It's interesting that this project is benefiting from Pathless Less Pedaled chat. I've been a little too verbose in commenting on the complaining videos (buying and discovering a new incompatibility. My point has been look at the products along the lines of design intentions - your project will benefit this I hope. I've suggested trying to understand range and limits from the actual parallogram design layouts. Ex. Old 7 sperd derailleurs could be pushed to cover 8 but not 9 speed gear ranges. Just trying to understand these physical limitations and appreciating what parallelogram design configurations had to emerge in order to cover 10 then 11 speed cassettes can shift the mindset from complaint to appreciating the realities of newer products. I'll keep watching. Currently Cues (and Linkglide) is an unknown to me. It has been interesting watching Shimano try to keep legacy pull ratios in their product line like RD-3100 9sp. Or making a 10speed system that was upgradeable to 11sp - RD 4700. Flow charts certainly help with confusion but you are going deeper into the understsnding of product designs. Bravo, I hope Russ is watching and Bikesauce too. Sorry - too wordy again.
Pretty much all of the shifting happens from the ramps on the cogs. Guide pulleys have a little bit of slop in the bushings in my experience, I think it is so that the cog can shift the chain smoothly and the g pulley will follow it into place as the chain climbs towards its final position over the cog. I think if they made the der travel bang on, it would interfere with the smoothness of the shifts, because it would be pushing the chain to the center of the next cog before the ramps have time to guide it there with a smooth power transfer over the ramps. So by designing their derailleur movement on the low side, I think they help make sure that a little bit of wiggle room appears for the g pulley to loosen up as you start ramping up into those larger cogs. Just a thought
Idk though, actually the distances we're talking about here, on the order of half a mm, there is probably that much wiggle rooms at least just from the loose chain mesh over the pulley. On second thought, I wonder if you're just observing the losses in the cable winding through the shifters, flex in the components, friction etc that would inevitable cause a loss of efficiency of derailleur movement compared to what's predicted by the design spec. 1-4% seems totall within the error for that kind of thing?
Here's what you do ; Ditch the index, ditch the wireless, the electronic, the whole fancy complicated controls. I just finished an 1800 km tour, fully loaded, on northern Ontario hills and challenging highway. My gears are friction. Downtube mounted. From 1979. Japanese. I use one derailleur, the rear, a pre- index Shimano Altus. Works perfectly. You tune in the gears just like you tune the strings of a guitar. You always have complete control and I wouldn't ever consider anything more complicated. My shifter has about five parts. No springs, notches, batteries, wires, charger... Riding a bike doesn't have to be so over thought and over designed. That is, if you can manage the skill that it takes to do something simple, and do it well.
If it works for you, great. I used contrafriction shifters (Simplex, the best made) for decades. With the change in chain and cog tooth design over those decades, friction shifting no longer worked the way it used to. I got tired of grinding my shifts, and when the hard-to-find spring in the right side Simplex shifter failed, I begrudgingly moved to a Shimano SIS six plus one, shifter. The option is to use all friction, or 6 indexed, the seventh being friction. I like getting the gears right almost every time, I use the indices. For some odd reason, grinding one's chain across the cogs is not good form. Edit to add: I still use the friction shift for my 3X at front. Stronglight 99s for the record, chainwheels still readily available from SPA Cycles in the UK. I'm in Toronto.
@@stephensaines7100 I just re-read your text there. Lots going on! I am lucky to chance upon old great vintage bikes in new condition. They're like finding gold ingots, maybe with a bit of gold filed off, but still gold. I'm a huge fan of 70's -80's Japanese- everything in bikes. I've had my simplex rear changer blow apart miles from home. I never loved the simplex except for how it looked. The way I see it, a philosophy, any philosophy or aesthetic, needs to survive being stripped down to it's essentials. So, like the simplicity of chain drive, friction shifting works in pure physics with no added assistance. I did work in a few bike stores and I saw the new stuff just when it came out. Some developments were good. I thought that SRAM gripshift, indexed, was the be all and end all for gearshifting. I have a Cannondale recumbent out in the shed from 2003 that I bought new at the store, and it's gripshift still works fine. Just the rubber handle grip wore off with sweat and sun and sand and rain, going clear across Canada and lots of other trips. Cog teeth, chain design, yes, those things may have changed, but still for me the objective is to get down to the most simple lay- out. I was years in toe clips and straps, racing and touring, then got the clipless, for decades, now I'm on rigid soled cycling shoes on a platform BMX style pedal and that's it. Most of my bikes are single chainring bikes. I think seven gears on back is enough. I like rim brakes. Cables out, where I can maintain them. I think all the modern stuff is messing with people's minds. Sometimes tradition is there for us so we don't have to confront trendy fads and other baloney. Cheers mate Stay safe
Excellent work! Is the problem here not that the shifter pulls 2% less cable than expected, but more in the way the "average" pull ratio is calculated? Does an average of a non linear curve really make sense? Don't you need to do something like integrate the pull ratio curve equation, to get the total mech movement for the range of cable pull you're interested in (or just measure it)? Or is that how you did it? Maybe it amounts to the same thing.. and I'm probably misunderstanding everything!
I think you're understanding it correctly. I'm not doing any integration. It's a bit tricky, but I'm figuring out the amount of cable pull needed to get you to the second biggest cog and subtracting the amount of pull to get to the second smallest cog. I then take that number and divide it by the distance between those two cogs on the cassette. You're right that this doesn't account for the nonlinear motion of the derailleur, but it should work fine since shifters compensate for that nonlinear motion too. I do have the actual curves and shift points, so it's straight-forward to figure out the jockey position for each shift point. When I plug in the numbers I get similar results. I'm starting work on a website to try to do these calculations and show you what it looks like. I want to make a video when I get the website to a presentable state.
@@boothsbikeprojects Ah I see. Will be very interested to see your website when it's live! I've also been working on a site to compile a list of pull ratios etc. Do you have some ideas as to why shifters pull slightly too little cable? I was wondering if it could be to compensate for very slightly bent mech hangers, or worn mech linkages as the mech ages?
Thank you! I finally understand what the pull ratio means and its real-world limitations. I'm guessing that the reason for moving the jockey wheel slightly less than the cog pitch is because it reduces the lateral kink in the chain as it turns around the jockey wheel. Is that your understanding as well?
I'm still trying to understand why they're designed this way. I've got a number of ideas, but I haven't put it all together yet. I'd like to look at some older 9 speed groups to see if they work the same way.
Great stuff!
It's interesting that this project is benefiting from Pathless Less Pedaled chat. I've been a little too verbose in commenting on the complaining videos (buying and discovering a new incompatibility. My point has been look at the products along the lines of design intentions - your project will benefit this I hope. I've suggested trying to understand range and limits from the actual parallogram design layouts. Ex. Old 7 sperd derailleurs could be pushed to cover 8 but not 9 speed gear ranges. Just trying to understand these physical limitations and appreciating what parallelogram design configurations had to emerge in order to cover 10 then 11 speed cassettes can shift the mindset from complaint to appreciating the realities of newer products.
I'll keep watching.
Currently Cues (and Linkglide) is an unknown to me. It has been interesting watching Shimano try to keep legacy pull ratios in their product line like RD-3100 9sp. Or making a 10speed system that was upgradeable to 11sp - RD 4700.
Flow charts certainly help with confusion but you are going deeper into the understsnding of product designs.
Bravo, I hope Russ is watching and Bikesauce too.
Sorry - too wordy again.
This channel is pure gold. Thank you!
Pretty much all of the shifting happens from the ramps on the cogs. Guide pulleys have a little bit of slop in the bushings in my experience, I think it is so that the cog can shift the chain smoothly and the g pulley will follow it into place as the chain climbs towards its final position over the cog. I think if they made the der travel bang on, it would interfere with the smoothness of the shifts, because it would be pushing the chain to the center of the next cog before the ramps have time to guide it there with a smooth power transfer over the ramps. So by designing their derailleur movement on the low side, I think they help make sure that a little bit of wiggle room appears for the g pulley to loosen up as you start ramping up into those larger cogs. Just a thought
Idk though, actually the distances we're talking about here, on the order of half a mm, there is probably that much wiggle rooms at least just from the loose chain mesh over the pulley.
On second thought, I wonder if you're just observing the losses in the cable winding through the shifters, flex in the components, friction etc that would inevitable cause a loss of efficiency of derailleur movement compared to what's predicted by the design spec. 1-4% seems totall within the error for that kind of thing?
Here's what you do ;
Ditch the index, ditch the wireless, the electronic, the whole fancy complicated controls.
I just finished an 1800 km tour, fully loaded, on northern Ontario hills and challenging highway. My gears are friction. Downtube mounted. From 1979. Japanese. I use one derailleur, the rear, a pre- index Shimano Altus. Works perfectly.
You tune in the gears just like you tune the strings of a guitar. You always have complete control and I wouldn't ever consider anything more complicated. My shifter has about five parts. No springs, notches, batteries, wires, charger...
Riding a bike doesn't have to be so over thought and over designed. That is, if you can manage the skill that it takes to do something simple, and do it well.
If it works for you, great. I used contrafriction shifters (Simplex, the best made) for decades. With the change in chain and cog tooth design over those decades, friction shifting no longer worked the way it used to. I got tired of grinding my shifts, and when the hard-to-find spring in the right side Simplex shifter failed, I begrudgingly moved to a Shimano SIS six plus one, shifter. The option is to use all friction, or 6 indexed, the seventh being friction.
I like getting the gears right almost every time, I use the indices. For some odd reason, grinding one's chain across the cogs is not good form.
Edit to add: I still use the friction shift for my 3X at front. Stronglight 99s for the record, chainwheels still readily available from SPA Cycles in the UK. I'm in Toronto.
@@stephensaines7100 I've probably seen you. Ridden Toronto roads since 1970!
@@stephensaines7100 I just re-read your text there. Lots going on! I am lucky to chance upon old great vintage bikes in new condition. They're like finding gold ingots, maybe with a bit of gold filed off, but still gold. I'm a huge fan of 70's -80's Japanese- everything in bikes.
I've had my simplex rear changer blow apart miles from home. I never loved the simplex except for how it looked.
The way I see it, a philosophy, any philosophy or aesthetic, needs to survive being stripped down to it's essentials. So, like the simplicity of chain drive, friction shifting works in pure physics with no added assistance.
I did work in a few bike stores and I saw the new stuff just when it came out. Some developments were good. I thought that SRAM gripshift, indexed, was the be all and end all for gearshifting. I have a Cannondale recumbent out in the shed from 2003 that I bought new at the store, and it's gripshift still works fine. Just the rubber handle grip wore off with sweat and sun and sand and rain, going clear across Canada and lots of other trips.
Cog teeth, chain design, yes, those things may have changed, but still for me the objective is to get down to the most simple lay- out. I was years in toe clips and straps, racing and touring, then got the clipless, for decades, now I'm on rigid soled cycling shoes on a platform BMX style pedal and that's it.
Most of my bikes are single chainring bikes. I think seven gears on back is enough. I like rim brakes. Cables out, where I can maintain them. I think all the modern stuff is messing with people's minds. Sometimes tradition is there for us so we don't have to confront trendy fads and other baloney. Cheers mate
Stay safe
@@davidrobinson9507 Malcolm at Biseagal claims he has replacement parts for Simplex shifters. When they work, they are the best.
@@stephensaines7100 go for it. Long live the retro age !
Glad I found this channel👍
Excellent work! Is the problem here not that the shifter pulls 2% less cable than expected, but more in the way the "average" pull ratio is calculated? Does an average of a non linear curve really make sense? Don't you need to do something like integrate the pull ratio curve equation, to get the total mech movement for the range of cable pull you're interested in (or just measure it)? Or is that how you did it? Maybe it amounts to the same thing.. and I'm probably misunderstanding everything!
I think you're understanding it correctly. I'm not doing any integration. It's a bit tricky, but I'm figuring out the amount of cable pull needed to get you to the second biggest cog and subtracting the amount of pull to get to the second smallest cog. I then take that number and divide it by the distance between those two cogs on the cassette. You're right that this doesn't account for the nonlinear motion of the derailleur, but it should work fine since shifters compensate for that nonlinear motion too.
I do have the actual curves and shift points, so it's straight-forward to figure out the jockey position for each shift point. When I plug in the numbers I get similar results.
I'm starting work on a website to try to do these calculations and show you what it looks like. I want to make a video when I get the website to a presentable state.
@@boothsbikeprojects Ah I see. Will be very interested to see your website when it's live! I've also been working on a site to compile a list of pull ratios etc. Do you have some ideas as to why shifters pull slightly too little cable? I was wondering if it could be to compensate for very slightly bent mech hangers, or worn mech linkages as the mech ages?
Thank you! I finally understand what the pull ratio means and its real-world limitations. I'm guessing that the reason for moving the jockey wheel slightly less than the cog pitch is because it reduces the lateral kink in the chain as it turns around the jockey wheel. Is that your understanding as well?
I'm still trying to understand why they're designed this way. I've got a number of ideas, but I haven't put it all together yet. I'd like to look at some older 9 speed groups to see if they work the same way.