Two Spoke Forums banner

1 - 5 of 5 Posts

·
Completely Human
Joined
·
177 Posts
Discussion Starter · #1 · (Edited)
The Smitty Report

Gear Head…

Thanks to the generosity of a friend I have had the opportunity to ride a Giant “Seek 0” For a few trips and I must say that I like the bike very well, for what it is made to do. It is a 20 to 25 pound bike, aluminum frame, hydraulic disk brakes and a Shimano Alfine 8 speed hub. This gear head Giant is really a nice ride, quick and capable, but also leaving me wanting in some areas, but, going in that direction would be getting into a bike review, which this essay is not.

What I am writing about in this post is about internally geared bicycle transmissions. I have been interested making the jump to an internal hub transmission for many years now, ever since the first Shimano 7 speed hit the market.
Through the years I have ridden many of the interesting internally geared hubs made by Bendix, Sturmey Archer and Shimano. These came in flavors of 2, 3, 4, 5 and 7 speeds. I personally had a kickback Bendix and a few S/A 3 speeds. I always liked the S/A 3 speed as a kid. It was a good selection of ratios arranged in a serviceable spread, especially for a 12 year old shredding 1970’s, middle American sidewalk pavement.

So for many years I have been amenable to geared hubs, and in some ways envious even, especially when it came to maintaining them. I clean and tune my ride a minimum of 3 times a year. This includes a once a year complete tear down and thorough cleaning and greasing of bottom bracket bearings and rear end bearings and truing of rims. All of this led me to start looking into an internally geared hub upgrade. Justifying an upgrade was simple to do and was based solely on simplicity, dependability, durability and the fact that the components are now affordable, well almost.

If you are concerned about “System Efficiency”, as any human power plant would, there are many studies and resources to be found on the web and I will recommend a few sources to use as a good start,

The web log “Hubstripping” is an invaluable resource and that site has volumes of reading and reviews to peruse. I would start with the,

Gear Hub vs. Derailleur
Hubstripping : Gear Hubs vs Derailleur

Two articles from “Human Power” the tech journal of the HPVA,

The mechanical efficiency of bicycle derailleur and
hub-gear transmissions
- Chester Kyle and Frank Berto
http://www.ihpva.org/HParchive/PDF/hp52-2001.pdf

and

Efficiency Measurements of Bicycle Transmissions - a neverending Story?
- Bernhard Rohloff and Peter Greb (translated by Thomas Siemann)
http://www.ihpva.eu/HParchive/PDF/hp55/hp55p11-15.pdf

I will also recommend an article from CORE77.COM,

Reinventing the Wheel : Pushing the limits in high performance bike design

- Steven MacGregor
Core77 - Reinventing the Wheel: Pushing the limits in high performance bike design

And of course the late Sheldon Brown’s site, specifically his page entitled,

“Gear Theory for Bicyclists”
- Sheldon Brown (revised by John Allen)
Gear Theory for Bicyclists.


Know what you got and how you use it!
I decided the place to start would be to fully study my existing setup and improve my knowledge of bicycle gearing schemes. I also needed to understand exactly how I use my current transmission. I also needed to educate myself on the new equipment now available in the market place.

The first for me step was to analyze my current ride and get an understanding how the configuration of my current transmission influences the way I ride or how the way I ride influences the way I use the transmission, if there is a difference.

The arrangement on my daily commuter is a standard Shimano Deore 9 speed free hub cassette with a triple chain ring, In the following tooth counts,



This combination yields the following ratios and range,



I’m an “input vs output” type of guy, so I really like “Gain Ratio” metrology best when thinking about bicycle systems as a machine. If you think about it a second, all we really are concerned with here is,
“What gear configuration gives us the desired pedaling experience”
So all we need to consider is “pedal to wheel” relationships and how any given gear configuration changes that relationship.
Gain ratio does just that by giving us an understanding of what 1 pedal revolution will yield in terms of tire rotation.
This is done by dividing the rear tire radius by the crank length (radius) and then multiplying by any of the gear ratios.
Doing this yields the following “gain ratios” for my set up,



So now for instance in the lowest setting I can pedal one revolution and the tire will spin 1.41 revolutions, and at the highest setting 1 revolution of the crank will yield 8.02 revolutions of the tire.

Of these possibilities there are basically 11 useable gear ranges determined by the physical limits of the chain to reach between sprockets without loosing efficiency and stressing the chain components causing strain that results in excessive wear and or premature failure. Furthermore, I use my bike as a commuter and a tourer and I do not have a need to double shift at all so I avoid it altogether opting to make only two range changes throughout the shift pattern.
The “Logical shift pattern” for my bike is as follows,



There will certainly be some disagreement as to when to make the shifts between ranges. I choose to take them at 3 and 7 not for any reason other than it seems to natural for the way I ride, sometimes I may change ranges earlier or later depending on the situation but for the vast majority of my commuting and touring this is the path used most.
This pattern then yields the following useable range progression,



I have done all of this to demonstrate in quantifiable terms what my pedaling experience is in my daily commutes and this chart pretty much sums it all up. I use my bike like a 7 or 8 speed with 3 or 4 specialized use gears. I must point out that typically I will be in mid range pretty much all the time with occasional jaunts into H7 and H8 when I need to maintain speeds in excess of 15 mph. I rarely use low range unless I am climbing and my commute does not call for that. I rarely use H9 because I rarely exceed 17 miles an hour. For 95% of my riding I use mid range. I can easily maintain 10 to 15 mph with those 5 gears and still have room to go faster or crawl slowly up a hill, which happens only 5% or less of my riding time. This reveals a new more specific chart that shows how I use the bike 95% of the time,



The Shimano gear set is good “mid range” 11 speed cruiser with 5 mid range gears, three “grannies” and three “Tall” gears over a 470% spread. Knowing how I ride it shows me that I typically only use 7 or 8 gears 95% of the time.

The ratios between shifts with the Shimano drive are also serviceable. The average gap between gears in range is 15.1%. The largest step percentages are shifts between ranges at 38% for the “Low to Mid” shift and 33% for the “mid to high” shift.

These charts also give points of reference to understand how any gear hub might perform compared to my current setup and useage. Armed with this information we can now “Clip out” the Shimano Deore 9 speed and triple chain ring and insert any gear ratios we like from any gear box or gear hub. All of the engineering data needed to evaluate the gear hubs is available on the internet, listed primarily at respective manufacturers websites.

Nearly all manufacturers specify a “Minimum input gearing” and it ranges from 1.7:1 up to 2:1. Additionally each hub is only available with limited and specific sprocket tooth arrangements so each test is dependent on those tooth counts to determine the minimum chain ring size. In many cases I calculated multiple comparisons of the same gear hub, given by Sprocket tooth count (all calculations were performed in the ubiquitous MS Excel).

Manufacturers also list “Maximum Torque Input” and these are ranging around 125nm to 250Nm. I know That I can easily produce up to 250Nm at the crank myself so how much you will input is dependent on you’re the force you apply to the pedal and crank length, with the maximum being your entire body weight multiplied by your crank length (Of course recumbent bicycles are a special case here). In any case you will have to look up those requirements yourself and make your own calculations, as this study is a general meta-comparison of gear hubs to my current Deore Cassette and chain ring transmission.


X-RF8 Sturmey-Archer
The first gear hub to compare is the cheapest entry on the market, The venerable Sturmey-Archer 8. I have ridden many an SA in my time. The SA hub is very durable and dependable if employed within manufacturers specifications. That being said I have seen my share of trashed hubs as well and every one of them were destroyed by reckless usage.

The new SA-8 Shows up on this list simply because it is very, very cheap. 8 speed units can be purchased from $150 to $200 from multiple vendors. There are three models, The X-RF8, X-RD8 and X-RK8. The X-RF8 is a brake free model with the other two sporting a 70mm drum brake in X-RD8 and provisions for a disk brake on X-RK8. All of the models now have polished aluminum housings, instead of the 1970’s chrome plated tubing.

According to the manufacturer the unit weighs 1790 grams, which is about 4 pounds.
The unit is available in 32 or 28 spoke, 185mm or 175mm axle lengths and is available in 20, 23 and 25 tooth count sprockets.

In the meta-comparison I use a 2:1 drive setup that yielded the following results,



This analysis is comparable to the Shimano Deore 9 speed in high range,



It is safe to assume that using this hub in place of my current setup would be much like having 3 mid range gain ratios that I know and love so well in my Shimano Deore and 5 “tall” gears to fly with.
At $170 per unit a person might be tempted to drop the tooth count on the chain ring to yield a ratio below 2:1 and risk premature failure. Such risky behavior can yield a lower overall gain ratio per gear. If we then plug in a 36 tooth chain ring to get a 1.44:1 input ratio we can then tune the results to,



This configuration is actually very similar to my current setup and shift path. With this configuration A person could easily be tempted to buy one just to test and play around with.



Shimano Nexus 7 and Alfine 8
All of the Shimano hubs come in standard and non-standard configurations with and without brakes. Shimano also supplies multiple input sprocket configurations and axle lengths. Units are priced from $175 to $400 new and can be found for seriously cheap bucks on line, used and on ebay. I have yet to find a maximum torque input or a minimum ratio input for these two Shimano gear hubs. The data is run with 2:1 input for this example and the spreadsheet yielded the following numbers,



Both of these gear hubs share similar range and step percentages with my most used gears and pattern. I am certain I would like either of these hubs for general commuting and could be serviceable as touring hub as well.

I have in depth experience with both of these hubs. I own a SG-7C18 and employ it on a Pacific steel frame cruiser. The SG-17 is an awesome hub for the money and transformed the Pacific battleship from a single speed hippopotamus into a very useable commuter. Both are great units but the Alfine 8 is a better unit than the Nexus 7 in my opinion. Shift quality of the Alfine 8 S501 is very good especially under power.


Shimano Alfine 11
This hub is the newest entry in the market from Shimano. It boasts advanced design and an increased range. The results were calculated with a 2:1 input ratio.



This chart shows a fantastic range and great spacing and from what I have read it is reasonably priced as well. I have yet to ride this unit but I must admit that it is an excellent range for the way I use my bike and I can’t wait to ride one to know first hand.

Rohloff Speed hub
This hub is the holy grail of the geared hub world. I must admit that it has the range, step percentage, design and reputation that is worth paying for, but let’s face it, the price is simply too much for the average Joe. In fact the price of this hub is more than the cost of most entire bikes and now with the advent of the 11 speed Shimano at less than ½ the price, I think the Germanic tribes will be loosing market share at the high end of the gear hub market…

The numbers generated from a simple 2:1 input speak for themselves,



I am fairly certain I would be comfortable with this hub nearly anywhere on the face of the continental United States.

i-Motion 9
This SRAM / Sachs creation is the current road racers favorite. It is cost effective and proven and has a good following. The numbers were run with a 2:1 input ratio.



Nuvinci CVT
Here is another hub that interests me intensely. I find the idea of a CVT for a bicycle to be a natural progression of comfort and simplicity. The chief advantage of this system is without a doubt the ability to fine tune a gain ratio seamlessly and instantaneously. In other words, no more searching for the right gear only to find your current gear ratios are not tuned to accommodate your needs at that time. Simply change the input ratio on the Nuvinci to find the pedaling effort that matches your desired speed.
The chief disadvantages are weight and efficiency. The hub weighs in at over 8 pounds and, in an absolute travesty of marketing policy, the manufacturer, Fallbrook technologies, refuses to release efficiency data. From my engineering experience I would guess the efficiency numbers to be in the mid to upper 80’s which is an acceptable number range, but not having the data from the manufacturer makes one wonder what exactly Fallbrook is hiding…. The refusal to supply this engineering data also demonstrates a distinct willingness to obfuscate the facts with marketing hype, which Fallbrook has no lack of, and just plain irritates me. Because of this terrible marketing decision I refuse to give the manufacturer my money and will wait to see how long they last in the marketplace, and judging from the overt willingness to hide facts about their unit, I suspect this organization's durability and longevity will be brief.
Regardless of Fallbrook’s failings as a business the technology behind the unit is sound and well engineered, it is the marketing and sales team who are the ones who need replacing. I have not had a chance to ride one of these units but have spoken to many who have and the consensus I gather tells me it may be worth considering.
As you can see, the range at a 2:1 input ratio is very serviceable. In fact if you look closely this unit covers my 95% gear path almost exactly, with the added ability to accommodate ANY gain ratio between 1.89 to 6.63, now you see why the Nuvinci has kept my attention.




So, which one will it be?
Keeping in mind that none of these results have been weighted for individual gear efficiencies per unit, I have come to many conclusions. After making all of these calculations and considering the results I discovered that nearly any of the hubs were capable of delivering something very close to my desired pedaling experience quite easily and some of them are capable of reproducing my current experience almost exactly or even better.
The units that were most suited for my commuting experience were the Shimano Line of gear hubs with the new 11 speed unit being the closest and most desirable among the three.
The Rohloff unit is nearly perfect, especially for Mountain bikes and commuters/tourers but again, priced out of range, almost to the point of absurdity.
The Sturmey Archer 8 speed would function but I am afraid the step percentages and high ranges would not be a good match for my usage.
The I-Motion 9 is worth considering, but comes in limited configurations and has a specialized following. I am also not so certain SRAM support is as sound as the huge Shimano distribution organization can muster, and I say this based on past experience with SRAM.
Fallbrook technology’s Nuvinci CVT is a boon to cycling and comfort, and I want one badly just to play with, but until the manufacturer comes clean or is outed by an independent tester I just cannot in good conscience give them my money.

Bottom line, I can see a Shimano 11 speed in my very near future. And if that unit comes to market at the touted $400 - $500 range, maybe even more than one! But lets be honest here, as a bicycle geek I want one of each!

Now, what would it be like to pair the Shimano Alfine 11 with a Schlumpf bottom bracket and crank? Time to crank up Excel again.........
 

·
Registered
Joined
·
2 Posts
Thanks for the report, 'Smitty'. I'm also struggling with it. I want to get rid of my derailleur systems and switch to an IGH. You're talking about the lack of efficiency data from the NuVinci guys, but do you have data from other IGH's? I found eff data at Rohloff's site talking about 96% in gear 1-7 and 98% in gear 8-14, but this seems simply impossible to me. Any data from SRAM and Shimano? I can't find anything about that. I had the chance to test the new NuVinci N360 ( ~5.3 lbs ) and it felt very efficient. Also tested teh new Alfine 11, but that's too expensive for me.
 

·
Completely Human
Joined
·
177 Posts
Discussion Starter · #3 · (Edited)
Thanks for the report, 'Smitty'. I'm also struggling with it. I want to get rid of my derailleur systems and switch to an IGH. You're talking about the lack of efficiency data from the NuVinci guys, but do you have data from other IGH's? I found eff data at Rohloff's site talking about 96% in gear 1-7 and 98% in gear 8-14, but this seems simply impossible to me. Any data from SRAM and Shimano? I can't find anything about that. I had the chance to test the new NuVinci N360 ( ~5.3 lbs ) and it felt very efficient. Also tested teh new Alfine 11, but that's too expensive for me.
Glad I am not the only one in this "club" 29r.
I am with you on reasons for changing. Much of the efficiency data is available on line, after a little digging of course. I did not go into much of it here because it would have made my post 30 pages long. I will gather what I know and post it here in the next few days. This report was meant more to show how a person could quickly model their own system and compare new transmissions to current setups. If we really want a more detailed model we really should adjust every gear of every model by multiplying the end gain ratio by the efficiency coefficient as established by the manufacturers.
96% to 98% efficiency in mechanically geared transmissions is not unreasonable at all. These planetary gear systems really only loose energy to heat generated by the meshing of the gears during operation so loosing 4% or 5% is really very reasonable. From what I have read and studied derailleur transmissions are no different and test in the same efficiency range, that is until they get dirty. Some of the data I have seen suggests that a dirty derailleur can degrade to significantly lower efficiencies once road dirt and accumulates.

What were your impressions of the Nuvinci hub?

I really like the 11 speed Shimano as modeled in this analysis. I was very excited a few months ago when I learned the new Shimano would enter the market at the $500 range, BUT, I have not found even one vendor who has listed the hub for less than $900. Unfortunately this excludes the hub from my range and is wholly unfortunate for Shimano as well because the unit could easily have become a standard in the industry. I am affraid that it will become yet another "exclusive" component in limited supply. $500 was a fantastic price mark, its too bad the marketing hype didn't hold up.

As far as starting to gather data for efficiency on certain models of gear hubs, I really do recommend reading the links I provided as a start, especially these two,

The mechanical efficiency of bicycle derailleur and
hub-gear transmissions
- Chester Kyle and Frank Berto
http://www.ihpva.org/HParchive/PDF/hp52-2001.pdf

and

Efficiency Measurements of Bicycle Transmissions - a neverending Story?
- Bernhard Rohloff and Peter Greb (translated by Thomas Siemann)
http://www.ihpva.eu/HParchive/PDF/hp55/hp55p11-15.pdf

I will post more links as I re-discover them.
 

·
Registered
Joined
·
2 Posts
NuVinci N360 CVT

Ive tried the NuVinci N360 (~5,3 lbs) for ~ 20 minutes at Intermot in Cologne. The NuVinci guys had a trekking bike with a double chain ring in the front which gave a total ratio range of 560 %. I could take the bike outside on several parking areas with different levels. It's amazing how easy the new hub shifts. I've been told (never tried it) that the previous ones (~8 lbs) were hard to shift under heavy pedal load. This one always shift very easy. The hub felt very 'direct'. With other IGH's you sometimes find gears where you almost feel that you have to 'draqg' all these little planetaries. I've thought a coupl eof days about it, but I'm going to buy one soon for my 'winter' road bike. :)
 

·
Completely Human
Joined
·
177 Posts
Discussion Starter · #5 ·
Data update!

For some unknown reason all of the tables and figures are missing from this post.

For that reason I am offering the original Word document used to generate this report to anyone who would like the complete post, with all the technical data, tables and links.

For those who are interested post here or email me!....
 
1 - 5 of 5 Posts
Top