Scout!

Are internally geared hubs strong enough for pulling heavy loads up steep hills?

I'm considering a sram i-motion 9 hub for a touring rig. The documentation for the hub says things like "not suitable for tandem use", and the minimum front chain ring / rear cog ration (1.73) gives a net gear ratio of about 1:1. I'd like have something a bit lower when I'm trundling my own 100kg plus 50kg of gear up the side of a mountain. My worry is that the resulting torque would damage the hub.

I could spend a lot more money and get a rohloff hub, but I can't quite justify the price.

I could go with the sram (or some other hub) and use it only within the specified range, but that would mean walking up some steep hills. I would rather pedal along at a slow walking pace than be off the bike pushing.

I could just stay with a derailleur system and enjoy a newfound respect for its durability.

But what I really want to do is to abuse the hub by using a smaller than spec'ed front chain ring. Does anyone have experience with over-torquing internally geared hubs? Is it a really stupid thing to do? If it were to damage the hub, would there be a gradual degradation, or a sudden catastrophe?

LWaB

Over-torqued IG hubs break = BANG. You've read the info from the manufacturers, you know the answer already.

Scout!

Thank's for the info. I was hoping that the manufacturers' specifications would be conservative, allowing some safe leeway.

Also, just to clarify, my plan would be to use a double front chain ring. Most of the time I would use a big one (within spec), and only occasionally drop to the out-of-spec small one.

rodar y rodar

Somebody here (I don`t rememember who) posts occasional reports on how his touring Nexus is doing. Seems to me it had quite a few loaded miles, but I don`t know what kind of terrain. Try a site search on Nexus and you`ll probably find some of those posts By my understanding, the Srams are tougher than the Nexus ones. Also, I think BF uses Dualdrives on some of their tandems- again, I don`t know if that`s similar toughness to the iMotions.

vengeful_lemon

I sure would like an I-9 with coaster brake set up as a 34x20 teeth for a low of 22gi and a high of 76gi. That sounds like a really fun setup.
Hmm, an input ratio of 1.5..
There doesn't seem to be much experience with the I9 around. Exceeding the input ratio seems to be unknown territory. But maybe we could think of it this way: you'd only need to break and replace one I9 to almost pay for a rohloff...

Scout!

I did see that thread about the nexus hub. It was encouraging, but it sounded like he wasn't going as low on the gearing as I'd like to.

It's also encouraging that Bike Friday uses dualdrives, but they have 20" wheels, which means less torque on the hub than with 26" or 700c.

It could be that the only way to find out is to do it. Or maybe I'll just stay with a derailleur system. Or maybe I'll buy the rohloff, and justify the price not out by practical considerations but simply out of admiration of the thing.

LWaB

Of course manufacturers' specifications are conservative and allow some leeway but you won't find anybody to say, 1.5 x or 2.0 x specified is fine. If you want to run outside the manufacturer's spec, you run the risk of failure.

I've not tried killing a SRAM 9 speed hub but have blown two SA 3 speeds (a tandem and a MTB). They were brittle failures. Some more complex internal hubs lose a couple of gears when they break and get very noisy but it isn't a gradual failure either.

rhm

I have a Nexus-8 on my touring bike. I have a 19T cog and 38T chain ring that, with 700c wheels, gives me a low enough gear for everything I've tried yet. Pretty happy all round but I have to mention I have not used this bike for loaded touring since I put the Nexus hub on it. Rode it 140 miles last Saturday, but wasn't heavily loaded.

As for over-torquing the hub: no idea, sorry.

I read somewhere (in Bike Forums, but where?) that the hub rated for the highest torque is the NuVinci. I have one of those too and like it a lot; wonderful way to shift, somewhat wider range than the Nexus-8, and much cheaper than a Rohloff. A bit big and heavy, but very cool.

Allen

That was probably from me.
The NuVinci is a tank.

On my assisted Xtracycle the NuVinci and the Rohloff are the only two hubs that could handle it.
I snapped a S-A, and two SRAMS (5 speed and 7 speed drum brake versions).

rodar y rodar

That might be low enough for you, but I have the same tooth ratio (21/42) with a Nexus on my commuter but with 26x1.5, so effectively even lower and it still doesn`t cut it for me unless I keep to my little valley. As soon as I find a good buy on a nice rigid mtb that I like, I`m going to build a more versitile (for me) bike that I`ll be comfortable riding into town on. The Nexus will go on my wife`s bikepath bike- she loves it and has less ambitious riding plans.

Scout, I don`t think that having 20 in wheels in itself has any efect on the torque- it just needs a different tooth ratio to get the desired gearing. And since I think the manufacturers count teeth instead of gear inches (or whatever that other measure is that S.B. liked), smaller wheels give you an advantage as far as what you can get away with under warranty. Not that you`d get any benefit from that unless you`re thinking of a little wheeled bike.

kuan

Funny this thread came up. I was considering a SRAM on my Big Dummy but after reading this I might just save my pennies for a Rohloff. I'm a wimp too. I need 16 gear inches.

lifeguardsteve

Here is a gear inch chart I ran across that may be of interest.

rhm

I have no problem with the Nuvinci hub, but that gear inch chart is bad. I haven't checked the numbers, but in the graphic Nuvinci's 481 inches appears much larger than Rohloff's 526.

Allen

It's a misleading chart.
They aren't all starting from the same point, but are graphed along a range. And it is showing the Rohloff with uneven gear spacing.

Still the NuVinci is a really nice hub.

vik 

+1....I had a laugh about that...the marketing dept clearly made that graph - not the engineering dept...

The Alfine on my Surly Pugsley is geared way lower than recommended and so far no issues under some hard use.....even with 4 panniers full of beer!

My Bike Friday NWT is geared at the minimum 2:1 ratio on a Nexus 8. I haven't put it to the test yet, but I get a decent low gear and only run a single chain ring on that bike. Time will tell if I'm happy with that gear range, but looks good so far.

I'm running my Rohloffs at the minimum gear ratio as well and that gives a huge and very useful gear range with no fear of damaging the hub.

HardyWeinberg

Nuvinci's 481 also wider than mtn triple pushing 600% in the picture

Speedo

The 481 and 526 are percentages, not gear inches. Take the highest gear in gear inches, divide it by the lowest gear in gear inches and multiply by 100. It's a number that adds some honesty back into the chart. That percentage gives a better indicator of the wideness of the gear range than the width of the line on the chart, which goes from the lowest gear in gear inches to the highest gear on a linear scale.

If the chart was plotted on a logarithmic scale it would make more sense. The width of the line would then be consistent with the percentages.

Speedo

Speedo

When plotted on a linear scale the Rohloff gearing isn't even. If this were plotted on a logarithmic scale, the Rohloff gears would be evenly spaced. And many of the other systems would look more evenly spaced.

It is, after all, a NuVinci marketing chart. So anything that makes systems with discrete gears look bad, makes NuVinci look good.

Speedo

rhm

Right. But bad marketing and misleading graphs notwithstanding, the Nuvinci hub IS good; probably the nicest shifting mechanism ever invented for a bicycle. I don't know how tough it is; mine has had no maintenance issues at all, though it probably has less than a couple hundred miles on it.

Speedo

Fair enough. I don't even think of it as bad marketing. It's just marketing. They are emphasizing a point of their design that other contenders don't have.

Speedo

Scout!

Thanks, Allen. It is an Xtracycle that I'm using, and my philosophy about cargo is, "If I can lift it, I can strap it onto my bicycle and carry it down the road." So it could be that even within the specified gear ratio I could end up destroying most internally geared hubs.

For heavy cargo use, it sounds like the gear choices are lots of weight (NuVinci), lots of money (Rohloff), or the old cassette and derailleur.

And thanks lifeguardsteve for the chart. I think it's quite good. Just don't pay attention to the percentage numbers and it makes sense.

rodar y rodar

Or mount one Nexus wheel and stap another on your rack as a spare . Heck, you`ve got plenty of room and I believe two of them cost about the same as a Nuvinci, less than a Rohlhoff.

kuan

According to Vik's blog where he quoted someone from Rohloff, the new Rohloff is not going to be suitable for tandems. Any word on the new Rohloff?

vik 

I wouldn't hold my breathe for a new Rohloff any time soon. From Thorn Rohloff forum:

"Hello Justin,

We are indeed working on a newer, lighter version of our ever popular SPEEDHUB 500/14.

It is not possible (as with other bicycle components) to simply replace the material from which the SPEEDHUB is constructed with that of a lighter metal. Instead we must try to reduce the surface area of the entire system and this means completely revising the construction of the hub. This process as I am sure you can understand, is not exactly a light task.

To date, the designs themselves are not satisfactorily completed, for this reason we have no prototypes which we can start testing and therefore will most probably take a good couple of years before we are ready to release something into the open market place.

Again, because we have not yet completed the designs of the newer SPEEDHUB version, I am momentarily not able to say anything more about the construction.

The current version (SPEEDHUB 500/14) weighs approximately 1760g and therefore the weight lies only a few hundred grams over that of a derailleur gear system of the same quality. For this excess weight you receive a product which saves maintenance and repair costs, time by cleaning and adjustment, and by shifting the entire weight of the shifting system to the rear of the bicycle, the total weight of a bicycle will be evened out and centrally balanced (balancing point over the BB).

More information over the current SPEEDHUB can be found on our internet site www.rohloff.de .


Best Wishes from Fuldatal, Germany.

Stewart Stabik,
Technical Support Manager
Rohloff AG, Germany
__o
_`\<,_
(*)/ (*)
-+-+-+-+-+-+-

------------------------------------------------------------------<
* contact:
* Rohloff AG, Stewart Stabik,
* Mönchswiese 11, D-34233 Fuldatal
* phone: +49 561 510 80 0, fax: +49 561 510 80 15
* https://www.rohloff.de, mailto:stewart.stabik@rohloff.de

* Aufsichtsratvorsitzende: Iris Doepfer * Vorstand: Barbara Rohloff
* Handelsregister: Amtsgericht Kassel HRB 4472
* Sitz der Gesellschaft: 34233 Fuldatal
* USt-ID-Nr. DE 113 084 808"

Scout!

I'm gonna try to make a quantitative answer to my own question. In fact, I'm gonna make the question more quantitative, and then try to answer it.

It has been a long time since I took physics 101, and I didn't pay attention very well when I did, so I am not confident this is correct. But here goes, anyway.

Let this be the question: How much weight can a NuVinci hub propel up a 10% grade?

I chose NuVinci because their documentation specifies the maximum sustainable torque allowable for their hub. According to https://www.fallbrooktech.com/09_LEV_Kit.asp, it's 48 lb-ft.

I'm supposing one is not sprinting up the hill, but just pedaling enough to overcome the force of gravity. We need to figure out the force from gravity under the given conditions, and then translate that into torque on the wheel.

I figure the force of gravity to be weight * sin(arctan(slope)). The arctan(slope) changes the slope into an angle, and the sine of that angle is the fraction of the weight that gravity is fighting you for.

Now for translating that into torque. This is the part I'm must unsure about. I looked up "foot pounds" on wikipedia, and found that "...one foot-pound is the torque created by one pound force acting at a perpendicular distance of one foot from a pivot point." So the question becomes, how does the force pushing the bicycle back translate into radial force one foot from the hub?

My approach is to think of the wheel as a kind of lever. To make things easier, let's suppose the wheel is 24" in diameter. I'm imagining the wheel acts like a third class lever, with weight * sin(arctan(slope)) amount of force pushing at the center, and one end "fixed" (against the ground). In that situation the leverage given at the end is twice the force pushing at the center.

So my formula for hub torque comes down to

torque = weight * sin(arctan(slope)) * leverage

We have a 10% slope, and sin(arctan(0.1)) is about 0.1. With a 24" wheel, the leverage is 2. So the torque on the hub is approximately

torque = 0.2 * weight

Turning this around gives

maximum weight = 5 * maximum torque = 5 * 48 = 240 lbs

Some issues to keep in mind are 1) my reasoning may be incorrect, and 2) it sounds like some people have exceeded the specified limits and come out ok. All the same, this result makes me leery of trying carry much more than my own self when using an IGH. Unless it's a Rohloff. Or, relying on AllenG's experience, a NuVinci.

Anyone see any problem with my analysis?