Custom Ti Frame Thoughts

Turboferret what was the reasoning behind the slightly flattened chainstay – looks cool but is it also to help with heel clearance/crank clearance?

That is for clearance for the belt-drive sprocket. If you check the Gates beltline document you’ll see what clearance you need, and the Di2 sits further inboard due to the size of the shifter unit.

Curved chainstays and to a certain extent seatstays are generally for heel clearance.

Cheers, Rich

Perfect, I’ll stick a Brompton wheel on the front

I’m riding 650B x 44c at the moment. It’s great. Same wheel radius as 700×23, but with more cush. Tongue in cheek name is B-

rich,

I saw these today and thought of your sharp edged holes:

Actually think this is a nice way of doing the cable holes – saves trying to explain to Chinese manufacturer about nice brake inserts and the size of these inserts. Although these do seem to be for Di2 and Mechanical only – not hydraulic.

So, I finally received all of my stuff which has been shipped home from India, among which was my bike stand. This has made all things bike fettling related much easier 🙂

I had thought that there was considerably more drag in my drivetrain than there should have been, and with the bike in the stand it was much easier to diagnose.

View from the top where you can see that the Di2 actuator isn’t sitting parallel to the belt, with some light visible towards the back of the bike and none at the front.

Sprocket looking very polished

Pretty clear where it has been polished from 😮

I haven’t really spent any time deciding how to rectify this, whether taking an angle grinder to the sprocket to reduce the width slightly is a good idea, or perhaps a better solution is to determine why the Di2 actuator isn’t sitting straight. There may be a clearance issue with the chain/seat stays and perhaps a small shim would alleviate the issue.

The good news is that I should be able to liberate a little more free speed from the bike very easily 😀

Cheers, Rich

At least you found the source of the drag. Not sure why it wouldn’t be set straight but then I’m not familiar with any of this gear…..yet. Although I think I would be inclined to determine why the actuator isn’t sitting straight to start with before resorting to angle grinder.

I’ve been thinking about the bottom bracket getting ready to touch base with china. If I’m not doing belt drive do I need to worry about any specifics to the BB – I want a shell compatible with PF30 – do you think this will be ok with alfine di2?

Some of the drag at any rate….

I’m not sure I want to angle-grind the frame at the dropouts, as that is the cause for the actuator not sitting straight! I think a small spacer will be the long-term solution sitting between the actuator and the dropout.

I haven’t been converted to PF bottom brackets, apart from the BB86 which came on my tri bike. Having said that assuming you aren’t needing an EBB, the type of interface shouldn’t really be of any concern, assuming you are happy with it constantly squeaking 😉

Alfine Di2 shouldn’t be any different from any other drivetrain really, apart from having a fixed chainline. The chainline isn’t dependant on the BB interface, so you are at liberty to go with whatever you like.

Cheers, Rich

Rich,

Have you had that much squeaking with your EBB?

No squeaks at all from mine. Perhaps I wasn’t hugely clear above – it’s the PF BBs which are renowned for squeaking, not related to the EBB.

I have had older EBBs slip, but this Niner Biocentric is a nice design and works very well.

Cheers, Rich

Have you been able to sort out the clearance now? What’s the drag like now?

I did spend a few minutes with a file the other day and that seems to have improved things slightly, but I think a small washer between the DI2 actuator and the frame is the long-term solution so that it will sit totally parallel with the hub. Normally you only have an 11/12t sprocket running that close to the frame so clearance is totally different to a big plastic DI2 hunk, perhaps I should have considered the chainstay/seatstay profile slightly more carefully.

It’s still a bit draggy compared to a normal bike, but slightly improved I think.

Cheers, Rich

On a similar idea, but without concealed cables, and a lot more £££: Indy Fab Alfine di2 with belt

Can I ask how you got the dynamo cable past the Di2 battery in the seatpost?

I’m wondering for a custom build if I could do something similar, having the light mounted on the back of the seat/rails. But i’ve got an internal battery to contend with too.

Change the wheel size

Fork rake?

cyclosis – not sure if I took a photo at the time, but I cut a small slot in the rubber bung/battery mount going inside the seatpost. This made it fairly straight-forward to fit a cable (or 2 in my case) up in addition to the battery.

This is the battery mount and you can see there is a fair amount of meat to play with.

Cheers, Rich

Last week I got caught out for the 2nd time with a flat battery on the way to work. This left me in a spinny little gear as I was going uphill as it shifted it’s last. Quite frustrating for the remainder of the ride, but fortunately after being sat in the sun for 8 hours the battery recovered enough to get a couple of shifts and get me into a middlish gear for the ride home.

The issue is that I had hidden the gear/battery indicator and charging port in the seat tube as I didn’t really want it on display on my bars. This does mean you don’t have much warning for a dying battery, and to charge it you need to remove the seat-post.

Therefore I decided to do the sensible thing and put the indicator where Shimano intended from the start!

I don’t think it looks too bad and doesn’t interfere with holding the tops of the bars.

Cheers, Rich

Rich, now you’ve had some time and miles on the bike, what are your thoughts on Alfine and belt? Are the weight and difficulty compromises worthwhile for reduced maintenance?

The weight isn’t an issue day-to-day, although my route to and from work isn’t particularly lumpy, so this might be more of a consideration to others. Having said that, it’s mainly the weight distribution which makes it feel heavy at the back. Had I wanted to save lots of weight I wouldn’t have fitted aluminium mudguards etc.

The drag isn’t negligible. I had expected there to be a bit of drag with the dynamo/belt/hub gear however it is a bit more than I anticipated. The road bike does feel lovely by comparison, and I don’t try and keep up with speedy roadies on my commute when I’m on the commuter.

Having said that, it is just for pottering to and from work, I certainly won’t be doing any sportives or particular long rides, so all it means is that I get more training daily 🙂

Knowing that I never need to lube the chain after a horrid wet ride, or charge any batteries for lights (ignoring the 3-monthly Di2 charging) is great. The fact that the brakes work perfectly whatever the weather too is another bonus.

There are a couple of minor things I would probably do differently were I to do it again, firstly being toe overlap, another drain hole in the bottom bracket shell, and probably reinforcing the cable holes to reduce the chance of stress-riser cracks sometime down the line.

I would also probably have spent a little bit more on a lighter front dynamo hub as the Shimano unit is a hefty old thing, albeit cheap.

Braking on the front over bumpy surfaces can get some juddering from the fork, so I might need to tweak my headset.

All in all, I’m happy. I was aware of the compromises I was making and I’d do the same again, but your mileage may vary.

Cheers, Rich

Having said that, it is just for pottering to and from work, I certainly won’t be doing any sportives or particular long rides, so all it means is that I get more training daily

That’s exactly how to look at it, I use my heavy racked ‘n guarded touring bike for commuting and round town load lugging and trailer pulling, it makes every ride on the light road bike seem like I’m not even trying 🙂

Have really enjoyed this thread as you’ve kept coming back with updates, which is nice, often threads like this start off well and then die off leaving many unanswered questions!

On the Toeverlap issue, do you think you’d really bother dealing with this if you were to do this again or is it a minor issue you’d ignore (I’m not tall and all my bikes have TO issues once guards are fitted to some degree). If you were going to solve it do you think you’d do it with offset, a wheelbase/toptube adjustment (and consequently shorter stem) or consider 650B wheels similar to older (mostly french) rando bikes?

Glad some people appreciate the updates, it often seems I’m just replying to myself!

On the toe overlap I’m not sure how I would address it, if at all. The bike feels identical to my road bike from a geometry/contact point perspective, which I really like, so I might just ignore it. It isn’t often too much of a problem, just the occasional tight low-speed manoeuvring when I forget that I shouldn’t have my feet at the 3 + 9 o’clock position. I doubt smaller wheels would be my solution as that would significantly reduce my choice of rubber.

Cheers, Rich

@turboferret, It’s been great reading through this thread. I’ve been looking for a bike that is almost the same as the one you’ve built.

* Carbon belt drive
* Internal gear hub
* Internal cable routing
* Hydraulic disc brakes
* Drop bars
* Mounts for fenders
* Mounts for rack — not sure I’ll use these, but I want the option

As near as I can tell, Di2 is the only option for internal gear hub + drop bars + hydraulic disc.

I’d like to see if it’s possible to route the cables from the shifters, inside the bars, inside the stem, down into the steerer, and out the back of the steerer into the down tube. I’m assuming this will mean I need a Ti fork so that it can have the appropriate holes machined without loss of structural integrity.

Have you managed to work out the source of the drag in your system? The biggest concern I have about the list I have there is the drag introduced by the hub gear & hub dynamo. I ride an aluminum road bike right now, so I’m worried I will get frustrated by the extra drag.

I like the suggestion about cutting Di2 cables and using only half of them. I assume that also means that you didn’t need to use the whole set of junctions. Out of curiosity, did you bother attaching the cables for the left shifter?

Is there anything you would change? Or any advice for a first-time custom build?

I think you’re right, the cable pull for non-Di2 IGH isn’t the same as typical Shimano road. You either go with the Versa lever or similar and don’t get hydraulic, or go with the 105 hydro levers and have the wrong cable-pull. I’m pretty sure I’ve seen some cunning devices which you can put in-line to adjust cable pull with cams/levers etc, but that doesn’t necessarily lend itself to a nice clean fully internal routing.

I wanted Di2 anyway, so that wasn’t a problem, although it did mean buying the most expensive levers around at the time!

Routing cables inside bar/stem/steerer would involve drilling some holes at contact points, where you have clamping forces. I’m not shy of drilling holes in things, but in the centre of my bars where they’re clamped – not so sure…. My one-piece stem/bar combo would in theory be easier, apart from the fact that they already have internal guides for the cables to pop out the bottom of them, so I’d have to destroy the guides to do so.

You would also have to drill the steerer, which wouldn’t be quite so much of a concern strength-wise, but would probably be tricky in terms of a star-fangled-nut or internal clamp being where the hose/cable would come out of the stem.

Also bear in mind that you would have to remove the hose/cable before making any adjustments etc. Routing the cables to the down-tube would potentially reduce the angle which you could rotate your bars before damaging the hose or pulling it tight. I know Cervelo have internal hydraulic hose routing on the P5, so it must be possible, but I also know that they are a nightmare to work on, with lots of proprietary components, so not necessarily straight-forward.

While elegant, personally I’m not sure whether the benefits would be worth the extra effort required during the build, future maintenance/adjustment issues, and compromises in strength in critical areas.

I did have some additional drag to begin with between the rear cog and the Di2 actuator, which was sat very close and not quite straight. A few minutes with an angle grinder created a little extra clearance, and all has been rose since. Having said that, the belt, IGH and dynamo hub all create more drag than their standard equivalents. Each in isolation isn’t significant, and probably wouldn’t be particularly noticeable, but combined, the bike does feel a little sluggish. My carbon road bike does fly by comparison. However, this is for pottering 20km each direction to and from work, not for racing, and adds a maximum of 5 minutes to each journey. It can be a bit frustrating when I’m tired or fighting a headwind, or when someone fast overtakes me to be on my ‘slow’ bike, but it’s all good training for when it matters 🙂

I didn’t use any standard Shimano Di2 junctions in the build. I made my own from servo cables (the same cables I used to extend the Di2 cabling) as there isn’t anything fancy in the junctions which needs to be replicated.

I don’t have anything to connect to the left shifter, so it’s completely unplugged. Perhaps there is potential down the line for something exciting like lasers or canons….

I would make a couple of very minor changes were I to do it all again, firstly a little more tyre clearance for the rear. There is sufficient width, but the mud-guard sits very close to the top edge of the tyre which means I can’t run anything bigger than 28mm. I also didn’t consider toe overlap with the guards. Changing that would have altered the geometry away from what I wanted, so I might not have bothered anyway, but it was something I overlooked. Not fallen off as a result of it yet!

Advice – go for it! Think carefully what you want, and what is feasible. If you are after anything deviating from the norm, and are having fabrication done at a distance, be very clear what you want, and make sure you can communicate that without interpretation issues. All the strange bits I could solid-model and send snap-shots and detailed drawings to ensure no confusion.

Also this thread is an example of my brain-storming and crowd-sourcing in terms of ideas/feasibility checks. Be prepared to iterate a few times to get exactly what you want/what is realistic.

Cheers, Rich

turboferret – Member
On the toe overlap I’m not sure how I would address it, if at all.
POSTED 2 MONTHS AGO

Before that ^ post I had already suggested replacing the fork for same model with increased rake (ime forks can be supplied as same model with different rake dims) or a different fork, longer rake.

I have a bike with 50mm rake forks and it feels and rides just swell.

Also, I seem to have missed the crank arm length you’re using, perhaps replace for slightly shorter ones?

Just a suggestion
🙂

My fallback for hydraulic brakes is versa levers with TRP HyRD brakes. It looks like the price difference between the ST-S705 + 2 TRP HyRD and ST-R785/BR-R785 is minimal, so if I decide to go down the Di2 road, the ST-R785/BR-R785 combo is a given.

Strangely, due to the startlingly high price of Versa 11 speed shifters, it looks like kitting out a bike with Di2 Alfine 11 AND ST-R785/BR-R785 is actually less expensive than Versa 11 + Alfine 11 + TRP HY/RD. That makes a pretty easy decision.

I had already been wondering about how to prevent the steerer from shearing the cables if there were a hole between the steerer and the down tube. I think you’ve managed to convince me that for the marginal improvement in aesthetic, it’s not worth the hassle.

My bike would be used primarily from commuting, some hauling of kid’s trailer, and running to shops. A little extra drag in exchange for lower maintenance overhead is fine, but I wonder if i can save a little drag by running a 6V battery instead of a hub dynamo. I suspect that for the weight of an alfine hub, I can insert an awfully large 6V battery. The ideal, of course, would be lights that connect to the Di2 bus! I don’t think Shimano’s caught on to that one yet, nor recharging your Di2 battery from your hub dynamo–which also seems like an obvious optimisation.

Thanks for the pointers about guard clearance and toe overlap. I’d already made a note to watch those after reading through your thread.

It is a little bit crazy to have a dynamo hub powering lights, but have to charge the Di2 battery separately, but that’s what I do at the moment. LiPo batteries can be a bit fussy about how they’re charged, so I didn’t want to take too many risks there, especially as the Shimano batteries aren’t quite as cheap as a typical 18650 cell from Hong Kong!

I would be a little wary about running lights etc from a battery wired into the Di2 loom. I’m sure you’re aware that the new Shimano system runs CANBUS and has signal and power on a pair of wires. Not sure how it would react to having other components also wired in. Who knows, it might be fine – you can pioneer it for us all 🙂

You are right though, the Alfine dynamo hub is a pretty heavy lump, and something which I would probably have swapped out for something a bit lighter had I fully appreciated, but I don’t think it’s worth rebuilding my wheel now to do so.

Not having the nasty graunchy grinding chain on a Monday morning after chucking the bike away wet on a Friday evening is certainly a bonus of a belt!

Cheers, Rich

I don’t have anything to connect to the left shifter, so it’s completely unplugged. Perhaps there is potential down the line for something exciting like lasers or canons.

my vote goes to a Raleigh Wildcat/Vektar style box of sound effects 🙂

Dear Rich TurboFerret,

I’ve been admiring your bike build and I’m so impressed that I feel itchy to build one like yours. You’re very detail oriented, your bike looks like a piece of art. Love all the details as you go above and beyond with the beautiful internal cable routing and reliability.
If you don’t mind, I wanted to ask you a couple questions before so I could avoid any mistake.
I believe you’ve been with your bike for more than 6 months. What do you think it will be next bike build ? Do you feel there is any room for improvement? I’ve read through the whole thread so I already understood you would fix the toe overlap issue, and have a lighter dynamo hub, besides that anything else ?
How’s the belt reliability, is it really maintenance free or you would still consider a chain ?
Should I worry about chainstay stiffness as Gates recommends and Rohloff requires ? Frame Stiffness Test
Do you have any issue with chains slippage ? Tension ? Would you beef up the chainstays ?
How about belt alignment, I don’t know much about frame geometry, how did you ensure your rear sprocket aligned perfect with front belt ring ?
I’ve read about your EBB choice and the change from 68mm to 63mm to compensate, I’ve thought about using this EBB: Wheels EBB . Should be similar to the one you got from niner. What do you think ?
Does your EBB have enough play to give chain tension on multiple sprocket combinations or is that a concern ? What’s your BB Shell width ? Are you using the Niner EBB Biocentric II ?
How do you like the Alfine Crank, would you recommend

The super-commuter I aspire, would be more based on a cyclocross geometry. I feel that I could try cyclocross on it besides commuting. While the main goal being commuting.
[list]Few changes I find interest:
[*]I would sacrifice the dynamos for rechargeable batteries. I don’t mind leaving it charging as the bike parks inside the apartment at night. So it should save a couple watts of resistance.[/*]
[*]Add mount for rack, if ever needed.[/*]
[*]Add tire clearance for 35mm mud tires.[/*]
[*]Everything else would be like yours. Alfine Di2, hydraulic, internal cable routing, mudguards.[/*]
[/list]
What would you do to reduce weight/resistance while keeping reliability ?

I don’t think I would do anything about to overlap, that would change the geometry, it’s just something to be aware of during low-speed manoeuvring.

Belt seems to be running fine, according to Garmin Connect I’ve ridden ~3500km on the bike this year with no belt-related issues at all. Sprocket wear doesn’t seem bad either. No maintenance on the belt at all. While a chain would certainly reduce the drag a bit, I would need to lube it occasionally, so pros and cons.

I didn’t really worry too much about frame stiffness at the time of speccing the frame, but the rear triangle is all plain gauge tubing of fairly beefy wall thickness, so it certainly seems rigid enough. I’m not a massive powerhouse of a rider, being only ~63kg, so I’m not stressing it as much as some riders might. I’m generally riding at a relatively sensible pace, but certainly don’t hold back at times, and I’ve never had any issues sprinting up hills etc.

I knew alignment would be critical, so I checked the belt-line on the Gates website for the Alfine Di2 sprockets. They have a specific Di2 sprocket which changes the alignment to keep the belt away from the actuator, which does bring the belt quite far inboard. I went with the Alfine crankset as this was a known factor which they design the front sprocket around. Due to the belt-line being so low I needed a very narrow drive-side chainstay.

The Biocentric was suggested by triple_s on page 1, and that’s what I’ve used, albeit slightly modified to fit the 63mm BB shell. The Wheels EBB looks pretty similar, working on the same principle of clamping on the outside of the shell.

The EBB gives just the right amount of adjustability based on the Gates website, but I did tweak the chainstay length slightly to suit the belt length over my original spec. I can’t use any different size sprockets, but I have the full range of my road bike with what I have fitted, so I can’t think of a scenario where I would want to change.

The Alfine crank isn’t particularly exciting, but I have no complaints. It does exactly what I want, is cheap and was guaranteed to work straight out of the box.

Without the belt and dynamo, the IGH might be an advantage for really muddy CX racing where normal gears might struggle, but I would have thought that most of the time a dérailleur would be faster.

I do have rack mounts, so can add a rack later if I want.

Tyre clearance would be something I would certainly change, along with reinforcement rings around my internal cable holes. Nothing much else springs to mind.

A fair amount of weight could be saved with butted frame tubing and choice of components, at a cost. I think the only change to the spec I would make were I to do it again would be a lighter front hub. Everything else is fit for purpose. It’s a utilitarian machine to get to and from work, and it does that pretty well.

Cheers, Rich

I’m in the process of building my n+1 formula busting bike and I’m really interested in what all the knowledge on this older thread has to say.

I’m wanting to run hydraulic shimano disc, thru hubs, internal cables and rack/guard fittings. I’m pretty sure I’ve shoe-horned it all in but am hinged on two nulloptions.

1. Run the r\derailluer and brake along the top tube then seat stay with either r785 or rs785 calipers. But will this foul on my guard and rack mountings?

2. Run the rear derailleur through the top tube then seat stay and then the brake through the down tube and chain stay. My concern here is the weakening of the CS at the hole next to the BB?

Can anyone shine any experience on this or any other tweaks I may need to do on my frame?

Thanks in advance.

Will

This forum should support pictures. Magrela is ready, rides great and fits like a glove.