[macsi]

The story of this tuning project is the following: I got a pre-production LX Hollowtech II crankset last September for testing. I was astonished how great value it represented: for 85 euros you get an incredibly stiff crankset with bearings that even 100 kg guys you can use on hard terrain for the same weight as my old Hollowtech II XT!

I rode it all winter in snow and salty conditions and the bearins remained smooth. The chainrings were now worn: 3500 km is enough for the cheap aluminum rings. But despite of all the good properties, there was one I hated - the same reason why I gave back the XT Hollowtech II from the previous year (2003) - it had too wide Q-factor for my liking...

I was continuosly thinking about how to reduce the Q-factor. I could get rid of one of the three 2.5 mm spacers if I didn't use the plastic bolt that you use to mount the left arm. (I had to get rid of the plastic tube as well, but I seal the seattube, so nothing enters the BB shell.) But this is only 2.5 mm and my favourite setup, a RaceFace Turbine with 108 mm ISIS BB has some 14 mm (!) narrower Q-factor, so there was a lot more to do to get the crankset the way I like it.

You have to be aware that wide Q-factor only bothers some cyclist. Some actually like to pedal with their feet spread out... I'm not one of them, though I'm not an extreme case like Thomas Frischneck, who uses road-wide pedal-to-pedal distance (155-160). I'm satisfied with the 160-165 mm widths for MTBs.

It is also important to know that I live in a country, where labour costs are low, and you can get a machinist to work a full day on a part like this for 20 euros. (This is how much the whole tuning project cost me + two beers for the guy who polished the arms...)

So here are the steps:

1. The easiest way to reduce the Q-factor is to move the peadals closer to the crankarms. Since I use Eggbeaters, I can't adjust the cleat position sideways, so material had to be taken out of the pedal thread. In the case of the LX, it's not a problem, since the first 3 mm of the hole has no threads: the reason for this is that it shares the design with the Hone, and the latter will have a steel sleeve from 2006, so they had to make room for it. (Why didn't it figure on the 2005 model? - Go and ask Shimano...)

So I had 2,5 mm taken out of the shoulder that had no thread, so the Eggbeater axle perfectly fits into this and makes the connecting point look real nice.

 

[macsi]

2-3. The next step has nothing to do with Q-factor, "only" saving weight. The part around the axle had also way too much material. It's pobably for aestetics, and to make the Hone stong enough for 100 kg freeriders. 2-3 mm could easily be filed away without any structural problems. I was more careful on the left arm around the bolts as you can see.

 

[macsi]

4. If the foot moves closer to the arm, and I already polished parts of the arm with my winter neoprene shoe covers, some material had to be taken from the arm surface. Fortunately Shimano used a shape where clearly there was excess materail on the edges of the arrow (the inside of the arm has an oval shape - I know, since we cut a Hone arm in half!), so these could be filed away with no ill effect.

It also has be noted, that if the same design in the case of the Hone is recommended for light freeriding, than the LX must be overbuilt for XC! So even reducing the wall-thickness of the arms by 1 mm should not cause a failure for a 72 kg rider on XC terrain...

 

[macsi]

5. The next most important step to reduce the Q-factor was to cut the aluminum axle 10 mm shorter. I measured the spline lenght of the XTR and my LX still has a longer splined section. Actually most of the power is not transfered through the splines, but the tapered part of the axle above the splines.

 

[macsi]

6. Since again the axle of the Hone and the LX is the same, the LX must be overbuilt for its purpose. I had the inside of the axle drilled from 14 to 15 mm. THe axle had been heat-treated from the outside surface, so the inside was softer and hence structrally less important. I had the end of the axle shaped so it looked nicer than before.

 

[macsi]

7. As all the spacers are now eliminated, the cups holding the bearings were to long, so I had it shortened to 10 mm, which should be enough for my weight and riding style. THe cranks will go into a frame which has perfectly machined, strong BB shell, so extra material here is not needed. Another alternative would be to get Dura Ace bearings, but machining around here costs next to nothing...

 

[macsi]

8. Again, if you have problems with water entering the BB shell from the seat tube, you need the plastic protecting shell, but my new frame is completely sealed from the top, so I simply ommited this from the system. Alternatively you can use the shell provided for the Ultegra or Dura Ace bearings, since they are exactly 7,5 mm narrower than the LX/XT/XTR models.

 

[macsi]

9. From the following spreadsheet you can see how much weight was saved. If you use cheaper chainwheels, you will probaly add another 20 gr, but you can file them as well to get almost as low as the FRM ones.


The reasult: 15 mm reduction of Q-factor, which is now 163 mm. The crankset is 12 gr lighter than an XTR and cost 85 Euros (this is how much you can get the crankset with BB on the internet +20 Euros for the machining plust two beers for the polishing, which you might want to try yourself...

-b

 

[mosquitos]

very nice work
where do you live to have such low machinig price (eastern europe???)

 

[- Jeremy -]

Wow! That's an awful lot of work, but it's quite impressive in the end!

- Jeremy -

 

[nino]

AWESOME POST!

that's a nice work - thanks for sharing the info!

 

[Mr.P]

Quality post! Thanks!

The cranks look sweet with the polishing too! It seems to upgrade the bling factor beyond "LX level".

Mr. P

 

[heatstroke]

Can I order a set ???...

You should set this up, churn out 5 pairs and I am sure you will have a market for it. Dont bother with new FRM chain rings, just get you machinist a crate of beers, and do those crank arm mods and sell them off.

I'll give you $100 for the set ( + shipping)

9. From the following spreadsheet you can see how much weight was saved. If you use cheaper chainwheels, you will probaly add another 20 gr, but you can file them as well to get almost as low as the FRM ones.

The reasult: 15 mm reduction of Q-factor, which is now 163 mm. The crankset is 12 gr lighter than an XTR and cost 85 Euros (this is how much you can get the crankset with BB on the internet +20 Euros for the machining plust two beers for the polishing, which you might want to try yourself...

-b

 

[bike_freak]

Great post!

Love the pictures!

 

[smudge]

Fantastic post macsi.

It seems that you're intimate with Shimano cranks. Can you divulge how the spider is attached? I'm working on a project for my trials bike and it seems that the spider is pressed on and then some material is "smooshed" (sort of like a rivet) to keep the spider from wiggling off under stress. I'd love to hear that it's easier than that and all I'll have to do is spend a few minutes with a lathe to cut off what appears to be a lockring.

 

[macsi]

Fantastic post macsi.

It seems that you're intimate with Shimano cranks. Can you divulge how the spider is attached? I'm working on a project for my trials bike and it seems that the spider is pressed on and then some material is "smooshed" (sort of like a rivet) to keep the spider from wiggling off under stress. I'd love to hear that it's easier than that and all I'll have to do is spend a few minutes with a lathe to cut off what appears to be a lockring.

Hollowtech I cranks had removable spiders, Hollowtech II has pressed ones. I saw a good post once on this forum of how someone made a trial-specific Hollowtech I. Deore crankset...

I have the pics but not the link...

-b

 

[WOY]

Nice work! I have a one question though. From my understanding you have removed all 3 of the BB spacers right? I think by removing all three spacers you would end up with a crankset that is slightly offset to the NDS (by 2.5mm).

I have done something similar to 2 sets of my 760 XT cranks. Have a look at this post.

M760 XT

However I have forgotten to mention a couple of things I have done

1, I retained the BB shell between the cups however I had to remove the rubber O-ring on one side, trimmed ~3mm off that end and shaved off the shoulders next to the O-rings so that the shell will fit inside the reduced space between the cups. I used a bit of greased to seal the interface between the shell and BB cup.

2, About 6mm of the main shaft was removed.

One note, I don't think one should consider running M760 XT cranks with no spacers at all. The diameter of the shaft is larger at where the shaft meets the BB cups (when install as per instruction). If all three of the spacers are removed the about half of the NDS BB cup would fall into the shaft where the diameter is smaller and sooner or late you wil wear through that plastic sleeve shielding the bearing.

 

[macsi]

Nice work! I have a one question though. From my understanding you have removed all 3 of the BB spacers right? I think by removing all three spacers you would end up with a crankset that is slightly offset to the NDS (by 2.5mm).

I forgot to mention - but you can see from the pics - that 2,5 mm of the left arm interface contact area has been removed, so the arms are perfecly symmetrical...

Yes, you're right, the XT is a different story: that's why I chose to do the project on the LX!

-b

 

[smudge]

Hollowtech I cranks had removable spiders, Hollowtech II has pressed ones. I saw a good post once on this forum of how someone made a trial-specific Hollowtech I. Deore crankset...

I have the pics but not the link...

-b

I've seen that too and that was a great project, but I'm working on the HollowTech II 760 cranks so it's not exactly pertinant to what I'm trying to do. Plus, I don't plan on using a threaded FW.

I was hoping that you might have some insider manufacturing knowledge that would confirm the production method I have assumed and might help me remove the pressed on spider without having to take off too much material from the crank.

I know it's pressed onto the crank arm, but beyond that it's being held in place by another method...either a third seperate piece, or material from the crank arm which has been "riveted" over the spider and then machined down. You can see the machining marks (they're faint) in my picture as well as evidence of what appears to be some kind of broaching tool (it's keyed).

If I can't figure it out and remove it delicately, I'll just have to turn it down on a lathe. My personaly problem is finding someone who has one big enough to clear the rotating crank arm.

 

[macsi]

I've seen that too and that was a great project, but I'm working on the HollowTech II 760 cranks so it's not exactly pertinant to what I'm trying to do. Plus, I don't plan on using a threaded FW.

I was hoping that you might have some insider manufacturing knowledge that would confirm the production method I have assumed and might help me remove the pressed on spider without having to take off too much material from the crank.

I know it's pressed onto the crank arm, but beyond that it's being held in place by another method...either a third seperate piece, or material from the crank arm which has been "riveted" over the spider and then machined down. You can see the machining marks (they're faint) in my picture as well as evidence of what appears to be some kind of broaching tool (it's keyed).

If I can't figure it out and remove it delicately, I'll just have to turn it down on a lathe. My personaly problem is finding someone who has one big enough to clear the rotating crank arm.

Sorry, we only cut a section out of the Hone's arm, so I have no idea, of how Shimano attached the spider. If the cut-apart Hone is still available, I can make another "slice", but I have to check...

Since you're interested in trials modifications, here is a crude one for front freewheeling arrangement. It works - at least the owner was second with it at the 2004 national championship!

-b