Winfried

Hello,

The page on Nexus 8 gear hubs on Sheldon Browns's site says that low/high-end hubs differ on two points:

• Ratchet+pawls vs. roller clutch
• Plain bearings vs. needle-bearing pinions vs. roller-bearing pinions

The page shows how bearings differ, but I'm curious to see how different the ratchet+pawls and roller clutches look. Would someone have close-ups?

Thank you.



--

Edit : After more reading, it looks like…

1. Only rather old Shimano Nexus hubs use pawls + ratchet on the left instead of a roller clutch
2. Roller clutches are what transmit the energy to the hub ultimately and, thus the wheel, so it's not a good idea to grease the surface of the rolllers — or only slightly — as they will then lose traction.

sweeks

Roller clutches work fine with grease or oil on them. Besides, how likely do you think it is that the lubricant wouldn't spread to the rollers? The Shimano Alfine 11-speed hub has four roller clutches... and runs in an oil bath. Alfine-11
Roller brakes, also, run in a greased environment and function (reasonably) well.

Winfried

I agree, although in this article, the user complained of malfunction after applying too much grease (just on the rollers? Also the gears?).

Still, am I correct in understanding that the rollers are indeed how the force goes from the cog to the ground through the gears and the hub?

I'm surprised it all works simply by having the rollers push against the walls of the hub.

--
Edit: It appears that the Nexus 7 has no roller clutches, and relies on two sets of pawls to rotate the hub:

Dan Burkhart 

Roller clutches are sensitive to the film strength of the lubricant which is the reason Shimano formulates their own semi fluid grease. When someone experiences these issues, I suspect it is because they use a more viscous grease which will hinder the engagement and traction of the rollers.
If you use either Shimano Nexus grease, or oil, you should experience no such issues.

Winfried

Thanks for the feedback.

tcs

Roller clutches are an old idea and have been used in niche applications in bicycles for a long time. They're silent, but have a little more parasitic drag than ratcheting pawls and have a spongier take-up feel.

Shimano introduced a 'silent freehub' a quarter-century or so ago. It was met with a yawn by the derailleur crowd*. I believe it's still available as part of their parts groups for law enforcement bikes.



*There are even boutique freehubs that use the amount of noise they make as an advertising feature. Go figure.

tcs

Ratcheting pawls are alive and well in current production Nexus hubs.

Ross200

The "silent free hub" did find a following in the IceBike groups for cold weather riding. Shattered pawls are no fun at -40!

Winfried

Apparently, the amounf ot noise made by pawls can be lowered depending on the material used as well as the number of pawls involved, among other parameters.

I do like how silent Alfine/Nexus hubs are on my bikes. I find the SG-8R31 more draggy than the SG-C6001, possibly because of the use of plain bearings vs. needle-bearing pinions.

tcs

Possibly, although plain bearings vs roller or ball bearings on sun gears in bicycle gear hubs has been studied and debated (along with materials and lubrication) since at least 1908, with the general consensus that the ball/rollers for the sun gears is more of a marketing feature for upmarket hubs than an actual efficiency gain.

Winfried

Happened on this page:

"Why does the Shimano Nexus-7 has such a great friction?"

sweeks

I'm not arguing with this, but I have been using a non-Shimano lubricant on a Nexus-8 and an Alfine-11, both of which have roller clutches. The Nexus hub gets its innards dipped in a synthetic gear oil once a year, and the Alfine gets an oil change about every 800 miles or so. Each of these hubs has about 10,000 miles on it, and both work perfectly.
The oil is Royal Purple 75W-140 Max Gear.

Interesting explanations of roller clutch operation and .

It would seem that it's not the function of the lubricant to actually transmit the force passing through the clutch.

GamblerGORD53

My SA XL-RD5w has easily done up to 46 mph. I can feel the woosh on any hill.
Rohloff14 has done 43 mph.
Nexus 7i won't top 38.1 mph. But that's with a boat anchor roller brake.
SA RD3 feels the same as the 5w and went 37.18 mph, mostly coasting because only 92.5 GI.
The SA pawls show maybe 1/1000th wear after 9,000+ miles.
I don't remember any of my 50/11T defaileur bikes going over 40. But 2 of them had a mechanical speedometer.

My SA hubs use 10-30w non-detergent lawnmower oil thickened with 25% Wynn's old car oil conditioner.
NOT dipped. SA hubs are easy to take apart fully.
I haven't taken the Nexus apart, except for the left bearing.

base2 

36 hole, 3 cross...on what? A 20 inch folder? Dang. That's some serious lacing on the OP's wheel. Edit: never mind. That pic is lifted from Aaron Goss's shop in West Seattle.

Roller vs pawls. Who cares? Neither is better or worse. Roller is how car starter motors are done now. It is clearly the better design for low maintenance reliability & longevity. Pawl are best for firm instantaneous engagement. Simply different design priorities.

IME neither makes a hill of beans worth of difference beyond the margin of error.

I like my Nexus 7 with roller brake. I like my Rohloff. When fully broken in, I guarantee you neither is distinguishable from the efficiency of a well lubricated derailleur system in terms of system efficiency in actual use.

If you can not go fast, look to other factors beyond the IGH. The IGH is not the reason.

3% of 200 watts is 6 watts. It's the amount of aerodynamic drag caused by exposed cable housing at 40kph. IOW: Meaningless.

What's the OP's question? How to rehash Harris Cyclery friends site pages (avdweb.nl) & Sheldons IGH observations? Lame.

Winfried

What gear oil do you use?

So there's no significant difference between the Nexus/Alfine hubs in terms of efficiency, and the use of plain/needle bearings and pawls/roller clutches is just marketing?

Once its new cog gets here (hello covid), I'll open up the SG-8R31, see how lubricated it still is, experiment with either fresh grease or oil, and see if it makes a difference.

sweeks

The Royal Purple 75W-140 I linked to. It's basically a 75W oil; the 140 part is not relevant at bike hub temperatures. A dip or two, followed by a few minutes of draining, works well. See the images below for a suggestion to reduce the amount of oil needed.
If you want to separate the big (drive-side) bearing from the gear cluster, you have to disassemble the right side. This is not necessary, in my opinion, but does allow for a more thorough cleaning of the bearing.
After the oil dip, I lubricate the bearings with Shimano Nexus Hub Grease, which looks like a molybdenum disulfide type of grease. It eventually bleeds into the gear cluster, probably helped along by the gear oil. I don't see any wear on the bearing surfaces, so things seem to be working. Also, the grease/oil combination has no adverse influence on the roller clutches' performance.
One of the things I don't like about the Nexus hubs is that they aren't well-sealed against the oil leaking out, especially from the drive side. So if your bike spends any significant amount of time on its side, it should be the *non-drive side* or you may have a small pool of oil. Since these hubs are originally lubricated by grease, not oil, the hub design can't really be blamed for the oil leakage. The bottom line is that if the bike spends most of its time in the upright position, oil leakage is minimal. A longer draining time after the dip helps; there's still plenty of oil coating the gears.

I have two Nexus-8 hubs. One is the "standard", which has plain bearings on the planetary gears, and the other is the Premium ("Red Band"), which has the roller bearings. I swap them on the same frame, and haven't noticed any obvious difference in performance. I'm not sure how much load is on the axles of the satellite gears; they seem to be held in place mainly by the sun and ring gears. If that's correct, the needle bearings probably don't do much to improve the hub's efficiency. Please note that I am an "arm chair" hub mechanic, and will defer to any/all of the experts on the forum!



There is a syringe epoxied to the bottom of a plastic ice cream jar.



The syringe accepts the axle of the gear cluster, greatly reducing the volume of oil required to submerge the cluster.



This can makes a nice base for the modified jar. The level of oil is sufficient to completely cover the cluster. I re-use this oil, adding more as needed.

Winfried

Thanks mucho for the infos.

While I'm at it, I'm thinking of drilling an oil port on the hub, and make it a breeze to add oil later à la Alfine 11.

sweeks

I wouldn't. You can't fill the hub with oil, as you can with the Alfine 11 because the seals aren't meant to retain oil. The added oil, depending on where you drilled the hole, might not be well-distributed among all the gears. It probably wouldn't reach the bearings on the non-drive side. It's easy to open the hub and dip it, and you get a chance to inspect for early signs of wear. Just my perspective.

Winfried

From what I read, recent models of the Nexus 8 don't leak much provided you don't use more than ~15ml of oil. The Alfine 11 uses 25ml.

As for resting the bike, especially a folding bike, it's not clear it should rest drive side up or down. I read both.

https://www.astounding.org.uk/ian/hubgear/oilport.html

sweeks

The seal on the non-drive side is better (though not perfect) than that on the drive side. I've had better luck keeping the bike drive-side up, with respect to leakage. My Alfine 11 leaked a few drops when laid on its drive side overnight.
I tried wrapping the drive-side seal in electrical tape after thoroughly degreasing the outside. The gear oil found its way through in a few days. What seems to work reasonably well is to fill the groove in the hub into which the plastic sealing cap fits with water-resistant wheel-bearing grease.

Winfried

Thanks for the tip. I'll grease up the cog-side.

Someone else tried the electrical tape:

Winfried

Since the hubs are compatible, it was a snap swapping the 8R31 for the C6001.

Thanks to its needle bearings, I confirm the latter runs better, especially the previously crappy 4th gear, even while still running its original Shimano grease.

https://www.sheldonbrown.com/nexus8.shtml
https://www.2rad.nrw/project/shimano...abe-demontage/
https://wanderingdanny.com/oxford/20...-8r36-redline/



--
Edit:

Stats: On average, after giving the crank a one turn spin by hand with the Brompton (ie. 16" wheel) on a stand…
- the 8R31 with the original Shimano grease stopped after about 5s
- the 8R31 after being bathed in ATF oil would stop after about 10s
- the C6001 with the original Shimano grease stops after about 15s.

I guess the C6001 will spin a bit more the day I give it an ATF job.