Loading…
Brother wrote:RX800 and RX812, with a max 34t sprocket for the former and minimum 40t sprocket on the latter.
pkram wrote:The chain wrap is different from the max sprocket supported. It relates to cross chaining: if you would ride 34Tx11T (and a have chain length that works with 50Tx36T) there is a lot slack on the chain that the rear mech needs to pick up. Again, most likely not an issue in real usage (especially if you avoid cross chaining). But make sure you cut the chain with this in mind though: maybe you can optimise the chain length.
Another thing on my mind: you plan to buy a left crank arm with power meter? or you will send in your own left crank arm? I ask because you might be able to use the 105 crank arm with power meter on a GRX crankset. In that case you can keep to the same chain ring sizes (46-30, or 48-31), in turn also the crank arm length (if you would prefer) and use your current cassette.
Lastly: make sure you lower you saddle slightly with the new crank arm length as the pedal will be lower on the down stroke.
Just sharing all my thoughts, good luck!
The problem with the RX812 is that it uses an offset upper pulley. While Shimano did move to this design even on 2x rear derailleurs in the MTB world, i'm not sure how it would work in a 2x setup.
IMO the 36T is a no-man's land because Shimano doesn't have an 11spd 11-36... When they come up with their specs, they most likely only consider their own cassettes... The 10spd GRX uses an 11-36, because Shimano does have (had) 10spd 11-36 cassettes.
OP, take a look at the 10spd GRX RD.
https://bike.shimano.com/en-AU/product/component/grx-10-speed/RD-RX400.html
It's just an old MTB RD from the 10spd days with a new logo. It uses the 11spd pull ratio like Tiagra 4700.
It's cheap, it has a clutch, 41T capacity, 36T max cog.
But really, plenty of people have exceeded the factory specs on a lot of RDs.

HollyBoni wrote:The problem with the RX812 is that it uses an offset upper pulley. While Shimano did move to this design even on 2x rear derailleurs in the MTB world, i'm not sure how it would work in a 2x setup.
IMO the 36T is a no-man's land because Shimano doesn't have an 11spd 11-36... When they come up with their specs, they most likely only consider their own cassettes... The 10spd GRX uses an 11-36, because Shimano does have (had) 10spd 11-36 cassettes.![]()
OP, take a look at the 10spd GRX RD.
https://bike.shimano.com/en-AU/product/component/grx-10-speed/RD-RX400.html
It's just an old MTB RD from the 10spd days with a new logo. It uses the 11spd pull ratio like Tiagra 4700.It's cheap, it has a clutch, 41T capacity, 36T max cog.
But really, plenty of people have exceeded the factory specs on a lot of RDs.
dodge77 wrote:
Would the Q-factor be the same on the 105 power meter left crank as the GRX? or is q factor set by BB? As in I wouldn't want my left foot to be closer inboard than my right etc. bar that its just hollowtech II 105 power, on hollowtech II grx crank i guess?
pkram wrote:Also...
https://store.stagescycling.com/Stages-Power-L-Shimano-GRX-POWER-METER-Factory-Install
Cmtgravel wrote:Easy solution: keep the gearing, lose the power meter.
dodge77 wrote:
I sorely miss my power meter
dodge77 wrote:
I'm getting the stages shipped to a US address then shipped across to the UK, so don't want to have to post a new crank arm to them, but from what you've said i'm hopeful about the 105/grx mix 😃
pkram wrote:
i made a comparison and posted some pictures below. The GRX left crank arm has a different angle outward compared to the 105 arm. So that’s how the Q factor is achieved on the left side. Which means you can’t solve it with a spacer between the arm and BB...
it does mount on the spindle as per usual, so you could ride it but then the Q factor would be different on the left side compared the right side.
I am sorry if I got you too excited by giving an invalid suggestion...
-----
The following photos of the crank arms show the 2.5mm difference in outward position, while both cranks were installed flush against the BB. So the difference in Q-factor is simply achieved by having a more angled crank arm. (ignore the absolute values on the ruler and just compare the difference)
GRX:![]()
105:![]()
