How to avoid another spinout?

[Qspawn]

Hi - I'm a noob with limited DIY skills and couldn't find a post about this, please point me in the right direction if I missed them.

I have a front wheel DD that just spun out (and wheel came off - ouch). Irony of ironies i was on my way to the hardware store to by more metal twist-ties for my torque arm as I had cut off the ones I had to work in the wheel.

In looking at the dropouts I realized that there's a few problems, and my question is how to fix them and avoid another spinout in the future?

1) dropouts (steel Surly LHT fork) are spread open so the axle turns somewhat easily when seated. Is there a workaround or do I need a new fork?
2) the washers with the little non-turn piece that is supposed to slot into the dropout (what are they called?) are too wide diameter - the piece just sits above the dropout, not in it. Is there a source to buy different sizes of these? Are they vital or could I just go with two torque arms and tight locknuts and not fret?

Thanks!

 

[wesnewell]

1. Squeeze them back as much as possible.
2. File them down so they fit in the dropout. Torque axle nuts to 100 ft lbs. Make sure they are flush. If there;s room for a jam nut on the inside of the dropout, use one on each side.

I've got 10,000 miles without torque arms in steel dropouts.

 

[dogman dan]

I think that if you can bend them back without seeing a crack, you can keep riding. Steel dropout forgings are somewhat malleable.

What you need is a set of C washers. Then put the TA back for best results, and insurance against a crack you cannot see.


http://endless-sphere.com/forums/viewtopic.php?f=4&t=15163&hilit=Ebikes+ca+c+washers

Your real problem was that you used the oversize tabbed washer in a fork with a quick release cup. This causes nut pressure to spread the forks, once the forks spread, the nut loosens and away you go. You can just file down your tabbed washer to fit if you like. If you do a homemade, get that fit perfect !!!!

Don't get too paranoid, and take a 24" wrench to your axle nuts. You might strip the axle if you give them 100 ft pounds. Just tighten them firmly with a short wrench.

 

[Drunkskunk]

I think the fork can be repaired, but I think you may need to replace them anyway. I've seen the LHT with 2 different dropouts. one is an extra beefy forging that would be ideal, the other is a lighter, thinner version that has a cup shaped half recess on the axle nut side. if yours is the second type, then propperly seating the axle and fitting the torque arm would be a problem.
If you have access to a welder, then cutting off the dropouts and welding on a pair of Doc's torque arms would be an over the top but effective fix.
If not, forks like the Surly 'check or Big Dummy have a dropout that allows for better torquearm fitment. Your old form may be salvagable for a non motor application.

I agree with Dogman, 100lbs torque is WAY too much. A car's lugnuts are normaly are around 80 foot pounds, and if you've ever had to change a tire, you know how hard that is to get off. An ebike motor axle is made of some pretty soft metal, and the threads are cut, not rolled (that makes them weaker) and to top that off, they only extend around 50% of the axle, so there is only half as much surface area for them. you would strip the threads trying for 100lbs torque. When they are snug and just a bit more with a hand wrench, they're tight enough.

Since you had a problem once, you might want to post some pics of your install before you ride it, for peer review. one of us might catch something you missed before any damage is done. It can help to have another set of eyes on any project.

 

[100volts+]

I try to use a mechanical solution (ie...drill and tap, nuts and bolts) to locking torque arms in place.
Pictured below is a rear hub with .375 cromoly horizontal dropouts and pinch type clamp.

 

[wesnewell]

Don't get too paranoid, and take a 24" wrench to your axle nuts. You'll strip the axle if you give them 100 ft pounds. Just tighten them firmly with a short wrench.

You need to do some research on this. 100 ft. lbs is on the low end of the scale for a M14 nut. Max is ~200 ft. lbs. if you strip them with 100 ft. lbs., they are worse than garbage. They should be class 10 -12 nuts.
http://yetmans.mb.ca/kohler/page3/page3.html

 

[teklektik]

Don't get too paranoid, and take a 24" wrench to your axle nuts. You'll strip the axle if you give them 100 ft pounds. Just tighten them firmly with a short wrench.

You need to do some research on this. 100 ft. lbs is on the low end of the scale for a M14 nut. Max is ~200 ft. lbs. if you strip them with 100 ft. lbs., they are worse than garbage. They should be class 10 -12 nuts.
http://yetmans.mb.ca/kohler/page3/page3.html

No research required - personal life experience trumps research. The issue is not the nuts, it's the hollow mild steel axles.

There are endless posts on the forum about the crappy steel used in axles -- low torque on the nuts is an expected side effect. Further, hollow/slotted axle ends deform easily (fold in) making the axle diameter small enough for the nut threads to strip off the tops of the axle threads. This is certainly an entirely different situation than a 14mm solid medium carbon steel bolt which is the expected matching 'proof' mate for the nut spec you quoted. Without a bolt of equal proof, the nut torque spec is inapplicable.

As dogman says - use a short wrench...

 

[wesnewell]

I've installed my hub motor at least 6 times and never striped an axle. I've also installed at least 3 others without striping an axle. I put more than 100 ft.lbs on them every time. Maybe mine are just better axles, but I doubt it. 100 ft. lbs isn't really that much compared to lug nuts that get up to 300 ft. lbs.

 

[teklektik]

Maybe mine are just better axles, but I doubt it.

As I said above - personal experience - I stripped an axle at barely over 65 ft-lbs - done with a torque wrench no less. You can doubt it if you wish...

I don't understand the reluctance to even acknowledge the possibility that axles may not all be grade 10+ which is what is required for the grade 10 nut torque spec to apply. If the possibility exists, then advising high torques should be given with caution because the consequences are very unpleasant.

I'm not saying higher grade axles don't exist, but particularly in the case of a modestly powered front hub motor, IMHO there is no reason to risk 100ft-lbs - let the lawyer lips and TA do their job.

FWIW - Here's a thread about stripping/repairing threads.

 

[wesnewell]

All I can do is provide my experiences and the torque specs. If you had an axle strip with only 65ft. lbs of torque then it was defective, either that particular axle or by design. Even into aluminum, the spec calls for 70ft. lbs for a M14 fastener. Now I saw a kid strip one, but that was because he cross threaded it. No matter how much torque one uses, they should be checked before very ride to see if they've come loose. At least for a few rides. Then at least a visual check before every ride.

 

[dogman dan]

Depends on the axle. Crystal clear the man never frocked up an old school Crystalyte axle, or even worse machined aotema one.

Point taken though, so I changed the post to say " you might strip the axle".

I have found my genuine 9c axles pretty good, and my newer 5304 axle is not too bad. But man, those others were soft as butter, and not even close to fitting the nut. I have no idea what he has, but I just wanted to make clear to him that you don't want to grab the air gun next time, max it out, and let er rip on your motor axles. Or grab that huge breaker bar.

Don't want to torque it down to just 10 ft pounds either. Get it good and tight, and if you can strip it with a wrench only 10" long, you were snafu'ed from the start anyway. I'm pretty strong in the arms too, so for me I can put close to 100 ft pounds on a crescent wrench, but use less on motor nuts. My guess is I get at least 50 ft pounds on the nuts.

The point is, his problem was not that he didn't tighten enough. The point is, his problem is the poor fitting washer, that spreads the dropout more and more the tighter he cranks it.

 

[chvidgov.bc.ca]

One thing I noticed is that if you use a lug nut right against a C washer, it can deform the C washer as the surface of the lug nut pushes on the surface of the C washer, at high torque. It's better to have a torque arm piece (which won't turn) on the outside of the C washer - it will only push down on the C washer, and then the lug nut on the outside of the torque arm piece, perhaps with another washer between the torque arm piece and the lug nut.
My .02 cents.

 

[teklektik]

You need to do some research on this. 100 ft. lbs is on the low end of the scale for a M14 nut. Max is ~200 ft. lbs. if you strip them with 100 ft. lbs., they are worse than garbage. They should be class 10 -12 nuts.
http://yetmans.mb.ca/kohler/page3/page3.html

All I can do is provide my experiences and the torque specs.

To make the point again - your 'spec' is inapplicable. It is based on an assumed grade of 10-12 for the nut/axle which is completely arbitrary. You are simply choosing a grade that when mapped through the 'spec' gives the torque you wish. Chose a different grade and get a different torque - this isn't a 'specification', it's an assumption about axle composition and shape wrapped in a bit of tabular sleight of hand; there is no specification involved that is based on factual physical characteristics of the axle... and this is ignoring the differing configuration of the solid bolt and hollow axle (possibly with side cutaway).

Here is a post by Justin "Thread/Bolt Stripping Test" describing a nut finally letting go at 125Nm (92ftlb).
FWIW - his comment regardling 92 ftlb (lower than your recommendation of 100ftlb):

I couldn't imagine anyone tightening a nut on a bicycle that way,...

In the same post, he tests a Crystalyte axle with conflicting results:

...all of Affliction's (Ben's) comments seem to be with regard to Crystalyte axles. Well we have some of those lying around too. So same test as above was done on a slightly used 400 series rear axle. The peak torque was 70 N-m (52ft-lbs) when the nut started to spin freely. Definitely stripped threads:

We then flipped the axle around to repeat the test on the other side to get a few more data points and get a sense for the repeatability. And guess what, on THIS side of the crystalyte axle, we had it all the way to 130 N-m when the wrench slipped and rounded the nut as above with the eZee axle.

So, while one end of the Crystalyte axle stripped all it's threads at 70 N-m (52 ft-lbs), this side withstood 130 N-m (96ft-lbs), with the threads still almost perfectly in tact:

Not totally sure what to make of this. Either there is a huge variability in the temper on the Crystalyte axles, or more likely I think it is the mechanical tolerances on the axle threads, with one side being a looser fit in the nut and hence much more prone to stripping.

Justin has posted graphs of results of No-TA Axle Spinout in steel forks showing results using (30,60,90)Nm or (22, 44, 66)ft-lb of torque. The upper limit of 66ft-lbs is very much in line with dogman's recommendation above. The entire page (thread) make interesting reading.

If you had an axle strip with only 65ft. lbs of torque then it was defective, either that particular axle or by design.

Indeed. I have two motors with this situation and others have experienced similar stripping due to a design issue as discussed in the thread referenced earlier. These then are production axles that you agree cannot be torqued to 100 ft-lbs... design flaw or not - that's the way they come from the factory.

I have no problem with claims of personal experience but you didn't qualify it as such (initially). Continuing to present your torque recommendation as a 'specification' is IMHO at best misleading. There are many folks reading this stuff who will take such statements and claims of 'specification' as gospel and who may pay an unfortunate price.

IMHO the more conservative torque of 66ft-lbs max used in Justin's tests (and which dogman has similarly recommended) is very much more in line with reasonable practice. Justin's graphs show axle spinout without TAs - certainly using axle nut torque even less than 66ft-lbs is not unreasonable with a TA. This bears out my experience as well (7300mi with a single TA and less than 50ft-lbs at up to 2kW and absolutely no loosening or dropout deformation). A good flush fit for washers, etc without bridging dropout indents/recesses will help minimize loosening.

 

[wesnewell]

Ok, I'll agree that many hub motor axles are of poor quality steel and you probably shouldn't buy them. If you do, use less torque and other means of securing the axle nut from coming loose. Nordlock washers, jam nut. thread lock, or something else.

 

[dogman dan]

I can't say I've seen all motors, but I do believe some margin of improvement has been seen in the quality of the axles. They may still be pretty soft, but the one's I've owned seemed to fit the nut a LOT better than early clytes and aotema's. Those loose fitting threads barely grabbed, and you could strip one at 20 ft pounds easy.

Justin's data showed me how good just the nut can hold. I don't dispute that, and rode about 2000 miles with a front hub before I ever owned a torque arm. Long as the nut never loosens, you don't really need the TA with steel. But it's nice to have if shit does happen, since two torque arms will help keep the wheel on the bike long enough to stop.

Definitely, the C washer is not intended to be used with just a nut. Dropout, c washer, washer or TA, then the nut. You might have a washer and the TA with the washer on top of the TA. If you have lots of axle, you could even stack a washer between the C washer and the TA.

 

[fechter]

It helps to get thicker nuts that will have more threads engaged (if the axle is long enough). I've found car lug nuts that work nicely.

 

[teklektik]

Justin's data showed me how good just the nut can hold. Long as the nut never loosens, you don't really need the TA with steel. But it's nice to have if shit does happen, since two torque arms will help keep the wheel on the bike long enough to stop.

Yep. I dragged up that data not to recommend that folks try driving without TAs, but to put the required torques in perspective. Something around 50ft-lbs might work without a TA, but with a TA (highly recommended), it's really only necessary to tighten things up sufficiently to prevent the nuts from working loose - the TA will stop the axle from spinning so killer torque isn't required.

What I was going after was that as long as there are issues with axle strength in some available motors, it's not possible to tell ahead of time how much torque can be applied safely without stripping - so a torque recommendation should embrace the low end of the strength scale. Poor nut fitment, although detectable, may not readily be apparent to inexperienced mechanics. There is no point is using excessive torque when a much lower amount will suffice with substantially minimized chance of failure. More is not always better...

My remarks were really aimed at amateur builders/mechanics like the OP with what are typically modest powered builds and who are looking for assembly tips - telling them that 100ft lbs is 'to spec' is potentially putting builds at risk for no reason - a fraction of that will work fine with or without a TA. On the other hand, if you are an experienced mechanic and familiar with the characteristics of a particular motor design and want to put a breaker bar on it, have at it.

 

[dogman dan]

I agree totally. It's not the axle nut on the rear end of a volkswagon. Don't use the wrench you have from when you had a bug, or the one you use to change CV's on your car.

Some hubmotor axles can take 100 ft pounds. I don't dispute that. How you know ahead of time if it can is definitely the problem. Even with a brand you have not had problems with before, the things are still made to Chinese quality control eh? I tend to try a soft torque first, then after about 15 min for the washers to settle and deform all they will, I go for the last bit of further tightening.

50-75 ft pounds is plenty, if you can get that much on it. Lug nuts make a good replacement nut, particularly if your axle has a loose fit with the nut. More threads engaged is good when the fit is loose as hell.

I have to say though, I don't know what tension I put on motors. I don't own a torque wrench, and I'm pretty strong. I do tighten them very hard, but with a short wrench. I wouldn't be surprised to find I actually torque them down a lot more than 50 ft pounds. I don't really think I get to 100 with a 6" wrench. With a long breaker bar, you can pretty easily exceed 100 ft pounds if you are a strong guy.

I'm just trying to tell folks that a torque arm, along with the proper fit of the washers, will hold fine without getting out the huge wrench. When you give it all you have with a long breaker bar or giant adjustable wrench, it's just too easy to blow out the axle. Personally, I can feel it better with a short wrench, when a bolt is just short of a full strip out or twist off. With the long bar, it just feels the same as you twist a bolt in half.

As many know, the nut is suppose to be slightly softer than the bolt, but with hubmotors, the axle is usually the soft part. And the motor is much more expensive to replace than the nut.

 

[wesnewell]

I use a 12" adjustable jaw (Cresent type) wench and pretty much put half my 270lbs on it on the side with a jam nut on the inside. I gave all my torque wrenches away years ago when I stopped working on aircraft, but I used them a lot back in those days and can judge it pretty close still I think. On the freewheel side with the hollow axle, I use less torque, but still probably closer to 100 ft lbs, than 50. There is a point where more torque does no good. Once tight, I try and pump the wrench. When it stops turning I stop pumping. i crushed the left side aluminum spacer when i first installed the wheel. That's when I put a Jam nut inside and threw the crushed spacer away. If you can put a 12" wrench on it and still turn it, it's not tight enough.

 

[999zip999]

60 - 75 ft lbs is fine and look and feel the nut on the axle's threads as some from the factory match like shit. On a front hub I would put a torque arm on for peace of mind at 25mph.

 

[alsmith]

I asked Crystalyte about the torque for my 5300- reply was:

Info Crystalyte info@crystalyte-europe.com
28/05/2011

to me
Hi Alan

40 Nm is good for the motor axle nuts.

(29.5 foot pounds
354 pound inches)

Greetings

Thijs


The 5300 uses a 14mm spindle, not the more usual 12mm.
The thread on it is M14 fine, I use nyloc steel nuts.