Please advise on best toque arm options please...

[Brummie]

Hi.
I have a specialized sirrus elite with carbon front forks and part carbon rear stays. Here's a link with a clear picture to zoom in on https://www.evanscycles.com/specialized-sirrus-elite-2009-road-bike-00125611#modal__product-zoom-modal

I'm looking to fit a 250W rear hub kit and want to make sure I don't do any damage!
I've looked at various installations of torque arm and some just look wrong to me with crank/pivot links that would make the jubilee clip either push up or pull down the tube, some would push against the tube & some pull away. I've even some that would cause the bolt to slide down an adjustable slot and just look mechanically wrong. Take a look here http://www.torq-reaction.co.uk/communities/8/004/012/703/138//images/4618760242.jpg

Any advice on how to ensure robustness on my bike would be very much appreciated please.

 

[Woosh]

you don't normally need a torque arm for a 250W rear hub motor.
The main thing is to make sure that the motor axle sits deeply in the dropout and the anti-rotation washer is also properly inside the jaws.
This is in fact straightforward, in most cases, just visually make sure that when you drop the motor in, it's properly in, like this:

those motors that require torque arms are usually the heavy direct drive motors, stay away from them.

 

[Brummie]

Thanks Woosh. My main concern was because of the aluminium/carbon joint you can see in my 1st link and whether it was more susceptible than your average installation?

 

[Deleted member 4366]

The seat stay is designed to be in compression, but when you fit a motor, the motor torque will put it in tension. I guess that the aluminium spiggot that joins it to the drop-out casting is only glued into the carbon fibre, so it could be a bit dodgy. I think I'd go for a crank-motor on that one.

If you still want a rear motor, it needs to be a cassette one, which will often mean that you need a large dish in the wheel to get the rim central. On a normal bike, I just force the drop-outs apart to stretch the frame a bit and add some spacers to left side, but you can't do that with carbon fibre.

 

[Woosh]

here is the hi-res pic, good enough to zoom in the rear hangers if you want to have a look.
the rear triangle seems pretty solid to me.
I agree with d8veh that a crank drive would remove any worry.

 

[Brummie]

Thanks (both).

If you remember from my previous thread, my main kit will be a 350/500 rear switch kit on an MTB. I've now tried a similar panda kit on an MTB like mine, although it didn't have the swytch 'shopper basket' battery/controller. Swytch are now offering me an extra kit (minus the power pack/controller) for only £105. This is why I'm considering a 250 kit for my hybrid but they only do front or rear hub options, and I have carbon forks, hence considering a 2nd rear kit for my sirrus bike

Looking at the bottom of the seat stay, there's no evidence of pinning so I guess it's glued. Looking at the picture, I could put an arm from the axle to bottom stay but it wouldn't help the tension problem on the seat stay insert. However, If I formed a torque arm triangle (pointing backwards), with a jubilee clipped around a rubber bung 2" up the carbon seat stay, it might keep the carbon/insert in compression or am I dreaming?

What do you think?

 

[Deleted member 4366]

The Swytch rear motor is a freewheel one, which will mess up your gears.

The torque arm only does anything when the motor would overwhelm the friction in the axle nuts. Normally, the axle nuts friction is what is resisting the torque, so the torque goes through the whole rear triangle, which means that the seat stay is in tension and the chain stay is in compression. Basically, the torque arm only stops the axle turning in the drop-out and it doesn't have any effect on forces in the frame. The only way to solve that would be to leave the axle nuts loose and take a tie-bar from strongly fixed torque arms all the way to the seat tube, but that's not really feasible.

 

[Brummie]

Thanks d8veh but I'm now confused...

The swytch rear kit is freewheel as standard, or $30 extra for freehub. They advised me that they are interchangeable between the 2 methods providing you have the an equivalent number of teeth and the same speed (max 8spd) and my bike is 8spd, although I'm opting for the freehub anyway. You may require minor derailleur adjustment (as per any cassette/wheel change).

Could you please explain why this kit would 'mess up' my gears because I could have misunderstood their claims?

I'll have a think about the torque arm for a while...

 

[Deleted member 4366]

A freewheel can take a max of 7 speeds before you get problems. Normally, I'd say pay the extra for the cassette version, but the Swytch kits I saw had Xiongda YTW-06 motors, but they don't have a cassette version. That means that Swytch must be using a different motor for the cassette one. I'd guess at a Q100C, but that's just a guess. The Q100C is 142mm wide, which brings its own problems. You'd need to confirm with Swytch, which cassette motor they supply - if they'll tell you!

 

[Brummie]

I will check if the motor is the same/different between freehub/wheel but reading all the info & my survey, I'm 100% certain they say freewheel includes up to and including 8spd and if you want 9spd+ you have to use freehub at $30 extra. BOTH types are 135mm.

 

[Deleted member 4366]

They can say what they like and you can fit any number of speeds on the freewheel, but it will bring complications.

 

[Brummie]

d8veh... I have not found what motor is used for each power, front, rear, freehub, freewheel etc but this is what they say...

Wheel Size and Speed Selection Please confirm the motor wheel size and top speed you require. Our rear wheel has an 135mm dropout width, with a freewheel fitting for gears. Note, if you only want to receive a hub-motor with no wheel, this option will be given under the 'Additional Requirements' section
Cassette Fitting Upgrade The standard fitting supplied with the rear motor is a "thread-on freewheel" fitting, the traditional method of fitting gears to bikes. It allows for up to 8-speed gear clusters to be fitted. The alternative "cassette" fitting is available for an added cost of $30 which will allow you to use cassette-type gear clusters and up to 11 speed gear cassettes.

I've paid for the $30 upgrade to use an 8 spd cassette (even though I could have used a freewheel). What could possibly go wrong?

 

[Brummie]

The torque arm only does anything when the motor would overwhelm the friction in the axle nuts. Normally, the axle nuts friction is what is resisting the torque, so the torque goes through the whole rear triangle, which means that the seat stay is in tension and the chain stay is in compression. Basically, the torque arm only stops the axle turning in the drop-out and it doesn't have any effect on forces in the frame. The only way to solve that would be to leave the axle nuts loose and take a tie-bar from strongly fixed torque arms all the way to the seat tube, but that's not really feasible.

I was thinking of how to mitigate against the issue you raised on your previous post with your concern that the seat stay should be in compression but the motor torque will pull the glued end in tension...

Looking at the picture, the motor would tend to twist the local axle bolt area of the triangle anti-clockwise, thus pulling on the carbon glued joint. My idea was to point the short part of a 2-part torque arm behind the bike (like an engine crank and main bearing), then connect the bearing (nut & bolt) to a connecting rod that that's little end is clipped 2" up the carbon seat stay. It would look just like a crank, con-rod and piston (the piston being wrapped around the seat stay in this case).
You may be aware that a piston sweep v's crank angle plot is like a sin wave. Its largest linear piston displacement for a given delta crank angle is half way up/down the sweep. If the crank is perpendicular to the seat stay, the slightest torsional twist of the axle area will pull down on the piston (jubilee clipped rubber), thus keeping the bottom of the seat stay in compression during a positive torque condition.

Take a look at the sketches at the end of this link and it may become more tangible. Imagine the crank rotating ant-clockwise. https://sites.esm.psu.edu/courses/emch12/IntDyn/activities/kinematics/slider-crank1/

I would make the angle between the crank & connecting rod more acute to avoid side thrust on the carbon and ensure the force on the carbon was straight down, thus not causing bending.

Now, there are some complications at the jubilee clip end to explain but I'll come to that at the next stage of the convo...

 

[Woosh]

I was thinking of how to mitigate against the issue you raised on your previous post with your concern that the seat stay should be in compression but the motor torque will pull the glued end in tension...

I think no matter how hard you try, adding more unsprung weight to the rear triangle is not going in the right direction, which is: a light rear hub motor with low reduction ratio (eg Bafang SWX) or a crank drive kit.

 

[Brummie]

the right direction, which is: a light rear hub motor with low reduction ratio (eg Bafang SWX)

Is the swytch 1.4kg xiongda ytw-06 not similar?

 

[Woosh]

Is the swytch 1.4kg xiongda ytw-06 not similar?

d8veh is the expert on all those xiongda ytw-06s.
I haven't tried one yet.

 

[Deleted member 4366]

The Xiongda's torque is similar to other motors because it has a big internal reduction ratio. I don't see mass as relevant because it only affects compression forces, then it only reduces them.

As I said earlier, a crank motor is the solution anyway. A small hub-motor would probably be OK regarding the torque, though that also means possibly not, but I'm more worried about the width. You need a cassette motor, which are nearly always wider. You could test to see how far you can spring your drop-outs.

 

[Woosh]

IHMO it's not only the motor's torque that presents a risk to carbon fibre components, the unsprung weight adds to the risk by its inertia.
Imagine a hammer handle made with carbon fibre...

 

[Brummie]

It was Woosh that showed concern of the weight on the rear triangle, and then said that a light rear hub motor was one of the ways to go (and it is a light rear hub).

The crank motor is not really an option. It's a bargain £105 extra kit on offer to anyone who's bought a full swytch kit (everything but the power controller) which I can just swap over from my full priced kit. They don't do a mid drive kit, its front or rear only or I just don't buy it!

No matter how many times I re-iterate my question to swytch, they keep telling me that there 135mm rear hub is always 135mm with up to 8 spd on freewheel & 11 spd on freehub! Are you sceptical about this?

Just tried... I can effortlessly spring the stays 20mm further apart.

Do you think that my torque arm explanation overcomes your seat stay tension concern (post 13)?

 

[Deleted member 4366]

I've never seen an 8-speed freewheel on a motor at 135mm wide. That's not to say that there isn't one, though. Personally, I'll believe it when I've seen it.
7 speed is 135mm. The stack-height of an 8-speed is only a few mm more, but it overlap the axle nut, so it would lock against the frame. You therefore need to add a spacer onto the axle, which is basically a thick washer with 20mm O/D and 12m hole. That's a non-standard size. You can get them, but they take some hunting around to find.

Those few mm don't make much difference to the width, but it's all on one side and the wrong side because that side is already longer than the other, which means that you have to put a bigger dish in the wheel to get the rim central. The wheel won't be built like that, so you have to do it yourself, which requires a bit of skill and experience.

When you put a dish in the wheel, the spoke tension is different from one side to the other. A small dish is no problem, but as you go further, the difference goes up exponentially until you have zero tension on one side and astronomic tension on the other when the spokes on one side are perpendicular to the hub. Obviously, you can't dish any further than that. The dish you need for an 8 speed is in the area where the difference in tension is too much.

You can solve the dish problem by adding spacers to the left side in order to bring the rim in the centre, but that increases your overall width.. That's OK on an aluminium frame because you can stretch it. I've taken a 135mm frame to 150mm, but you can't stretch carbon fibre.

I know some people are quite happy running around with 8 speed freewheels and their rim off centre. It can be a problem with rim brakes, though you can pay around with the spacers to mitigate it.