Disc Brake Rotor by Shimano

In this article, we’ll be discussing a Disc Brake Rotor by Shimano, US patent 10,920,839. The publication date is Feb. 16th, 2021 and the filing date is June 21st, 2018.

Brief Summary (tl;dr)

Shimano are introducing a rotor that contains non-metallic (resin) inserts that are poured into recesses of the inner portion of the rotor. A magnet can also be place inside the resin. They say this can change ‘mechanical characteristics’ of the rotor, whatever that means.


Modern bike brake rotors are pretty incredible little pieces of engineering. There’s a lot of time and money being poured into developing a rotor that’s lightweight, has good thermal properties, and has a long life. As with any engineering process, as you turn on dial, other dials turn. You want a rotor with a long life? Well, it’s going to weigh more. You want a rotor with good thermal properties? It’s not going to last very long.

Rotors, or brakes in general, aren’t something I go cheap on. These manufacturers have defiantly improved their design and manufacturing of rotors in the last few years with very serious R&D. I’m currently running SRAM centerlines and I’ve never once had an issue, and never once questioned if my bike would stop from full speed. All of these manufacturers make sure these things work nearly 100% of the time, and this invention is just another step toward improved performance.

I wrote an article a while back about a fastener-less rotor by SRAM, where the center part is aluminum and the outer part is a spray-on stainless steel. SRAM are doing this for less weight and better cooling. As far as I’m concerned, this particular rotor appears to have the most potential compared to any other rotor I’ve seen, either in a patent or on the market.


Shimano are introducing a slightly different method to “change the characteristics” of a brake rotor. The rotor is question is a two-piece rotor, attaching the center portion to the outer portion with a set of fasteners. This particular rotor looks awfully similar to the current XTR RT-MT900-S, which will set you back an $85 each. Lordy. I’d bet this rotor will cost even more.

The unique part of this is Shimano are pouring a non-metallic insert inside recesses of the inner portion of the rotor, and a magnet inside the non-metallic piece. The non-metallic piece is designed to change the mechanical characteristics of the rotor, such as rigidity. They don’t state anything about temperature, but I’d be willing to be this would affect brake temperatures, too. The magnet …is used to sense rotation of the disc brake rotor 10.

I don’t have much more than that, as far as sensing rotation. Reading between the lines; Shimano may be working on an integrated wheel speed system and this rotor can be used with that system.

Edited 2/24: Some sleuth readers have brought to my attention that this magnet is currently being used for E-bike wheel speed systems, thought it may be used for other reasons that haven’t been publicized yet. The image below shows this exact rotor (RT-EM910) with the magnet integrated into one of the arms of the rotor (Fig 8 below). So, this patent is a variation of this design with resin inserts that change ‘mechanical characteristics’, along with maintaining the magnet.

Intended novelty

The intended novelty here is a two-piece rotor, where the center portion is a separate member to the outer (brake track) portion, where the two-pieces are attached with a fastener. Then, there’s an insert is at least partially mounted inside the center portion without using a screw connection, and the insert may contain a magnet, though the magnet may not be necessary for the legal novelty. There’s also a limitation that I’m wondering if anyone can figure out what it means:

…the [insert] has a radially outermost edge that is radially inward of a radially innermost edge of the fastening member.

They typically add limitations like this to get patents through the USPTO.


As with every other Shimano patent, they’re super vague about the reason they’re developing these new products. The lines below are the only ones I could find that suggest anything.  Shimano state that they’re doing this to change mechanical characteristics, save weight, and improve rigidity of the rotor.

…it is possible to adjust mechanical characteristic of the disc brake rotor by changing a material of the second member…it is possible to save weight of the disc brake rotor with adjusting mechanical characteristic of the disc brake rotor…it is possible to maintain or improve rigidity of the first member with adjusting mechanical characteristic of the disc brake rotor.


Figure 2 shows the rotor in question. Here, we’re focusing on the center portion 16, recess 24, and the insert 20.

FIG. 3 shows the insert in an exploded view. Each of the arms of the center portion have a recess and can accept the insert. So, in the case of this particular rotor, there can be 4 inserts inside the arms of the center portion. Additionally, component 30 is the magnet, located inside the insert itself. Figure 4 shows the rotor with all the inserts inside the center portion of the rotor. Each insert can contain a magnet, or none of them can contain a magnet.


Figure 5 shows a side cut-out view of the rotor. I’m fairly certain Shimano are stating that the insert is made by pouring a non-metallic resin into the rotor itself, but it’s hard to tell. They’re very vague. Component 28 is the resin and component 30 is the magnet. This is what I’m reading:

…the [center portion] 16 is made of an aluminum alloy or a magnesium alloy. The non-metallic part 28 is made of a resin material. The non-metallic part 28 is formed by casting a melted resin material to the recess 24 in a molding process (injection molding process) of the [insert] 20.

Figures 6, 7, and 8 are other possible configurations of this system. Figure 6 shows a removable insert, where the insert can be attached to the rotor with snap-fits 20S and a magnet 30. Figures 7 shows the insert only, without a magnet. Figure 8 shows the magnet only, being held inside the inner aluminum portion.

Figures 9 and 10 show the insert in a more detailed view.

…the arm 18 includes a projection 38 provided in the recess 24. The [insert] 20 includes a groove 40… the projection 38 is provided in the groove 40 in a state where the [insert] 20 is provided on the recess 24.

The insert is kept in place using a coupling part 20C of the through hole 26. In other words, the non-metallic insert is poured into the rotor and the beveled edges of the coupling portion keep the insert in place.

The second member 20 includes a coupling part 20C provided in the through hole 26. The coupling part 20C couples the second member 20 to the first member 16 to be non-detachable from the first member 16 without substantial damage.

Shimano state one of the novelties they’re trying to claim. The insert is not connected to the rotor with a screw, bolt, nothing. It’s just sitting in there, held in with the beveled edges.

The [insert] 20 is secured to the first member 16 without using the screw connection.


Because Shimano are so vague on this one, it’s very hard to tell why this is being developed, and what the use case is. Is this for road or mountain bikes? Are we making the rotor stiffer or more elastic? Is this insert going to help thermal properties? What’s the magnet for? I can make assumptions but I’d rather present the idea and let the reader tell me what they think on this one.

6 thoughts

  1. The magnet would be for e-bike use, where magnet sensor style speed detection is required. Magnets on rotors are the standard method.
    I imagine that as shown, there’s not much to it. Slight stiffness gain? Keeping dirt out? Some of their rotors are notoriously difficult to clean.
    But what if it’s more like their outer chainrings where it’s a hybrid hollow structure? Or the aluminum spider is so thin that it is not strong enough without the plastic (or carbon reinforced plastic)? That could save weight.

  2. My reckless speculations:

    • Checking for wheel lock-up for ABS
    • Monitoring speed for active suspension
    • Noise and vibration damping

      1. Yeah, a really good blog I read tipped me off 😉

        Technologies like those always trickle down from cars to motorcycles to bicycles. At the consumer level, cars have had ABS for over forty years, and motorcycles for about thirty. Cars have had active suspension for almost thirty years, and motorcycles for almost ten. K2 / Noleen had a primitive active damper almost 25 years ago.

        Both these things are surely coming, but the timeline is impossible to know and the performance of first implementations usually feels like a “public beta test”.

    1. My first inclination would be noise and vibration as well. If you change the stiffness you change the frequency of vibration which changes the pitch of the squeal? Damping it with resin could eliminate the “goose call” noise altogether? For ABS I think you would need greater resolution like you would find with an encoder.

Leave a Reply