Technical - Q-Axle Crank Interface

Unicycles and Equipment
1

It’s been a while since Qu-Ax released their Q-Axle interface and I’ve not really seen all that much discussion about it on here.

Now that’s probably a good sign, but the tech’s also on a far smaller number of unicycles compared to the alternatives so perhaps that means nothing at all.

So, who’s got a Q-Axle based unicycle, and what issues (if any) have you had with it?
What do you like? What do you not like?

Not looking for anything to do with the unicycle that the hub’s attached to, just discussion about the interface itself.

Personally I’ve so far had no issues with the interface, but I do have some thoughts:

  • The lightweight hollow axle so far has been used with a pressed on aluminium shell which has the potential to shift if spacers are not used/one crank is forced on with the end cap before tightening the other (QU-AX Q-Axle 29 Disc white)
    It also could possibly rotate if the tolerances are not absolutely fantastic - Nimbus and KH multi-part hubs had this issue in the past, sometimes leading to creaking.
    If this did appear to be a problem, is there anything that would stop the hubs being made one piece?

  • The tolerances between the crank and the axle have to be absolutely perfect to ensure that the splines are in solid contact on both forward and backward motion. On a bike this isn’t really so much of an issue - if the cranks slip slightly then it’s unlikely they’d ever really move back and forth as the force is mostly all in one direction.
    From my personal experience although I’ve not experienced any slipping, the tolerances do not appear to be such that the splines quite contact in both directions (on my 145mm Zero-Q cranks) - it seems like the clamping force is doing some of the work in stopping them moving.
    On an ISIS spline as you tighten up the crank (to a correct width spacer), the interface typically deforms very slightly ensuring a perfect fit between the crank and axle. If/as the interface wears, you can generally use a slightly smaller spacer to get more life out of the crank.
    Any wear on the Q-Axle interface would seem to lead to potential creaking forever if the clamping force alone is not enough to stop movement.

  • The bolts on the end of the crank require either large cranks to get a circle, or have an uneven bump. In reality this appears to not be an issue at all for roll tricks (with the zero-q cranks) from what I’ve seen.

  • The larger axle requires smaller bearings to fit into a 42mm bearing holder which will wear faster, and from personal experience do not appear to be as good quality as the basic ones supplied with other unicycles.

  • The interface forces a single angular crank location so they cannot (unlike with other interfaces) be rotated to even out tyre wear/wheel force location.
    Now with ISIS regular removal and re-installation of cranks is a good way to wear the interface, but it is at least possible. Rotating the tyre is somewhat more difficult.

2 Likes
2

There would be no issues with making it one piece. But a press fit like that isn’t too uncommon in the engineering world, lot’s of products can pull it of without issues - the tolerances have to be good, but it’s nothing very special. I think one rotating or moving side to side on the hub is highly unlikely.
I’ve seen one flange break on a Q-axle hub - on a wheel someone has built himself, with I suspect too much spoke tension. Breaking flanges somehow seems to be something few unicycle hubs are ever 100% safe of…

In the (admittedly limited) time I had to try out a Q-axle flatland unicycle with Zero-Q cranks and their roll disk, I’ve found them to be “surprisingly ok” for rolls. The bump is noticeable, but you can do crank rolls on them much better than I initially expected - without the rolodisks they are terrible.

If you are a flatland rider focussed on rolls - you likely won’t be happy with them. If rolls aren’t the focus of your riding, the “Rollmops” rolo disks allow you to do them - it won’t be the ideal setup for rolls, but I’d say it’s similar to either Isis Zero-Q or Mad4one without rolo disks, which is pretty decent.

Bikes and especially mountainbikes do have forces in the opposite direction about as much as a unicycle I’d say - every time you stand up on your pedals, which on a MTB is quite often.

The setup Q-axle uses is a much more “common” thing to do in engineering, just straight splines, no taper. Why is it prefered? Because it’s a lot simpler to get the tolerances right, Isis cranks have to be first broached, then coined to create the taper - which is relatively tricky. Q-Axle can be just broached, which can be done much more accurately.
Of course, the fact that an end user is supposed to be able to install cranks at home makes life difficult again, which is why they’ll need a loose fit, which is then reduced by the clamping. A correctly manufactured crank interface shouldn’t wear at all, so I personally wouldn’t give Isis any extra points for the ability to adjust for it. (And the spacers aren’t even a part of the Isis standard, I think they are mostly a result from the size of bearings unicycle manufacturers decided to use)

Experiences/Issues
I’m personally a bit of a “if it’s not broken, don’t fix it guy”, and not really a weight weenie, so when Qu-ax came out with Q-Axle, I wasn’t super excited by it. But so far it seems to me, for everything except roll tricks, it works at least equally well as Isis while being lighter.
I don’t own one, but I know a number of people that do, and their issues are: caps falling off, and some of them do get some creaking which can be fixed with cleaning, greasing and retightening the cranks (or usually just retightening). Which to me seems to be roughly equal amount/severity to what people experience with Isis cranks.
The main concern for a buyer remains of course the obvious lack of cross-compatibility and crankchoice. (Depending on how much you change around cranks, that may or may not be an issue)

2 Likes
3

For me, it’s this. I haven’t swapped cranks around recently, but having set/s of dedicated cranks for just one uni which I can’t interchange and use on other unis is the problem. So it may be slightly superior, but every other uni I have is either cotterless or ISIS. To have the QuAx standard introduces a third standard, and third different kind of crank.

4

Well, it’s already happened at least once to a Q-Axle hub, and was a common occurrence on old Nimbus/KH multi-part hubs.

This seems to be why most hubs have moved back to single piece steel affairs - they hold up to a lot more abuse than the aluminium ones. Aluminium flanges do definitely have a habit of stress fracturing over time, but as you say, that’s nothing to do with the interface used.

This is a curious one to me. The standard specifies 16mm to a stop - the spacers on a unicycle being that stop - as otherwise you wouldn’t be able to get the bearings on. I don’t know whether any unicycle manufacturers at all follow this spec in design though. If not, I wonder why not.
Personally when I purchased my Impact Gravity a few years ago I discovered that as new, the cranks would touch the supplied spacers immediately - that kind of poor manufacturing/design from my experience sadly seems common and perhaps is a cause of more issues than the ISIS interface would have were things as they should be.

Thinking about this, it depends on which foot forwards you stand up. If you’re left foot forwards, I don’t think the force direction ever changes (as you have to remember that the sprocket is directly attached to the right hand crank rather than the bottom bracket itself).
On a unicycle for most people the force direction changes every single pedal stroke though so any slight movement is going to occur far more often than on a bike.

Oh, and aside from e-bike versions, aren’t the right hand crank and the axle normally bonded together on bikes, with only the left hand side being removable with pinch bolt tightening?




Does anyone know whether Qu-Ax has a torque spec listed online for their Q-Axle (or in the booklet the unicycle is supplied with? Mine's been recycled already).

Shimano seemingly state 9.9-14.9Nm (88-132 in lbs) for the pinch screws, and 0-5-0.7Nm (4–6 in lbs) for the caps.

I’ve never seen a recommended torque spec for any ISIS interface bolts either, with the recommendation generally being a big hex key and plenty of force.
Truvative say 43–47Nm (384–420 in lbs) for their ISIS drive setup.

5

On a bike that stop is just integrated into the spindle itself, or the bearing - it’s not changeable by the end user. The spacers on a unicycle are a good design, don’t get me wrong. I’m pretty sure Koxx one were the only ones not using a spacers, with a crank stop integrated into the hub and larger bearings (on their last, “welded together” series of hubs). It’s manufacturing issues with either the hub or cranks that leads to cranks not fitting well, not the spacers themselves. (Gotta love overseas manufacturing and the resulting QC, but on the otherhand without it we wouldn’t have affordable unicycles…)

Anyway, the point was that if there were major problems with opposite forces on a particular crank interface, you would find those on a bike too. You are correct though, that direction change happens more often on a unicycle. (But splines generally speaking are good at handling direction changes)

2 Likes
6

Speaking of one piece hubs, UDC has just revealed that they’re getting in 32h one piece aluminium Q-Axle hubs in their next container which might be a nice option for a super light RGB/carbon rim 36er.

2 Likes
7

It has been available on the Qu-Ax website for a while now. If you compare the prices, you know why they still produce the unis with the two piece hub.

2 Likes
8

Qu-ax proudly and rightfully presents the light weight of their hubs. And a number of riders seem to think that the Q-axle system is quite a lot lighter.

But is it really (that much) lighter?

— Slightly unfair hub weight comparison:

For ISIS hubs the clamping mechanism, i.e. the bolt on each side, is counted as the hub’s weight.

For Q-Axle the clamping mechanism is “externalized” to the cranks, i.e. two smaller bolts and a bit extra material on each crank arm.

Thus a direct comparison of ISIS and Q-axle hub weights is slightly unfair.

— Actual weight difference for muni hub+cranks:

Let’s go for some rather robust mountain unicycling setup with about 127mm cranks:
Q-axle: 870g (ZeroQ; 380g+487g)
KH: 960g (Spirit; 580g+380g)
Mad41: 970g (Agile; 575g+393g)

So your actual weight saving is about 100g. (And strangely, I consider it “non-rotating mass”.)

Congratulations! Now treat yourself to an extra 100ml of your favourite drink before each ride, because that’s as much as you saved here. :wink:

Further notes: Of course lighter options for either hub system are available – by sacrifising sturdiness and/or more of your money. To be precise, sacrifising sturdiness is mandatory if you go lighter on Q-axle – unless for ISIS/KH, where you can “Exceed”. We are deliberately neglecting all pros and cons here. This is about the (somewhat silly?) weight question only.

If you are a gram counting degenerate like me, you can do your own hub+cranks weight comparisons for trials, freestyle, light muni, touring/commuting or your favourite style. I guess results won’t be too shocking either.

2 Likes
9

To preface this, I’m pretty much an anti weight weenie, my setup weighs about a ton and I’m perfectly fine with it.
But a 10% weight reduction assuming equal strength (I’d say the Q-Axle setup is the most sturdy of the setups you listed, or at least the zeroQ cranks are), and for I’d guess similar money is actually really good. On most other components, you will pay for those 10% in price, or strength by a lot.

3 Likes
10

100g is 100g!

I’d be interested where you’re getting the weights from though - I’ve never found quoted weights from manufacturers to be very accurate…

I’ve thought for years that it would be nice to have a database of actually weighed parts - people can add the weights of what they have, and it’ll show the lowest, highest, and some averages.

2 Likes
11

Yes, I quoted weights from manufacturers there. You might want to take these numbers with a grain of salt.

I found a nice list of weights from anno 2009: weights of Uni parts - #21 by olarf Apparently unicycles did already exist at that time! :open_mouth:

I admittedly weighed a fair number of parts myself. That was until I discovered that the scale, I used back then, was quite inaccurate itself. :persevere:

12

I own a Qu-Ax RGB 27.5" unicycle with the Q-axle, almost two years now. Haven’t taken anything near the hub apart yet - cranks came mounted when the uni arrived. The hub performs just as it should, not creaking and feeling solid. I ride relatively light offroad most of the time, on this uni at least.

As to weighing of parts: I’ve weighed a lot of my own parts on a good weighing scale, every time I take something apart. If there is a database somewhere, I’ll gladly add my data. It would be nice if the database allows multiple values per parameter, so that each person can enter their own data. The database could calculate the average per parameter (or perhaps the median, to prevent outliers from influencing the ‘result’).

4 Likes
13

Each weight should include an identifier of the actual measuring instrument used. Then it could potentially recognise patterns that suggest the particular scales were not accurate.

14

The database should also include precisely where in the world the measurement took place, so as to account for local variations in the earth’s gravitational field.

2 Likes
15

People use hundreds (if not more) types of scales, so I don’t think your pattern recognition would work.

Seriously??
Besides variation with latitude, there is variation with altitude. But the deviations from the typical value of g (at sea level and 45 deg latitude) are less than 0.5%. Not worthwhile to take into account. The inaccuracy of many ‘household’ scales is probably more than 0.5%.
Besides, who knows if scales sold near the equator aren’t actually calibrated for the somewhat lower gravity in the area. (Similar to automotive fuels that have differing formulations depending on region and season.)

1 Like
16

I am pretty certain that @Ulrik was just being sarcastic. :rofl:

1 Like
17

Be that as it may

18

So before we someone mentions :last_quarter_moon_with_face: tidal forces’ effect on weighing …

The grey Q-axle QX cranks can be “heel kickers” :foot:

qaxleqx

You ride a bumpy muni section. Your foot slightly changes position on the pedal. The rear end of the crank (where the two bolts are) catches your (shoe’s) heel from below. It levers your heel up a bit and subsequently your foot off the pedal.

It might also happen after a jump, drop, fast pedalling or even relaxed riding when your feet aren’t firmly planted on the pedals and move a bit.

Having your foot suddenly lifted off the pedal can be annoying or even lead you into an unplanned dismount. :anger:

The grey QX cranks are particularly prone to that due to their particular edgy shape at the rear end. The rear end works like small lever and is even bent outwards very slightly.

Noteworthy this might not be (much of) an issue at all for you! :relieved: – depending on the crank length, riding style, your shoes and your lower extremities’ anatomy.

Personally I had way less issues with other straight cranks, even longer ones, e.g. Eiffel Tower 140mm or some aluminium 145mm. Thus the grey QX cranks appear to be the worst in this regard.

Also noteworthy is the other end of the spectrum, KH Spirit cranks, the shape and bending of which will lead your misplaced foot securely into place. :relaxed: And I presume that this is no or less of an issue with the black ZeroQ cranks, as they seem to be better shaped.

1 Like
19

Air pressure is a thing too, and humidity, and temperature, and whatnot.

20

As much as I agree with others that this sounds a little silly on first glance, it’s something I’d thought about.

If you were weighing a whole unicycle and you used bathroom scales, it’s definitely worth taking that info account if possible.

Perhaps “precision of scale” could be a good option to allow some weighting based on a user’s weighing apparatus.