XM Carbon Speed Race Frame - The wish list

[kubackje]

Hand drawn rendition of some changes I'd make. Mostly focused on dropping weight by giving the rear some Incolor SSR and Scott Foil treatment.

This looks amazing 🤩

[aeroskiii]

Hand drawn rendition of some changes I'd make. Mostly focused on dropping weight by giving the rear some Incolor SSR and Scott Foil treatment.

This is nice, dont like the transition from headtube to top tube, I think the slow taper looks really ugly and dated.

[toxin]

You mean the transition from the back of the head tube into the top tube? I don't necessarily disagree, I used to really hate it too, but I've come around to being in favour of it as a pragmatic choice.

It's better structurally, but more importantly, I think it helps with aero because that part still sees a decent bit of clean air. I think the best way to design a fast head tube is to make it taper relatively sharply, which will improve low yaw performance. Then you use some of that 8 cm alloted for the top tube to effectively deepend the head tube, increasing the side profile of the bike and (hopefully) the area that contibutes to sailing effect, to improve high yaw performance.

The worry I have for this combination of features is stability in crosswinds. The sharp taper should contribute to earlier separation of flow and worse aerodynamic stability. However, as your ground speed increases, flow attachment improves and the apparent yaw angles you experience lower. So as you go faster the bike becomes more stable and starts performing better aerodynamically.

[Sakizashi]

You mean the transition from the back of the head tube into the top tube? I don't necessarily disagree, I used to really hate it too, but I've come around to being in favour of it as a pragmatic choice.

It's better structurally, but more importantly, I think it helps with aero because that part still sees a decent bit of clean air. I think the best way to design a fast head tube is to make it taper relatively sharply, which will improve low yaw performance. Then you use some of that 8 cm alloted for the top tube to effectively deepend the head tube, increasing the side profile of the bike and (hopefully) the area that contibutes to sailing effect, to improve high yaw performance.

The worry I have for this combination of features is stability in crosswinds. The sharp taper should contribute to earlier separation of flow and worse aerodynamic stability. However, as your ground speed increases, flow attachment improves and the apparent yaw angles you experience lower. So as you go faster the bike becomes more stable and starts performing better aerodynamically.

I don't fully understand why you think a sharp taper is faster? This does not make sense or align with the data unless I am confused about what you mean by that, which is most likely the case. Most of the shapes used for an HT would be truncated aerofoils where you can control the whole profile, so really different from wheels, where, at the rear, you need your design to account for the tires as the leading and trailing edges of the bike.

Regardless, I still like the sketch and agree with the principle of focusing on the fork crown, head tube and the fork blades. The fork blades are likely deeper than they need to be for this kind of design, but its hard because this isn't just a yaw problem, its an angle of attack problem. I think you get a lot less for the depth on the road than you would on a velodrome or even most TT courses, as you need to account for more variance on that AA vs. just the yaw (because you turn the wheels more).

[toxin]

Okay "taper" wasn't the technically correct term to use, I just prefer to keep it to more broadly understandable language on public forums. What I mean is that for a profile with the same leading section, a truncated airfoil that terminates in a narrower trailing edge is faster at low yaw, so effectively a lower aspect ratio airfoil.

The problem with desiging a head tube is that you have to fit things inside, so you don't have too much freedom in designing the leading section. Ideally, you'd probably place the steerer closer to the middle of the head tube (Pinarello Bolide F TT), to give you more freedom in design. I placed the steering axis 45 mm behind the front edge of the head tube. I can't say with any true certainty, but I think that's on the far side of the spectrum compared to most of what's on the market.

Yesterday I took the widest leading section of my HT design (a 45 and 26 mm radius half ellipse), and ran cfd tests with trailing edges of the same lengths and different width trailing edges. I tested the 38mm I originally used, 28, 20, 15 and 10mm (and a 2 mm radius fillet on all of them) at 40 and 50 kmh from 0 to 12°. As the trailing edge narrowed, the result was consistent improvement at low yaw and, conversely, deterioration at rising yaw.

[Sakizashi]

Ah, I get it now. The narrower "taper" rule is highly dependent on the constraints you are dealing with here and would hold if you dramatically changed the depth of the HT. Since the optimization should follow a curve, you could also move to a narrower nose radius without the correct depth, and the shape would get slower. You can see this in some of Hunt's wheel white papers; though those also need to be taken in the context of the tire needing to be roughly round in shape and located where it is.

It's too bad that the speed sniffer making sense doesn't make it any less ugly...

[toxin]

I made a second iteration. This one actually had some planning behind it so it turned out much better than last time when I was making it up as I went. There are still some errors (I wish I remembered what I was doing when I sketched out the fork crown 2 mm further forward than the head tube) and some things I didn't have the foresight to plan around, but for the most part it's alright enough that I can put it through some cfd. The fork even turns a full 360° without contacting the frame. Should still clear 40mm tyres and geometry is a touch more aggressive

When I posted the first design, a user messaged me that they saw some prototypes from XMCS, so they are probably far along enough in production that if there are any similarities, they probably arrived there separately.

I won't clog this thread further so it'll probably just be a private passion project from now on.

[bmrk]

I made a second iteration. This one actually had some planning behind it so it turned out much better than last time when I was making it up as I went. There are still some errors (I wish I remembered what I was doing when I sketched out the fork crown 2 mm further forward than the head tube) and some things I didn't have the foresight to plan around, but for the most part it's alright enough that I can put it through some cfd. The fork even turns a full 360° without contacting the frame. Should still clear 40mm tyres and geometry is a touch more aggressive

When I posted the first design, a user messaged me that they saw some prototypes from XMCS, so they are probably far along enough in production that if there are any similarities, they probably arrived there separately.

I won't clog this thread further so it'll probably just be a private passion project from now on.

I really like this design and the intended geometry. Please keep sharing it perhaps in a separate thread?
I belive it can inspire both users and manufacturers in the forum(s).

[rhenders]

I made a second iteration. This one actually had some planning behind it so it turned out much better than last time when I was making it up as I went. There are still some errors (I wish I remembered what I was doing when I sketched out the fork crown 2 mm further forward than the head tube) and some things I didn't have the foresight to plan around, but for the most part it's alright enough that I can put it through some cfd. The fork even turns a full 360° without contacting the frame. Should still clear 40mm tyres and geometry is a touch more aggressive

When I posted the first design, a user messaged me that they saw some prototypes from XMCS, so they are probably far along enough in production that if there are any similarities, they probably arrived there separately.

I won't clog this thread further so it'll probably just be a private passion project from now on.

This is so cool. I want to ride that.

[Takiyaki]

Man I was considering trying my hand at frame design in Fusion. You beat me to the punch with basically exactly what I want in a frame. Aero in the front, climber in the back, ~35mm tire/38mm wheel clearance. I think I would prefer a conventional headset situation but other than that it's basically perfect. Looking forward to see what XMCS comes up with.

[PeterXu]

So, we are making two new framesets for 2026, haven't named them yet.

[Freddy__HH]

Looking Good - any indication, when these will be available to Order?

[KRoser]

PeterXu: Looks pretty good, even in my XXL size geometry looks spot on. Only thing if I may to say my wish, take a look at offering seatpost with forward facing offset maybe 5-10mm.
I love it on my Aeroad to have seat further forward and in my size I would have to have seattube angle 75-76° that offset seatpost is pretty good solution.

[Sander2177]

So, we are making two new framesets for 2026, haven't named them yet.

The second on with deep head tube and S5 esque seat tube is looking noice!!
Need a medium though

[patliean1]

I am super excited for that S5/Foil aero frameset. And the geometry for a size large is spot on! Would be interesting to test this against the Scott Foil and QuickPro ER:One if I get my hands on one.

-Frame
-Handlebars
-Wheels
-Carbon crankset
-Carbon chainring

All that's remaining for Peter to design is a carbon saddle and bottle cages 🤣