As an engineer I really like your videos. If you find a combination of hydraulic brifters and clutch rear derailleur able to run 10 or 11s LG cassettes would be very nice. People that are using them for MTB are claiming how good they shift under load, so it would probably be good for gravel too. By the way, just for future records, may you want to differ the HG cassete to the new LG ones cause they probably will become common in entry level bikes very soon.
Still greatly appreciate your work here. I'm curious - in the spirit of reverse engineering, would mapping derailleur pivots and body dimensions (diagram of parallelogram) provide any worthwhile info to the database you are building?
It would be incredible for someone to try and precisely measure the parallelogram. I could then benchmark my empirical measurements against a more analytical method. I didn't go that route because it requires more precision than I have access to. It wouldn't be so hard for 1x derailleurs since the parallelogram only moves horizontally. You'll also need to figure the position of the cable stop with a good amount of accuracy. The cable stop can't be at a right angle to the lever arm throughout the range of motion, which causes the effective pull ratio to be lower than simply measuring the parallelogram and the lever arm. 2x derailleurs are trickier. They require that you measure the "slant" angle in the slant-parallelogram design. I remember trying to do some measurements on a GX derailleur and found that small amounts of error in that angle can greatly change the resulting pull ratio calculation. Additionally, derailleurs end up having all kinds of quirks. For example, SRAM derailleurs have that "shark fin" on the lever arm, which would require modeling itself.
@@boothsbikeprojects Thanks for answering. As I thought you could and have added to my understanding of why it would be difficult. It would be interesting to do a kind of historical thread of derailleur designs. Even as this thought emerged, I realize that would be daunting. It was 60 years ago when the slant parallelogram was introduced. The fact that even the cable stop location introduces changes are beyond what I'd thought of. That and the slant's effect might explain Suntour's indexing failures 35 years ago. The SRAM "shark fin" that ramps-up cable pull is something Ratio Technology Co seems to grasp as well as the mechanism that indexes inside the shifter. I've always guessed that their being based in Germany (where the SRAM engineers are) might have given them some insider info. Again, all these dynamics beyond what I'd thought of - I appreciate your work as is.
You are amazing, thanky you so much for your work
Great video!, thanks for doing this :)
As an engineer I really like your videos. If you find a combination of hydraulic brifters and clutch rear derailleur able to run 10 or 11s LG cassettes would be very nice. People that are using them for MTB are claiming how good they shift under load, so it would probably be good for gravel too.
By the way, just for future records, may you want to differ the HG cassete to the new LG ones cause they probably will become common in entry level bikes very soon.
Still greatly appreciate your work here.
I'm curious - in the spirit of reverse engineering, would mapping derailleur pivots and body dimensions (diagram of parallelogram) provide any worthwhile info to the database you are building?
It would be incredible for someone to try and precisely measure the parallelogram. I could then benchmark my empirical measurements against a more analytical method.
I didn't go that route because it requires more precision than I have access to. It wouldn't be so hard for 1x derailleurs since the parallelogram only moves horizontally. You'll also need to figure the position of the cable stop with a good amount of accuracy. The cable stop can't be at a right angle to the lever arm throughout the range of motion, which causes the effective pull ratio to be lower than simply measuring the parallelogram and the lever arm.
2x derailleurs are trickier. They require that you measure the "slant" angle in the slant-parallelogram design. I remember trying to do some measurements on a GX derailleur and found that small amounts of error in that angle can greatly change the resulting pull ratio calculation.
Additionally, derailleurs end up having all kinds of quirks. For example, SRAM derailleurs have that "shark fin" on the lever arm, which would require modeling itself.
@@boothsbikeprojects Thanks for answering. As I thought you could and have added to my understanding of why it would be difficult.
It would be interesting to do a kind of historical thread of derailleur designs. Even as this thought emerged, I realize that would be daunting. It was 60 years ago when the slant parallelogram was introduced.
The fact that even the cable stop location introduces changes are beyond what I'd thought of. That and the slant's effect might explain Suntour's indexing failures 35 years ago.
The SRAM "shark fin" that ramps-up cable pull is something Ratio Technology Co seems to grasp as well as the mechanism that indexes inside the shifter. I've always guessed that their being based in Germany (where the SRAM engineers are) might have given them some insider info.
Again, all these dynamics beyond what I'd thought of - I appreciate your work as is.
it's noteworthy that 2.50 cable pull should be the same as shimano 9 shifters (road-mtb, good old times when they were the same).