Chainring Choice or Shifting Ramps?

The chainring choices of our René Herse cranks are not limited by dedicated shifting ramps, so you can use any gear combinations you like. We have optimized the chainring tooth profile to shift well at all times, and not only twice per crank revolution when a ramp and pin are aligned correctly.

During the development of our cranks, we spent a lot of time testing different prototype chainrings (above), as well as the ramped-and-pinned cranks of other makers (below).


Modern chainrings have ramps and pins on the backside below the teeth. The ramps and pins are located where there is an optimal path for the chain from one ring to the other. For that reason, modern chainrings only work in sets of two (or three for triples).

If you put a small chainring with a different tooth count on the cranks shown above, then the ramps of the big ring no longer line up where they should. That is why the large rings are marked not just with their own size, but also with the size of the small ring that is part of the set. In the photo below, the large ring is a 50/34 ring, with ramps that don’t work with other small chainrings, like a 39-tooth ring.

Most component makers offer very limited chainring choices, otherwise they would have to develop a multitude of chainring pairs. Most double cranksets today are available only with 53/39 and 50/34 chainrings.

Some smaller manufacturers offer ramped chainrings that are not designed in pairs. (They are easily recognizable, because they don’t specify for which small ring they are designed.) These ramps are not very effective and mostly serve to reassure customers who see ramps as an important asset of chainrings.

For many decades, chainrings did not have ramps and pins, and yet they shifted fine. Ramps and pins serve only as “insurance” against bad shifts that occur when the rider doesn’t push the lever far enough, or when they forget to let up on the pedals during the shift. Most of the time, the rider initiates the shift when no ramp is aligned correctly, and the chain just moves to the big ring without the help of ramps and pins. (An exception are Shimano STI triples, which don’t work without ramps and pins.)

For the new René Herse cranks, we had to make a choice: Design a few chainring combinations with ramps that offer insurance against bad shifts, or offer almost unlimited chainring choices without ramps. (The third option, to provide “cosmetic” ramps, was not considered.)

It would be prohibitive to provide ramped chainring pairs for each of the dozens of chainring combinations possible with the new René Herse cranks. We would need to develop no fewer than six 48-tooth chainrings, depending on whether you want to use a small chainring with 32, 34, 36, 38, 42 or 44 teeth. And so on for each chainring size! (Now you can understand why even big makers offer only very few chainring choices.)

Instead, we focused on the tooth profile to make sure the chain has an easy path onto the chainring – not just in a few places where there are ramps and pins, but at any spot in the pedal stroke. We tested a number of tooth profiles with a variety of derailleurs to determine how to optimize the shifts without ramps. In the photo above, you can see how the chain runs diagonally between the teeth at the onset of the shift. We use an asymmetric tooth shape that provides more room for this shift. (The teeth bear the chain load only on one side, so there is no need to have as much material on the other side.)

Here’s a bad shift, just what we don’t want, where the chain rides up on the chainring at first, and only engages after half a chainring revolution! This “prototype tooth shape C” was not selected for production…

The downshift to a smaller chainring is relatively simple (above). The chain simply drops down onto the smaller ring. It works every time, without any ramps, pins or special tooth profiles. Small rings wear faster than big ones, so ours use a different tooth profile from the large ones, one that is optimized for durability.


You probably will not notice the optimized chainring tooth profiles when you install your René Herse cranks, but we hope you will notice the difference once you ride them on your bike. Click here for more information about the René Herse cranks.

Further reading:

About Jan Heine, Editor, Bicycle Quarterly

Spirited rides that zig-zag across mountain ranges. Bicycle Quarterly magazine and its sister company, Compass Cycles, that turns our research into high-performance components for real-world riders.
This entry was posted in Product News, Testing and Tech. Bookmark the permalink.

24 Responses to Chainring Choice or Shifting Ramps?

  1. Bob says:

    Great explanation Jan. I’m violating the manufacturer’s (FSA) recommendation by running a 34/46, but like you said, ramps and pins are either cosmetic or serve only for insurance. It helps that I swapped out STI levers for bar end shifters (front is friction). A well adjusted derailleur of the right design, coupled with basic operating skill in shifting, is more important than having ramps and pins. I would also be interested in a technical analysis (if not already done) of the shape of rear cogs (twist tooth vs the older flat design).

    • Conrad says:

      I’m also interested in a technical analysis of rear cogs. Any of the currently available cassettes ( I use shimano or SRAM 8 or 9 speed cassettes) have profiled teeth on the cogs. I find that even if I replace chains at the 0.075 mark using a chain checker- the cassette still needs to be replaced after every 4th or 5th chain- or the chain skips under load. I wonder if an older cassette or freewheel with non profiled teeth would be more durable and still shift decently?

      • You make a good point. The more cutouts you have on the teeth, the faster they wear. Especially the old “twist-tooth” Shimano Dura-Ace and 600 freewheels shift very well even with mediocre rear derailleurs. With an excellent derailleur like the Nivex, even straight-tooth freewheels like old Reginas offer excellent shifting. The old 5- and 6-speed freewheels are very durable – I usually get at least 40,000 km (25,000 miles) out of them before a few cogs need replacing.

      • Steve Palincsar says:

        We used to have to change those old-time freewheel sprockets fairly often, too, else they’d skip under load. The biggest difference is that in the days of the cog board, you could replace individual sprockets, changing only those that were worn.

      • I think a lot depends on how often you change your chain. When I was racing, my cogs lasted 25,000 miles, and then I changed them not because they skipped, but because they looked a little worn. This was on a bike without fenders, so there often was a lot of grit on the chain. As you say, it was easy to replace cogs in those days…

  2. rodneyAB says:

    Being unwilling to embrace new shifting technology, I’m quite pleased with the Rene Herse crankset 48/32 used with a 1980’s super-record FD and matching downtube shift lever and cable. I do use a 9 cog cassette with friction mode bar-con, and I do need to tweak lever position for the FD after a couple cog changes, but overall the chain-ring shifts are just what I want.

    • The trimming of the front derailleur is simply a function of the front derailleur design and gearing you use. At least you can trim the front derailleur after rear shifts – many modern systems don’t allow this, and you get the chain to rub on the derailleur cage in certain gears.

      • msrw says:

        “…..many modern systems don’t allow this, and you get the chain to rub on the derailleur cage in certain gears…..”

        I’m curious which modern systems you’re referring to. I’m not a great fan of brifters, but have ridden and raced Ergo and STI extensively. The adjustment needs to be more precise particularly with Shimano, but neither system, when properly adjusted, imposed chain rub on the front derailleur cage in any gear combinations.

      • SRAM and Shimano only have a single half-click, which often isn’t enough to keep the chain from rubbing in all gears. With the high-end systems from Campagnolo, you have multiple intermediate clicks, which allow you to fine-tune the front derailleur. The setup for Shimano and SRAM would be easier if there was a barrel adjuster on the front derailleur, too, not just the rear. Then you could set up the derailleur that the half-click falls so that the gears you use most often don’t rub.

  3. Paul Ahart says:

    Very good essay on shifting. One aspect of front shifting you didn’t mention is the effect of different kinds of chains. I’ve been using 46x34T Salsa non-ramp/pinned chainrings on a Ritchey cyclocross crankset, with a 9spd bar-end shifter setup. Front derailleur is a Shimano R443 “Flat Bar” derailleur. I’ve long used Connex-Whipperman chains, which have long life and shift well. However, with my current setup, shifting form the 34T to the 46T was very poor. Since Connex chains have non-bulged sideplates, I wondered if switching to a SRAM chain with bulged side plates would make a difference. The effect was instantaneous and amazing! Front shifts are now flawless. The bulged sideplates allow the teeth on the larger ring to catch faster, resulting in a perfect shift.

  4. stig says:

    “We use an asymmetric tooth shape that provides more room for this shift. (The teeth bear the chain load only on one side, so there is no need to have as much material on the other side.)”

    I’ve always felt that one of the advantages of unramped, unpinned chainrings with a high tooth profile, is the ability to flip the ring and wear the teeth on the other side. I guess this is not a recommended practice with the Rene Herse chainrings ?

    • You cannot flip chainrings (of any brand), because only one side is recessed for the chainring bolt and nut. (Unless you put the outer ring in the inner position…) In addition, you are correct about the asymmetric tooth profile: The René Herse chainrings should not be turned around. An outer chainring should not be used in the inner position and vice versa.

  5. Rich Freeman says:

    With all the different tooth profiles in the world, developed over several decades for derailer shifting, what were you looking for that wasn’t already available? I would think that non-ramped rings were fairly well understood by the time ramps were introduced. Did you see room for improvement, or more to the point, have you made improvements to the state of the art back then?

    • Before ramped chainrings were introduced, most chainrings were symmetric. Front shifting wasn’t a big issue. With STI and Ergo, the demands on a front shift are higher, since the rider doesn’t have quite the same control over the movement of the derailleur.

  6. Vlad says:

    As good as the Herse cranks and rings are (tooth profiles and all), their versatility is limited by the fact that they are offered in one crankarm length. 171s are just too short for some riders.

    • You are right. There are riders who may benefit from 190 or even 210 mm cranks. However, few manufacturers offer them. As to cranks that are 1.5 or 4 mm longer, I wrote about this earlier. I’d like to see a double-blind test before I believe that a 2.8% difference in length really makes a difference.

    • Andrew says:

      And too long for others. I have tried every crank length from 155 to 175, and 165 is my length of choice. 170 is ok for short rides but they give me sore knees on long rides.

      • It’s hard to please everybody… I wonder where you’d consider the cutoff. If 165 mm is perfect, but 170 mm doesn’t work, is 167 mm still OK? What about 168 or 169 mm?😉

        Jim Papadopoulos once said that cyclists tend to identify potentially meaningful variables and then argue over meaningless differences.

        The hard part is establishing what is a meaningful difference. Generally, it appears that 5% or more can be meaningful. For fork offset and trail, this means about 2-3 mm. For handlebar width, it means about 2 cm. For cranks, this would mean about 8.5 mm. To be meaningfully different, our cranks would have to be 180 mm or 162 mm long. That matches my own experience.

        Generally, I find the width (tread/Q factor) more important than small variations in length. I prefer cranks with less than 140 mm tread, but I am happy to ride cranks that are spaced 5% further apart (147 mm). Much wider, and it becomes noticeably sub-optimal.

      • Andrew says:

        I’m not sure where my sensitivity threshold for cranks is exactly, but the most common increments are in 5mm, so that’s mainly what I’ve tried. And 5mm differences feel very different to me. I also think that shorter riders are probably more sensitive than tall people because a long crank results in an overly tight bend at the knee which I think is responsible for the pain that I feel at times. Taller riders may feel that they lack leverage with a short crank but are less likely to actually experience discomfort.

        For me, the difference between a 160 crank and 170 is huge. The 160 doesn’t give me enough leverage and I feel like I’m having to shift into small gears all the time. The 170 creates a noticeably tighter bend in the knee and feels awkward.

        But of course for a small manufacturer of a fairly niche item like the Rene Herse cranks it wouldn’t make any sense to cater for outliers, so your reasoning is sound. I’ll just have to stick to the TA pro 5 vis…

      • It’s definitely a topic that warrants further study. Maybe I’ll come to Australia some day, and we’ll do a double-blind study with a bunch of TA cranks.

      • Andrew says:

        Good idea. It’s a shame I sold all of the other lengths!

  7. Vlad says:

    I ride 180s on a bike that’s set up on a computrainer and it works fine — on the computrainer. I’ve been riding bikes with 185s most of the past year, then a few days ago I went back to a 190 I’d used for 7 years previously and it felt just wonderful. I’ve tried the Herse 171 — it’s a drop-dead gorgeous crank — but I find it too short for my liking, alas.

  8. Vlad says:

    Further to the above, the difference between 171 and 190 is more than 10% — statistically very significant. TA makes 185mm cranks and I can think of three other companies that make cranks over 20cm. Wistfully wishing RH would jump on the longer crank bandwagon — it could use a few more passengers!

    • Vlad, I agree that if you need 190 mm cranks, then our cranks won’t work well for you. Neither will Campagnolo, Shimano and SRAMs. There are some small makers who offer extra-long cranks. For us, the cost is prohibitive, since we use a net-shape forging process to ensure the strength and reliability of our cranks.

Comments are closed.