Spun front axle and flew over the bars @ 25mph

[Affliction]

He has drilled the axle outside the dropouts. Thus the axle to dropout strength is maintained; the external portion of the axle must transfer torque only.

The one potential failure mechanism I see would be in tension, if the bolt is tightened too much. The normal force on that shaft from the nut _is_ through the now-weaker area.

This is not entirely correct, Tension is also needed to prevent rotation, not just the keyway of the dropout. Too little tension will alow movement in the keyway and gradually mushroom the metal of the dropout. As the metal mushrooms on the dropout it actually increases tension on the clampload and will break at the weakest point, this now being at the drill point on the axle. Steel is very maluable and if you have ever done some old-school blacksmithing then you know what I said is true.

 

[methods]

The chopper is repaired and back in action.
Controller works perfectly.
I will write it up later, but I want to put a few cycles on the bike before I put my foot in my mouth.

It would suck to make a nice long post all about how there is no problem then go ass-over again

-methods

 

[methods]

Someone contacted me about the resolution of this issue.
I have positively identified the problem.

* First *

This had absolutely nothing to do with the kit. This was 100% my own doing.

To make a long story short:

One of the contacts in the anderson connectors that tie the phase wires to the controller was loose
This loose connection backed out to the point where it was intermittent.
The controller was a pedal first so it relied on feedback from the phase wires
In the situation where I snapped the throttle from Full to 0V the controller would lose synchronization
When the controller lost synchronization the fets were shorting the phase wires causing the great regen effect.

All adds up perfectly.

Since repairing the wire I had many good runs

I am 100% positive that this was all caused by the loose wire.
I even simulated the effect manually to be sure.

So, we can put it to bed.

-methods

 

[stevero2001]

It was me that contacted methods. I had the same thing happen with my new 2806, which i just got two days ago, except I didn't spin the axle, but I get the same heavy regen action he described. My anderson connections look fine, so the problem is either I have a faulty controller or there is a bad connection back in the controller itself. This is a very annoying problem. I got a torque arm on mine, so it appears unharmed so far. Justin at ebikes is looking into this also, since I'm the second one who has contacted him regarding this behaviour. Also, I have the 20 amp 72v pedal first controller, but I find it peaks at around 17 A at a steady load.

 

[docnjoj]

This problem with brushless motors and lockup goes way beyond front wheels (although ass over teakettle is a bad way to flip). My tadpole had brushed motor rear wheel lockup and flipped me at 20 mph into 9 broken ribs. I won't ride a non-freewheel motor ever again. It appears now that this problem can occur in all non-freewheel motors, irrespective of A/C or D/C. That sucks!
otherDoc

 

[ZapPat]

This problem with brushless motors and lockup goes way beyond front wheels (although ass over teakettle is a bad way to flip). My tadpole had brushed motor rear wheel lockup and flipped me at 20 mph into 9 broken ribs. I won't ride a non-freewheel motor ever again. It appears now that this problem can occur in all non-freewheel motors, irrespective of A/C or D/C. That sucks!
otherDoc

Just remember that just using two or three appropriate fuses on your motor (phase) lines would protect you from this, or using a sensored controller would also avoid this potential problem.

Personnaly I think there are some shortsighted evil design flaws in any controller that will ever do this anyways, sensored or not... They must have been forced to market the thing before doing any safety testing (or just don't give a hoot), because it's pretty easy in code to know what commutation sequence will produce such a lock-up, and thus is easy to avoid.

One good thing I've learned from these stories though is that it is really important to simulate all possible wire break situations while testing a controller, so it is on my "to test" list now for sure!

 

[methods]

I took a series of classes in my post grad studies where we had to design a compiler (i.e. software compiler) from scratch
The easy part was parsing all the text and building the executable code.
The hard part was trying to account for every possible sequence of inputs that you would have to handle

At the end of each week we would submit our progress
The TA's had a program that they would run (that they had been refining for years) that would find every flaw in the compiler
They would not tell you what or where your flaws were, they would only say that: " you passed 39 out of 152 test cases"

ARG!
It was maddening trying to foresee every possible situation

At least with a hub motor there are only so many permutations possible.

-methods

 

[docnjoj]

Thanks Pat! What are "appropriate"fuse sizes for the phase wires? It would be like pants with a belt and suspenders( with a freewheel motor), but I really don"t ever want to have that kind of accident again. No warning and no chance to even apply brakes! I'm too old for that sh*t!
otherDoc

 

[John in CR]

Thanks Pat! What are "appropriate"fuse sizes for the phase wires? It would be like pants with a belt and suspenders( with a freewheel motor), but I really don"t ever want to have that kind of accident again. No warning and no chance to even apply brakes! I'm too old for that sh*t!
otherDoc

I always thought the weight distribution was too far forward on those tadpoles, seems more and more like leaning delta is the way to go with 3 wheels. Thank goodness it wasn't a front motor on an upright bike.

FWIW I think Methods is letting the alignment off the hook, because I believe direct drive hubs on a front wheel (or apparently any wheel of a SWB recumbent) is fatally flawed as a basic alignment. All we need is a million average Joe's installing them, and just a few hundred going over the handlebars, then the gov't is going to step in, not to mention the bad press.

John

 

[docnjoj]

Actually, John, I have about equal weight distribution, but cornering understeer on my tadpole. I also have a skid under my pedal boom to allow the nose to slide in event of forward tip up. The 2 front discs really allow very quick stops and it is near impossible to over the front even in front lock up braking. That being said, for the rear lockup (no rear brake at all) I have no answer, other than to always minimize the chance for it ever to happen again! Cornering on a tadpole is phenominal, more like a go-cart! I'm afraid I do agree on the short wheelbase high recumbents! They seem too easy to go over the front on. The regular recumbents (long and medium wheelbase) have weight distribution heavily biased to the rear, so front lock up usualy ends in a straight away skid with no spin or flip! My wife has experienced this several times when she hits the front brakes on her Bike-e hard, but she has learned to pump them to gain steering control again.
otherDoc

 

[John in CR]

Doc,

I guess I don't understand exactly what happened in your accident, other than hub motor locking up, which I assume was on the rear wheel. Does a rear lockup do something strange on a tadpole? BTW, I fully appreciate your handling advantage, but I think a leaner will win every time.

John

 

[docnjoj]

Oh yes! John in CR! It did what most vehicles do with the rear locked up. At about 20 mph downhill it locked and immediately went into a spin sideways. This action plus the grade down (about 8%) caused it to roll and flip, thus depositing me on my side on asphalt! Luckily, I could ride it one block to my hospital ER (using very slow electric only, couldn't pedal!) My wife was about 20 feet behind me and saw the whole thing! I'm still recovering with crunchy ribs, but at least I can ride. Since I couldn't reproduce the lockup in vitro, I trashed the brushed motor and controller (both old and cheep) and rebuilt my Bafang with Infineon controller! Never again!
otherDoc

 

[MidniteTweeker]

Most tadpoles don't have a rear brake just for this reason. To appreciate the situation once the rear wheel locks up, imagine the tire sitting on a sheet of oiled glass. If any sideways tilt exists (such as road camber) the wheel immediately drifts towards the low side. Countersteer all you want, it doesn't matter. The rear wheel acts just like a pendulum swinging towards the lowest point. If you're headed downhill, then suddenly downhill is the lowest point and it will gladly swap ends. When it's skidding, even a little sway in the road surface that gets your weight shifting a bit is enough to launch the rear wheel sideways.

The launch itself isn't what gets you. It's what happens when the tire stops sliding. It doesn't necessarily need to be enough traction to cause the trike to flip. If the trike isn't completely sideways, when the wheel grips again the trike will happily take off in whatever direction it's now pointed. Problem is, it goes on the new heading quicker than your body mass can respond. That is, you keep going the same direction you were a second ago, except now that direction is sideways and off the seat. Naturally enough, you cling to a handlebar to save yourself and end up pulling the trike over with you. Because the center of gravity of a trike is usually pretty well centered front to rear, when they begin to slide out they rotate about the CG amazingly fast.

MT

 

[John in CR]

I got it now. Leaning Delta is definitely the way I'll go with three wheels (that is unless I do an inline 3 wheeler). I've controlled rear wheel lockups with 2 in the back many times, and I'd prefer a longer wheelbase anyway. Sorry Doc, but no tadpoles for me. No laughing or coughing during the rest of what I hope is a speedy recovery.

John

 

[docnjoj]

Thanks John! I now can laugh and cough with the best! Ribs in old guys dont heal like in you youngsters. They cartilage over at the breaks and usually stay flexible, which helps breathing! They are more prone to re-breakage, however.

Thank you MidnightTweeker for that scary accurate description of my accident! That's why I will now only ride with freewheel motors. It's not a perfect solution, but good enuff for me!
otherDoc

 

[MidniteTweeker]

Doc is right, though. A recumbent tadpole plays with corners like a go-cart. For pure excitement it's hard to beat flying through a set of S-curves with your hip pockets 8 inches off the ground. Of course, that's also what gets you in trouble! When you're playing Indy racer, it really helps to be on a familiar road!

I have a freewheel on my tadpole and no brake or regen, so slide-outs from braking aren't a problem. Just the opposite. I have 1/4 mile gravel lane with an S-curve and uphill to the road. In the curve the surface is cambered to one side for drainage. If I get too ambitious going uphill in the curve, either pedaling or motor-only, and hit some loose gravel, the rear wheel can break loose and begin to spin. At that point it's the same as skidding with a rear brake lock-up, and it immediately begins to slide towards the low side of the lane, leaving me sitting there nearly sideways. But no damage done, other than losing a bunch of valuable momentum.

MT

 

[swbluto]

These horror stories make me think direct attachments aren't the best.

 

[methods]

Nah

Just wear full motorcycle gear like I do every day no matter what :wink:

-methods

 

[docnjoj]

I'll tell you, Methods, that wearing leathers on a 98 degree F 95% humidity day could really help my weight loss program Of course, it would only work until I croaked from heat stroke!
Hey Tweeker! I set my trike up with slight understeer so if i slide out (which is quite often!) the bike maintains fairly neutral handling. Unfortunately, I didn't account for rear wheel lockup!!!! No one expects the Spanish Inquisition! Usually as long as all wheels can rotate, the trike straightens out. Oh well, bah humbug!
otherDoc

 

[docnjoj]

Well, here we go again! My trike had rear wheel lock up again. But this time I saw the result and lived to tell about it. It turns out the chain caught in the deraileur and locked up the Bafang by pulling the deraileur down TOO THE GROUND like a rear skid! The wheel didn't touch the ground for about 10-15 feet. Luckily I was on the brakes early when the chain slipped off the front chain ring and only skidded about 8 feet in a straight line! This also could have been the cause of the first lock-up that broke my ribs and flipped the bike! I will spend next week re-engineering the drive line with a Singualtor type tensioner and a new chain! Wow! Oh the Huge Manatee!
otherDoc

 

[John in CR]

Doc,

No wonder we feel different about helmets (btw I do wear one now...pesky laws), you're accident prone. It sounds like you need one of the 30lb wheels I have back there, which should also give you enough traction to steer your way out of these issues like a car. Also, if a derailleur catching can lift the tire, how can that tadpole weight distribution be correct? Looking at them always did make me feel the weight is too forward, and apparently that was correct. All it will take is moving the front wheels far enough forward to get your body all the way in the vehicle...Limbs don't belong out there anyway, it's a 3 wheeled go-kart, not a bike.

We (Jester and I) just need to come up with a leaning delta that out handles and out circles your tadpole, to get you riding a safe EV.

John

 

[docnjoj]

Well, chain suck sucks! When the chain wraps around anything its not supposed to, it will pull it irrespective of the weight distribution! I really think this was the cause of my first accident also, but there it locked up the rear wheel. It can happen on a 2 wheeler also, and if you are going 20-25 mph you will lose control! More weight in the rear just makes things happen with less control. Think old VWs and Porsches! When they broke loose, you spun! Engine in the rear! Actually my favorate trike, Auzzie, is a delta called the T.Rex with a cool air suspension. Originally developed by APEX tech in Canada. Don (Battboy) tried to get it manufacturered, but it was just too complex and costly!
otherDoc

 

[MidniteTweeker]

@John:
The thing to remember about a tadpole trike (or most recumbents for that matter) is where your weight is heightwise relative to the height of the front axle(s), especially compared to an upright. My rear end is actually sits closer to the road than the front axles, and my body center of gravity is correspondingly low as well. I have disk brakes on the front wheels (i.e., plenty of stopping power) and have never lifted the back wheel even when I've tried to. Part of this may be due to my trike having a bit longer wheel base than some.

As far as having legs out in front, I'll take that setup any day compared to an upright. Brake too hard or hit anything with an upright and your head will be the first thing that hits as it goes over the bars. To me, that's scary.

MT

 

[John in CR]

I understand your point about the legs out front vs an upright bike, but I don't like either. My answer would be a longer wheelbase. To me a great ebike doesn't have to start with a great pedal bike, so I don't feel restricted to what is already for sale. I mix with traffic regularly, so being as low as you is out of the question for me on a daily rider. I do know some smooth surfaced mountain roads with almost no traffic where your tadpoles would be an absolute blast to ride. I'm going to come up with a leaning delta before it's all done though...hopefully with some kind of removable canopy to make it an all weather vehicle.

John

 

[docnjoj]

Absolutely agree with John and ET! My trike also has discs on the front but will tip forward if I lock up the brakes downhill. I have a skid but am thinking of using a small wheel so as to have the boom nose just roll in the rare instantce it tips up. It has only happened once in 7 years of riding. John, we have a traffic free 3/4 mile track with many loops as our primary "high speed" zone (hospital parking lot with a fountain and many trees!). We do about 5 warm up miles at about 20 mph and then go for another 5-10 miles on the city streets. Really cool!
otherDoc