately been collecting MC riding gear from thrift stores as I find it, because as I age (44 now) I will get closer to more easily crashing and hurting myself even without outside help, and traffic here can be really unhelpful at times. Nothing wrong with caution.

So far on CrazyBike2 the only significant breakage was the frame under the seat in a sliding crash, but due to the "design" (happy accident like much of my stuff) even that didn't keep it from being perfectly rideable. :lol:

 

[Punx0r]

Regarding your brake, you seem to have run into the same issue I did - having to use washers to space out the calliper from the bracket.

It seems you're supposed to have pairs of special cup-and-cone washer twixt calliper and bracket. These apparently allow the calliper to better align with the disc.

 

[thickhead]

It seems you're supposed to have pairs of special cup-and-cone washer twixt calliper and bracket. These apparently allow the calliper to better align with the disc.

Wow, thanks Punx0r. Do you have more info on these washers? I think I get the idea: does the conical surface allow for angular adjustments? That would be helpful. Are they Tektro parts? More details please?

 

[docnjoj]

My Avids have them and they really make adjusting a lot easier. They are available as separate parts from most good bike shops if Tectro doesn't use them.
otherDoc

 

[thickhead]

My Avids have them and they really make adjusting a lot easier. They are available as separate parts from most good bike shops if Tectro doesn't use them. otherDoc

Thanks docnjoj! Not sure if they'll work with Tektro geometry, but shouldn't be too expensive to just try. Sure sounds easier and more solid than fiddling with making my own shims.

 

[Punx0r]

Sorry, I don't know any more about them; it was something pointed out to me recently in another thread. I'm still using the plain washers as the alignment happens to be ok.

 

[thickhead]

Sorry, I don't know any more about them; it was something pointed out to me recently in another thread. I'm still using the plain washers as the alignment happens to be ok.

Despite my best googling, I'm having trouble finding that thread. Could you give me the link?

 

[amberwolf]

If you have any old rim brake pads/sets around, you have some cup-cone washers right there in those. Might be too thick for what you want, but you may be able to file them down on their wide flat sides.

 

[thickhead]

If you have any old rim brake pads/sets around, you have some cup-cone washers right there in those. Might be too thick for what you want, but you may be able to file them down on their wide flat sides.

Doh! Thanks, amberwolf. I am very late in life becoming a serious (e)bike mechanic. Lots of details to learn from the masters.

 

[pdf]

I think either your or my W-hr/mile is wrong. I have an Xtracycle with a Stokemonkey. There are bound to be differences, but I am seeing more like 16-17 Wh/mile, which seems to be too different from yours. I also have a lot of hills. While the Stokemonkey is more efficient on hills to be sure, I can't believe that explains all the difference. I also have a hub motor on a non-Xtracycle and I get about 18 or so Wh/mile on the same route. The non-Xtracycle is lighter and I go faster on it so I expect the two might more or less even out.

I think one of us (or both) might have the shunt resistor calibration wrong. Your numbers in the high 20's sound way high. I've never gotten higher than 22 Wh/mile for any significant distance no matter how loaded I was and it is pretty hilly where I ride. I usually do around 20 mph, higher where I can (hard to go too fast on greenway).

I could be wrong though. My non-Xtracycle bike has a Lyen controller also. It is an 18 FET. Don't recall the shunt number; I emailed Lyen for it. I could check if that would help.

 

[thickhead]

I think either your or my W-hr/mile is wrong. I have an Xtracycle with a Stokemonkey. There are bound to be differences, but I am seeing more like 16-17 Wh/mile, which seems to be too different from yours. I also have a lot of hills. While the Stokemonkey is more efficient on hills to be sure, I can't believe that explains all the difference...

I know what you mean, pdf. It seems high--not outlandishly high, mind you, but definitely on the high end. On the other hand, yours is pretty darn low. I do think the mid-drive makes a big difference, or can make a big difference if you shift correctly. Also, if you look at power consumption numbers on ES, these no-pedal WOT consumption numbers of mine, in the high twenties, seem to be what many other people get.

I am having some issues with my CA power metering for aux DC power to my Lumenator, but when running the motor only it appears that the Rshunt value I got from Ebikes SF, 1.315 mOhm is at least very close to correct. I have a Watts Up meter also, and the power and energy measurements of the CA and WU are within 3% of each other. Lyen confirms that my shunt modded controller should have an Rshunt of around 1.333 mOhms.

If both of our readings are correct, that is an impressive comparison of hub vs. mid-drive, with the clear advantage toward the latter.

I'm only in the 'break-in" period for my bike. When I can I will be doing some more systematic tests and I'll post revised numbers to the efficiency thread. I'm doing this before I start playing with controller programming, to establish my "stock" baseline. I want to do a bunch tests for two basic conditions, WOT and long-range cruise.

 

[pdf]

If both of our readings are correct, that is an impressive comparison of hub vs. mid-drive, with the clear advantage toward the latter.

I'm only in the 'break-in" period for my bike. When I can I will be doing some more systematic tests and I'll post revised numbers to the efficiency thread. I'm doing this before I start playing with controller programming, to establish my "stock" baseline. I want to do a bunch tests for two basic conditions, WOT and long-range cruise.

I don't think the difference is mid-motor vs hub, but I could be wrong. I went back through my email and the number for the shunt resistor I got from Lyen was "...about 1.4 Mohms.." so that will make my current a little smaller than yours for a given voltage which, if I have it right, will make my numbers a little on the low side. However, not very much. Also, note that the Lyen is on the hub motor bike. The Xtracycle/StokeMonkey uses some other controller. It came with the setup so I assume the shunt value is correct, but it might not be. Let me know if you find out anything different. It would be interesting to use another method for the current so it could be compared. I suppose it doesn't matter really but people do often compare efficiencies of their bikes and I often wonder to what extent people's energy per unit mile numbers are "correct".

By the way, I am frequently at WOT on my hub motored bike also, but it is a high wind count at high volts and will only do about 26 mph WOT. It still gets less than 20 Wh/mile however, if I have it calibrated correctly.

 

[thickhead]

I don't think the difference is mid-motor vs hub, but I could be wrong. I went back through my email and the number for the shunt resistor I got from Lyen was "...about 1.4 Mohms.." so that will make my current a little smaller than yours for a given voltage which, if I have it right, will make my numbers a little on the low side. However, not very much.

I agree. That is not enough to make a big diference.

It would be interesting to use another method for the current so it could be compared. I suppose it doesn't matter really but people do often compare efficiencies of their bikes and I often wonder to what extent people's energy per unit mile numbers are "correct".

That's why I sprung for the Watts Up. It was 35 bucks I think. There might be cheaper ones on ebay. But it gives me an independent reading that matches my CA pretty closely.

By the way, I am frequently at WOT on my hub motored bike also, but it is a high wind count at high volts and will only do about 26 mph WOT. It still gets less than 20 Wh/mile however, if I have it calibrated correctly.

Well... that is odd. Or is it? Not enough experience to speak authoritatively on it. But if you look at the sticky system efficiency thread, my efficiency numbers seem in the ball park (if on the high side).

 

[dogman dan]

Your wh/mi is spot on for quick riding of a hubmotor. The stokemonkey, provided the correct gear is selected has potential to greatly increase efficiency, particularly on hills. And apparently does for him.

Good decision to just make it a pure surly. More money, but I bet it lasts long enough to pay for itself many times. Again, nothing wrong with a first class approach.

When looking for old bikes to hack, I have it made, the swap meet is less than 2 miles from my house. I'll buy a high quality bike with half the parts missing for $20 any weekend. Only caveat, it must have rust, indicating it wasn't fresh stolen. My franken bike longtail has less than $100 in the bike before tires. The motor is pricy now though, I just installed a 5304 in it. Sweet ride now, but I sure kissed 30 wh/mi goodbye. Lucky to get less than 50 wh/mi now with a 48v 40 amp controller and 5304. But so nice to blast along at 30mph on it on another trip to the swap meet.

 

[thickhead]

Your wh/mi is spot on for quick riding of a hubmotor. The stokemonkey, provided the correct gear is selected has potential to greatly increase efficiency, particularly on hills. And apparently does for him.

OK. That's what I thought. Still, the efficiency improvement for the mid-drive was much greater than I thought it would be.

Good decision to just make it a pure surly. More money, but I bet it lasts long enough to pay for itself many times. Again, nothing wrong with a first class approach.

That's what I keep telling myself, dogman!

When looking for old bikes to hack, I have it made, the swap meet is less than 2 miles from my house. I'll buy a high quality bike with half the parts missing for $20 any weekend. Only caveat, it must have rust, indicating it wasn't fresh stolen.

very funny... in a "so funny it hurts" sort of way. I live in a college town, and we have annual flea market type things where the students (and others) sell their stuff for charity ("Trash to Treasure"). Also, Craig's List and the classifieds get good stuff at the end of each semester. There are also really good flea markets sporadically in the region. I've just never been in that scene very much. I will be now!

My franken bike longtail has less than $100 in the bike before tires. The motor is pricy now though, I just installed a 5304 in it. Sweet ride now, but I sure kissed 30 wh/mi goodbye. Lucky to get less than 50 wh/mi now with a 48v 40 amp controller and 5304. But so nice to blast along at 30mph on it on another trip to the swap meet.

Wow. That's pretty zippy for that big bus of yours! The efficiency is fun to play with, but at 15 cents or so per kwh, it really isn't that critical. As is always the case with "alternative energy", it's the capital equipment costs, not operating costs, that are the major issue.