Ellis 953 One Year Later

[August]

http://farm7.static.flickr.com/6195/6160181370_811230ea91_o.jpg

Eleven months later to be precise. I've been meaning to write a post on this topic for a little while now, and since I just freshened it up with a new saddle and tape, now's the time. I've written about this bike before, its aesthetic appeal and the somewhat unusual way in which I ordered it, so now I'll explain its performance and durability. To say the least it has been a pleasure to own it. I used it as a training bike, strange as that sounds, in preparation for a 100K mountain bike race, and since I needed to be comfortable riding for up to six hours at a time, this bike and I have covered a lot of ground.

http://farm5.static.flickr.com/4132/5077467334_53d5efb5a8_z.jpg

A few things stand out when I recall all those various moments riding. Most obvious was the effect that such light-weight, thin-walled tubing had on handling: I can definitely perceive the steering column and the seat tube moving in different planes. This is not as strange as it sounds, nor does the bike fail to follow a line. It's just more pronounced now that I've ridden bike with ver stiff, too stiff, I think, front triangles. Sprinting in the drops exaggerated the sensation of course. Those who claim there is no difference between OS and XL tubing are wrong. It's a matter of taste, intent and the rider's numbers that decides between them.

http://farm6.static.flickr.com/5007/5260930276_c22b66e98f_z.jpg

When riding with Cat 1 and 2 racers, this bike proved its handling capability, and on twisty descents it stuck to bad road surfaces when the carbon bikes where stuttering and shaking away their traction. The relatively long stays made for confident descending and cornering, and the rear wheel stays planted under power. And the bike is perfectly comfortable. The steel fork was a revelation after carbon, so much so that I feel safe saying every steel road bike ought to have a matching fork.

http://farm3.static.flickr.com/2321/5703278246_ac3ea5480f_z.jpg

My complaints are one. First, steel tubing in OS diameters seems to be reaching its limits in terms of wall thickness. The bike wobbles if I'm not paying attention. It doesn't scare me or ruin my ride, but it is an attribute of this bike that is married to it, and to get rid of it, one would have to build a completely different frame. I like the bike exactly as it is, and that phenomenon can no more be called a flaw than steel's much lauded compliant ride. I suppose I could say the bike is not as forgiving of a daydreaming rider as some others I've ridden, and in that way it could be seen as flawed; but it also gives the bike personality, and for that, I'm thankful. It wants all my attention, all my effort.

As components go, I haven't changed anything except the chain (once), the tires (twice), and the aforementioned bar & tape. Perhaps I'll put some deeper carbon wheels on in the future, but for now, the 101s are doing the job just fine.

The bike looks like I just built it up last night, and that's a testament to the resilience of polished 953, Jason Sanchez's paint and Super Record parts. Nothing, not salt, ice, melted asphalt or dead worms can unpolish the mirror-like tubes. It may be flashy but its quite practical besides.

http://farm6.static.flickr.com/5181/5832810807_7847f3ecdc_z.jpg

Finally, I want to address myself to Dave Wages. The bike is far and away my favorite of all I own or have owned, and I am proud every time I get to ride it. I've ridden it as hard as I've ever ridden a bike, and it just won't give. Modern Classic is Dave's motto, and if you take a look at his picasa page, you'll see that no other custom builder, of steel bikes or otherwise, has made a personal commitment to the integration of new, modern technology, specifically Di2, while retaining that which has always made classic bikes classic. He is a consummate craftsman. He is a conscientious business man. There are only a few builders I would even consider if I were thinking of another bike, and Dave is among them. Thanks again, Dave, for a bike I can appreciate over and over, that renews my love for cycling and the personal discovery riding affords.

[avalonracing]

Beautiful bike and good and honest write up about it. Enjoy.

[Pete Serotta]

and excellent craftsman and builder. PETE

[rnhood]

Yes, good honest write up and that bike needs no introduction of craftsmanship. Stunningly good looking.

[rounder]

Everything about the bike is beatiful.

[Lovetoclimb]

What a beaut

[phcollard]

OMG a work of art in its own right. I like the minimalist look (although it's packed with high end equipment). Thanks for sharing your thoughts as well. It must be a joy to ride and own.

[Bob Ross]

First off, that is a gorgeous bike!

I've loved every Ellis I've ever seen, and yours may be one of the finest examples. Everything looks right. I swear, an Ellis Modern Classic is on my short list. (It may also be a bucket list, but that's not the point.)

But a number of statements from your ride report caught my attention:


...Most obvious was the effect that such light-weight, thin-walled tubing had on handling...
...Those who claim there is no difference between OS and XL tubing are wrong...
...steel tubing in OS diameters seems to be reaching its limits in terms of wall thickness. The bike wobbles if I'm not paying attention...


The "conventional wisdom" -- or perhaps rather, the new conventional wisdom that seems to be gaining traction amongst custom framebuilders and their aficianados, and which appears to be replacing the old conventional wisdom from what I can tell -- suggests that those qualities you describe aren't necessarily properties of the material, but rather of the execution.

So I'm curious: do you (or anyone who'd care to pontificate here) think it's possible that a skilled framebuilder could build a bike that evinces the perceptions you describe above if it were made out of a different material? Or are you convinced what you're feeling is an inherent quality of 953?

Thanks. Did I mention that's a gorgeous bike?

[August]

Bob, it's the builder and the materials both. Mine is a big frame with OS tubes. If Dave had chosen tubes with thicker walls, the bike would ride differently. If the tubes were the same wall thickness but of larger diameters--as those of my partial 953 Di2 bike are--the ride would likewise be different. The worst speed wobble I ever experienced was on a mass-produced carbon frame made overseas for an American brand. The wobble came on out of nowhere and nearly dumped me at 50 miles per hour. The Ellis is completely stable at that speed; that is, when I have both hands on the bars. Its tendency to oscillate is compensated for by less weight and excellent cornering traction (read Dave Kirk's monograph on frame flex for more insight).

The phenomena I described and you quoted back to me applies more or less to every bike I've ever ridden. It's odd to say so, but I think I like the 953 bike the most because it's so lively, because it insists that I keep my head. That said, it has never held me back nor made me apprehensive. It has made me a better, more experienced rider, one who can appreciate the nuances different bikes have to offer.

[Lionel]

August, the wobble you are talking about is a high speed wobble then ? I am not clear why you attribute this to the tubing?

[dd74]

Gorgeous bike. Congratulations!

[soulspinner]

Stunning :beer:

[toaster]

Yes, why is speed wobble blamed on tubing?

[rustychain]

Great bike and good write up.
I agree with others however on the " nature". of 953. I own an IF 953 and I'd have to discribe it as a full tilt stiff race bike ( thats what I asked for btw). Had it well over 50 mph and tracking was spot on, carves corners, climbs etc. Size 58 the bike is 15 lbs. A good builder can give you the ride character you ask for. Sound like you got a great bike for your purpose of doing lots of mile with comfort. Hope your race went as well

[odin99]

amazing bike and good write up... one day i hope to own something of that sort.

[August]

It's not a speed wobble like the one I experienced on the carbon bike. It can happen at any speed, but only if I'm riding with both hands off the bar. I suppose I should have been more specific.

I don't think there is anything wrong with 953 or that a bike is fated to wobble if it is built with that tubing; as I mentioned, I have another Ellis whose top- and downtubes are 953 and it displays no such behavior. A tube's wall thickness, butt lengths, over all length and especially its diameter are what make a bike handle one way and not another. OS tubes with an .8mm-.5mm-.8mm butting profile are, I think, very different from a builder's perspective than 953, the thinnest of which, like those of my bike, are .5-.3-.5. Take a digital caliper and have a look at what .3mm is--not much of a wall.

rustychain: there are a variety of 953 tubes with different diameters and butting profiles, so without knowing which your bike was built with, I can't say why it is so solid. I suspect the builder knew what "full tilt race bike means," though. You're right, a builder can provide exactly what one asks for. That is to say, 953 is a great material, and a builder can build many different frames with it; but my point was simply that at certain diameters, the butting may be approaching the limits of what is possible. I'm imagining a .25mm wall--would such a tube, strong as the material may be, resist all the forces a cyclist can generate (or even deflect a small rock without denting)?

[benb]

Did you have this bike built with a longer top tube then past bikes?

I had this same kind of wobbling with my Serotta Concours.. when I had gone in to be fit for it they had put me on a longer top tube then usual. Mine could even have a hint of this kind of wobble if I was on the bar tops and hit a bump... it would definitely do it no handed. And yet it definitely did not have speed wobble at any speed up to ~50mph.. because if I got in the drops I was putting enough weight forward.

I went back to a shorter TT like I had ridden on other bikes this year and there is none of this wobbling, even riding no hands. My new bike is stiffer then the Serotta but not by much, and other then TT, front center, and wheelbase it is just about exactly the same geometry. The front triangle on my Concours was very stiff and it's handling other then this weird wobble was fairly spectacular...

I never blamed the frame, just me + the frame, I wasn't big enough for it.

I can kind of duplicate it on other bikes now that I understand it simply by sliding the saddle back too far.. it unweights the front and the bars start to get "lively" and you have to "pay attention".

Damn good looking bike BTW.. :D

[Lionel]

Yeah, so many variables enter this wobbling: geometry, weight distribution, frame alignment, wheelset, fork, etc.... But I doubt that tubing wall thickness is one of them.

[Dave Wages]

Yeah, so many variables enter this wobbling: geometry, weight distribution, frame alignment, wheelset, fork, etc.... But I doubt that tubing wall thickness is one of them.

I feel like I'm a bit late to the thread, but suffice it to say that I'm flattered by all the nice compliments and also really pleased that Chris is enjoying his bike!

Lionel, to address your comment, I have to disagree, I think that wall thickness has as much to do with this as any of the other variables you list. When I set about designing a custom frame, tubing diameter, and wall thickness play an integral role in how the bike will "ride" and perform. While it's hard to quantify this precisely, think of the latest wonder bike ("it's 12% stiffer and 7.5% more supple")????, it clearly does make a difference. Why do you think that old school builders used Columbus SP (1.0 x 0.7 x 1.0) on larger bikes instead of SL (0.9 x 0.6 x 0.9)? Back then, there weren't nearly as many tubes to choose from, so it was a pretty simple equation, but now I can use tubing diameters and wall thicknesses to really find what I hope will be the perfect combination of stiffness and comfort for a given rider.

On Chris' 953 frame, I used some of Reynolds thinnest walled 953 tubes. They are advertised at (.55 x .35 x .55) (sorry Chris, you're off by .05mm!), and they lend the liveliness that he describes. Chris is a tall guy, but not a bruiser by any means, so I chose these tubes based on our conversations about what he likes, and I'd like to think my choices were right on, he seems pretty happy. If I had a 6'5", 230lb guy who specializes in crits and hammering out of the saddle, I'd certainly have made different choices in both the tubing diameters and wall thicknesses.

To me, this is just another level of custom frame design. So many folks focus on custom geometry, and that's important, but the reality is that most folks (probably 95%) could ride some form of "stock" geometry without too much trouble. What I see as the extra benefit of the custom is this tube by tube choice of material. Let's say that a product manager is designing a stock 56cm frame, he doesn't know if the end user is going to be 125lbs or 250lbs, so there's usually some overbuilding that has to be factored into a stock frame. When I design a frame, it's with one specific rider in mind, and I try to find out their ride preferences, what they might be carrying on the bike, and where (front rack, rear rack, panniers?), in addition to simple things like height, weight and basic geometry.

Wow, you really got me going there. Hope that is enlightening?

Cheers,
Dave

[Ahneida Ride]

There are builders whose work is impeccable.
Mr. Wages is one.

[tuscanyswe]

I really really like this bike! Lovely colour n component choices and apparently rides great too. A winner!

[Lionel]

I feel like I'm a bit late to the thread, but suffice it to say that I'm flattered by all the nice compliments and also really pleased that Chris is enjoying his bike!

Lionel, to address your comment, I have to disagree, I think that wall thickness has as much to do with this as any of the other variables you list. When I set about designing a custom frame, tubing diameter, and wall thickness play an integral role in how the bike will "ride" and perform. While it's hard to quantify this precisely, think of the latest wonder bike ("it's 12% stiffer and 7.5% more supple")????, it clearly does make a difference. Why do you think that old school builders used Columbus SP (1.0 x 0.7 x 1.0) on larger bikes instead of SL (0.9 x 0.6 x 0.9)? Back then, there weren't nearly as many tubes to choose from, so it was a pretty simple equation, but now I can use tubing diameters and wall thicknesses to really find what I hope will be the perfect combination of stiffness and comfort for a given rider.

On Chris' 953 frame, I used some of Reynolds thinnest walled 953 tubes. They are advertised at (.55 x .35 x .55) (sorry Chris, you're off by .05mm!), and they lend the liveliness that he describes. Chris is a tall guy, but not a bruiser by any means, so I chose these tubes based on our conversations about what he likes, and I'd like to think my choices were right on, he seems pretty happy. If I had a 6'5", 230lb guy who specializes in crits and hammering out of the saddle, I'd certainly have made different choices in both the tubing diameters and wall thicknesses.

To me, this is just another level of custom frame design. So many folks focus on custom geometry, and that's important, but the reality is that most folks (probably 95%) could ride some form of "stock" geometry without too much trouble. What I see as the extra benefit of the custom is this tube by tube choice of material. Let's say that a product manager is designing a stock 56cm frame, he doesn't know if the end user is going to be 125lbs or 250lbs, so there's usually some overbuilding that has to be factored into a stock frame. When I design a frame, it's with one specific rider in mind, and I try to find out their ride preferences, what they might be carrying on the bike, and where (front rack, rear rack, panniers?), in addition to simple things like height, weight and basic geometry.

Wow, you really got me going there. Hope that is enlightening?

Cheers,
Dave

Dave, this sounds great but I still fail to see how a low speed wobble, riding no hands, can be attributed to wall thickness. The wheel out of true, the frame not aligned or the weight too much on the back wheel I see all that.

[staggerwing]

Stiffness of a particular tube in bending is directly related the moment of inertia of the tube; which is:

pi/4*(outer_radius^4-inner_radius^4)

If unfamiliar with the notation ^4 means to the fourth power.

Of course, stiffness is also related to elastic modulus. And, as tubes get really thin, buckling behavior has to be considered. The quoted wall dimensions have to be getting close here.

These modern super steels are much stronger than more traditional tubing although their elastic modulus (ie spring constant) is the same as for any other steel. In short, a 953 tube can handle a much greater deformation before taking a permanent set, or in engineering speak, a plastic deformation.

Mr. Wages, based on a conversation with the original poster, built a lively, somewhat flexible ride. I hate to use the "f" word, as many seem to take it as a negative, which it really isn't. I have an early Nova Special X, and at 200lbs, may be a little large for the lightweight tubing. I can feel that bike move under me, soaking up everything from high frequency road chatter to lower frequency undulations. I can feel it "load up" as tilted into a high speed sweeper and spring back as the road straightens. It feels "alive", in the very best sense of the word. Perhaps that what the OP was trying to convey.

[Mark McM]

Dave, this sounds great but I still fail to see how a low speed wobble, riding no hands, can be attributed to wall thickness. The wheel out of true, the frame not aligned or the weight too much on the back wheel I see all that.
Dave, this sounds great but I still fail to see how a low speed wobble, riding no hands, can be attributed to wall thickness. The wheel out of true, the frame not aligned or the weight too much on the back wheel I see all that.

Tube thickness can contribute to a wobble because it affects the stiffness of the frame. A frame is not a perfectly rigid member - every load that is placed upon it will result in some amount of flex. For very stiff frames, the flex is negligible and mostly insignificant. For less stiff frames, the flex may have noticeable consequences.

We steer a bike by adjusting lean and steering angles. However, if a bike frame is very flexible, flex in the frame may also produce small changes in the lean and steering angles of the at the front of the bike. These angle changes due to frame flex are small, but can contribute to the handling feel of the bike, and even result in the wobbles mentioned.

In particular, you mention no-handed riding. When riding no handed, we rely on lateral weight shifts and leaning, and the affect these have on the front end geometry, to steer the bike. The original poster mentions that he can perceive the head tube and seat tube moving out of plane due to frame flex. Since riding no handed requires lateral inputs at the saddle/seat tube to control the actions of the steering geometry (at the head tube), you can see how out-of-plane flex between the seat tube and head tube can affect no-handed handling.

While I have never ridden large sized 953 frame, I do have first year Lightspeed Ghisallo frame, which has oversized diameter but very thin walled titanium tubing (my sized ML weighs only 865 grams). This frame was notorious for its flexiness, and my copy is no different. I too can feel the head tube flexing out of plane with the seat tube, and it too can be very wobbly, especially when descending in a cross wind.

[maximus]

that is a stunner. thanks for the write up.

we should all be so lucky :beer:

[54ny77]

i remember seeing pics of this bike awhile back.

stunning. absolutely stunning.

glad to hear it's fun to ride as well. :beer:

[David Kirk]

Dave, this sounds great but I still fail to see how a low speed wobble, riding no hands, can be attributed to wall thickness. The wheel out of true, the frame not aligned or the weight too much on the back wheel I see all that.

As odd as it sounds wobble is rarely caused by alignment or even an out of true wheel. In fact if you take a bike that wobbles and make the alignment 'bad' by clamping the wheels into the frame/fork crooked it will often stop wobbling. The reason being that the frame acts like a spring and it tends to wind up in one direction and then spring back past 'home' and go in the other direction and this sets up the pattern and the only thing that will stop it is some outside force to damp out the vibration and change the rate at which it cycles.

So if the frame is crooked it will often just stay on one side of 'home' due to the load the wheels rolling down the road place on the frame. They load it and keep it loaded. If the frameset if dead straight it is much more prone to move a bit one way and then back and so on and so forth.

This is why wall thickness has a large effect on the wobble - it changes the spring rate of the frame which changes the frequency that it will oscillate. This is true regardless of the speed at which the wobble occurs.

One thing we noticed a very long time ago was that when aero bars and/or STi shifters became popular that the number of bikes that had an issue went way up. This is because there is that much more weight out front of the steering axis to swing back and forth and set up the wobble.

If you have a bike that wobbles you can often change/fix it by doing something to change the damping of the vibration/oscillation. Different wheels, wider tires, less air in the front tire and/or more in the rear........etc can all take the problem over the edge and make it go away.

Good luck.

dave

[August]

Stiffness of a particular tube in bending is directly related the moment of inertia of the tube; which is:

pi/4*(outer_radius^4-inner_radius^4)

If unfamiliar with the notation ^4 means to the fourth power.

Of course, stiffness is also related to elastic modulus. And, as tubes get really thin, buckling behavior has to be considered. The quoted wall dimensions have to be getting close here.

These modern super steels are much stronger than more traditional tubing although their elastic modulus (ie spring constant) is the same as for any other steel. In short, a 953 tube can handle a much greater deformation before taking a permanent set, or in engineering speak, a plastic deformation.

Mr. Wages, based on a conversation with the original poster, built a lively, somewhat flexible ride. I hate to use the "f" word, as many seem to take it as a negative, which it really isn't. I have an early Nova Special X, and at 200lbs, may be a little large for the lightweight tubing. I can feel that bike move under me, soaking up everything from high frequency road chatter to lower frequency undulations. I can feel it "load up" as tilted into a high speed sweeper and spring back as the road straightens. It feels "alive", in the very best sense of the word. Perhaps that what the OP was trying to convey.


I feel pretty inadequate after reading your expurgation, but, yes, exactly.

Chris--

[Lionel]

As odd as it sounds wobble is rarely caused by alignment or even an out of true wheel. In fact if you take a bike that wobbles and make the alignment 'bad' by clamping the wheels into the frame/fork crooked it will often stop wobbling. The reason being that the frame acts like a spring and it tends to wind up in one direction and then spring back past 'home' and go in the other direction and this sets up the pattern and the only thing that will stop it is some outside force to damp out the vibration and change the rate at which it cycles.

So if the frame is crooked it will often just stay on one side of 'home' due to the load the wheels rolling down the road place on the frame. They load it and keep it loaded. If the frameset if dead straight it is much more prone to move a bit one way and then back and so on and so forth.

This is why wall thickness has a large effect on the wobble - it changes the spring rate of the frame which changes the frequency that it will oscillate. This is true regardless of the speed at which the wobble occurs.

One thing we noticed a very long time ago was that when aero bars and/or STi shifters became popular that the number of bikes that had an issue went way up. This is because there is that much more weight out front of the steering axis to swing back and forth and set up the wobble.

If you have a bike that wobbles you can often change/fix it by doing something to change the damping of the vibration/oscillation. Different wheels, wider tires, less air in the front tire and/or more in the rear........etc can all take the problem over the edge and make it go away.

Good luck.

dave

Interesting. Are you saying that the more flexible the frame is the more it "springs" and the more it can wobble?

[David Kirk]

Interesting. Are you saying that the more flexible the frame is the more it "springs" and the more it can wobble?

Not necessarily - I don't think I said this in my previous post but softer frames will wobble at a lower frequency and stiffer frames at a higher frequency.

I do think that softer frames are more likely to wobble as are larger frames (because they are in effect softer due to the length of the tubes).

dave

[Lionel]

Thanks for the details Dave !

[cuda2k]

OP,

Enjoyed the review, and always loved that Ellis frame since the first photos of it that I came across.

To both Dave's - always enjoy reading well thought out responses and honest answers from frame builders about what goes into frame design.

[soulspinner]

Great stuff from the builders and a gorgeous bike :beer:

[staggerwing]

Not necessarily - I don't think I said this in my previous post but softer frames will wobble at a lower frequency and stiffer frames at a higher frequency.

I do think that softer frames are more likely to wobble as are larger frames (because they are in effect softer due to the length of the tubes).

dave

A little more gearhead speak, but the natural frequency of a spring-mass system is proportional to:

(K/m)^(1/2) ;where K is spring constant and m is mass

Reality is a little more complicated, but the basic concept holds true. Quadruple stiffness, and natural frequency doubles; quadruple mass of system, and natural frequency halves.

Something 3-d, like a bicycle has multiple degrees of freedom, which means it will have multiple vibration modes.

If interested, checkout the Wiki Vibration Page (http://en.wikipedia.org/wiki/Vibration). Many equations to look at, but the pretty pictures cover the concepts qualitatively.

Even a great deal of math and engineering speak will not decide what works best for you. I'm comfortable suggesting an experienced builder will likely have a better understanding of what to change to achieve the required balance than the average engineer.

[zap]

Great looking bike and well written review.

[cmg]

nice to read a thread where issues are discussed in a mild manner. the Ellis is a thing of beauty, just love the paint job.

[kgreene10]

Huh, I hadn't heard of Ellis Cycles until now. My first child was born just five weeks ago and we named him Ellis. I guess I had better get him one of these (in my size, of course, so he can grow into it).

[David Kirk]

Huh, I hadn't heard of Ellis Cycles until now. My first child was born just five weeks ago and we named him Ellis. I guess I had better get him one of these (in my size, of course, so he can grow into it).

That sounds like a good plan!

dave

[oliver1850]

Yeah, name the next one Kirk.

[pdmtong]

Huh, I hadn't heard of Ellis Cycles until now. My first child was born just five weeks ago and we named him Ellis. I guess I had better get him one of these (in my size, of course, so he can grow into it).

Congrats on your baby Ellis.

This bike is truly awesome. Serrotta alumni kicking serious ass. FWIW I loved the Di2 show bike but had no coin for it

[kgreene10]

Come to think of it, he would get automatic personalization with his name on the down tube.

And I can already tell that he's going to be a great cyclist -- his suck-to-weight ratio is tremendous.

[jr59]

Yeah, name the next one Kirk.


Bravo... well played! :beer:

[Ahneida Ride]

Huh, I hadn't heard of Ellis Cycles until now. My first child was born just five weeks ago and we named him Ellis. I guess I had better get him one of these (in my size, of course, so he can grow into it).

Kelly, Kirk and Wages were THE builders at Serotta for many many moons.

Call em what you want, Master Builders or the 3 stooges .....

Bottom line, these guys all have impeccable resumes, eons of
hands on experience. Their work is top notch. The quality of their work
is undeniable.

[neiltron]

Beautiful bike, wonderful write-up. Thanks for sharing.

[BumbleBeeDave]

Call em what you want, Master Builders or the 3 stooges .....

There went YOUR discount! :p

BBD

[Ahneida Ride]

There went YOUR discount! :p

BBD

Now I am on triple secret probation .... :crap: