The Great Internal Cable Routing Debate

[Daniel_Y]

@crowe-molybdenum thanks for doing all the digging. Aside from the TINY bearing, it is honestly not the worst idea:

• 42mm OD 34.1mm ID bearing
• 5.4mm brake
• 1.125in (29.8mm) steerer

There is just enough room to fit two brake cables:

However looking at their wedge, it looks D shaped and eccentric:

Those sharp corners make me a bit nervous, that was the whole reason why specialized had to recall their SL7:

The first revised part is a considerably lengthened aluminium expander wedge that now runs from the top of the steerer tube all the way to below the height of the top headset bearing. […]. The second part (Figure 2, B) replaces the culprit of the issue, the metal compression ring itself. The new version is now two pieces, with a stainless steel sleeve sitting between the compression ring and the carbon steerer tube.

Well, we can only hope that these designs were lab-tested and tested in the real world. The strada system could turn out to be a good option for metal bikes. So far the Columbus fork is the only player in the small-upper-bearing-market: Futura Trefoil Road Fork – Columbus

[mark_pmw]

Thanks to all who replied, now I’m getting it.

@crowe-molybdenum

…but is offset rearwards from the steering axis…

However, the steering axis is defined by the steerer tube, so as I see it, the headtube simply has more room towards the front of the frame for cables/hoses.

@crowe-molybdenum

…the fork and compression ring having bearing interfaces which are not concentric to the steerer.

This is simple enough, but in practical terms, how are the top and bottom bearings aligned? If one bearing is installed 180 degrees out of phase, it seems as though it’s an inadvertent adjustable headset. 90 degrees out of phase is really good at turning in one direction. Unless the bearings and headtube are mechanically keyed, some unfortunate mechanic is going to miss the point and botch the installation.

@Daniel_Y

Aside from the TINY bearing, it is honestly not the worst idea:

A tiny bearing at the top should not be a problem, as the load here is almost entirely radial. Headsets are rarely replaced because the top bearing has failed.

From a manufacturing perspective, a simple solution for metal bikes is difficult.

Thinking out loud: How about external cup headsets with eccentricity in the bearing cups? The actual bearing can be an off the shelf item, with the eccentricity built into the cups where they press into the frame. The indexing of the cups is crucial; I don’t have a fool-proof answer. External cups keep the OD of the headtube to a reasonable size, which is a strong positive for metal frames. If internal routing is promoted as an aero feature, are EC a non-starter?

Just what we all need, another headtube, EC37/44!

[Neuhaus_Metalworks]

EC37/44 could easily replace 34/44 if headset manufacturers would buy in since EC37 does everything EC34 does.

[crowe-molybdenum]

the axis around which the steering mechanism rotates, needn’t be the steerer tube. in a normal system, the fork pivots around the centre of the bearings, and this won’t change, but as I see it there’s no reason the steerer tube HAS to be centred in the bearings.

the point of this excercise (for me) would be to keep the headtube small, but have enough space between the inner edge of the upper bearing and the steerer wall to fit all the cables through the top bearing without messing with the steerer tube.

I see it working with a reasonably normal IS headtube a normal steerer tube and the upper bearing described above. the head tube would reamed and faced with normal IS tooling.

the sneaky parts are reasonably simple;

special ‘crown race’ which in my mind is simply machined in place with the steerer in a four jaw chuck.

a compression ring for the upper headset not dissimilar from what we see above, (though maybe not designed by Niki Terpstra’s sleep demons like those ones) . yes, you’d need to get it in straight, but its slip fit; I think that’ll be doable, and it’ll be super obvious its eccentric to anyone holding it, given one side will be 5mm thick, and one will be nearly 0…

the fork would need to be built with its desired offset measured from the steering axis, not the steerer but I don’t see that being problematic.

the EC idea I think I need to ponder… that pressed in part of the cup takes up some room inside the headtube around the steerer, which is going to be pretty hot property…

im going to try get my hands on a couple of those 3t bearings, and have a play around/ check some actual dimensions.

im not sure if im allowed to ask this here , but was there a bearing-industry guy keeping active on #whispers# V-salon? does anyone remember who it was?

[mark_pmw]

I did not see that coming! Maybe I’m thinking too conservatively, I never thought of having the steerer not centered in the bearing.

Sneaky parts:

@crowe-molybdenum

special ‘crown race’ which in my mind is simply machined in place with the steerer in a four jaw chuck.

This is good for one-offs, but eventually there needs to be an interchangeable standard part that is serviceable by the rider. We’re back to proper indexing of the crown race with the compression ring. I have to be honest, it really worries me that it can be correctly done by even a seasoned mechanic.

@crowe-molybdenum

it’ll be super obvious its eccentric to anyone holding it, given one side will be 5mm thick, and one will be nearly 0…

Right, it’s easy to see that it is eccentric, but by eye, it’s difficult to spot the very thinnest part. 10 degrees of radial movement is about .001" in thickness variation of the compression ring. Maybe this simply means that exact radial alignment is not critical, or if there’s an mark at the thinnest point, it can be eye-balled within a couple of degrees. Count me as a skeptic at this point, but I’d be happy to be shown that exact alignment is not an issue. Maybe I just wrote the reasons why it could work.

I may be wrong on this:

@Daniel_Y

Aside from the TINY bearing, it is honestly not the worst idea:

Within a minute of my last post, a headset manufacturer I’ve been consulting with let me know that they are nervous about small bearings, and that they would not be trying them anytime soon.

@crowe-molybdenum

im not sure if im allowed to ask this here , but was there a bearing-industry guy keeping active on #whispers# V-salon? does anyone remember who it was?

This may be Matt Harvey with Enduro Bearings. I haven’t talked with him for awhile, but I’ve found him to be very knowledgeable and friendly. Might be worth a try to contact him.

[Spencermon]

@mark_pmw

This is simple enough, but in practical terms, how are the top and bottom bearings aligned? If one bearing is installed 180 degrees out of phase, it seems as though it’s an inadvertent adjustable headset. 90 degrees out of phase is really good at turning in one direction. Unless the bearings and headtube are mechanically keyed, some unfortunate mechanic is going to miss the point and botch the installation.

I’m just thinking out loud, but most of these upper headset bearing compression rings are keyed to interface with the proprietary headset spacers. I know that’s the case with FSA. Those spacers are also keyed to interface with the stem. Would it be possible that the upper headset compression ring is aligned by aligning the stem with the front wheel?

On a related note, with the steerer tube offset rearward of the steering axis, the labeled stem length would actually be a touch longer than the measured stem length from the steering axis. I wonder if this is accounted for in Geometry charts or if, since the stems are proprietary, they account for the length difference and label the stems differently?

[SoyWater]

@mark_pmw

How about external cup headsets with eccentricity in the bearing cups? The actual bearing can be an off the shelf item, with the eccentricity built into the cups where they press into the frame. The indexing of the cups is crucial; I don’t have a fool-proof answer. External cups keep the OD of the headtube to a reasonable size, which is a strong positive for metal frames. If internal routing is promoted as an aero feature, are EC a non-starter?

You’ve basically described a reach adjust headset. I was thinking of this when I first started lurking in this thread but figured I was out of my depth. Might be a starting point if someone wanted to springboard off it.

Reach Adjust Headsets

[mark_pmw]

Well, great minds think alike! I did realize a flaw, in that the cup is stationary, and the cables must move with the steerer. That means a big slot for the cables, which means it’s difficult to keep water and other debris out. I’m thinking of going deeper into the idea that the steerer is eccentric to the ID of the bearing.

Even the Wright Bros. spent years talking about stuff before they flew!

[mark_pmw]

@Spencermon

Would it be possible that the upper headset compression ring is aligned by aligning the stem with the front wheel?

Exactly! And the crown race is aligned with a witness mark to a cardinal point of the fork, such as straight ahead.

I’m way too into this, as my intention has been to come up with stem components that work with internal routing. That may be solved by having the correct headset interface, such as King or FSA. I’m going to take a break from this so I can my other work done!

[Spencermon]

Again, thinking out loud. I wonder if there is a way to make the headset spacers interface with the rear expansion slot of a stem. I think most stems align the stem bolts centered in the rear. From a production standpoint, you would want the system to work with any aftermarket stem or in a way that any frame builder could make a compatible stem easily. The fewer proprietary parts the better.

This was just my 10 second CAD explanation.

[SoyWater]

@mark_pmw

Well, great minds think alike!

And, strangely, it seems like we do too

@mark_pmw

I did realize a flaw, in that the cup is stationary, and the cables must move with the steerer. That means a big slot for the cables, which means it’s difficult to keep water and other debris out.

I had not considered that. You know, I really didn’t wanna spend any time thinking about headset routing, but looks like that’s happening. Damn it.

[Deep]

So, I got my alloy custom bike and have been riding it for a few weeks now. Took me a while to nail the fit. Couldn’t understand why I kept getting lower back pain until a fitter checked everything and told me I had a hip impingement and was using using cranks that were simply too long for my legs.

Anyway, everything is sorted and I am riding happily now and after following this entire thread, I have to say, I don’t think fully internal/integrated cable routing is worth it. Completely agree with Daniel that a “forever bike” should permit changes in fit and components as we go along.

So I guess the Ti bike is going to have “normal” guided routing through the downtube.

[Daniel_Y]

@Deep

So, I got my alloy custom bike and have been riding it for a few weeks now. Took me a while to nail the fit. Couldn’t understand why I kept getting lower back pain until a fitter checked everything and told me I had a hip impingement and was using using cranks that were simply too long for my legs.

Very cool! What crank length did you go down to?

@Deep

I have to say, I don’t think fully internal/integrated cable routing is worth it.

You never know until you try.

@Deep

So I guess the Ti bike is going to have “normal” guided routing through the downtube.

Now that you have a good baseline, you can design a better bike for yourself. If you know you are running shorter cranks, you can lower the BB or bump up the stack accordingly.

[Deep]

So, the geometry of this bike was based on the Trek Domane, which I had thought was pretty comfy. The one thing about that bike I had noted was the BB drop - 80mm. That’s what my custom bike has as well. The lower centre of gravity means I am much more stable at high speeds, can descend and corner with much more confidence, and don’t get wobbly at really slow speeds.

About the cranks, bike fitter Rick Schultz shared a formula for caluclating your ideal crank length. According to that, my ideal crank length would be 155mm. But my fitter suggested that I go for 160, which he thought would be a nice balance. It has worked out well so far.

Also, the hip impingement means that the typical narrow road bike q factor is very uncomfortable for me. So, Bike fitter suggested I add pedal spacers. Got a pair of 20mm steel ones off amazon. Have worked out extremely well for me.

No more knee pain or back pain and I can ride for hours.

[PineCycles]

Daren Baum blowing my mind once again.

F Y X O on Instagram: "Do you even engineer, bro? Absolute wizardly in cable...

[crowe-molybdenum]

alright, a bit of thread necromancy here, but here’s an off the shelf headset that moves the steerer CL off of the steering axis, called a “virtual pivot headset” the intended purpose is to reduce fork offset (for absolute pinners on DH rigs), but it accomplishes what we’ve talked about, though no internal cable porting.

VPH Virtual Pivot Headset – EMRG

its only available for 44/49 head tubes or larger, but its a purchasable item that speaks to some aspects of an earlier point in the discussion that was somewhat unresolved.

in other news i’ve bought the front end assembly off a new cervelo S5 (which is as pictured above) and am going to press on with the “steerer-delete” prototype; I appreciate this isn’t really a functional model for further production, but I have a crush on Wout, I like the swoopy shapes, and ill do what I want.

[Batcountry]

I don’t know if it’s better to bump this or start a new thread, but here goes!

I am building a few more identical BMX frames, and I would like to do internal routing. Usually I would not, but it solves another problem, so I’m seriously considering it…

Based off of the FEA @Daniel_Y posted, it appears that the side of the tube is the best from a strength perspective. My question is, in my particular situation, is there going to be any overly adverse strength impacts? I think it will be ok, but I just want to get some other opinions before I do this. These are bmx bikes that will occasionally take big impacts w/ no suspension. Blue dot represents where I am considering having the tubing port- leaning towards the Cinelli type…
Thanks for any input, 509mm length, 1.5 x .049 straight gauge downtube:

[crowe-molybdenum]

@Batcountry

but I just want to get some other opinions before I do this

purely opinion here, and i’ll welcome some contradictory opinions or data, but I would pretty strongly suggest, most of the considerations discussed and taken around drilling holes in main tubes and then reinforcing them with weld/braze/solder-in ports are pretty notably not from a “crash-survivability” perspective, they’re from a ride-the-frame-without-it-failing perspective.

in ten years i’ve had two mountain bikes and a road bike of my making, notably fail in crashes, and they each failed where I had added something to a downtube, some in better places, some in worse. ive seen a bunch of other wrecked bikes in my time and pretty overwhelmingly feel like stuff on the downtube is disproportionately involved.

I know BMX is a relatively wide net these days, and my opinion of it is likely based on some tacked-on-the-end segments in skate videos from 15 years ago… but if people are going to be rough on those bikes all, I wouldn’t flirt with the downtube.

[Batcountry]

I really appreciate your thoughtful reply. I was on the fence about this, and having talked to another trusted builder I already went for it

[chris_n]

Moots, apparently: