Accessibility Widget: On | Off

How stiff are dual crown forks?

Create New Tag

4/11/2021 10:56 AM
Edited Date/Time: 4/11/2021 11:17 AM

Consider this part two to the ‘Are Upside-Down forks really flexy?’ topic from around the turn of new year, where the inevitable question was asked of how dual crown forks fit into the stiffness was asked. At the time I said I’ll get onto it and to expect some results around the march territory. We are deep into April, the pandemic is not letting up, just like my procrastination. But considering the original question and intention of doing some analysis was pencilled in about a year ago and it took most of 2020 to get the first results, I consider hitting April with this set of results a complete win!

Compared to last time, I’m guessing this time we won’t have a mother of all long posts (the previous thread opening post was, pictures included, 27 pages long inside Word), but it will nevertheless be the daughter of that post. And like the last time, the end of it will include some punchlines, so scroll down if you’re not in the mood to read.

Compared to evaluating a bunch of variables the last time, the analysis performed this time is less broad, but not worse in my opinion. The original post covered results including stanchion diameter, stanchion wall thickness, axle diameter, axle thickness, axle clamping width, crown thickness and travel. This time I only checked the different stanchion diameters, as we saw the last time that other variables don’t have that much of an influence. I did perform the analysis at two different travel settings across the range, but the travel data point was collected into a single pair of graphs, but as always, more on that later.

The original analysis, covering only single crown forks, used a 160 mm, 35 mm stanchion diameter ‘29er’ fork that was based on a Lyrik. To include dual crown forks, a different premise was needed. Cue modern ‘SuperEnduro’ bikes with 180-ish mm of travel on single crown forks (mostly the Fox 38 and the RS Zeb, both sporting 38 mm stanchions), where a lot of people have asked why not just throw a dual crown fork on it? It turns out that question is not as far fetched as it might seem at first, as the 180 mm variants of the 38 and the Zeb are roughly within 5 mm of the 200 mm Boxxer and the Fox 40 when it comes to the axle to crown distance. So, the change in geometry, going to a 200 mm fork, will be minimal on a bike like that.

The analysis is thus based on the question ‘What do we gain by going from a 180 mm travel, 38 mm stanchioned single crown fork to X?’. All forks in the analysis are designed around a 600 mm Axle-to-Crown distance with 180 mm of travel for the single crown forks and 200 mm of travel for the dual crown forks. As previously, the dropouts, axles, clamping geometries, the bushing topologies, etc., are the same, so are the lowers on the right side up forks, the crowns are designed to be similar between the single crown or the dual crown pairs, and so on.
The stanchion diameters checked are 35 mm, 36 mm, 38 mm and 40 mm – this covers all major single crown and dual crown forks currently on the market, from Intend to Fox, Rock Shox, DVO, SR Suntour, Xfusion, Ohlins and Marzocchi (even if the old stuff would be compared) and maybe some that I missed.

Onto the details:

The models:
Single crown right side up:

Single crown upside down:

Dual crown right side up:

Dual crown upside down:

These four pictures show how the models looked overall with the details of each part of the structure covered separately.
The crowns and steerer tubes:
Unlike with the previous analysis the crowns are much more realistic this time with the stanchions (or outer in the case of upside down forks) inserting into the cylindrical holes within the crowns. It’s also not a matter of copy and paste between the single and dual crown variants, just extending the two tubes and adding another crown, the structure is quite different between the two. The single crown forks also include a tapered steerer this time.

Single crown variants:
This time we have a much more shapely crown that is also more properly hollow, with both the stanchions/outer and the steerer inserting into the crown through cylindrical holes. The steerer tube is tapered and of varying thickness, it has a 5 mm thickness on the bottom, larger diameter with it decreasing through the taper to 2 mm wall thickness at the top, where the stem would mount. The diameters are 38 and 28 mm if I remember correctly, which is roughly 1,5 and 1 1/8 inches. The crowns are very similar between both right side up and upside down variants, except the lowers insertion geometry, which adapts to the current variant keeping a clamping thickness around the outer/stanchion of 5 mm in all cases.
The ‘stack height’ of the crown itself is 60 mm and thus covers the 20 mm difference in travel compared to the dual crown variant (which has a lower crown height of 40 mm).
The sectioning of the steerer is there for support definition in the analysis, which will be covered in detail below.

Dual crown variants:
Compared to the single crown variants, the dual crown variants have a much simpler crown design, as is the case in real life too. The lower crown has a height of 40 mm (some of the material is removed in the middle to add some ‘lightness’), as mentioned above, while the top crown has a height of 20 mm, which is roughly in line with the dimensions of a Boxxer (I had the dimensions checked by a friend of mine). The steerer has a constant diameter of 28 mm (roughly 1 1/8 inches) and a constant thickness of 2 mm. Both the steerer and the stanchions/outers insert into cylindrical holes in both crowns. The overall design is very simple, the larger the diameter of the stanchion or outer inserted, the wider the crown is, as the 5 mm wall thickness surrounding the inserted tube is kept here as well. This does have some effect on the stiffness, but I suspect not that much. Thus, the difference between right side up and upside down forks is mainly in crown width (length, when looking at it on the bike?), other than that the two parts are identical. And the crowns in this case are not hollow, as the shape doesn’t facilitate that. Real world crowns do have some more material removed and include webbing to make them stronger, but the effect of that would be negligible.

The dropout:
The dropout is the same as in the previous analysis, I even kept the same picture. And the same part essentially is used on all variants, regardless of sidedness or the number of crowns.

The dropout is mirrored between both left and right side of the fork and is mostly always the same. Since a single axle diameter and a single clamping width were used, the geometry of the part stays the same across all analysis variants.

The bushings:
The axial position within the outer and the height of the bushings is something that has been carried over from the previous analysis and was based off the Lyrik. It was kept the same for all variants (except for the diameter of the bushings of course). As covered in the previous thread, additional bushing overlap (longer distance between the two bushings) has a small effect on stiffness itself. The effect on performance will be higher, as the friction between the bushing and the stanchion will be lowered, the tolerances can be opened up without it causing a high clearance between the lowers and uppers (that can be felt by the user), etc., but it’s not significant for this case.

Single crown, right side up:

Single crown, upside down:

Dual crown, right side up (40 mm stanchions are shown here as that design point was chosen and I didn’t notice it until taking this picture...):

Dual crown, upside down:

As can be seen from the pictures, the bushing topology is identical across all variants.

The bridge:
The bridge connecting the outer tubes is something that is present only on the right side up fork and is, for obvious reasons, missing on the upside down fork. It is crude, but it is present and that is the important thing for this analysis. It is also a ‘component’ that has been kept the same dimensionally through all variants. This was the same for the previous analysis and I mentioned then that it could be a point in the investigation even back then, but considering the smaller scope of this analysis (as variables go), I have of course kept it the same. It does its job and that is the most important part. The picture used is again the same as in the previous analysis (therefore the incorrect crown), so please excuse that, but you can see on the overall fork overview that it is in fact exactly the same.

The axle:
Again, this is a carry over description, including pictures, from the previous analysis. The axle is just a straight tube of constant diameter, thickness and the correct length to fit within the dropouts. The same part is used in all variants considering a constant diameter of 20 mm (also in single crown variants!) and a constant thickness of 3 mm with a constant 25 mm clamping width.

The axle itself is constructed from 5 parts, but only to form additional distinct surfaces to apply the loads to a defined section of the axle (roughly where the hub bearings would be).

The stanchion:
Since this part is mostly unchanged compared to the previous analysis, I have reused the picture in this case too. So, just a straight tube of a constant (but varied, depending on the parameter) diameter and constant thickness of 2 mm. The only variable with the stanchion is the right side up dual crown fork, where the stanchion length is much larger to fit through both forks, of course.

The outer:
Since this part is mostly unchanged compared to the previous analysis, I have reused the picture in this case too. The outer is a mostly straight tube of varying diameter with the addition of the bushings. The bushing locations and heights are the same as in the previous analysis and are based off a Lyrik. There were comments that a larger bushing overlap would be better, but I’ve checked that and the difference on stiffness is marginal. The performance will be improved much more through lower friction between the bushings and stanchions, but not the stiffness. The thickness of the outer also was not changed for the analysis. And the same outer is essentially used for all variants, except for the upside down dual crown fork which requires the outer to fit through both crowns and thus requires a longer outer tube.

Base configuration:
As noted in the intro, the base configuration is a single crown fork of the right side up variety with 38 mm stanchions.
All forks use a 2 mm stanchion thickness, 20 mm axle (not what the market offers, but check the previous analysis on what the effect of a 15 mm axle is on a single crown fork) with 3 mm thickness and 25 mm of clamping width. The steerer thicknesses have also been kept constant, but further investigation could include varying that as well.
The travel was set to 20 % (roughly the sag value for most forks) with additional analysis performed at 60 % of travel for comparison purposes. This means either 36 mm for single crown forks or 40 mm for dual crown forks (effective axle to crown lengths of 564 and 560 mm) or 108 mm for single crown forks or 120 mm for dual crown forks (effective axle to crown lengths of 492 and 480 mm).
Hub spacing is 110 mm. The fork trail was set to 37,5 mm, less than the 42 mm of a 29er fork, but this is one of the assumptions and simplifications. The value is the same as in the original analysis.

The material for all components was set to aluminium 6061, T651. It doesn’t matter much which grade and temper it is. It is the ultimate tensile strength that differs between grades Aluminium, while the stiffness characteristic (Young’s modulus of elasticity) is more or less equal across the range. And since we are dealing in relative terms, as long as the materials are the same across the variants, it doesn’t really play much of a role – we could have used steel and still get the same relative differences between variants.

Analysis parameters:
This time it’s much simpler than the first time:
Stanchion diameter: 35, 36, 38, 40 mm
Travel: 20 % (36 or 40 mm), 60 % (108 or 120 mm)

Finite element analysis details:
This part is again the same as in the original analysis. Therefore, the description is kept exactly the same (except for the number of variants).
The model was assembled with all contacts using the ‘bonded’ method, which essentially makes one single part of the model. The exception is of course the stanchion-to-bushing interface where a frictional contact with a very low friction value was used (enabling any axial sliding of the two bodies caused by deformation). Regarding the fact that it is not as realistic as using a hub axle, axial preload for the dropouts, frictional/interference fits between the tubes, etc., this has been covered in the previous analysis, but I’ll mention it again. All of it is true, but we are not dealing with real world products, so a lot of variable are completely unknown. Plus, a lot of other factors will play a huge role and this is just a comparison analysis regarding stiffness without touching other factors (for example friction and/or suspension performance). So, the concerns are noted and known, but intentionally ignored.

The mesh applied to the models (if anyone is interested) was the same as before, applied to all faces with the element size set to 3 mm. This gave a fairly quick calculation (~3 to 5 minutes per design point) with relatively low memory usage (~10 GB ) and good accuracy. I did run some tests for the previous analysis going down to 0,6 mm element size and the difference in the results was 1,3 % at 0,6 mm, but that required 2,5 hours per design point (there were 64 in total...) and ~80 GB of RAM. At 0,8 mm and up the difference was within 1 %. And in any case the error will be more or less the same in all variants, so the comparisons are still valid.

Loads and supports:
All design points were checked for their bending stiffness with the force applied perpendicularly to the axle (perpendicularly to the steerer tube) and torsional stiffness with the moment applied to the axle around the vertical axis, parallel to the steerer tube. The force that was applied was 100 N, while the torque applied was 50 Nm. This is again the same as in the previous analysis.

Unlike the previous analysis, the fork was not fixed in the same manner. Previously the complete cylindrical surface of the steerer tube was fixed in place, which meant it did not have an influence on torsional stiffness. This is fine when comparing only single crown forks, but does not work when dual crown forks are thrown into the mix. A different way was needed.

As noted in the crown section, the steerer tube was sectioned. Just above the(lower) crown a 10 mm high section was cut out and 130 mm above that another 10 mm ring was cut away. On the dual crown variants, the upper crown was mounted just above this ring (crown spacing was thus 150 mm). These two rings were used for a revolute joint between body and ground, effectively simulating an infinitely stiff head tube, but enabling the steerer tube to twist and bend in between. The upper part of the steerer tube was then fixed in place, similarly to the previous analysis.

The loads are applied to the two sections in the axle, roughly corresponding to the locations of the bearings of the front hub. This is also the same as in the previous analysis.

The loads and supports are only shown on the right side up, single crown fork, but they are identical for all four variants. The way the steerer is supported at the top means the top crown hole for the steerer is supported, but this is very much inline with direct mount stems anyway. The first picture shows the fixed support of the steerer and external loads on the axle, while the second picture shows the joints, simulating headset bearings.

The results were taken identically to the previous analysis - the maximal deformation value of the middle part of the axle in the bending mode and the maximal deformation value of the right dropout for the torsional mode.
Both deformation modes are shown in pictures below, again only for the right side up, single crown fork. The deformation was scaled by a factor of 10. The torsional deformation is shown in two pictures with the second showing just how much the steerer actually affects torsional stiffness. This is also the area of most gains for dual crown forks of course.

The results:
Finally, the results. The results will be normalised to a 38 mm fork and in the case of multiple variants, it will be either the single crown variant, the right side up variant or the single crown, right side up variant. The effects of going deeper into the travel will be covered in a specific graph.

The results will be shown as a percentage of stiffness of the base variant. The base variant is of course 100 % stiff with a fork that has 200 % of deformation being 50 % as stiff, while a fork with only 50 % of the deformation being 200 % as stiff.

To make some more sense of the graphs, the right side up and upside down variants are colour coded (orange and blue), with single crown variants presented in solid bars with dual crown variants presented with a checkerboard pattern.

Single variant comparisons:
Single crown, right side up:

The results in bending are similar to the results of the original analysis, but the gains through larger diameter stanchions are much lower in torsion compared to previous results due to the different loading setup. In this case a large part of the deformation is caused by the steerer tube, which is the same for all variants, thus the differences turn out to be small.

Single crown, upside down:

For the inherently stiff upside down fork the gains in bending are even smaller than for the right side up fork, as we are not strengthening the weak points of the fork as much as with the right side up variant. The results are also very similar to the original simulation, as with the right side up fork. In torsion the gains are higher than with the right side up fork, but this is mostly due to the fact that an upside down fork is much less stiff in torsion from the start, so the effect of the steerer is lower than with a right side up fork. Any gains thus show themselves more clearly in this case.
With numbers like these, if you are going for an upside down fork, it makes sense to go for the 35 mm diameter, as larger diameters will only add weight and friction through a larger surface area with minimal gains in stiffness.

Dual crown, right side up:

The differences between stanchion diameters are very similar, just somewhat exaggerated in right side up variants with the addition of a second crown as we are not depending only on the crown stiffness in bending mode and on the steerer in torsional mode.

Dual crown, upside down:

Again, the inherently stiff upside down fork shows higher gains with increased stanchion diameter with the addition of the second crown, but compared to a dual crown right side up fork, the differences between diameters are not as pronounced in bending, but are more pronounced in torsion.

Comparison within crown variants:
Single crown variants:

As we’ve seen in the original analysis, comparing single crown forks shows that the upside down variant is roughly 66 % more stiff than the right side up variant in bending, but roughly 60 % less stiff in torsion. The stanchion diameter mostly affects the bending stiffness, which, as mentioned is a consequence of the effect of the steerer in torsional mode, which is present and the same in all variants in torsion, minimizing the differences in other parts of the forks.

Dual crown variants:

If the 66 % increase in bending stiffness of the single crown upside down fork was immense, what about a 125 % increase in the dual crown variant? Sadly, comparing different varieties of apples between themselves does bring up the sour apple again, as the torsional stiffness of the dual crown upside down fork compared to the right side up dual crown counterpart gives a very similar picture to the single crown comparison. A dual crown upside down fork is thus all bending stiffness, no torsional stiffness gain. Larger stanchion diameters do pay off a bit more than with single crown forks, likely due to thicker outers connecting the two crowns giving more stiffness in that area as opposed to the thinner stanchions connecting the crowns with right side up forks. This might be the reason why Intend can skip the full length steerer tube on the dual crown Infinity fork.

Comparison across crown variants:
Right side up variants:

Now for the main course of the analysis – what would happen if you would mount a dual crown fork to a super endure bike? Well... you would gain roughly 5 to 10 % of bending stiffness, if you kept the stanchion diameter the same. At 35 mm the gains are even lower than that, on the order of 2 %, which could almost be a rounding error. On the other side (Totem vs. 40) it’s just over 10 %. The main effect can be seen on the torsion side though, where the dual crown forks are roughly twice as stiff.

The question now is, which school of thought you prefer – if you want steering precision and think upside down forks are too twisty, then by all means, go dual crown. If you’re of the opinion that a twisty fork (with more bending stiffness) is more forgiving and tires you less, for the love of god, don’t swap out your Zeb/38 for a dual crown fork.

Upside down variants:

With upside down forks the bending gains are much higher by adding the second crown, as the thin stanchion near the crown is not the weak point as with right side up forks, but due to the stiffer stanchion-outer layout the crown is the one suffering the most. Adding a second crown and tying both of them with a thick outer tube gives a lot of gains in bending with upside down dual crown forks being roughly twice as stiff as their single crown brethren. Torsion gives quite similar results to right side up forks, except for the fact that a thicker stanchion gives even more results with dual crowns than with right side up forks, likely due to a stiffer crown area compared to the right side up variant.

Tying it all together:
Comparing all four diameters and all four variants we get a much clearer picture of what is stiff and what isn’t.

Bending stiffness is the mecca of upside down forks, as we’ve seen in the original analysis, which is just exaggerated with the addition of the second crown. We’ve seen similar numbers in dual crown comparison, but simply because adding a second crown adds little to the right side up fork in bending. An upside down 40 mm stanchioned dual crown fork would be almost three times as stiff in bending compared to a 38 mm single crown, right side up fork.
In torsion the dual crown, right side up fork is the clear winner with steering precision for days. The dual crown upside down fork also isn’t looking as bad anymore, compared to single crown right side up forks. The question is, can you handle the stiffness in bending that it brings with it? Is there even the need to say to write your opinions in the comments?

Going deeper into the travel:
We’ve covered the travel portion in the original analysis, so this will be quick. I compared each variant only to itself when going from 20 % to 60 % of travel to see the increase a single fork will have deeper in the travel. I then averaged the stiffness gains across the four different variants in both modes.

Right side up forks gain the most in bending as the distance between the anchoring of the stanchions and the loads gets lower. The gains are not as high for the dual crown forks since they have enough stiffness in this regard as is. In torsion the gains are again higher for dual crown forks due to their stiffer crown-steerer setups (including stanchions/outers) and the upside down forks in general gain more due to their lower torsional stiffness – similar to what happens with right side up forks in bending compared to upside down forks.

-upside down forks are still torsionally much more flexy than right side up forks (about 2,5-times as much)
-adding a second crown is not the upside down silver bullet – torsional stiffness gains manifest themselves in bending stiffness too
-adding a second crown does wonders for torsional stiffness
-adding a second crown does a lot more for upside down forks than it does for right side up forks
-to reiterate, the bridge makes all the difference in right side up forks and is the silver bullet to gain torsional stiffness without adding tons of bending stiffness.

Final thoughts:
As I’ve noted in the conclusions of the previous analysis (and used a lot of it here too), these numbers shouldn’t be taken as the be all, end all or as a basis to say ‘yeah, well the 40 is xx % stiffer than the 38’ or ‘the Boxxer is yy % stiffer than the Zeb’ (or is it vice versa?). Real life products have different parts of their whole optimised to achieve different results, as can be seen by a much different crown and bridge on the Zeb compared to the Lyrik, by the ovalized steerer tube on the 38, etc. We have seen now that with single crown forks, the steerer tube stiffness plays a very large role on torsional stiffness and we can also see why a tapered steerer is not needed with dual crown forks. The steerer tube length could be something to check further into as well.

And as stated before, while the geometry of both right side up and upside down variants was kept the same in this case (and the effect of bushing overlap on stiffness was found to be very small), upside down forks open certain possibilities not available to right side up forks, like mounting a bushing much closer to the crown of a dual crown fork with the stanchion passing into the outer above the lower crown of a dual crown fork. While not adding a lot of stiffness, it does gain other performance benefits, as noted before.

I’ve previously said I don’t see upside down forks becoming popular outside niche brands and checking what happens with the additional crown just confirms this. Maybe we could see some more dual crown upside down forks down the line if Downhill become even more of a speciality as it already is (number of bikes sold on the market and all) with some manufacturers experimenting more, but I for sure do not see upside down forks come to market in higher numbers in single crown form.

As for dual crown forks on Enduro bikes? Honestly, I don’t see much of a benefit with the added weight and torsional stiffness, even more so given the comments in the previous analysis’ thread, where there were cases where riders prefer less torsional stiffness. Want more stiffness? It might make sense to invest the weight difference between the single crown and dual crown forks into making the steerer tube stiffer torsionally and net the same gains.

This is based purely on stiffness, other performance benefits might sway it the other way, for example a dual crown fork can have a lower axle to crown for the same travel, lowering your front end – this could be a benefit with 29” wheels for example. Air spring volumes can be larger with a longer stanchion, etc.

Comments? If anybody is interested in any specifics results wise, fire away, just as in the previous topic.

(And I said this won’t be the mother of all posts, yet it came out even longer than the previous one.)


4/11/2021 1:39 PM

I'm skeptical about some results.
A 2.5 factor between USD and no-USD forks is huge. More than in real life even with smaller tubes on the USD forks.
Maybe your arch is a bit overkill as it seems plain.
A single crown bike fork is about 20 to 30 Nm/° of torsional stiffness. Are your results in this range?


4/11/2021 1:49 PM
Edited Date/Time: 4/11/2021 1:54 PM

Ballparking/back-of-the-enveloping it, single crown right side up forks had a max deformation on the dropout of just over 2 mm at 50 Nm of load on the axle. If said deformation occurs at 70 mm off-axis, this gives ~1,6° of rotation at the bottom or almost exactly 30 Nm/° of stiffness.

If you take a look at the original analysis, a right-side-up fork without a bridge is half as stiff as an upside down one.

Also, the 2,5 factor is a dual crown, upside down fork vs. a single crown, right side up fork in bending mode. Upside down forks are, by design, stiff in that mode and the dual crown design adds to this as well. It's a bit of an apples to oranges comparison. And the 2,5 factor area is when you go to a 40 or a 38 mm stanchion, while all the upside down dual crown forks on the market are in the 35 to 36 mm range - 36 mm for the Dorado and Emerald and 35 mm for the Intend. So the more realistic factor is *2.

Another also, some results were a bit surprising, but looking at how the models deformed, it did make sense. If you want more details I'd be happy to share them of course.


4/13/2021 9:36 AM

What a great technical post.

Thanks for doing this.


4/13/2021 11:19 AM

You did a good job putting this together. Though some of the constraints could be refined, as a piece of comparative analysis I'd say this is perfectly valid. The relative qualities of the 4 major fork categories come as no surprise really. The ensuing discussion will of course center around what qualities equate to a faster bicycle. After all, that is the purpose of engineering bicycles in the first place!

I still maintain that a stiffer fork is always going to be able to be ridden faster, comfort and fatigue be damned. Compliance may be nice for the weekend warrior, but the highly-skilled professional will be able to carve off a few tenths and train their body (mostly their wrists!) to handle the abuse. However, I can never support this argument until someone gets some testing done. Until then, we won't know which of the engineering parameters equates to goodness.

I recently got a reply on a PB forum from a rep at Vorsprung. He said that the SC fork provides enough strength, at the lowest weight. Since most customers care more about weight than anything else, the industry has made a product that seeks to be durable enough, and as light as possible. There is no competition driving performance and investigation into stiffness like there has been in the case of motorcycling. So now that we know what all of these different designs do, we won't know


4/13/2021 12:09 PM

freaking awesome! nice work. homepaged and a fun photo to add to it


4/13/2021 12:37 PM

Great work! Interesting read.

A single crown fork also has less aero drag so maybe another benefit there wassat


4/13/2021 12:42 PM

@TheSuspensionLabNZ you, sir, wint he internet!

@sspomer many thanks for the homepagedness and the absolutely awesome and representative photo!


4/13/2021 12:54 PM

justinc5716 wrote:

You did a good job putting this together. Though some of the constraints could be refined, as a piece of comparative analysis ...more

Regarding constraints, of course, this was covered in the original thread as well. The stanchion to bushing interface is ideal and is aluminium to aluminium. Using a realistic DU-like bushing will change the local stiffness, clearance between the stanchions and bushings is a must and would influence the results a bit (then varying clearances between different forks of the same models give another layer of influence), the axle usually has some clearance in it as well, mostly seen with 15 mm axles, that only tighten axially and without pinchbolts, there is no axial preload taken into account here, etc. etc.

But like you said, it's a comparison and the mistakes are the same across all variants. And the intention wasn't to show a real world state, only actual testing using force and torque gauges would achieve that (and to be honest, to be sure in the results you'd need to test at least 5 to 10 forks of the same model to see how the production varies between them, just to be sure). The intention was to at least get a feel of what to expect when marketing departments tout a 2/3 mm larger stanchion diameter as the next big thing. And the original intention was to see if upside down forks REALLY are that torsionally flexy.

Regarding stiffness and better for racing, dunno, there has been a lot of discussion about this also in the previous topic and Jeff Brines had a lot of things to say about a year ago somewhere on this forum (forgot the topic). The problem is that it's REALLY hard to measure these things except by using a stop watch or using highly biased test devices - humans.

As for the Vorsprung rep, looking at the results here, yeah, I can see the point. I mean seeing these graphs, I see no reason to go dual crown on anything other than a DH bike. And even then... a 200 mm single crown fork? It might just work.

As for any other analysis, venturing into real world territory, sadly outside winning the lottery I can't really see how to add anything to these simulations. Though I would love to do that of course smile


4/13/2021 1:02 PM

My takeaway, which may not have been intended, but logically follows from the data, is that the slacker your head angle, the more advantage an inverted fork has.


4/13/2021 1:28 PM

Primoz wrote:

Ballparking/back-of-the-enveloping it, single crown right side up forks had a max deformation on the dropout of just over 2 mm ...more

I was mainly answering the 2.5 factor mentioned in the "TL:DR part". But graphs are showing more than a factor two in torsion stiffness, for SC forks, so it seems a bit huge, as in real forks a factor 2 between a stiff 35mm SC and a flexy 32 mm SC USD fork is a big difference. That's why I think the arch you designed is a bit strong, as it seems plain. real life arch are thin (and magnesium) and there is variations in them, translating in variation in fork's stiffness that covers other parameters (Rockshox' forks are stiffer in torsion, or at least as stiff than fox' even with 2mm less stanchions, because more meat in general and more webbing in the arch.)
It may influence some results. Or not that much, I don't really know (I'd say a flexier arch would reduce differences with USD forks but will also reduce differences with dual crown forks because more overall deformation= an absolute reduction of it makes a smaller relative difference).


4/13/2021 2:38 PM

groghunter wrote:

My takeaway, which may not have been intended, but logically follows from the data, is that the slacker your head angle, the ...more

Or maybe not. The slacker the angle, the more inline with the stanchions bumps act upon the fork, thus transferring more energy into bump absorption than bending. So the bending benefit of upside down forks is... Not taken advantage of. The story is opposite on relatively flat, smooth landings, where you load the fork more perpendicularly with a slacker headangle.

faul the bridge has the most effect in torsion, it has very little effect in bending (it does, as it prevents the lower outers from rotating around the stanchions - not having a bridge would mean all of that load would be carried by the axle).

I wouldn't say it's overly strong, it is full, while real life bridges are webbed, but that is for weight vs. stiffness reasons (webbing takes away a lot of material, but little stiffness, for example). And the bridge that I have is both not blended into the chassis as nicely as real life products have it done (because that would just add finite elements that have little effect on the final result, while real life products, besides looking good, must also be manufacturable, so need smooth transitions when casting parts), but also not extended forwards as much as real life products have it. The Zeb has this pointed out even more strongly than other forks with the intention of the bridge to clear the crown and the headtube. In my case the bridge juuuuuuuust gets past the crown (maybe in some cases it even doesn't), but I don't care, as the model doesn't need to be functional in a rideable sense. The more offset the bridge is, the stronger the structure will be as you're using that bridge in a more 'off-axis' way, making it effectively stronger for the same amount of material (you're not only twisting it in the horizontal part, but bending it, or if you're twisting it, you're loading the stanchions much more). Look at Magura, they used two very thin arches, which is a similar principle.

And, like I said, a bridgeless right side up fork is half as torsionally stiff as an upside down fork. The bending difference was negligible.


4/13/2021 2:57 PM

groghunter wrote:

My takeaway, which may not have been intended, but logically follows from the data, is that the slacker your head angle, the ...more

Primoz wrote:

Or maybe not. The slacker the angle, the more inline with the stanchions bumps act upon the fork, thus transferring more ...more

TBH, I think that this is one of those situations where preconception does not agree with reality. On anything smaller than a full DH bike, I think we underestimate how much of the time we are not at an angle of attack that would place bump forces more inline. I think that perpendicular load is a lot more common than people account for.

The picture that sspomer added is a good example, actually. fast rider, dual crown fork, on a fairly steep section, and we're still seeing visible deflection... and bushing friction that negatively affects suspension performance probably starts long before you can see the the stanchions bend visibly.


4/14/2021 2:01 AM

Excellent job. Thanks for concluding and answering the questions raised by the first article. The best piece of MTB analysis available on the internet!

I can't see a need to rush out and buy dual crown for my Enduro bike for the sake of stiffness, but it's a good way to go if you want 200mm travel!


4/14/2021 5:58 AM

Test a Lefty smile It's supposed to be pretty decent:

And a Trust Message.

4/14/2021 7:58 AM

Would love to see a comparison to USD thru-axle, RSU thru-axle, and RSU QR9.

I would choose torsional stiffness over bending stiffness, all day. I went from a Fox 32 150 9mm QR, to the same fork with 100x15 thru-axle, to a Fox 36 160 with 110x20 thru-axle. Yes, it's amazing how much more the 36 resists bending (I'm ~100kgs, too), but the bigger improvement to me was the torsional stiffness increase when going to the thru-axle, and then again to an even stiffer thru-axle.


4/14/2021 9:01 AM

Awesome read. Why do motocross bikes use an USD fork? Seems like the forces applied to those forks would be way higher than anything seen on an MTB.


4/14/2021 9:14 AM
Edited Date/Time: 4/14/2021 9:55 AM

One part is definitely packaging, which was covered in the previous topic (I forgot who explained this, sorry, please claim your explanation) - MX bikes have more travel and use smaller wheels, which means a larger stanchion to wheel overlap. A right side up fork would need to extend below the axle to fit all that travel. With a dual crown, upside down fork they can drive the stanchions past the lower crown. Supposedly the older forks got so low, that riding ruts became a bit of a problem, and that was at a time when the travel wasn't as long as it is today.

As a benefit at least some forks use a traveling upper bushing, meaning it's mounted to the stanchion, not the outer and the bushing distance increases going through the travel. Which is by itself a good thing, but requires more expensive parts to make it happen (you need to control the roughness, quality and hardness of the inner surface of the outer tube too).

My theory was also that MX bikes maybe don't need as much torsional stiffness, as more compliance riding ruts might be a benefit (with more steering precision prefered for MTB due to a different nature of terrain that is ridden). There wasn't much agreeing with that at the time, but thinking about it now, the overall stiffness levels are higher for MX bikes due to larger and heavier stanchions and outer and so on. MX bikes can handle heavier forks, as they are powered by motors and the system weights are much higher anyways.

Regarding system weights, it seems plausible to me that torsional loads aren't particularly higher than on mountain bikes, while flexural loads, due to much higher speeds and especially higher system weights, are much higher. So upside down forks for MX bikes might be in the torsional and flexural stiffness sweet spot due to the demands and use case, while the same is true for MTBs, but for right side up forks.

To add other answers, @Jacquers a lefty would be doable, I've got some comments on that and I might look into it, but it would require a bit of a modification and research to prepare the model. I was also thinking about modelling a rough guestimate of the DVO Emerald brace/fender to see how much of an effect it has on torsional stiffness. It might be much harder to hit a representable stiffness since the actual product is carbon, but anything would be better than nothing.

As for the video, I smell bullshit. I bet the Fox was fiddled with to be that rough. More or less just putting hands on my Lyrik moves it a bit when it's under pressure. I'm fairly sure it would move easily if even lightly greased.
Sure, the needle bearings of the lefty will be smooth as hell, but still, it can't be THIS bad.

As fort he Trust fork, it's carbon, I'm not touching that, layup has an IMMENSE effect on stiffness when it comes to composites. Sorry smile

@just6979 also not touching a QR interface, as that would require realistically modelling the fork-axle interfaces, both for the QR and the through axle. I cheated here with a bonded contact (essentially welded or acting as a one part), which is not realistic, but it's the same for all variants. A QR is a different animal altogether, it's not just a 9 mm axle. You would need to model the hub (as the actual hub axle has a larger diameter than the QR and the 9 mm dropout) and then model the axial preload force, the contacts between the hub and fork, etc.

Considering that there are no more QR forks out there (at least new ones), I don't think it's worth the hassle. Similar goes for the Trust since the company folded...


4/14/2021 10:23 AM

Has anyone quantified just how much torsion force there is on a fork during real world riding? Just how much force can the contact patch of the tire actually exert on the fork?


4/14/2021 12:11 PM

krisrayner wrote:

Has anyone quantified just how much torsion force there is on a fork during real world riding? Just how much force can the ...more

Very little (very very little) from the tyre itself, when you twist the forks riding along, but the torsional loads on the fork come mostly from sort of side impacts with stuff on the ground - roots and rocks. That and loading through steering, caused by trail, leaned bike, terrain shape and so on.

But it is also a good point. Some strategically located strain gauges on the handlebars could do the trick here


4/14/2021 9:26 PM

krisrayner wrote:

Has anyone quantified just how much torsion force there is on a fork during real world riding? Just how much force can the ...more

Primoz wrote:

Very little (very very little) from the tyre itself, when you twist the forks riding along, but the torsional loads on the ...more

This is why the torsional "give" on a USD fork is a good thing. It smooths out the side impacts and lets you track better.
It's a little weird when you first try them, but they track waaaaay better on rough off camber sections. You can hold a line without fighting your bike.


4/15/2021 7:54 AM

The idea of torsional compliance being good is strongly misguided. Sure you are providing a means of isolation for some side impacts on the tire, but remember that these impacts are now imparting a directional change to the tire rolling axis. The result is that body inputs and extra steering inputs are needed to keep the bike upright and on-course. Its not actually a win-win. Furthermore, any intentional steering inputs will now be delayed. This creates something called a phase lag in the dynamic system, which degrades control-ability. I could dive into this in painful mathematical detail, but please just take my word for it.

Quantifying exactly how much torsional deflection occurs on trail is an engineering pitfall that you don't want to go down. Firstly, the task of defining tire loads for any vehicle is difficult because they are highly statistical in nature and very hard to calculate without lab data. Secondly, there is no standard for how much deflection is too much. We just end up with another comparison metric. Stiffness on the other hand, is a parameter within any dynamic system guaranteed to increase control-ability. That is why you hear engineers talk about stiffness all the time, because that is the parameter that can be easily measured, changed, and accounts for the lion's share of vehicle handling.

There may be a corner case at some point where, like in Motocross and MotoGP, we discover that a controlled amount of torsional and bending compliance actually benefits grip by smoothing out the loads on the tire contact patch. However, we are in a stiffness regime currently that sits far below this point. Hopefully discussions like this will lead to that eventuality where engineered fork stiffness is a thing, but first we need stiffer forks!

What I've been trying to get across in my recent posts to this thread is that we've been living within a paradigm dominated by strength-to-weight ratio, which has resulted in very low stiffness forks being offered for sale. That has created handling problems as speeds and loads increase. However, because this is what we are used to, no one realizes that fork stiffness is too low unless they spend a lot of time going back and forth between trail bikes, DH bikes, and motorcycles. If we shift our paradigm by changing the focus to stiffness-to-weight, then other solutions start to make more sense, like dual-crown forks for example.

I hope for a future where customers demand more of their forks so that dual-crown forks become the norm, with more USD options being available for DH applications.


4/15/2021 9:25 AM

>I hope for a future where customers demand more of their forks so that dual-crown forks become the norm, with more USD options being available for DH applications.

"more USD options" @justinc4716 didn't you just say most MTB forks are already too compliant, especially torsionally? Unless you go down the roller bearing route (high manufacturing and QA costs, because of higher tolerance requirements), I don't see any way to increase USD torsional rigidity without dramatically increasing weight.

On that subject, @Primoz, thoughts about the carbon arch lowers that DVO used on their USD forks? They claimed it increased torsional stiffness, but I have doubts since it only connected at the dropouts, and didn't transmit load to the stanchions via bushings.


4/15/2021 9:41 AM

I did say that stiffness is too low, and nominally USD forks suffer from reduced torsional stiffness. This seems like a contradiction, but as the results suggest, going to a dual crown USD fork allows for stiffness-to-weight ratios that surpass those attainable with RSU forks. There is a break-even point for this to be true. Below a certain stiffness, RSU forks offer an advantage in stiffness-to-weight (S/W - so that I don't have to keep typing it every post).

As a case study, consider trials motorcycles. They have largely reverted back to RSU forks after partially adopting USD forks for a short while. The stiffness was simply not important for their low-speed shenanigans, but the weight was. Also, some like to talk about the reduced likelihood of smashing a fork tube against a rock. Maybe we will find this point in mountain biking as well, and the USD fork with its rigidity and its relative weight penalty will simply be out of our usage envelope. But then back to my original point, we are so far below the stiffness values that would approach this optimization point, that wholesale changes in architecture are warranted.


4/15/2021 12:15 PM

Primoz wrote:

One part is definitely packaging, which was covered in the previous topic (I forgot who explained this, sorry, please claim ...more

Not sure if I brought up the RSU MX forks and ruts in the last thread but thats been my understanding. Dynamic bushings aren't necessarily the answer IMO, as you are trying to push the top bushing in to the bending zone. Static bushings with a large overlap makes more sense to my mind.

As for the lefty video - it is an F32 and he's loading it quite a lot so I wouldn't be surprised if thats a totally standard fork. Bushings have massive stiction once loaded, it's a real weakness for sure

@justinc5716 - I would agree that more stiffness is a good thing, again in terms of reducing friction (thats my perspective). If we need to introduce compliance, that can be done with the wheel, which Crank bros and Zipp have already cottoned on to and taken a stab at


4/15/2021 12:53 PM
Edited Date/Time: 4/15/2021 1:02 PM

TheSuspensionLabNZ wrote:

Not sure if I brought up the RSU MX forks and ruts in the last thread but thats been my understanding. Dynamic bushings aren't ...more

Not sure who it was honestly.

Yeah, it's an F32, but most importantly, I'd hazzard a guess both forks are dry. Dry needle bearings will have a lot less friction than lubricated bushings. Add lubricating oil to both of them and the playing field does get a lot more level. Not the same, far from it, but more level still.

Regarding stiffness, this is what I tried to imply in the original post too, this only gives stiffness results without any thought about how changes in stiffness affect performance, mainly through friction in this case. BikeRadar did have a podcast and Seb Stott (now at PB ) did say that he thinks Zeb's advantage over the Lyrik is added stiffness which makes suspension performance better through the fork actually moving through the travel rather than bending backwards when hitting a bump. But by that metric, you could say upside down forks will be worse because of that as they will twist torsionally, which also increases friction.

@justinc5716 more stiffness will give more precision, but on the other hand the fastest driver/rider is a comfortable driver/rider. Could more torsional compliance also cause less fatigue for the rider and thereby make him faster? Maybe over a longer period of time? (EWS? 24h racing? etc.?) There have also been cases where riders took some stiffness out of the bike (Athertons cutting the rocker brace on V3 Supremes). But I do agree that it starts to get complicated and that stiffness is an easy thing to measure. Though for important answers I think a more whole picture needs to be look at, not just and only stiffness.

As for customer demands, customers don't know what they want. Most of us aren't even in a position to try 3 different forks, let alone 3 different topologies of forks. Though for the single/dual crown, right side up/upside down comparison we could do it with a Lyrik, Boxxer, Intend Infinity and Intend Edge. All four are 35 mm in the stanchion diameter and both single crown forks do go up to 180 mm of travel. Maybe that would give some answers? But again, you have different spring and damper technologies with these four forks (okay, similar ones when you take into acount the sidedness).

@hamncheez2003 I agree on the costs issue with roller bearings, but yeah, it would indeed be a solution. And a fork like that actually does already exist, at least it did.

As for DVO... I don't think it does much, at least it's not the game changer some might hope it would be. The issue is that the part is open at the back side, which makes it a C section. Any closed section will be MUCH stiffer compared to a similar open section (even when there is only a narrow slit in the effectively open section). The part is also thin and relatively long. A benefit is that it has two bridges that are spaced fairly far apart, so that part could be relatively stiff given the thickness of it. It helps that it's carbon, but still. Having a closed section would probably help quite a bit, but assembly of the fork would be painful then. One option would be to insert each complete leg from the bottom of the fender up, bolt the fender to the dropout and then insert both legs into the crowns. That way the tubes of the fender wouldn't be overly large (to clear the dropouts) and you'd have a closed section.

I've been thinking about this for a while and I will probably try to simulate it, even if it's just to see if it's a single digit change or if it's more profound. And probably try to make it both a closed and open section while we're at it.

The fact that it's not loading the bushings doesn't influence it much. One part is that the legs of the fork try to move in separate directions. The length for the USD fork is more or less the complete ATC. An RSU fork has a brace mid way up, tying the legs together, so they have to bend in a more banana shape (if the bridge is fixed, the stanchion will bow the opposite way the dropout deforms to). Then there's the added issue of rotation of the stanchions in the outers, which is not prevented. With RSU forks the outers can't rotate as freely, as the bridge keeps them in place.

With the DVO fender, this bending connection is still present, the difference is the legs can bend inside the fender, sot hey are somewhere in the middle when to overall system stiffness. But the main issue I see is the fact that the structure is both thin and open, so it can in no way give the same stiffness as cast lowers have.

Dammit, the more I talk and think about that fender, the more intrigued and interested I am! :D

@justinc5716 part two, for what it's worth, Fox did trial an upside down DH fork, had it under Gee Atherton and I think Gwin and supposedly the racers said it's not stiff enough and the engineer said it would be too heavy to make it stiff enough for the racers.

As for stiffness to weight, if you're looking at torsional stiffness, an RSU fork will always have a better ratio. If just by the fact that the connection between crowns is made from a lighter, thinner part as opposed to the USD fork. Honestly I'm not sure I'd sell the bending stiffness advantage as one of the main points for USD DC forks, I think it'd try to maximise the benefits of the concept and try to make it as smooth as possible - by distancing the bushings apart as far as possible for example.

Rock banging is an issue, one that is easily solvable with fenders, but nevertheless. Yeah, RSU stanchions can be damaged too, but the handlebars protect them much more than USD forks just by the fact that the ground clearance (with the bike laid on the ground) is much higher for RSU forks - the stanchions are so much closer to the bars).

There is another point to take into account here. Headtube angle and the bushing loads that are experienced by the forks. And thus the friction that occurs there. The loads in the bushings should be higher for USD forks when using the same bushing spacing as both bushings are further away from the load itself. This is true for 0 travel and can be the opposite when going deep into travel actually. Plus having a much wider spacing also does have the chance to lower friction quite a bit.

Oh to have an unlimited budget, a good team and the option of a clean sheet design of a MTB fork...

EDIT: I'm looking at the DVO arch:
Kinda seems to me its something along the lines of sheet moulded composite maybe? It'd be possible to cheat by using a polymer with a high fibre content for the simulation then. And it would make sense production wise, it'd be much cheaper than a hand laid, optimized layup part.


4/15/2021 1:43 PM

This was interesting RSU fork from Noleen. It used the sliding bushing on the stanchion. You can see it in the 6-7 minute mark.


4/15/2021 2:36 PM

We've talked about this quite a bit in the previous thread. It does make sense from a lot of perspectives, but makes the construction MUCH more complicated and expensive, as you can see in the video - requires removable dropouts, requires controlled outer surfaces (current designs have much cheaper legs in that regard), etc.


4/15/2021 4:26 PM

justinc5716 wrote:

The idea of torsional compliance being good is strongly misguided. Sure you are providing a means of isolation for some side ...more

I'm speaking from actual experience here. Not just in theory.
I have Dorados on my DH bike. They track better. They beat you up less.
There is no issue with steering accuracy due to phase lag. The bike goes exactly where I aim it. It's not a wet noodle and it is in the sweet spot of torsional rigidity. For reference, I am not a lightweight either. I'm 95kg (about 210lb)
Yes, you need to adjust your expectation of how the bike feels and this might put a lot of people off.

I also disagree with the assertion that building compliance into the wheels is the way to go. When you take tension out of the wheel you lose strength, and they feel dead. It's fine for pros who have multiple sets of wheels to run weak low tension wheels to compensate for too much stiffness elsewhere but it's definitely not the optimum solution. Especially running super stiff carbon rims with low spoke tension. That is just dumb!


4/15/2021 4:29 PM

I would also be interested to see if you could model the effect of the Manitou hex lock axle. That makes a big difference to the torsional rigidity of the Dorado.