Shimano Alfine 11 - Longer term reliability

For discussions about bikes and equipment.
Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 22 Apr 2019, 6:07am

the hubshell (including the centrelock spline) is aluminium. There are hard steel inserts in the hubshell for the bearings and for the roller clutches to engage with. If the hubshell is welded, the heat of welding will inevitably run through the entire hubshell, and there is a real danger that the hard steel inserts will be tempered (softened), distorted and/or loosened.

So I don't think welding will be a good repair. I think that you would be best off getting another hubshell, and using a much smaller brake rotor on it. There is a suggestion that if the bearing adjustment is a bit tight, this may help the hubshell to fatigue too.

FWIW the A8 shell is a different design; it has a larger steel insert in the left side (which is used to take the drive in gears 1 and 5) and this may help it withstand the loads without damage.

A11 hubshells fail quite a bit less often than internals do, so I think finding a replacement hubshell oughtn't be too difficult.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

meandros
Posts: 59
Joined: 1 Jan 2018, 7:34pm

Re: Shimano Alfine 11 - Longer term reliability

Postby meandros » 22 Apr 2019, 12:47pm

Thank you Brucey.

I have no source for a S700 hubshell. I haven't even found one as spare new part. I guess I'll have to take my chances with having it welded back.

Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 22 Apr 2019, 2:29pm

there's a 26" wheel with hub, internals and shifter on ebay right now, for £50....?

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

meandros
Posts: 59
Joined: 1 Jan 2018, 7:34pm

Re: Shimano Alfine 11 - Longer term reliability

Postby meandros » 26 Apr 2019, 8:56pm

I appreciate it, Brucey, but it's not a matter of replacing the hubshell. I'm running A11 transmissions on my three bikes and this is part of the longer term reliability ;-) testing of these units. The challenge is to have them run for as long as possible without recurring to new parts, if the parts in question can be mended.

I must admit that having seen just how thin the material connecting the rotor wedge to the rest of the shell is, I was having some doubts. But the fact it's aluminum means it can be easier welded onto the hubshell and filed back into shape. Fact is it is now done. I had it welded and then filed the excess material around the centerlock spline. There was quite a bit that filled between the individual splines but managed to get it to a working shape that will hold the centerlock-6bolt Ashima adapter which I also had to file in order to take out about 2mm in order for it to be flush against the hubshell and maintain the rotor alignment.

Like you mentioned, Brucey, on account of the heat and the hubshell being mainly aluminium, there was a real danger of warping and it is true a bit of bubbling went on the inside, where the torch melted the aluminium but I filed that too back and tested for the hub scraping against it. I've now put a couple of layers of paint and a layer of clearcoat and am literally waiting for the paint to dry in order to assemble it all back together and hit the roads.

All I can say is having gone through this is that Shimano could have the whole centerlock wedge made of steel and have it machine welded onto the rest of the hubshell in a stronger manner. That would make these hubs truly trail worthy, in my book, as it would allow for larger rotors and heavier braking. I'm going to continue using the 180mm DH rotor on the back. It does seem, however, that more progressive, xc-trail rotors would be more appropriate.

Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 26 Apr 2019, 9:13pm

I shall be interested to see how you get on with the repaired hubshell.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

bsimulj
Posts: 1
Joined: 23 Jun 2019, 10:13pm

Re: Shimano Alfine 11 - Longer term reliability

Postby bsimulj » 23 Jun 2019, 10:40pm

Hello Brucey
I was running some test on Shimano SG-700 oil and in a home enviroment here are some results:
Density: 0.894 g/cm^3
Calculated viscosity: ~1500 cSt !!!

Tried to play with numbers to get some error margin 682<1500<2270 cSt where 1500 is most like result on 25 deg C

This looks to me as gear oil SAE 140 and greater.

You probably had opportunity to compare oils? Is this number real? All results show viscosity higher that SAE 90.
https://www.youtube.com/watch?v=Q7VZh0dI35M
Here is how i played with test.

Also do you have some measurements of the hub gears to determine contact surface to determine GL4 or GL5 application.

By the was we are riding two alfine 11. Not sure which version but both of them do not leak oil (at least no visible). Only problem we experienced on wife's bike is derailer hook was not hard enough and we had to machine a better one on CNC machine. Since then operation flawless.

To me shimano for sure uses gear oil as I compared it with ATF dextron II and SAE 90 but results show even higher viscosity.

By the was thank you for a wonderfull threads like this.


Brucey wrote:well it is more complicated than that even. Remember that Dexron is a GM standard and they have (I believe) withdrawn all the manufacturing licences for both II and III now. If you see some newly manufactured Oil with Dexron II or III on it, it isn't regulated to any GM standard any more. GM's product is now known as Dexron VI and it isn't always backwardly compatible. Some GM manuals originally listed Dexron II as a satisfactory alternate lube to (yet another) preferred GM lubricant but in some applications of this sort Dexron III isn't a suitable alternative. In most other applications (which is most of them) Dexron III can substitute for Dexron II and will be just fine. [ Ford specs list 'Mercon' fluids; the details are different but the story is the same; the latest Mercon fluids are not always backwardly compatible and there are no licences from Ford to manufacture older spec fluids to a maintained standard.]

It gets worse. Oils are complicated. Remember I mentioned that ATF fluids are designed to run hot and lubricate auto transmissions? And that this is a different thing from a cold bicycle hub? Well ATFs have friction modifiers in them that on the one hand allow the torque converter to work properly (which is only coupled via the fluid under normal conditions) and yet be slippery enough that the lockup and the valve bodies operate smoothly even when relatively cold. These characteristics require 'friction modifiers' that alter the viscosity vs the strain rate of the oil. In addition to this there are a load of other additives to reduce foaming, provide some degree of EP properties, prevent corrosion, give acceptable viscosity at high temperatures, help keep everything clean. Some of these additives have a useful function inside an IGH, some don't, and some would do but it never gets hot enough.

However one thing I wouldn't ever do for any length of time is use a Dexron II-D oil in a bicycle hub. The reason for this is that (if it really is Dexron II-D...) it will contain a friction modifier (believed to be Jojoba oil!) that turned out to be excessively hygroscopic, and this caused corrosion in some applications. II-D was fairly quickly replaced by II-E which was better and -III is better still in this regard (but probably still isn't as good as most gear oils). Even so if you run Dexron II-E or III in a bicycle hub it isn't unusual to find that what you remove after a period of use is an emulsion of oil and water; in a car gearbox it gets hot enough that the water may be driven off in normal use but in a bike hub this doesn't happen.

Gear oils meeting GL-4 or GL-5 are manufactured to an independently maintained standard. The difference is that these lubricants are designed to give the maximum protection in highly loaded gear trains, specifically including a sliding component of loaded motion, and at modest temperatures. This (for an off the shelf product) matches quite well with IGH requirements; most IGHs have plain bushings in them and many have tooth forms that are (through design, manufacturing, or wear) imperfect and sustain sliding contact, rather than rolling contact as would be found with a perfect involute tooth form. You can buy ATF fluids that meet GL4 and GL5 specs but only the most expensive specialist ATF grades made with fully synthetic base stocks (costing about x5 as much as normal) are ever like this. The penalty for using a gear oil is that the viscosity is usually somewhat higher than an ATF grade oil.

So I'd use ATF as

1) a cleaning oil for short term use

2) in combination with an extant shot of grease (that could be dried) again in short term use. [This combination allows for a mixture inside the hub that isn't excessively thick and gloopy.]

3) In longer term use provided it was combined with better anti-corrosion additives as well as better EP additives

If choosing an off-the-shelf oil to run in an otherwise clean IGH, I'd choose a GL4 or GL5 spec gear oil. If the extra viscosity was a concern (eg if there was grease inside the hub already), I'd spec a synthetic gear oil with the same EP performance but a lower viscosity at low temperatures.

cheers

Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 24 Jun 2019, 9:06am

oils are still an ongoing area of research. You can see why when you consider the net effects of friction and wear on energy consumption and equipment life, world-wide. Potentially there are savings to be made that might equate to the entire GDP of a small country. However many of the tests that are done to characterise lubricants are pretty 'broad brush' and are carried out under conditions that you might not find easy to replicate at home; in particular oil viscosities are often tested/quoted at elevated temperature. Also, additives can alter the viscosity in a way that varies with shear rate.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

meandros
Posts: 59
Joined: 1 Jan 2018, 7:34pm

Re: Shimano Alfine 11 - Longer term reliability

Postby meandros » 22 Sep 2019, 4:42pm

Is it possible to lace the Alfine-11 in 4-cross or is the hubshel too big to allow for such extreme spoke lenghts?

Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 22 Sep 2019, 5:17pm

meandros wrote:Is it possible to lace the Alfine-11 in 4-cross or is the hubshel too big to allow for such extreme spoke lenghts?


You can lace it four cross but

a) in 32h especially, the spoke heads may foul adjacent spokes
b) the rim may or may not allow the nipples to articulate fully; choose your rim with care

Small rims invite bigger nipple angles and are less easy to build this way.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

meandros
Posts: 59
Joined: 1 Jan 2018, 7:34pm

Re: Shimano Alfine 11 - Longer term reliability

Postby meandros » 22 Sep 2019, 6:34pm

Brucey wrote:You can lace it four cross but
a) in 32h especially, the spoke heads may foul adjacent spokes
b) the rim may or may not allow the nipples to articulate fully; choose your rim with care
Small rims invite bigger nipple angles and are less easy to build this way.
cheers


Thank you much, dear Brucey, for the swift reply!
I was rather entertaining the idea for I have a bunch of spokes lying around that don't fit any other way. Until now, I've laced A11s 2-cross and 3-cross but was reading up on 4-cross lacing and was wondering whether it as possible with the A11. For what I read into it, Brucey, I'll stick to what I've been doing. Was even thinking about mixing up patterns, 1/2-cross for the drive side and 3-cross for the brake side...

Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 22 Sep 2019, 8:03pm

https://www.kstoerz.com/freespoke/

allows the prospect of a spoke head/spoke clash to be evaluated.

https://www.ebikes.ca/tools/spoke-calc.html?hub=cust_d95_s60_o0_n36_l135_h3_p16.6&pair=false&rim=700DM18&cross=3a

gives the angle (in one plane) between spoke and the rim, but (annoyingly) doesn't seem to allow x4 to be used.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

meandros
Posts: 59
Joined: 1 Jan 2018, 7:34pm

Re: Shimano Alfine 11 - Longer term reliability

Postby meandros » 30 Sep 2019, 8:35pm

I laced an A8 4-cross and it's fine, the spoke heads don't touch the spokes. I used Sapim 2.0mm straight spokes. They way it looks, more likely that with more heavy-duty 2.3mm spokes the heads would touch the spokes but other than that, it's doable.

Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 30 Sep 2019, 8:52pm

it makes a difference if it is 32 or 36 drilled

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~

meandros
Posts: 59
Joined: 1 Jan 2018, 7:34pm

Re: Shimano Alfine 11 - Longer term reliability

Postby meandros » 30 Sep 2019, 9:56pm

Brucey wrote:it makes a difference if it is 32 or 36 drilled

cheers


Cheers Brucey!

Indeed, it was laced onto a stamped 36h 559-19 Ryde Zac Pro rim. I haven't tried a 32h but issues could appear as the angles would increase. as reading somewhere the more holes a rim has the more crosses it can take, like for instance a 40h rim would do 4-cross easy. The real advantage of 4-cross lacing on a 36h might be lost but I have laced the drive side as a 3-cross and only the brake-side 4-cross, so as to have a more even spread on the hub shell. The wheel looks nice, with pairs of 4 spokes spreading the load more evenly on the hub shell. Had the spokes lying around so wanted to have a go at it.
Have you mixed cross lacing from left to right like that, Brucey? Before this, I laced an A11 on a 36h 584-19 rim alternatively 1-cross on the drive-side and 3-cross on the brake-side. The 1-cross were black DTSwiss 2-1.8-2 spokes and the 3-cross silver Sapim straight 2mm spokes, creating quite the stiff but enduring wheel. The dude that got this transmission is now into an electric-bike ~5000km roundtrip. He overloaded that bike and as far as he reports, the wheel is good and the hub is, ofcourse, flawless (he still uses the 2-chainring front derailleur). He got a later edition SG-S700 series "O" hub (if that makes a difference to you, Brucey) and it performs great with a 250W pedal asist electric motor.

Brucey
Posts: 39899
Joined: 4 Jan 2012, 6:25pm

Re: Shimano Alfine 11 - Longer term reliability

Postby Brucey » 1 Oct 2019, 10:57am

I've built all kinds of weird wheels; you can mix the crossings up on an IGH like an alfine because the hubshell is stiff enough to transmit driving and braking torque between flanges and to share it pretty well. However how much benefit there is to be had from using odd combinations of crossings is debatable; you can often introduce new problems (e.g. nipple kinks, or flange loadings) in return for fairly slight gains.

cheers
-------------------------------------------------------------------------------
~~~~~~~~~~~~~~~~~~~~~~Brucey~~~~~~~~~~~~~~~~~~~~~~~~