[AdamR83]

I think this will mostly end up a 'photo dump' and a place for me to scribble down some stuff I'd otherwise forget, but a small intro...

This will be the 3rd frame I've built of this style. The first two were 26", one from cheap ERW (gave the bike to a mate and it's still going!) and the second TIG Brazed in Columbus (also still going): Steel Singlespeeder Home Build

V3 (below) will be 853 / Columbus mix and 29". I'd always thought 26" wasn't dead until I rode a mate's 29er down a known bit of techy trail and it absolutely blew me away. It definitely seems to be more about geo and parts selection than wheel size - except when it comes to rolling over lumpy stuff!

Anyway. I spent a while nerding out with BikeCAD (huge thanks to Brent for the great price on a 'diy-er deal' for the Pro edition) and some geo charts of other bikes I'd ridden and liked, came up with the following:

The plan is to end up with something very all-round-y - a quick blast up the woods with mates, an all day epic or even some short course XC racing / endurance racing. I have a SID fork I can swap out which slackens the static angles by about a degree, but I've built the bike around rigid forks as I love the direct feel (and low maintenance!) of a very simple bike.

I have ridden trials for about 25 years so wanted shortish stays, low standover and a huge dropper to get the saddle right out of the way. The dropouts have 9.5mm of movement which is enough to account for one chain link difference in chainstay length (eg. I plan to run 32:20 as it's hilly round here, but the frame should also take 30:20 for a lighter gear ratio, and 32:18 for a harder one, both with a 6.35mm / 1/4" increase in chainstay length). The rest of the adjustment is to allow for chain stretch. I just hope I can manufacture it accurately enough!

I am also a previously-recovered self-confessed weight weenie in relapse (and proud, ha). Interestingly I find reduced weight - especially on a rigid bike - helps more downhill than uphill. The bike really loves to skip around and get out of the way of bumps, sure it can get a bit lively but if you let the bike go on its merry way it's a lot of fun.

Parts-wise I've gone fairly pimp, so given a projected ~2kg / 4.4lbs frame weight it should be a smidge over 9.2kg / 20lbs with the SID fork (750g / 1.6lbs less rigid). I know that frame weight sounds very light but, as the previous frame, I have worked hard on reducing the bulk of 'lumpy bits' like BB shell, dropouts, etc - they hide SO much weight, especially in steel frames. I'm also not a heavy rider and don't really 'send' stuff, so I can get away with light bikes.

I'm not trained in any of this, but been in the bike industry over 20 years now and love sticking lumps of metal and plastic together in odd ways with various amounts of wheels hanging off the bottom. Such as...

And...

Oh yeah, working title is the ARSE 29er. Adam R Singlespeed Experiment: 29er.

So there we have it. Intro done. Tips from more seasoned frame-builders, questions, or anything else along the way are always welcome!

 

[AdamR83]

Photo dump round 1...

 

[Steel-Onions]

Wow!, what a great project, the chainstays and bb shell looks so clean,simple and beautiful !!, great work

 

[*OneSpeed*]

Cool project, subscribed.

The only thing that caught my eye was the short amount of chain tension adjustment in the dropouts. I would want more range, something like 20mm.

I understand the desire for short chainstays, and you'll still be able to achieve that, but if you buy nice cogs and need to change your ratio combination, or add a little chain stretch, you can run out of room quickly with less than 1cm. You'll probably still be able to make it work but it could mean swapping the chainring, and may need a new chain more frequently.

Just seems like more of a hassle than the minor weight savings are worth. (I am NOT a weight weenie and I need strong things, more metal please). Plus what if you want to pass the frame along to someone else in a few years, tiny adjustments may not be so awesome over the lifetime of the frame.

Other minor notes: The STA may be a little steep for a XC hardtail (for me), and I'm assuming you checked that the suspension fork crown clears the downtube? I like the BB drop though, not too low! Great pics too, thanks for sharing.

 

[AdamR83]

Thanks guys. Hope you find some other stuff in the thread interesting too!

I hear you about the dropouts, OneSpeed. I do agree that more movement would ultimately be better, and considered making that compromise. My current bike has 8mm of movement and I've never found I need more. Have used 3 different ratios and been through 3 chains from new to dead with no issues.

With the setup I've just built above, this will work with 10 different gear ratios between 87% (easier to pedal than) and 125% (harder) of my 'go to' 32:20, using 28T, 30T and 32T front rings. I haven't investigated 34T up front as I don't have one, but I would hazard a guess at that opening up another 2 or 3. (All providing I've built it right!)

The weight saving was worth it to me as it's in a critical area - chopping weight off the rear end of the bike makes such a huge difference to how the bike rides in my experience. I love feeling it popping and skipping all over the place. Big rocker / slider dropouts are at least twice the weight. I know that's not a lot in absolute terms, but if everything on your bike was twice as heavy...

Seat angle - yeah, I agree with you there. I would have liked a little slacker, but also didn't want to have a manipulated tube. I actually drew the thing up with a split / twin seat tubes to start with, and 406 stays, but decided against going too wacky, ha. I've toyed around on another bike with sliding the saddle right forward to get the same seat position relative to the BB axle in X and Y planes, and it feels ok. I have a short torso which helps. I figure nothing is perfect....!

Good call on fork clearance too. I can't envisage a problem, (fork is only a weedy SID) but will load up BikeCAD now to have a check - thanks for the heads up!

 

[AdamR83]

Today's photo dump. Some lucky welds (must be tig Tuesday) balanced out by a horrible tip n dip followed by a burn through one of the frame ends, and a horrible 'fix'. Ack!

Practice makes... slightly less crap.

The good...

The bad and the ugly!

Running too hot...?

Head angle adjustment...

BB drop adjustment...

Seat tube 'prop' to ensure adequate spacing from tyre.

Straight off the lathe...

Tidied!

Had some issues with head tubes wanting to make a break for freedom in the past, so as well as using a longer head tube and spacing the DT further away from the TT for more support, I am making sure this joint is as good as I can.

 

[dr.welby]

With the setup I've just built above, this will work with 10 different gear ratios between 87% (easier to pedal than) and 125% (harder) of my 'go to' 32:20, using 28T, 30T and 32T front rings. I haven't investigated 34T up front as I don't have one, but I would hazard a guess at that opening up another 2 or 3. (All providing I've built it right!)

I'm in the process of a similar concept - I don't really need every single combination of gears so I picked a magic gear range that works for me (30/18, 32/20, 34/22) and designed the dropout to only adjust for chain wear.

Kudos on the tooling too, lots of clever ideas there.

 

[Grinchy8]

Great work!

 

[#/er]

Love the perfect miters
Why put the dropout adjusters at the rear of the dropout? The chain will be pulling the axle forward away from the adjusters (unless I’m missing some sort of tug device).

 

[AdamR83]

I'm in the process of a similar concept - I don't really need every single combination of gears so I picked a magic gear range that works for me (30/18, 32/20, 34/22) and designed the dropout to only adjust for chain wear.

Kudos on the tooling too, lots of clever ideas there.

Thanks! You know, it's kinda comforting to know I'm not the only nutter out there who builds a frame around a gear ratio, ha. May I ask if you have any photos of your solution?

Great work!

Thank you!

Love the perfect miters
Why put the dropout adjusters at the rear of the dropout? The chain will be pulling the axle forward away from the adjusters (unless I’m missing some sort of tug device).

You're absolutely right - and that's a very good question. I toyed with having them in the front, but ended up with this solution for a few reasons:

1) I have some experience with trials bikes using the Hope hub with no tensioners on horizontal dropouts. The axle bolt washers and the end caps of the hub have steel serrations which bite into the frame pretty hard. One of these bolts done up at 40Nm (low end of torque spec) creates ~2 tons of clamping force.

A chain breaks at about 1000kg load. Given a 170 crank and a 32T ring, that means I'd have to put ~375kg in at the pedal (not happening!) to break the chain. However, if we work on that as a worst case / maximum load scenario, the coefficient of friction of this interface has to be around 0.5 for it to hold. This is based on 2000kg clamping force * 0.5 CoF = 1000kg forward sliding force (maximum breaking load of the chain) that has to be resisted.

Steel on steel CoF for flat surfaces, dry and clean, is quoted as anywhere between 0.5 and 0.8 - that's without serrations - so I figured it would work, and it does in practice.

2) I wanted the chainstay attached to the 'front' of the frame end (dropout) rather than around the 'bottom' so the pedalling loads aren't trying to rotate it, which would in turn add load to the weld and flex into the system. To have the tensioners in the front, I'd have had to drop the chainstays right down as the frame ends are 32mm diameter and the chainstays 13mm (see photo below), and then still had to do some funky stuff to make the tensioner screw clear the chainstay / frame end weld.

3) It works great on my current bike (you can just see the (black) bolt head peeping out there). In fact, when I sat down and throught about the best and worst bits of this bike, the dropouts came right at the top of the list.

This is how they work... Loosen the tensioner screws off, pull the wheel back by hand. Tighten the drive side axle bolt up when you have chain tension you're happy with. Drive the tensioner screw into the hub bolt (can just be done by hand). Then get the wheel in the centre of the bike by hand, and tighten the non drive side axle bolt up. Bottom that side tensioner screw into the hub bolt. Then set your rear brake up accordingly.

To remove the chain, just loosen the bolts and the wheel slides forward. Putting the wheel back in, it always goes back in exactly the same place because the hub bolts use the tensioners screws as a stopper. Simple.

With the bolts in the front of the dropout, you would have to undo them every time you want to take the wheel out - and getting them back in exactly the same place is tricky, plus it's more faff especially with the bolt head being sandwiched between the seat and chainstays.


One thing I'm not 100% certain about yet: the hub I'm using this time doesn't have the serrations, and the end caps are aluminium. CoF for this interface is said to be 0.61, so I think it'll still work just fine - but if it doesn't hold at least I have the option to add serrations to either end caps, under the bolt heads, or both; in which case I'm back to exactly the same setup as the above photo which I know works perfectly.


And the award for longest winded post goes to......

 

[Joe Nation]

I have some experience with trials bikes using the Hope hub with no tensioners on horizontal dropouts. The axle bolt washers and the end caps of the hub have steel serrations which bite into the frame pretty hard. One of these bolts done up at 40Nm (low end of torque spec) creates ~2 tons of clamping force.

A chain breaks at about 1000kg load. Given a 170 crank and a 32T ring, that means I'd have to put ~375kg in at the pedal (not happening!) to break the chain. However, if we work on that as a worst case / maximum load scenario, the coefficient of friction of this interface has to be around 0.5 for it to hold. This is based on 2000kg clamping force * 0.5 CoF = 1000kg forward sliding force (maximum breaking load of the chain) that has to be resisted.

Steel on steel CoF for flat surfaces, dry and clean, is quoted as anywhere between 0.5 and 0.8 - that's without serrations - so I figured it would work, and it does in practice.

A+ for showing your working-out

 

[sikocycles]

Great job!

 

[AdamR83]

A+ for showing your working-out

Always got into trouble at school for not doing it... seems it finally had an effect!

Todays photo dump...

Cheeky tip - I don't have a 46mm holesaw needed for the head tube and BB shell mitres. However, if you take a 44 and 'nudge' it off axis by 1mm (DTI works well to check this), you'll make a 46mm hole. You have to go a bit slow as it makes a very wide cut, but does the job perfectly.

Good enough for me...

First TT mitre done using the technique described above - 44mm holesaw and 46.5mm head tube.

And another little project within a project... I'm making some spokes. So I've brought my hubs home to tidy up the spoke holes.

I can see some confused faces, and rightly so. More info here... Make your own polymer (UHMWPE) spokes?

 

[dr.welby]

You know, it's kinda comforting to know I'm not the only nutter out there who builds a frame around a gear ratio, ha. May I ask if you have any photos of your solution?

Here it is, I haven't drilled it for the grub screw stop yet. Idea was a vertical dropout where you could let the wheel settle back a bit to take out some wear. I had some other stuff to laser cut so I threw it in with the job. And since someone is bound to ask, yes, it's designed for a 1/2" x 1" rectangular chainstay.

If I hate it I'll just get some normal horizontals cut.

 

[AdamR83]

Thanks! And that looks brilliant. I see no reason why that won't work as you intend. I guess alignment will need to be spot on since you're dealing with wheel movement in two planes rather than one when you adjust axle position, but it sure is neat!

It's pretty similar to my previous frame (I had a ~30 degree angle to the slot), and basically a miniaturised / modernised version of the old classic.

 

[Joe Nation]

basically a miniaturised / modernised version of the old classic

True, but we stopped using them a long time ago! Jk, do whatever you want, you'll know what's right.

 

[AdamR83]

True, but we stopped using them a long time ago! Jk, do whatever you want, you'll know what's right.

Haha. This is true. I wonder why? Industry obsession with making bikes as complicated as possible perhaps?

Today's dumping...

Phasing.

I've always found this joint very tricky. Glad to finally do a decent one!

Jig becomes prop...

Better than the last frame I did. I'll take that. Bit short on one of the stop-starts but I doubt the head tube will fall off because of it...

Almost looks half decent. A bit of paint will cover the worst of it!

Any opinions on whether this'll need reinforcing? It's 0.8mm thick 853, 34.9mm diameter, and the butting is where those two marker pen lines are (starts tapering to 0.55mm at 20mm from the top of the slot). I think it'll be ok but experienced eyes always welcome...

Had another look at geo this morning too - seat angle is now 74.5 rather than 76. Achieved this by offsetting 7mm more at the BB shell. Also gives 3mm more rear tyre clearance and sets the saddle further back by 15mm.

 

[mickuk]

Good progress - I'm going to have to get my arse round for a visit before you finish the damn thing!

I'm a coward so would probably silver solder a little patch around the dropper slot just in case. I often just use a MAPP plumbing torch for soldering guides and little stuff.

 

[AdamR83]

Yes, you might Mick! Been busy at it again this morning and it's starting to look like an actual frame.

Thanks for the heads up on the slot. I've got some silver soldering stuff ready for cable guides etc so will see how brave (or not) I'm feeling while those jobs are being tackled!

Back purging worked a charm. Little tidy with a file, post goes straight in.

Really good fit up vs not so good fit up (had to fettle the seat tube where it overlapped the existing weld). Can you guess which is which? A reminder of why it's worth putting time and effort into the mitre process!

There's about 40g to come off each seat stay. Add disc mount, braze ons... might just nick under the 2kg target (scale shows 2001g there).

Had a message from my mate the other day: "Do you have a 44mm reamer / facer?".

Pre-empting the next question, I decided it was a good time to perform said operations...

Obviously, like my last frame, I am not doing normal things with the BB... I've toyed with an angular contact setup on the current bike and it's worked out really well, despite being an afterthought / bodge, so this time the BB shell is cut from a 250mm length of roadie spec Columbus head tube - 46mm OD, 1.1 wall. Reamed out it's come up nicely.

The plan is to turn down some 'normal' headset cups which I have knocking about and squeeze them in with a tight press fit (need to check the numbers still). These will then be reamed out in situ to a Campag (45/45 degree, 41.8mm OD) fitment, and I have some King Dropset bearings to go in. Combined with the proper aluminium threaded lockring on my cranks and a couple of decent quality 7075 1 1/8" fork crown races, I am hoping it will stay tight, not creak, and spin nicely. Time will tell! Worst case I can squeeze in / weld in some steel sleeves and go to a BB30 style system.

Head tube came out nicely too, only a very small area untouched by the reamer. This was a Reynolds / Bear Frame Supplies one (631 material), seems better than the HT44 cromo version from Ceeway.

 

[dr.welby]

Combined with the proper aluminium threaded lockring on my cranks and a couple of decent quality 7075 1 1/8" fork crown races

Ha ha, best hijacking of one standard for another, ever.

I'm also being cheap/lazy and using 44mm head tube for the bottom bracket, but I'm just going to turn some quick acetal cups to fit a 41/24 bearing.