HiYoSilver

Well, 16 may be overkill. I was just working the numbers and not concerned about replacing gears.

I think others interested in tightly spaced gears are A- those who participate in any road or tri races, B- those who are regularly faced with winds, C- those who are getting older but don't want to slow down as much, and D- those who
face a variety of hill grades in their riding.

My thinking is that the rider would only shift from low-medium-high chainring to get in the range of gears
for current conditions and the rest of the shifting would be just progressing up and down their cassette. If don't want
as close gears but good range, they could shift to a 12-25.

HiYoSilver

I was wondering how much change in GI % is acceptable in lower gears. I have noticed that the larger jumps in lower gears does not generate a momentum hit like it does in the middle and high ranges.

Agreed the 9->10, 10->9, 18->19, 19->18 would require double shifts. Ideally it would better to have a single shift solution, but there is not such a system widely in use today.




Agreed about the derailer current limits, which is why started by asking for a re-engineering effort. I'm not a mechanical engineer and so don't understand the physical limitations. And yes, I ignored "cross chaining" rule because it is based on current engineering solutions. I haven't seen any creative solution to the common rider gearing experiences in the last 40 years. Thus I credit this to lack of vision instead of to lack of engineering ability.





I did look at that, but it seems to have problems with 1) weight, 2) transfer efficiency, and 3)ease to understand how to best use it.

fietsbob

Math help to understand it , : https://sheldonbrown.com/gears/internal.html

tatfiend 

The limit on small chainring size is the chain line from the top of the inner ring to the cassette. A front derailleur rear cage crosspiece needs to clear the chain stay which with normal frame and derailleur construction it would not do if trying to have a 53 tooth outer ring and 16 tooth inner. Also shifting onto the middle ring from the inner gets very bad if a greater than 12 to 14 tooth difference between middle and outer rings is tried due to the front derailleur being too far away from the middle ring to properly control the chain. This is discussed in the book I referenced in my prior post.

Here is a discussion of Rohloff efficiency printed in the IHPVA Journal.

https://www.ihpva.org/HParchive/PDF/hp55/hp55p11-15.pdf

At high power levels they claim similar efficiency to derailleur drivetrains which, depending on gear selected, are not as efficient as frequently claimed.

https://www.ihpva.org/HParchive/PDF/hp52-2001.pdf

As for weight the Rohloff was compared some years ago with a XTR group and the difference in weight between them was about 200 grams. Remember with the installation of an IGH you are losing the weight of two derailleurs, rear derailleur hub and cassette, two of three chain rings and one shifter. The current XTR group has been further lightened but I would still expect the difference to be only about a pound.

I am not a racer so do not worry much about drivetrain efficiency, about which there is relatively little published data any way. I figure a few % lower efficiency just means a better workout while riding.

The 100+ page Rohloff manual, available for download in English from their web site, discusses the various hub models and frame requirements for installation.

Derailleur drivetrains without duplicate gears, or with minimum duplicates, were designed in the days of 4 and 5 speed freewheels. They were advantageous then but with 20+ speeds (double) or 30+ (triple) now available I do not see the necessity now.

HiYoSilver

Thanks for the updates. I haven't had time to review the rohloff plan. I missed the basic explanation that it replaces both derailers and couple of chainrings. It would be have better if Rohloff had an image: "not this" but "this". The 13% may be a tad high, and I seem to vaguely recall an issue with span of total gear range.

fietsbob

13% in real life is not much, and as its 13% the next gear the difference
is smaller as the gearing goes down.
OTOH with the chain tensioner , you can get a second chain-ring, double crank, for about a 6 or 7% difference.
to half step the hub .. 3~4 tooth difference .. in the top 7,
as said the bottom 7 are already getting closer.

hub cog will take any 8 speed 3/32" chain.

I've got a 17" low and a 90" high and find it a fine range .. 38/16x26"wheel

recent acquisition a BikeFriday Pocket Llama its Geared 53/16x20"wheel

combine it with a Overdrive Schlumpf High speed drive crankset ,
and you could wheel suck behind mopeds.

HiYoSilver

I'm not familar with chain tensioners. In real life 13% may be too high for the upper gears. I currently run 6-8% for top 6 gears and 11 for next two lowest. The 14% in low doesn't seem to be a problem. But if I increase the % change in the uppers then there is an issue.

HiYoSilver

Still haven't had time to review Rohloff range. I can't find data on the GI for the different gears. The supplied link to the pdf did provide the efficiency data.

90.8% to 89.8% rohloff
94.6% to 93.3% shimano

That's about a 4% hit. With wider gearing of 13, I'ld never make that up by keeping pace level.

It was the lost of efficiency that led me to discard this engineering solution last time.

HiYoSilver

Ok, reviewed Rohloff with brain in gear. This is not a viable solution. It costs $1,200, approximately. It's range is a mtb range, 19 to 98GI. This does not work for road riding. It would mean giving up 3 top end gears.

BarracksSi

How fast do you really want to go, then? Is 35-40 mph too slow?

Booger1

On the internal hubs,why can't you raise the gearing?You don't need 19 inches on a road bike. They have a torque limit,but they don't have a speed limit.

fietsbob

And they make 13 15,16,17 t cogs too

rudypyatt

Interesting thread. HiYoSilver, I believe that the solution to your problem already exists: The hybrid IGH-cassette drivetrain. In my view, it's being overlooked because (1) the components involved aren't Shimano, Campy or SRAM (though SRAM does make at least one hub that allows it; to my knowledge, the Alfine and Nexus hubs don't have this feature); (2) because the system hasn't been used in racing; (3) fears about excess weight and mechanical complexity; (4) the sheer novelty of the thing puts people off; and (5), I think only Brompton has taken the plunge and put the system onto a production model.

Sturmey Archer makes at least two different internal gear (3-speed and 5-speed) hubs that have splines for adding 8- or 9-speed cassettes to the drive side of the hub:

https://www.sturmey-archer.com/products/hubs/cid/4/id/64
and
https://www.sturmey-archer.com/products/hubs/cid/8/id/58

As I understand it, a hybrid drivetrain essentially uses the IGH as to provide additional "rings", with the splined cassette serving its normal function. Thus, you could have 45 speeds if you combine the 5-speed IGH with a 9-speed cassette: Effectively, you get the equivalent of 5 chain rings for a 9-speed cassette. Of course, you'd have to play with the actual chain ring size to prevent the highest gears in each range from becoming too high to be usable, but this presents no real obstacle to implementing the system.

I've been messing around with the gearing calculator on the Sheldon Brown site, experimenting with the possibilities (some of which I've printed out, notwithstanding the warning on the page that the print function doesn't work. It does, and you can print a screen capture anyway). The program lets you run calculations on various derailleur (single, double, triple ring) and cassette drivetrain combinations. It also lets you experiment with gearing combinations for igh drivetrains and for hybrid igh/derailleur set ups:

https://sheldonbrown.com/gears/

Example: For the SA 3-speed, using a 39 tooth ring (the inner ring on a Trek Madone) and a 13/14/15/16/17/18/19/21/24 cassette, as the hub steps from Low (.75 of direct drive) to Direct (1.0) to High (1.33 over direct) (call it Low, Mid and High range), you'll get the following, in gear inches, rounded to the nearest whole number:

Low: 58/54/51/47/45/42/40/36/32
Mid: 78/72/67/62/59/56/53/48/42
High: 104/96/90/84/79/75/71/64/56,

with obvious possibilities for changing these values by using different rings or cassettes. Also note that CS-RF3 hub is compatible with an ordinary mountain bike left hand shift set-up as used for triple chainrings. I believe that a standard rear derailleur and shifter would operate the cassette.

As I recall, Sixty-Fiver has (had?) a bike with this hybrid set-up. He can give us an in-fact-not-theory evaluation of how it performs.

TurbineBlade

This thread sucks -- I'm trying to read why most cyclists are "bored" compared to pros. Maybe it's because they have to read internet threads as long as the Old Testament.

BlazingPedals

I got bored with the rant and stopped reading. I've been saying for years that compacts are for cyclists who need lower-than-standard gears but are too vain to use a triple. That sounds like you. If you want more low gears without giving up your high ones, get the triple. For hills, I run a 30/39/53 along with an 11-32.

chasm54

wut? Why not change the sprocket? Then you can have as high a top end as you like.

Rohloffs are pricey, it is true. They are very wonderful, though.

Monster Pete

A hub is just a ratio-changer. You can easily raise or lower the overall range by changing the chainring/sprocket ratio. There's a low-gear limit due to the problem of over-torqueing the hub, but no upper limit. You could run a 13t sprocket with a 53t chainring if you really needed it.

Doug5150

I was agreeing with you up to this point, but you lost me when you began to describe what would be a "better" gear system of an open chain still jumping around on sets of cogs.
You are nitpicking over very tiny faults, in a gear system that even with such 'improvements' would still have a number of glaring shortcomings.

My thoughts would mainly involve three features:
1) the maintenance should be minimal--say, for a casual rider, once every five years? Maybe once every couple years for a frequent rider. That rules out just about any exposed chain setup.
2) the inter-gear spacing should be reasonably small.
3) the (single!) gear shifter control should functionally place all the gears "in order", so you don't need to know which-combination-of-what controls to go to the next higher or next lower gear. (this also assumes that there will be no gear-ratio duplication, too)

Flying Merkel

I can imagine coming to a halt and toppling over on a hill trying to figure out what to shift........

Had a road bike with a 52-42 up front and a 12-28 8 speed cassette. Worked very well around here. Orange County Ca has surprisingly stiff climbs and open highways with a constant tailwind. You will never always have just the right gear. That's what legs are for.

fietsbob

Heck the Pros are part of Marketing the new stuff, the engineers do all the design
and prototyping,
spring classics they run the stuff around Belgium and such, to see if the strong,
Trained racers can break it , in one day of racing,
then summer .. TdF, getting people to think it can make them faster,
if they buy the latest stuff
then that fall the trade-shows take dealers orders for the bike companies
that bolt it on their brands.

rudypyatt

You make a good point about the length of the posts But I wouldn't say that the thread sucks. I find the idea of alternatives to current systems fascinating, and this probably the best sub-forum in which to discuss the possibilities. But I can't see any of the major pro teams making themselves test subjects for such experiments. And there's certainly no percentage in it for Shimano, Campy or SRAM to push something so far removed from conventional methods. As for me, I ride a single speed Trek Cruiser and an old Schwinn 3-speed. Simple is good, even if exotic is interesting.

Alexandria, eh? I grew up right across the Potomac, near what's now National Harbor.

rudypyatt

Then again, this is true, and the pros are already the test subjects; but I doubt the Schleck brothers would even consider racing anything that departs radically from existing systems.

Sixty Fiver

rudy,

The new dual drives that incorporate an IGH with an 8-10 cassette do provide a lot of options, particularly in regard to smaller wheeled bicycles where getting adequate high gearing for performance applications can be a challenge.

With regular sized bicycles a derailleur drive provides the most weight and cost effective option while with bikes like roadsters a simple dual drive can really add some range.

My dual drive set up a simple 2 by 3 to give me a 6 speed with a nicer range but when I want to get out and have serious fun I run a derailleur equipped road bike with closely stepped gears.

Jeff Wills

Y'all realize that HiYoSilver started this thread more than a year ago, doncha?

Artkansas 

Have you sent this to the concerned manufacturers? I'm impressed with your analysis, but we're not going to do anything about it here on BF.