Modern Needle Bearing headsets

[Michael Slater]

I am building a frame that will be a good candidate for shimmy -- relatively thin tubing on a relatively large frame with 42 Hetre tires.

BQ has said one antidote is to use a needle bearing headset.

Problem is, as far as I can tell they aren't made anymore (1) and they're generally for 1" threaded steerers (2).

Are there any 1-1\8" thread less headsets that are known as more resistant to this sort of thing? Is using a otherwise-needlessly-large 1-1/4" headset going to help?

[Andrew R Stewart]

Michael- I wouldn't give too much credence to roller bearings being the
solution. I've had three bikes with Stronglite roller headsets (both with
and without the "o" ring seals) that all shimmied. I think that the frame's
torsional stiffness is FAR more important.

What I did find that was specific to the roller designs that you might want
to hear is- the adjustment is very sensitive. Both to the frame/fork face
and chasing, rollers have no ability to handle misalignment (why do you
think that ball bearing elements have been used for so long on bikes) and
the rollers riding on secondary surfaces placed onto the Alu cups and cones.
This added layer has it's own bit of rock and fitting issues. Additionally
with the Stronglite is that these steel surfaces are relatively soft and
will dent easily if the headset adjustment is even a little bit loose. These
headsets are best run with a fairly high preload. Next are the lubricating
challenges. This aspect is one that I have yet to read about before. The
rollers fit into a plastic retainer that takes up a fair amount of space.
There's not too much empty space for the grease to fill. I didn't find much
difference between the "o" ring sealed version and the basic version as far
as the grime getting in. These headsets really like frequent cleaning and
lubing. After a number of years of using these on a few bikes I finally
moved on and up to Kings for most of my bikes. I still have a Stronglite or
two on my shelves (as well as at least one Campy) and there they will stay.
Andy.

> --
> You received this message because you are subscribed to the Google
> Groups "Framebuilders" group.
>
> Searchable archives for this group can be found at
> http://groups.google.com/group/framebuilders (recent content) and
> http://search.bikelist.org (older content).
>
> To post to this group, send email to frameb...@googlegroups.com
> To unsubscribe from this group, send email to
> framebuilder...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/framebuilders?hl=en
Andrew R Stewart
Rochester, NY

[Mathias Scherer]

Why would a needle bearing headset help to avoid shimmy? I know this was
presented as a possible remedy on earlyish Japanese motorbikes, where
comparatively powerfull engines teamed up with flimsy frames which led to
shimmy. Needle bearings did indeed help,not because they were needle
bearings but because they actually worked as bearings, which couldn't always
be said about the original bearings. Hence my solution to your possible
problem would be to build the frame strong enough for what you will be using
it for, and to use a decent headset known to work (Chris King being one of
the more obvious candidates, I'd say).


Am 22.12.2012 4:37 Uhr schrieb "Michael Slater" unter
<michael...@gmail.com>:

--
Mathias Scherer
Ludwig & Scherer GmbH
Wintringerstrasse 19
66271 Kleinblittersdorf

Geschäftsführer Mathias Scherer

[M-gineering]

The needles have a constant diameter, which makes it impossible for them
to roll on a circle without skidding. Thus they have much more
(friction) damping than a proper bearing

On 22-12-2012 9:33, Mathias Scherer wrote:
> Why would a needle bearing headset help to avoid shimmy? I know this was
> presented as a possible remedy on earlyish Japanese motorbikes, where
> comparatively powerfull engines teamed up with flimsy frames which led to
> shimmy. Needle bearings did indeed help,not because they were needle
> bearings but because they actually worked as bearings, which couldn't always
> be said about the original bearings. Hence my solution to your possible
> problem would be to build the frame strong enough for what you will be using
> it for, and to use a decent headset known to work (Chris King being one of
> the more obvious candidates, I'd say).
>

--
mvg

Marten Gerritsen
Kiel Windeweer
Netherlands

[Mathias Scherer]

You're a smart guy. Never thought of it that way.

[Alistair Spence]

On Sat, Dec 22, 2012 at 12:44 AM, M-gineering <in...@m-gineering.nl> wrote:
> The needles have a constant diameter, which makes it impossible for them to
> roll on a circle without skidding. Thus they have much more (friction)
> damping than a proper bearing

Martin,

I've never used a needle bearing headset myself, or serviced one in
some years and I'm wondering if there aren't some models that are
available with tapered rollers (instead of constant dia.)?

Even though they overcome the skidding problem you mention above, they
do still come with a friction increase disadvantage. The tapered
rollers tends to want to eject themselves from between the races and
against whatever keeps them captured in the bearing, increasing drag.

So maybe in the relatively light loading scenario of a bicycle
application they are considered overkill, and not worth the added
cost? Just wondering.

Alistair.

[Mathias Scherer]

I think war Marten said was a needle bearing headset may indeed help to cure
shimmy to some extent because they act as friction steering dampers.

The Stronglight A9 which sort of springs to my mind when talking about
needle bearing headsets was anything but expensive. I think they retailed
for around 13 Euro (17 USD). Not a lot of money, I'd say.

Still I'd rather beef up the frame by using a lager tube diameter.


Am 22.12.2012 18:21 Uhr schrieb "Alistair Spence" unter
<alspe...@gmail.com>:

[Alistair Spence]

On Sat, Dec 22, 2012 at 9:31 AM, Mathias Scherer
<Mat...@ludwigundscherer.com> wrote:
> I think war Marten said was a needle bearing headset may indeed help to cure
> shimmy to some extent because they act as friction steering dampers.
>
> The Stronglight A9 which sort of springs to my mind when talking about
> needle bearing headsets was anything but expensive. I think they retailed
> for around 13 Euro (17 USD). Not a lot of money, I'd say.
>
> Still I'd rather beef up the frame by using a lager tube diameter.

Right, but he referenced cylindrical rollers in explaining the
friction mechanism, which are cheaper. I was wondering about tapered
rollers, which are typically found in car and motorcycle applications,
and wondering if any headsets use those?

Alistair.

[Brian Smith]

This is a framebuilding list, and if your intended creation is expected to shimmy, it seems part and parcel to the list to recommend that you build a better frame. 

I should stop here.

Although headset tech is a fun topic, it's probably OT here, and at any rate if we're wallowing then how about doing it with a solution and using a steering damper for steering damping instead of a sub-par headset design which doesn't yet physically exist for your intimated steerer choice? 

Although I'm guilty of forgetting it too, I think when in search of your solution you identify that a "problem is, they aren't made anymore," there is often a good reason, and often it's an appropriate reason even for a planned output of one unit.

[Eric Keller]

On Sat, Dec 22, 2012 at 12:31 PM, Mathias Scherer <Mat...@ludwigundscherer.com> wrote:

Still I'd rather beef up the frame by using a lager tube diameter.

The only study I've seen where they analyzed shimmy showed a very weak, second-order effect of frame stiffness on shimmy.  There are tons of big frames out there that don't shimmy, and tons of smaller bikes that do.   My experience with shimmy shows that it is a rigid body phenomenon.  It seems intuitive to me that a vibration mode with a hinge in the middle is very unlikely to have much of a bending component at all, but I'm literal-minded like that.
Eric

[Alex Wetmore]

Stronglight made them for 1-1/8" threadless. About a year ago Fred Blasdel (easy to find on the internet) imported about 20-30 and resold them through the Rawland list. I don't know if they are still available, but I'd check the French internet supply houses.

Rene Herse/Boulder sells Miche ones. I use one of their 1" threadless headsets on one of my bikes and it doesn't shimmy, but I never compared with another headset to see if it should.

I'm about 185 and have had bikes with "standard" diameter tubes in 7/4/7 and 8/5/8 wall thicknesses and haven't had shimmy problems. The 7/4/7 bike (Platinum OX) was a 59cm frame and the only shimmy that I experienced was while loaded touring. Moving the rear panniers (containing about 10-12# of stuff, so fairly light) forward 1" eliminated the shimmy. Most of the weight was in front lowrider panniers.

My 8/5/8 bike is only used with a small handlebar bag and it's never shimmied. That bike is made of Ishiwata 019 and has a Chris King 1" threadless headset.

I personally think that 1-1/8" steerers look pretty silly on a bike with a 1" top tube, so I'd use 1" if the frame isn't built yet. There is no reason to burden a light frameset with a fork that has a heavy 1-1/8" steel steerer. 1-1/8" is fine and it's nice to have a nearly universal standard, just save it for aluminum and carbon steerers.

alex
________________________________________
From: frameb...@googlegroups.com [frameb...@googlegroups.com] on behalf of Michael Slater [michael...@gmail.com]
Sent: Friday, December 21, 2012 7:37 PM
To: hei...@earthlink.net; Framebuilders

Subject: [Frame] Modern Needle Bearing headsets

I am building a frame that will be a good candidate for shimmy -- relatively thin tubing on a relatively large frame with 42 Hetre tires.

[Jona Aal]

Can i just point people to this:

http://sheldonbrown.com/brandt/shimmy.html

Seems to be a spring thing, maybe the frame just needs to be from larger tubing?

FYI, I also had an experience where i was riding out and broke two spokes in the rear wheel, the wheel was still totally rideable for the journey home, but seemed to increase shimmy to the point where i could not take both hands off the bars at all, at all speeds.

Bike was a specialised stumpjumper steel model

Jona

--
Jona Aal

H 01434 345 620
M 07940 731 375

[Alistair Spence]

Getting back to the OP's question somewhat, I'll throw some numbers out there.

My most ridden bike is along the lines of the one Michael is planning to build.
1 1/8" DT, 1" TT, low trail, Hetre tires and a front rack carrying
typically 5-10lbs of whatever. Tubing is 9-6-9, and I'm 150lbs.
Headset is a Dura Ace cartridge model.

Definitely prone to shimmy with hands off the bars.

Alistair.

[pop's]

Back around the late 1970's - early 1980s the Italian firm Galli made both a bottom bracket AND a headset with tapered roller bearings.
They both had problems with tight spots and binding due to the sloppy tolerances and lack of a concentric mounting in both the headtube and the threaded bottom bracket shells of the day. Later, Tange offered a headset with tapered rollers on the lower cup and ball bearings in the upper cup; which I'm guessing was to allow for a smoother running bearing in a less-than-perfect mounting (i.e. the headtube).

The Stronglight non-tapered roller bearing design relies on the races being able to "float" a bit in order to cope with the typical real-world bicycle frame bearing mounts. The theory may predict all kinds of issues with bearing skid and lubrifrication, but I used and serviced many of those headsets over the years I worked in bikeshops, and I thought the design worked quite well in practice. The A9 sometimes needed a bit of extra tightness to eliminate rocking "play", but the real gem in my experience was the later design X12 type headsets where the roller bearings were opposed to each other, and the sealing was improved with o-rings. Those were simple, durable, and light headsets that IMO held up much better than the typical ball bearing headsets of the day. The bearings were cheap and easy to replace if  there was wear or damage to the unit.

I don't agree at all with the idea that the headset has much, if anything to do with damping shimmy on a complex set of springs and weights that comprise a bike and rider.

Mike Fabian
San Francisco.

[mallard]

For headsets the issue with them is contact-surface area, ball-bearings are normally given rpm ratings as that gives the dynamic contact area for most uses.

 

Obviously headsets don't spin so the older headsets would become "indexed" as facing forward gets pounded into the races, a reason for the try at tapered bearings, and the more practical move to smaller diameter balls to get more points near the major axis.

 

Modern ceramic sleeve bearings have far more contact area & distribute the stress well, so, the only issue with them is a brief break-in period & they should then operate like a quality bearing headset for feel according to the sales rep & got a catalogue & seemed some T styles made for thrust loads would do, these have lifetime lube.

 

Waiting till I have a studio-shop again but will give it a try on a tt frame with a 1" compact tiny ht to see if my hunch is correct, that is by having so much more surface area you don't need a larger tube up front & does a better job to stabilize the fork.

 

Consider this short 1" ht geometry, like 4" tall, if you take in deformation of the balls handling the impact and extend the line to the front axle vs at rest, it won't be in the same place by a distance that depends on the impact.

 

Sleeve bearings provide so many more times contact-area they'll have no significant deflection by contrast to using ball-bearings; therefore, they will stabilize the fork vs ball-bearings.

 

So, if that works it means it'll allow the trad 1" look on any size ht & any use for the frame & be std building by custom machining cups to insert them into the frame.

tom mallard

[Andrew R Stewart]

Tom- A couple of points about your post.

 

I always was told, and have seen from shop experience, that the load capacity of a ball bearing is more dependent on it's diameter (by the size squared) then the number of balls (rising only in a linier fashion) in the element. So if you halved the ball diameter but doubled the count you'd loose load capacity. My understanding to way smaller balls are used is everything to deal with packaging ((for headsets this is largely stack height).

 

Trek had a top head set bearing that was a sleeve design with some kind of low wear/friction material for the surfaces. It was a bit lighter. But it didn't adjust well. You'd get it right in the shop, no rock and still freely spinning. but after a few real rides it would loosen up WRT rocking. Our customers wouldn't accept a lower (in their eyes) performance for a few grams. Both now ride a conventional balled  upper stack and don't complain any more.

 

So as you experiment with your ideas don't forget that some have gone down that path already. Andy.

 

----- Original Message -----

From: mallard

To: frameb...@googlegroups.com

Sent: Sunday, December 23, 2012 10:09 PM

Subject: Re: [Frame] Modern Needle Bearing headsets

For headsets the issue with them is contact-surface area, ball-bearings are normally given rpm ratings as that gives the dynamic contact area for most uses.

 

Obviously headsets don't spin so the older headsets would become "indexed" as facing forward gets pounded into the races, a reason for the try at tapered bearings, and the more practical move to smaller diameter balls to get more points near the major axis.

 

Modern ceramic sleeve bearings have far more contact area & distribute the stress well, so, the only issue with them is a brief break-in period & they should then operate like a quality bearing headset for feel according to the sales rep & got a catalogue & seemed some T styles made for thrust loads would do, these have lifetime lube.

 

Waiting till I have a studio-shop again but will give it a try on a tt frame with a 1" compact tiny ht to see if my hunch is correct, that is by having so much more surface area you don't need a larger tube up front & does a better job to stabilize the fork.

 

Consider this short 1" ht geometry, like 4" tall, if you take in deformation of the balls handling the impact and extend the line to the front axle vs at rest, it won't be in the same place by a distance that depends on the impact.

 

Sleeve bearings provide so many more times contact-area they'll have no significant deflection by contrast to using ball-bearings; therefore, they will stabilize the fork vs ball-bearings.

 

So, if that works it means it'll allow the trad 1" look on any size ht & any use for the frame & be std building by custom machining cups to insert them into the frame.

tom mallard

[mallard]

Right Andy, talked to the rep about that, there will be no discernable wear was his take & experience from customers for this range of use, it's too hard a surface the reason given, so the test frame is to do the deed at the extreme & put miles on it to find out.

tom

[John Thompson]

On 12/21/2012 10:21 PM, Andrew R Stewart wrote:

> What I did find that was specific to the roller designs that you might
> want to hear is- the adjustment is very sensitive. Both to the
> frame/fork face and chasing, rollers have no ability to handle
> misalignment (why do you think that ball bearing elements have been used
> for so long on bikes) and the rollers riding on secondary surfaces
> placed onto the Alu cups and cones.

*Tapered* roller bearings are quite sensitive to alignment, but the
Stronglight uses cylindrical roller bearings. Furthermore, the
Stronglight uses floating races which further improves tolerance to
misalignment. The downside to cylindrical roller bearings is that the
difference in radius between the inner end of the roller and the outer
end means that there is always some sliding contact in rotation, but the
fact that headsets are not in continuous rotation when in use means this
is a minimal concern.

--

-John Thompson (jo...@os2.dhs.org)
Appleton WI USA