Longer cranks should be stronger


Our Rene Herse cranks are available in three length: 165, 171 and 177 mm. We chose 3.5% increments, because that is the smallest difference you’ll notice as you ride. That part is just common sense. What makes our cranks unique among small-production cranks is that we use different forging dies for each crank length.

Let’s first talk about why we forge our cranks: Forging strengthens the metal because it aligns the grain structure (above). By contrast, CNC machining just carves the part out of a big block of aluminum. You’ll still have the grain structure of the original block, which is now interrupted where the block has been cut away. On a complex shape like a crank, this creates a lot of weak spots. (Aluminum behaves a lot like wood in this respect, where you always want to work with the grain, not cut across it.)
To make up for the lack of strength, CNC-machined parts use more material, making them bigger and bulkier. If you want slender, lightweight parts that still are strong enough for hard riding, you’ll want to forge them.
3_lengths
To obtain the full advantages of forging, the forging die must be close to the final shape of the crank. Otherwise, you start cutting into the grain structure again, and you lose the strength advantages of forging. That is why Rene Herse cranks use different forging dies for each crank length. Above you see the raw forgings. To turn them into cranks, holes are drilled and threaded and the arms are polished. The grain structure of the cranks remains uninterrupted.

Forging dies are expensive, and that is why small makers either CNC machine their cranks or, if they forge them, often use a single forging for all their crank lengths (above, the final forging is at the bottom). The area where the pedal eye will be is elongated, so that the crank can be machined to the final length as needed. This saves money, but it means that the forging’s grain structure is interrupted in the highly-stressed area at the transition to the pedal eye, where many cranks break. Does it matter?
Years ago, the then-owner of TA told me that in the past, they had two forging dies for their cranks. Back then, most riders used 170 mm cranks, so they made a net-shape forging for that length, similar to the Rene Herse forgings above. This made sense, because it eliminated the machining, which was expensive in those pre-CNC days. But there was an added benefit: Very few of these cranks broke.
For the other arm lengths – and TA used to offer many – demand was not enough to warrant a net-shape forging die for each length, so they made the forging with the oblong pedal eye that you see above. This was then machined to the final shape. According to the owner of TA, those cranks were less reliable.

This matches my experience. Recently, I encountered a broken crank (above). Checking the length, I wasn’t surprised that it was a 177.5 mm crank. When I traced the shape of the raw forging on a piece of paper, I could see that the crank broke exactly where the oblong pedal eye started on the original forging, and where the material was removed. It makes sense – this is the most stressed area, because the pedal has the most leverage here.
This doesn’t mean that all cranks that don’t use net-shape forgings will break. Note the oxidation on the broken crank – it’s seen a lot of miles, and it was used on a commuting bike, where lots of starts and stops put great strain on the crank. Still, I sleep better at night knowing that Rene Herse cranks don’t have that weak spot.

When we developed our Rene Herse cranks, we decided that they had to be as strong and as reliable as the best cranks in the world: Our cranks had to pass the EN ‘Racing Bike’ standard, not the less-demanding ‘Trekking/City Bike’ standard that most other small-production cranks meet. The only way to pass that rigorous test is by using net-shape forgings, which require dedicated forging dies for each crank length.
new_dies
Using separate forging dies for each crank length has one added advantage: We can make the longer cranks stronger. If you look carefully, you can see that the arm on the left has a larger cross-section. This compensates for the longer lever of the 177 mm cranks and also for the higher power output and greater weight of taller riders. It’s logical, yet I haven’t seen any other cranks that are beefed up for the longer versions.
This also means that all our cranks – and not just the shortest ones – pass the test. In fact, we’ve tested each length several times to be sure. (A single test might just capture a lucky outlier.)

Making separate forging dies for each crank length triples our tooling costs, but it’s the only way to make high-performance cranks that match the performance and reliability of the best cranks from the big makers, while still offering unlimited chainring choices and an understated classic aesthetic. You don’t make the world’s best components by cutting corners!
Further reading:

 

27 Responses to Longer cranks should be stronger

  1. Neil Hodges March 6, 2019 at 9:24 am #

    Is it possible to order Rene Herse cranks in mismatched lengths?
    I saw a bike fitter a while back and he strongly suggested a 5+mm difference because my femurs are nearly an inch different in length. After years of knee problems a few instances of muscle damage, different crank lengths finally made everything fall into place and I haven’t had any issues since.

    • Jan Heine March 6, 2019 at 9:30 am #

      No problem. Just put a note in the comments that specifies the length (right and left). We custom-assemble each crank with the chainrings you order, so this can be accommodated easily. (And of course, without an upcharge.)

  2. Robert March 6, 2019 at 9:42 am #

    Any consideration given to longer lengths?

    • Jan Heine March 6, 2019 at 9:45 am #

      Yes, definitely considering it. We just need to make sure there is enough demand to warrant the investment in the forging dies.

      • snilard March 6, 2019 at 10:25 am #

        And for shorter?

      • Gert March 6, 2019 at 3:27 pm #

        Yes please
        I use 185 TA, but could live with 183 if You only will do 6 mm larger

  3. D lowrie March 6, 2019 at 11:27 am #

    The broken piece in the photo is a TA crank arm? Looks to me like corrosion failure then it lost strength and broke. Could this be the case with this one? One might have to examine TA crank arms under an electron microscope to determine if milling the end of the arm makes it weaker?
    Just trying to reason thru your statement that machining a cold forged crank arm to accommodate different arm lengths weakens it – by how much. .

    • Jan Heine March 6, 2019 at 11:48 am #

      When we first saw the crack, we wondered why it had cracked in the middle. It was only when we traced the original forging (see paper) based on a shortest length of 150 mm (which TA offered back then) that it made sense – it’s right where the milling starts. The information that the ‘multi-length’ forging was more susceptible to breaking than the net shape forging came from TA directly.
      Regarding corrosion failures, TA’s arms are made from 2014 aluminum – a material that isn’t very susceptible to stress corrosion cracking. Some other small makers use 7075, because it’s even stronger on paper, but the stress corrosion cracking makes 7075 ill-suited to cranks. If these were 7075 aluminum cranks, I’d suspect a corrosion failure, too.
      (Background: Stress corrosion cracking occurs when parts are stressed while they are exposed to moisture. 7075 suffers from stress corrosion cracking – it’s great for airplane parts that are protected inside the fuselage – but not for highly-stressed bicycle parts. We make our chainrings from 7075 for wear resistance – chainrings don’t break – but the crankarms use 2000-series aluminum.)

  4. Rick Thompson March 6, 2019 at 12:09 pm #

    I’m 200 lb and climb standing quite a bit. Good to hear my RH 177 cranks are designed for that. I’m also knock kneed and duck footed, enough that the low-Q you prefer does not work for me and I have 20 mm pedal extenders installed. That has to put more torque on the cranks, do you feel confident they are capable of handling this?

    • Jan Heine March 6, 2019 at 2:13 pm #

      I’m not worried about your weight and height. The pedal extenders… I checked the test procedure, and it puts the load 65 mm outboard from the crankarm. A standard pedal puts the center of the cleat at about 55 mm, so you have 10 mm of leeway. I suspect the extra 10 mm of your 20 mm extenders are fine – but we haven’t tested that, so I cannot guarantee it. If you want to be safe, perhaps you can replace the extenders with 10 mm-wide ones?

  5. Allen Potter March 6, 2019 at 12:36 pm #

    Customer testimonial! I have a crankset from your very first batch. What year is that, 2011? When I swapped it from my daily rider to my fancypants new rando bike a couple years later, I made a big mistake, not limiting the travel of my front derailleur cage. I cranked it out and so it clicked against the crank arm for some time before I figured out what I was doing wrong. By that time, I had creased a groove right out of the inside of the arm. I worried that I’d seriously compromised the integrity of the crank. But I was too embarrassed to ask anyone if it was a problem, and too broke to replace it. So I rode it like that. And rode it. And I’m still riding it. To say I’m satisfied with my Rene Herse crankset is a woeful understatement. What a terrific product!

    • Jan Heine March 6, 2019 at 2:09 pm #

      With a groove like that, make sure there are no sharp edges. File and sand it smooth, and it should be fine. The strength is very slightly compromised, but I wouldn’t worry about it.

      • marmotte27 March 9, 2019 at 7:49 am #

        Sh….ugar!!! I bought a couple of the latest version (70th anniversary reedition, 172.5mm) of the TA cranks. And I realise that the end of the derailleur cable has become bent the wrong way some time ago so that it has now worn a slight groove into the left crankarm.
        If I understand correctly, there’s no chance they made more than one forging die for this edition, and the shorter length cranks are probably cut down and milled from the longest 180mm version? And the groove has weakened the crankarm further?

        • Jan Heine March 11, 2019 at 8:57 am #

          You are right, the anniversary forging is shown in the photo – there was only one, which was then cut to length. I wouldn’t worry too much about it, unless you are somebody who tends to break components. Make sure the groove is smooth – it probably is, considering how it formed. And watch the crank – usually, cracks appear a bit before the crank fails. If you’ve broken cranks in the past, I would consider switching to a stronger crank…

  6. jasonmiles31 March 6, 2019 at 2:05 pm #

    I understand that RH cranks maintain a classic appearance and that is a major part of their appeal.
    However, other cranks have used topology optimization and more radical forged shaping to develop higher stiffness cranks while maintaining similar weights.
    https://solidthinking.com/customer-story/Race-Face/design-and-manufacture-bicycle-crank/
    Seems strange to focus so intensely on having length specific forgings but not to focus on other shaping improvements. Are RH cranks designed specifically to have flexible arms?

    • Jan Heine March 6, 2019 at 2:39 pm #

      The shapes of the Rene Herse cranks are the result of careful FEM (finite element model) analysis to optimize the shape. They look similar to the 50-year-old originals, but the details are quite different. For example, this is easy to see in the transition from the pedal eye to the arm. Mostly we sought to improve strength and durability by reducing stress concentrations. We also looked at stiffness, but decided not to worry about it.
      It’s interesting how often stiffness is treated as a desirable quality without anybody ever having tested whether it improves performance – whether it’s in the frame, the cranks or the fork. From all we can tell is that crank flex is so small that it doesn’t matter. If there was more of it, it might be beneficial – like the ‘planing’ of the frame or the flex you get in wide tires.
      Stiffness is useful in many engineering applications, but stiffness by itself isn’t always beneficial. An example from the car world: A stiff bodyshell allows the suspension to work properly, but this works only with independent suspension that has sufficient travel. Many great cars of the past used the frame as part of the suspension – off-road trucks still do… Racing motorcycles had a problem that the suspension stops working when the bike is leaning deep into a turn (the bumps no longer are in the direction of suspension travel). A stiff frame resulted in the bike skipping and losing traction.
      Human bodies are far more complex than cars, trucks or motorcycles, and it makes no sense to ‘optimize’ stiffness without first researching whether stiffness is a desirable feature.

      • jasonmiles31 March 7, 2019 at 4:34 pm #

        I totally agree that a stiffer crank may not be better. I think it is just interesting that the RH cranks are relatively thin I-beam shape. I cannot think of a more flexible stiffness to weight shape for a crankset.
        Your planing frame experiments have shown that stiffness can be very important for performance, it would be interesting to extend this test to other components.

        • Jan Heine March 7, 2019 at 5:38 pm #

          The groove in the front of the cranks isn’t very deep. Mostly, the cranks are an almost square cross-section, which is close to optimal when you consider that cranks are mostly stressed in torsion. Round would be better, but hard to integrate pedal eyes…
          If crank flex is desirable, we should try to use an optimized shape that is as small as possible, rather than make an inefficient shape that is bigger.
          Even if you want the stiffest possible crank, there isn’t much you can do: The width of the cranks is more or less fixed – basically the length of the pedal threads. Making the cranks wider would increase the Q-factor, which isn’t desirable for most riders… So most modern cranks are taller, but that adds a lot of material for little gain. That is why our Rene Herse cranks are still among the lightest in the world.

      • jasonmiles31 March 7, 2019 at 9:11 pm #

        I strongly disagree that there is not much that can be done to increase crank arm stiffness.
        Modern cranks use hollow crank arms that are much closer to the ideal shape to maximize torsional stiffness. These cranks may be hollow forged, two bonded clamshells, or bladder moulded. Additionally they use oversized hollow 30mm spindles to reduce spindle twist.
        Many crank arms are mfg using carbon fiber which has a higher stiffness to weight ratio.
        I haven’t seen a stiffness test for the RH cranks, but the data from fair wheel bikes is pretty interesting. http://blog.fairwheelbikes.com/reviews-and-testing/road-bike-crank-testing/
        The RH cranks are probably most similar to the 2006 Campy Record cranks. They have 45% more deflection than the stiffest cranks and 57% lower stiffness to weight than the crank with the highest stiffness to weight.

      • jasonmiles31 March 7, 2019 at 9:18 pm #

        That said all of these modern “improvements” do have their downsides.
        I would much rather run a JIS bb than some of the PF BBs. Additionally the RH cranks come with some really interesting CR combos.
        If you are interested in testing crank stiffness. I would be happy to supply a very stiff FSA crankset. Would be interesting if a flexible crank can makeup for a frame that does not plane well or if a stiff crank can heighten a frame that does.

        • Jan Heine March 8, 2019 at 7:31 pm #

          What I meant is that the shape of a crank is constrained by the need to fit into the space restricted by the chainstay on one side, and by the need for a low Q-factor on the other. You are right, different materials can solve that: A steel crank would be far stiffer. If somebody shows that stiffer cranks are better, perhaps we’ll offer the Rene Herse cranks in steel.
          Beyond that, most of the flex isn’t in the arms themselves, but in the BB spindle and the interface between crank and spindle. As you mention, a square taper isn’t the stiffest, but it’s got many other advantages, from fine-tuning the chainline and Q factor via different spindle lengths to its small size that allows such a slender (and lightweight) crank.

  7. Peter Stokely March 6, 2019 at 3:41 pm #

    Jan, I enjoyed the information about net shape forging. I would like to learn more about the process for small parts like crank arms. Can you point me to a video or other media to I can lean more? Is the process cold forging or hot? Is the aluminum injected or pressed into the die? Stuff like that. I am just a curious avid cyclist (and subscriber to Bicycle Quarterly) and I like to know how things are made. I have very much enjoyed Paul Components videos on how they stamp and cut parts.
    Thank you

    • Jan Heine March 6, 2019 at 3:49 pm #

      You’ll enjoy this blog post about forging the cranks. Once you see the size of the forging hammer, you realize why no bike company except Shimano can forge their cranks in-house.

      • Peter Stokely March 7, 2019 at 5:49 am #

        Thanks Jan, very interesting and now I more fully understand the process!.

  8. Nick Bull March 7, 2019 at 7:22 am #

    Great article, thanks, Jan. I’d be interested to read an article about how the EN testing is done complete with photos of destroyed crank arms and measurements of the torque that destroyed them 🙂
    Nick

    • Jan Heine March 7, 2019 at 10:12 am #

      You can see the test setup here. It’s a fatigue test: If your crank survives 100,000 loading/unloading cycles, the crank passes the test. The difference between the easier ‘Trekking/City Bike’ and the harder ‘Racing Bike’ tests is that the load on the cranks is much higher.

  9. Joe Torres March 9, 2019 at 10:45 am #

    Absolutely an interesting article and comments. Always wondered how forging was done. Quality products cost money.