[natzoo]

Happy New Year!

The past year I started working with Nick at Neuhaus Metalworks to design and build some pretty cool bikes. We have made a lot of progress, so I thought it would be good to do a recap to benchmark our progress and serve as a reference/inspiration for other framebuilders.

To give some background, Nick is the framebuilder, I am the armchair engineer/CAD monkey. We work together to figure things out. I bring his bikeCAD into the 3D CAD world, handle the 3D printing process, then he builds the frames.

3D Printing:
Over the past year, we spent several hundred hours and several thousands of dollars designing and printing prototypes. I found that the design freedom of 3D printing has really enabled some really cool workflows.

After all the expensive dust settled, I think 3D printing is the future of custom framebuilding. The only limiting factor is people’s CAD skills. Better fire up those Fusion360 tutorials. I think 3D printing is more accessible than most people realize.

How to build a rear end:

Chainring and tire clearance:
One of the most challenging aspects of designing a modern mountain bike is getting enough clearance for the tires. The worst pinch point happens at the chainstay, between the tire and the chainring.

These are the clearances I design to:

• Tire clearance: 6mm on each side
• Chainring clearance: 2mm

SRAM publishes their crank and chainring and dimensions: https://www.sram.com/globalassets/d...ons/mtb/2021-mtb-frame-fit-specifications.pdf
Shimano’s are not published but are floating around. They are slightly more conservative than SRAM's

Tire Clearances:
Go/no go based on 6mm tire clearance, 6.5mm thick yoke, and 2mm chainring clearance.

Wheel and Tire	Chainstay	52mm (boost)	55mm (mid boost?)	56.5mm (super boost)
29x2.6	420	yes	yes	yes
29x2.8	420	no (1.5mm short)	yes	yes
29x3.0	420	no	no (.3mm short)	yes
29x2.6 without yoke*	440 32t chainring	yes	yes	yes
27.5x3.0 without yoke*	430 32t chainring	yes	yes	yes
29x2.4 without yoke*	430 32t chainring	yes	yes	yes

*Your mileage will vary without a yoke, it all depends on how much you dimple the stays and the exact profile of your tire.

The chainstay is constructed with .75x.035in tubing with a single 11deg bend. I have found all pre-made chainstays are a trap.

Seat tube Design:
The seat tube is made with straight 1.375x.035” Chromoly tubing a welded on topper for several reasons:

• availability
• they allow for 31.6 droppers
• the thick topper distorts less and is easier to ream
• the thick topper reinforces the TT to ST junction

There are two things you need the seattube to do:

1. You want to clear the tire
2. You want the saddle to be in the right spot for pedaling

You get two parameters to fiddle with:

• offset
• actual seat tube angle

The combination of the offset, angle, and saddle height gives you an effective seat tube angle. This is what I have found works well:

Even in an edge case, it gives enough clearance for a 29x2.6

A few tips:

• It is better to minimize the offset. large offsets cause the seat tube to be slacker at full saddle extension
• Taller riders tend to need steeper seat tube angles
• shorter riders tend to need slacker angles

Seatstays:
I don’t have much to say about seatstays, except that the material is super limited at the moment. Single bends are cool, S bends are cool too.

Typical Seatstay Tubes:

• 1/2x.035
• 9/16x.035
• 5/8x.035

Design Examples :
Here are a few construction drawings of a Solstice. Hopefully, they can help shed some light on the design and fabrication process.

These are construction drawings, so some dimensions are missing. For example, the wall thicknesses and butts are specified by Nick’s experience and expertise.

Medium:

Large:

Moving forwards, I am going to make an effort to put more information and answer questions on this forum. I have learned a lot from the mid-school builders that have taken the time to document and publish their processes. I hope to do the same and encourage others to do so as well. Having a positive, open, and inclusive community benefits us all.

 

[neongreen]

Thanks for sharing. It gives a beginner like me a lot of good info to stew on.

Just for contextual information, how much would one expect to pay for a 3D printed yoke? And what material?

 

[natzoo]

Thanks for sharing. It gives a beginner like me a lot of good info to stew on.

Just for contextual information, how much would one expect to pay for a 3D printed yoke? And what material?

Great question. Parts are printed from 316 stainless steel. Printing prices are more or less by weight, and my yokes weigh ~130g.

The majority of printing fabs in the US view additive manufacturing as a means for prototypes or as a novelty. They have a more traditional workflow: sales rep > account manager > drawing > technician > print > metrology. The cost of prints reflects this: ~$4/g. The benefit is they handle all the potential issues with 3D printing for you. Yoke Price: $400-500

I think the fabs in Asia have a much better approach towards low volume production. The price is much cheaper, but you need to have a solid understanding of the printing process to design your parts properly. You give them a file, and they just print it. There are no guarantees about print warpage and support removal. I actually prefer it this way.

Jens at 3dpbs prints in Asia. He charges $1.14USD/g. Since he has a lot of experience printing bike parts, he is a great bridge to the printing process. That being said, you still need to design your parts properly for additive. Yoke Price: ~$150-200

Hope that sheds some more light.

 

[Kmccann137]

I would like to learn to do carbon work. From my perspective that’s where you can really innovate because very few are doing carbon work.

 

[Blatant]

Not a frame builder, but wanted to give a thumbs up for an excellent post.

 

[chomxxo]

I would like to learn to do carbon work. From my perspective that’s where you can really innovate because very few are doing carbon work.

Agreed. Look, great post informing what it takes to design a frame, I appreciate the detail and the amount of work it takes.

That said, as a potential buyer: what does Neuhaus Metal Works do that hasn't been done by custom frame builders all over? The Custom option is where I'd start.

The geometry and performance of the stock models is decent but nothing outside the box. There are some people who want to have a brand that nobody else has, but beyond that niche market I don't see the value.

Now, build me a carbon 36er and you'll be onto something revolutionary! I know a custom carbon builder and he was willing to take a shot at a project like this, except for the fork. I'm just saying, do something that stands out, if you're going to put so much work into it.

 

[AdamR83]

Great post, thank you!

I love the seatstay yoke. A damn sight easier than doing that same joint manually too!

Interested, if you wouldn't mind expanding, in your comment about pre-shaped chainstays being "a trap"... I've not built many frames but the two I have with Columbus stays (one single bend, one double bend) worked out nicely. I did perhaps have to design 'around' the stays a bit, but I do think I ended up with a better result overall using these rather than a plain round tube.

 

[Prognosticator]

I don't know jack about frame fabrication, but I appreciate the detail of this post. Very interesting! Wishing much success!

 

[Little_twin]

I did perhaps have to design 'around' the stays a bit

This is the “trap”. You compromise your design based on what is available.


Sent from my iPhone using Tapatalk

 

[natzoo]

Interested, if you wouldn't mind expanding, in your comment about pre-shaped chainstays being "a trap"... I've not built many frames but the two I have with Columbus stays (one single bend, one double bend) worked out nicely. I did perhaps have to design 'around' the stays a bit, but I do think I ended up with a better result overall using these rather than a plain round tube.

To me, modern construction is about building the bike that you want: the chainstay you want, the tires you want, the wheel size you want. I find a lot of people design bikes around what tubes are available. They fall into the trap of buying the "29er tube set" which was designed for bikes in 2010. You end up limited to 440mm long chainstays and 2.3in clearance.

To be clear, there is nothing wrong with a bike with long chainstays and narrow tires, but to me, you shouldn't let the tube someone had made 10 years ago tell you how to design your bike.

All the premade stays look great, they just don't give you enough freedom to build a modern bike.

 

[natzoo]

I realized the original post did not easily allow you to view the small drawings, so I am re-posting them here larger.

I'm also going to include an all-inclusive geo chart here. We call it "Full Spectrum Sizing". It is way more than a colorful chart. There was a lot of testing to establish the size range and there is an entire backend and digital workflow that allows us to fabricate the frames without being a total pain in the ass.

Medium:

Large:

 

[natzoo]

[place holder]

 

[a_j_p]

I have virtually no experience with additive metals but quite a bit with additive plastics. Your comment about knowing how to design for additive makes me curious: how do you account for additive material properties? Do you worry about part orientation during printing, etc.? Is this part of what your initial money spending was investigating?

 

[natzoo]

I have virtually no experience with additive metals but quite a bit with additive plastics. Your comment about knowing how to design for additive makes me curious: how do you account for additive material properties? Do you worry about part orientation during printing, etc.? Is this part of what your initial money spending was investigating?

Printing metals have the exact same design constraints and considerations as printing plastic:

• heat warpage
• orientation
• support material

Since you can print plastic, you can print metals too!

There are a few differences for metal prints:

• If a small part fails during a build, it will likely ruin the entire batch of parts
• You need to account for powder removal
• support material is much harder to remove (grinder, rotary tools, etc...)

You can see examples of the supports below, circled in red. These need to be removed manually by a technician. Notice how my yoke designes are self-supported, with minimal support material.

With regards to mechanical and material properties, the jury is still out on that one. It varies by printer model, print settings, print orientation, post-heat treatment, etc... There are several papers out there that try to nail that down. The general consensus and the testing we have done point to printed stainless is within 10% of the yield stress of bulk stainless, but reduced elasticity, and most likely reduced fatigue life.

One of the best pieces of advice my professors taught me is that strength comes from geometry, not material properties. With 3D printing, you have total freedom to create strong shapes and put material where you need it. This can easily overcome the limitations of printed materials.

That being said, will the printed parts I designed eventually break? No one knows for sure. Every new technology comes with risks. To give some perspective, traditionally constructed steel frames crack all the time. All my parts are pretty beefy, designed to weld easily (heat stress from welding is the real frame killer), and none of them are in cantilevered locations (stems, forks, bars), so I have reasonable confidence in their strength.

 

[a_j_p]

It's definitely interesting to see this - I'll have to read/investigate more. Printing has not ever struck me as practical/approachable for someone like me who just builds a few frames a year for myself... but certainly things will change as printing advances and maybe I am completely wrong to begin with.

Anyway, appreciate the response and you throwing this out there for everyone.

 

[AdamR83]

To me, modern construction is about building the bike that you want: the chainstay you want, the tires you want, the wheel size you want. I find a lot of people design bikes around what tubes are available. They fall into the trap of buying the "29er tube set" which was designed for bikes in 2010. You end up limited to 440mm long chainstays and 2.3in clearance.

To be clear, there is nothing wrong with a bike with long chainstays and narrow tires, but to me, you shouldn't let the tube someone had made 10 years ago tell you how to design your bike.

All the premade stays look great, they just don't give you enough freedom to build a modern bike.

Thanks for your thoughtful response!

I do agree to a point - certainly if you mindlessly stick together a full tube set from X manufacturer you will end up with the sort of geo / clearances you mention, but with some thought it doesn't have to be that way.

Not a lot of us have access to 3D printing, or can justify $500 for a BB yoke when that will buy a full set of 853 tubes, but it is possible to do short chainstays on a 29er (sub 420) with off the peg chainstays, clearance for a 2.5 tyre, without a BB yoke and without crimping the hell out of the tubes. Light, cheap, strong: all three. Holy grail.

Sorry for the thread drift - I'm stoked to see the tech I studied for my degree nearly 20 years ago making it to bikes!

 

[natzoo]

Not a lot of us have access to 3D printing, or can justify $500 for a BB yoke when that will buy a full set of 853 tubes, but it is possible to do short chainstays on a 29er (sub 420) with off the peg chainstays, clearance for a 2.5 tyre, without a BB yoke and without crimping the hell out of the tubes. Light, cheap, strong: all three. Holy grail.

Sorry for the thread drift - I'm stoked to see the tech I studied for my degree nearly 20 years ago making it to bikes!

No worries, there are multiple ways to build a bike, that's part of the fun with framebuilding!

I do want to make a slight correction though, a printed yoke is not $500USD. It is much less. You can print a single-sided yoke for as low as $60USD, or buy a CNC yoke for that price. The price is only going to come down over time.

One of the points I am trying got make is that the barrier to 3D printing is not the cost, it is the design knowledge and CAD ability.

 

[fuzz_muffin]

That ss/st junction is the tits!

Thanks a ton for the info dump. I do often wonder how much all this extra effort is worth compared to just compromising the design slightly to reduce cost, especially considering the ultimate limitation of the hardtail platform. Coming from a guy that rode a "modern" hardtail for a bit on all the things, whilst you can ride almost everything on one, they will always be limited capability wise compared to a good full squish.

I guess what I'm getting at is the goal of design effort seems to be more "Optimise the hardtail so it can be as 'good' as possible", rather than using the same resources to "Design the 'best' two wheeled vehicle for terrain x".
At what practical point is it just throwing money at it for the fun of design? Thoughts?

Cheers.
Also excuse the spelling and grammar, written on the throne.

 

[AdamR83]

Apologies Daniel, I must have skimmed your pricing post even more poorly than I thought! The 3dbps service sounds very well priced, especially for a one sided version.

Hadn't seen the Cobra yoke before, that is an excellent option too!

I agree about learning CAD and learning to design for 3D printing... Two different things and quite a time investment, even individually. Definitely more of a barrier than cost right now, but perhaps we will start to see more off the shelf options cropping up in the future.

This all makes me want to build another frame, dammit...

 

[BadgerOne]

After reading what you've written about the printing of stainless junction components, to me the next logical step and in my mind the most exciting is material advancement. Specifically, you mentioned stainless printed parts approaching the characteristics of bulk stainless. Of course there are many grades of stainless. I'd be interested in attempting to print maraging stainless. If this were possible, it could be combined with air hardening steel such as 853 or even another maraging steel like 953, and in theory the result would be the ability to join thin-walled steels of exceptional strength and toughness but with high compliance for comfort and liveliness. Combining these kinds of alloys actually increases the strength of the HAZ at the joins during welding due to precipitate hardening. Just something to think about.