2WD (dual motor, two wheel drive) Trike build.

Bah! The tilting frame is manageable without batteries, however with the weight of batteries it is unmanageable. Although a two wheeler works fine with heavy batteries, even off balance, the three wheeler doesn't. Rather than try to explain this, I'll leave it at this - once the batteries are on there, it flops over to one side and it is hard to right it while riding it. Occasionally it flops over to the other side out of control. The tilting frame was a noble experiment, but it isn't ready for prime time yet. Perhaps I'll make another crack at it at a future time. I know there are three wheel vehicles with tilting frames, but this isn't going to be one of them.

Much was captured on this thread about the physics of tilting vehicles and cornering physics of trikes, and I hope it will be of use to someone who will push this experiment further into reality. Hopefully the community has benefitted, I am certain I have. Even if the experiment doesn't work, the effort is time well spent.

Meanwhile, the project is still in forward gear. Just took a spin around the neighborhood, trying out some corners at speed to see how I liked it. Bike easily accelerates to 25 MPH, zips around pretty well, handles OK. Better tires just arrived, I'll try to get some pedal power added to the mix. I'll try some of the same experiments after I lower it to see how much difference it makes. Subjective, but I suspect my inner ear has a pretty good grasp of cornering physics, despite being useless at math.
 
If the batteries are below the axis of tilt, it shouldn't be a problem....but above that line, I can imagine it certainly would be!
 
amberwolf said:
If the batteries are below the axis of tilt, it shouldn't be a problem....but above that line, I can imagine it certainly would be!
Click to expand...

Yeah, I thot of that, but they'd drag on the ground in my case.

I dunno, maybe they'd strike sparks on the pavement and that would ignite the JATOs, making the bike go faster.
 
I don't know if you've seen http://endless-sphere.com/forums/viewtopic.php?f=3&t=49000 this, but it might be up your alley. I don't know if it would perform any better, but I'd be very interested to find out.
 
MAJOR ABDOMINAL SURGERY: I cut off the entire front end, or cross or whathaveyou, including the hinged mechanism that allows it to lean. Created a new "cross", welded to the main frame and triangulated with a couple of struts. Front end is now rigid. Whole thing is dramatically simpler and lighter, yet more rigid.

Geometry changes: Dropped the center of gravity 10". Set the wheels wider, from 43" to 48" outside-to-outside. I can already hear the comments that this is too wide, but relax, an old friend has been building trikes for many years, and he's using 48" width between the front wheels. Corners like a Ferrari, he says, and usually I am right behind him on a two-wheeler, so I can attest to this fact. According to the physics article referenced earlier, this thing should be able to take a corner twice as fast as the old geometry, given the same radius.

Of course being wide is not an asset in traffic, but there are defensive biking tactics that can be employed. The best Idea I have come up with is to buy a rubber brick from a gag supplier, put it on the end of an orange flagpole, and hang it about 2 feet to the left of my bike. Drivers won't realize it is a fake. "Hey, buddy, you gonna shave it too close, you'll get this brick in your headlight!" But I digress.

On critical welds, since I have a MIG welder and am using chrome moly, I am using an extra step. MIGs will cause chrome moly to stress crack at high stress areas unless the steel is then heat treated. So I am hitting the weld with the mig, tacking it really, then finishing the critical welds with oxyacetylene welding. I make sure the whole joint gets good and hot, then wrap it in a blanket (plumbing shops sell a sheet of high temperature insulation that plumbers use to keep from burning stuff with thier torches, I just wrap it in one of those), then let it set like that till it cools. It would be cool to heat treat the whole frame, but this trick has worked in the past on critical welds, after repeated failures without heat treating.



Pics coming soon.
 
Learned a new thing about Ackerman tie-rod steering.

As trike-happy builders know, there must be a tie rod between the front wheels. It must allow adjustment, to produce a slight toe-in. My trike-freak friend usually uses 1/2" of toe-in.

There is a critical dimension on Ackerman tie-rods. If the tie rod is too long, then the outside wheel doesn't follow the same curve center as the inside wheel, and one or the other wheel scrubs on corners, causing loss of control.

The old formula often misquoted is this: "Draw a triangle from the contact patches of the front wheels to the contact patch of the rear wheel. On this triangle, the two tie-rod ends must fall. If they are outside the triangle , or inside it, then one wheel will scrub."

This is almost correct. It actually works in a very simple front end. However, following this rule, I found my outside tire always scrubbing on sharp corners. Hmmm... What's going on here?

Wikipedia presents the Ackerman steering compensaiton idea like this:

Code: 
Design and choice of geometry

Simple approximation for designing Ackermann geometry
A simple approximation to perfect Ackermann steering geometry may be generated by moving the steering pivot points inward so as to lie on a line drawn between the steering kingpins and the centre of the rear axle.[

I realized the contact patch rule is inaccurate. My bike has a tilted front wheel kingpin for the two front head tubes (I need to draw this) so that the kingpin axis hits the ground right at the wheel contact patch. 12" up in the air, where the tie-rod-ends will be, the kingpin axis is 6" inside the contact patch. This makes the base of the triangle much thinner, and necessitates the tie rod ends being much closer together. To draw a line between the kingpin through the tie rod end and to the rear contact patch is the correct way to lay it out, in 3 dimensions, not in 2 dimensions on the ground.

In this case the rule should be: "Project a triangle in 3 dimensions from the rear contact patch on the ground, through your tie-rod-end, to the axis of the front kingpin up in the air. The tire rod end has to fall on a line from the kingpin axis to the rear contact patch, in three dimensions" This makes the tie-rod ends about 2.5" shorter on both sides than I had them before. Viola! no more scrubbing. I need to scan a sketch of this to illustrate it, nobody is going to get it from that lame description.
 
It makes sense just fine. :) (maybe cuz I've sketched up so many trike design ideas at lunch over the years, maybe cuz you're beter att descriptions athan you tinhk you are.)
 
Really feeling good about this iteration. Re-welding the frame made me have to take it completely apart, which meant I put it back together in a better way than original. Wire rats nest under the controllers is gone, more cables and wires are routed together, one of the two internal hubs is operational, instead of stuck in only one gear. Seat and handlebar position fits great, my arms hang naturally right where the handlebars ride. I should be able to take it on another test ride tomorrow.
 
Successful test run today. Pedal power train is working, dual internal hubs switching nicely through the complete range, able to pedal at a fine clip at top speed. Comfy on bumpy roads that I could not handle on my old hard-tail two wheeler.

It isn't as fast as I'd like, top speed is only about 25 mph, however the two-motor rig powers up steep hills at a pretty good clip and zips right up to speed. I am still running some old knobby tires, haven't switched out to the Schwalbe Marathon Plus tires which should make something of a difference.

Isn't very efficient either, running about 50 WH/Mile. So if I want efficiency, I can use the old bike. I didn't build this to break any efficiency records. Later I plan on working on more aerodymamics mods that will help incrementally. The two-motor design has some inherent advantages (non-overheating on hills is one) and disadvantages (not the most efficient way to go down the road). Trikes have 50% more ground friction than a two-wheeler, and quite a bit more frontal area, mitigated somewhat by the fairing. Trike cargo capacity is really huge, both in weight, volume, or inbalanced loads.

With current battery pack I should have a 20 mile range, which isn't enough unless I plan on a booster charge at my destination. Town is 20 miles away. Maybe I'll have to slow down a little until I expand my battery pack :cry:

Corners fantastically. I can zip a U turn in a standard cul-de-sac at 14 mph, which is as fast as I care to go, and probably faster than the same maneuver in a car.

Got a bit of a pull to the right, may have something to do with disc brake adjustment. Needs a seat cover, and the chain slaps against the chain guard on bumps. (yes, it has an old-school chain guard, cut off a 1950's 3-speed frame. Don't want to get grease on my pantaloons.) Could use another terracycle chain idler

Here are a few more photos

P1060163.resized.JPG



P1060165.resized.JPG


P1060166.resized.JPG
 
Another test run today, faster. I forgot I had a pretty low amp limit dialed into the Cycle Analyst. Set it up to 25 Amps and 35 MPH speed limit, and was able to cruise right up to 35 MPH no problem. That's what I'm talkin' about! I don't want to take it any faster than that yet, we are just at the checkout stage, and I am worried something important will fall off. This rig is really working well!

Up to 59 WH/Mile, that isn't good, considering that I need a lot of range and will have to buy batteries. MORE batteries. Oh well, still cheaper than a used car.


After the run, I started changing out to Schwalbe Big Apple tires from the knobbies that I had laying around. Maybe we will actually be able to measure a difference in WH/Mile for better tires on the same course?

All the rubber is Schwalbe: Super duper high pressure Schwalbe rim strips, ""heavy duty" schwalbe tubes, and Schwalbe puncture resistant tires. I am really impressed with all this stuff. For instance the tubes have a steel filler valve, not a rubber one. The steel filler valve has a little nut on it, so you can tighten it down snug to the rim instead of fishing around inside the tire to get it back out. This is really good stuff.

Schwalbe Marathon Plus tires are notoriously hard to mount on a rim. Schwalbe even admits this. The Schwalbe Big Apple tires, however, had no problem going on the rims.

Still debating whether to use slime in the tubes. Slime has ruined a couple of tubes, by getting into the filler tube and gumming up the works. Once that happens, you aren't ever going to fill the tube again without a lot of work and a new valve core. But slime has saved my butt time and again. To Slime, Or Not to Slime? Maybe I will run without slime for a while and see how it is. It is a lot easier to change a tire on this bike than it was on the last one, so I might just carry patch kits.
 
Wow That is wide. That is the secret to cornering, for sure. Our trikes have to ride into the house through the standard door, so no wider than 30" or so. Also means they are lower, although my wife's delta has negative camber rear wheels and separate axles for each wheell. It really helps the cornering. Keep on keeping on since it looks like you got it for the "wide track" bunch.
otherDoc
 
Wow that is wide! Where did I see that comment? :D

I like it and think you have done a terrific job. Nice to finally see someone put power on both front wheels of a tadpole. If I am not mistaken you have two identical hub motors and controllers running in parallel from the same battery pack and no special electronic differential stuff.

For my use I would be concerned about the width. No doorway issues, but some of the multi-use paths have posts to get by, narrow bridges, etc., and even my standard trike is sometimes borderline, particularly when there is other traffic.
 
Rassy said:
Wow that is wide! Where did I see that comment? :D

I like it and think you have done a terrific job. Nice to finally see someone put power on both front wheels of a tadpole. If I am not mistaken you have two identical hub motors and controllers running in parallel from the same battery pack and no special electronic differential stuff.

For my use I would be concerned about the width. No doorway issues, but some of the multi-use paths have posts to get by, narrow bridges, etc., and even my standard trike is sometimes borderline, particularly when there is other traffic.
Click to expand...

My bike-building buddy built a 4 foot wide trike about ten years ago. We've ridden everywhere together, except through doorways, usually ribbing each other about 3 wheels vs two wheels. Well, he finally wins the argument. We have yet to find anyplace where his wide trike won't go by hook or by crook. The bike landscape here isn't particulary well developed, compared to larger cities, and I might encounter more obstacles in other places. What the hell - I can just drive down the road AT THE LEGAL SPEED LIMIT if I want to.


With a wide trike you gotta think differently in traffic. Sometimes he just takes up a lane and force cars to pass, instead of squeezing into the last inches of pavement at the edge of the gravel.
 
Two more test rides:

1. Installed Schwalbe Big Apple Tires. A very unscientific comparison between two rides that were different roads and different wind conditions shows that the Big Apples are 2 WH/Mile better than the old knobbies. One day I'll ride the same routes I recorded with the knobbies and get a better comparison.

I'm getting about 57 WH/Mile, 35 top speed, 25 average speed, 25 amp limit, and pretty much all systems go. Now it is time for minor tweaks.

The Big Apples are fantastic tires! They really do smooth out the bumps! Just for fun, I ran on the rumble strip beside a road, you know the thing that wakes you up when you fall asleep at the wheel? I couldn't tell I was riding on the rumble strip. The tires, plus a really cushy seat made out of 3" thick upholstery foam. No hard-spring seat made out of mesh foam for me. Although hot, It isn't more than 8" wide, just supporting the spine, so it really doesn't become too hot overall. we've checked out this design for a homemade trike seat for years and it works surprisingly well.

2. Tried a really low speed run. Let my wife drive into town, and set the Cycle Analyst so that speed limit was 12 MPH, amp limit 15 amps. At that speed I could keep up on a spare pedal-powered bike, just barely. Golly, if she didn't get 12 Wh/Mile, just like the old Ebike would get under those conditions. This datapoint makes me believe that this bike isn't inherently less efficient than the old one at the same speed, but I am just going a lot faster and eating up a lot of juice.
 
tweaks: Added a tension side chain idler, because the chain was slapping the chain guard.

Changed the handlebar angles (they are highly adjustable) so that I do more steering with my bicep and tricep, rather than using a teeny little muscle on the side of my arm, to steer. Haven't tried this on the road yet.

Need to redo a few gear cables, and one of my shifters isn't adjusted yet. It is a new kind I am not familiar with, on an 8 speed internal hub, and the instructions are terrible. Need lights, and better cargo hauling gear. Need to assemble a stock toolkit that always goes with the bike.
 
I am having an annoying problem that the bike pulls to the right.

I haven't been able to decide if it is a steering problem or a motor imbalance problem. I may try switching the lead motor, and seeing if the bike then pulls to the left, and also switching the motor controller leads, to see if the problem follows ons of the controllers.

Might also be a mechanical problem, but I don't see how. The steering is basically using a tiller, you point it in the direction it needs to go.

I may adjust the tow-in. Last I checked it was about 1/2" which may be excessive. I saw a site that recommended 0.1" toe in, I'll check that reference and see if I believe it.

I am pretty sure I don't have a dragging brake.

Any ideas about what might cause a trike to pull to one side would be appreciated.
 
Other than what you already listed, misalignment of the wheels could do it, as could different amounts of air in each tire.
 
and one of my shifters isn't adjusted yet. It is a new kind I am not familiar with, on an 8 speed internal hub, and the instructions are terrible.
Click to expand...

Don't know which internal 8 speed hub you have, but I thought my Nexus was giving out before I read and reread the instructions and got a flashlight out so I could see. Turned out to be real simple, there was a little window back at the hub where you had to align something (a colored dot or line, can't remember for sure) by turning the adjuster at the shifter. Before I had just turned the adjuster until the shifting seemed right, which was wrong. :D
 
The toe in should be around 1-2mm so 0.1 inch seems good. Otherwise those Big Apples will disappear real soon from grinding rubber. It should still track fine but do not go to toe-out! Vewwy vewwy bad! I use 2 dowels and zip ties to make a sliding wheel allignment tool. Allign at the midpoint of the tire front and rear. Also center point steering (angle of camber) is quite important. The line through the cambering headpieces should go through the center of the tire as it meets the road. Makes steering much smoother.Caster should be about 5-10 degrees positive but don't remember exactly. That is the front to rear angle of the headpieces. Check, ( as always) with Julian Edgar from his column Autospeed. He built some amazing trikes and documented everything including the caster/camber angles and toe-in too. Also made one with a Di-Dion front axle suspension. Cool, helpful stuff.

otherDoc
 
amberwolf said:
Other than what you already listed, misalignment of the wheels could do it, as could different amounts of air in each tire.
Click to expand...

Thats a simple idea! Just bought a good tire gauge, so I could stop using the "pinch with thumb" method that has served well on two-wheelers with rock-hard tires. Check Simple things first, Stupid ! !
 
docnjoj said:
The toe in should be around 1-2mm so 0.1 inch seems good. Otherwise those Big Apples will disappear real soon from grinding rubber. It should still track fine but do not go to toe-out! Vewwy vewwy bad! I use 2 dowels and zip ties to make a sliding wheel allignment tool. Allign at the midpoint of the tire front and rear. Also center point steering (angle of camber) is quite important. The line through the cambering headpieces should go through the center of the tire as it meets the road. Makes steering much smoother.Caster should be about 5-10 degrees positive but don't remember exactly. That is the front to rear angle of the headpieces. Check, ( as always) with Julian Edgar from his column Autospeed. He built some amazing trikes and documented everything including the caster/camber angles and toe-in too. Also made one with a Di-Dion front axle suspension. Cool, helpful stuff.

otherDoc
Click to expand...


A picture is worth 2kbytes

I have approximately zero camber, probably too much toe-in, and caster such that the axis of the kingpin leads the contact patch by about 1.5 inches. When I next get out in the shop I'll measure these things more accurately, and set the toe-in lighter as well as getting the tire pressure even.

The line through the cambering headpieces should go through the center of the tire as it meets the road.
Click to expand...
I think I have this part correct. Will measure it again to be sure.

camber_caster_toe.jpg
 
You must log in or register to reply here.

Similar threads

marka-ee
    • Doin' Science
Observations from a new to tadpole trike rider.
2
Replies
26
Views
1,001
neptronix
neptronix
P
Dual Motor Build for an Ice Adventure Trike
Replies
21
Views
2,767
matmaxgeds
M
K
Full-fat tadpole tandem trike - build log
2
Replies
25
Views
3,236
kiltedcelt
K
B
Chicken... meet egg. My trike quandary
2
Replies
44
Views
2,486
dogman dan
dogman dan
Jaded
Surly E•D•LHT - Touring eBike Build with BBSHD!
2
Replies
36
Views
9,653
jonsully
jonsully
Back
Top