16T Mac motor splayed fork dropouts - what next?

kiltedcelt

100 W
Joined
Aug 19, 2015
Messages
158
Location
Chicago, IL USA
So my 16T Mac motor apparently splayed out the dropouts of the fork on my Surly Troll. I was de-electrifying the Troll and had intended to use the hub motor on a cargo bike I'm building. I found out when I went to mount a dyno hub wheel back in the fork that the dropouts were no longer parallel. I was using dual Grin torque arms and I'd thought I had the bolts snugged enough, but apparently there was just enough play for the axle to start rotating slightly, thus splaying dropouts. I'd intended to put this motor in a 24" fork on the front of a longjohn-style cargo bike I'm building. The 24" fork I'll be using is actually a trials bike fork. Originally I thought this would be strong enough for this motor but now I'm having second thoughts. Maybe I should rig this motor up as a mid-drive, which will probably be a colossal pain in the butt, or maybe just sell the motor, custom battery case, and V3 Cycle Analyst and instead use those proceeds to buy a Bafang mid-drive. As it is, I had to buy a replacement fork to get my Troll back on the road. This Mac motor seems to have so much torque I'm kind of gun-shy now of putting it on the front of another bike as an actual front hub motor. Ad hoc mid-drive might be a better option. Also, when I first bought this it was paired with a 36v battery, but now I run with a 52v, which seems to allow it to develop a lot more peak power. I used to get about 900w max out of the motor (25 amp controller), with that 36v. Now it can push out more than 1200w. Any thoughts? Has anyone tried running a Mac as a mid-drive? I've seen some machined bracket kits but they're super pricey so I'd likely be trying to machine/weld up something of my own if I went that route.
 
Invest in a stronger fork still sounds cheaper than other things you said.
 
That was my experience with my first hub motor, a Crystalyte 5305, and a Surly Karate Monkey fork. The axle never spun, but the fork tips were hosed. I wasn't using a torque arm, because the one that came with the X5305 was so loosely fitted that I decided it was useless. This was about 14 years ago and torque arms were a new thing to me.

I have to assume that your torque arms were either poorly anchored or else not closely fitted to the axle.

Surly fork tips seem to be laser or waterjet cut. That makes them substantially weaker than forged fork tips made from like materials. But if I were to choose a commercial fork dropout to use for a hub motor, it would be a sturdy BMX type plate dropout. The material isn't as strong as forged, but the shape is better.

IMG_20180805_220158.jpg
 
Another thought had occurred to me as well - are those Dr Bass torque arms still available? I've done some searching but can't seem to find any current mention of them as if they're no longer available. This is the actual fork I have that will be going onto my cargo bike.
large_insflowsteel.jpg
 
If they are available, they should still be in the thread in the items for sale - new section. YOu can search there for threads started by DoctorBass if it's still running.


If you can't find the torque arms you want, there is a nice thread Torque Arm Picture Thread with a bunch of DIY and other options.


Generally whenever I've seen an axle deform dropouts front or rear, it means the torque arms weren't fixed down securely enough, and still allowed it to rotate. Or the arms' axle holes didn't perfectly fit the flats, thus allowing it to rotate a little inside the arms, and then deform the dropouts, and eventually even the arms.

If the arms are snug enough on the axle, and fixed snugly to the frame or fork to prevent any rotation of the arms, then you could leave the nuts off and it'd still stay where it is. ;) It'll work better with tight nuts and locking hardware, but the arms along with the dropouts should be able to take care of the forces involved.
 
The Mac has a freewheel and cannot apply regen torque, so it is exclusively applying torque counter to wheel rotation. Unlike a DD with regen, it does not rock back and forth and is always applying force to one side of the flatted hole in the TA. This means that a loose TA fit really has little effect. A slightly sloppy dropout fit also has little effect. Making these fit snugly is a valuable goal, but the one-way nature of geared motors diminishes the importance.

Without a photo to reveal the exact nature of the failure, it's likely this is an installation problem.

Since this is a gear motor and particularly if there is slop in TA or dropouts, it's arguably best to clock the axle fully in the dropouts in the direction of wheel rotation using an adjustable wrench on the axle flats - use the wrench to hold it clocked when tightening the axle nut on the other side. This unloads the dropouts so they are bearing no torque. The TA should also be clocked fully in the direction of wheel rotation. Under load the axle will torque counter to wheel rotation and the TA will bear it all and dropout very little. Things will deform slightly over time, but you are starting out with the load being borne where intended.

Ideally for two-piece TAs, the TA arm attachment should be behind the fork so torque forces the shaft deeper into the dropout instead of throwing it out.

Most importantly, the arm of the TA must be securely fastened to the fork and unable to move. A non-obvious failure mode for two piece hose-clamp TAs is for the hose-clamped arm-component to slide down the fork allowing the lower axle-retaining part to rotate with the axle because the two TA parts rotate relative to one another about the joining bolt. This makes it particularly desirable to attach the TA arm to a fender stay mount or disk brake mount instead of using hose clamps. Clearly these kinds of TAs work great out of the box, but parts can creep over time and checking TAs may not be on the radar.

Looking at your new fork, I would recommend you abandon the Grin TA hose-clamp arm part and instead cut a short piece of 1/8" flat stock and accurately drill it to fasten the lower Grin TA part to that protruding mount. Do the clocking/preloading mentioned above prior to marking/drilling. You really don't need giant Doc Bass TAs - you just need to tend to the details and the Grin TA that you have will work fine.

Anecdotal evidence is generally worthless to prove the positive, but FWIW I have been running a BMC (same as a Mac) at 1600W on the fork of a cargo bike for 15000mi with zero dropout issues using a single 1/4" TA.
 
Thanks for the further, enlightened replies. As an aside - now that I've stewed on this a while, I'm pretty sure that the stock Troll fork was out of spec to begin with. I do recall that the dropout edges were not parallel right out of the box, and had an ever-so-slight splayed appearance. Things *might* have gotten worse over the last year by somehow *not* setting up the torque arms initially in the absolute bes-practices way. *Maybe* things got a little more splayed but at this point I'm not so sure. Might just be I don't remember how unacceptable the stock fork was. Sounds like the new fork won't need any overtly stronger torque arm of any sort, although I will need to buy new Grin arm(s) because the current ones I have are the two piece type intended to bolt to fender mounts, and of course a trials fork doesn't have those. I am intending to use hydraulic Magura rim brakes on the suggestion of forum user Chalo, from a post I made over a year ago. Anyway, since I'm not using the disc mount I imagine I could easily fashion a way for the Grin hose clamp bracket to attach instead to the disc mount on the fork. The only thing is, I was intending to attach my steering arm to the disc mount since that's a similar height to where Larry vs Harry has their tab attached on the forks of their Bullitt cargo bike (although they have an actual tab that's on the drive side). Does anyone see any issue with having both a torque arm attached to the disc tab as well as a plate that will have a ball joint attached to the steering arm? If that seems like too much business on one attachment point I can always weld my steering tab/ball joint attachment onto the drive side of the fork.
 
kiltedcelt said:
The only thing is, I was intending to attach my steering arm to the disc mount since that's a similar height to where Larry vs Harry has their tab attached on the forks of their Bullitt cargo bike (although they have an actual tab that's on the drive side). Does anyone see any issue with having both a torque arm attached to the disc tab as well as a plate that will have a ball joint attached to the steering arm?

There are two 6mm mounting holes on a disc tab, with plenty of room between them. You can use one for your torque arm and the other for the steering joint.
 
teklektik said:
The Mac has a freewheel and cannot apply regen torque, so it is exclusively applying torque counter to wheel rotation. Unlike a DD with regen, it does not rock back and forth and is always applying force to one side of the flatted hole in the TA. This means that a loose TA fit really has little effect. A slightly sloppy dropout fit also has little effect...

Braking force alone is enough to make a loose fit TA letting the axle rocking back and forth. Precision of the fit is worth a lot of material strength.
 
MadRhino said:
Braking force alone is enough to make a loose fit TA letting the axle rocking back and forth. Precision of the fit is worth a lot of material strength.
Braking force does not apply torque to the axle or TA and cannot reasonably create any rotational rocking affect to spread the dropouts.
 
Back
Top