Shimano's R9100-P is a chainset-based power meter that offers consistent overall data, and left/right balance figures that are now in line with those of other brands.
- Pros: Neat integration with chainset, very good overall accuracy, one rechargeable battery, excellent chainset
- Cons: Not easy to swap between bikes, broadcasts only over ANT+ at the moment, there are cheaper options
Shimano's R9100-P power meter uses strain gauges integrated in both cranks. Elements of the power meter are visible when the chainset is fitted but they're hardly noticeable, the only tell-tale signs being small rectangular sensors – where the strain gauges are housed – added to the inner face of each crank, and a control unit, featuring an on/off switch, that sits between two arms of the spider.
The non-driveside sensor extends inwards about 8mm while the driveside sensor doesn't extend beyond the inner chainring so there's no issue with clearance there. The control unit is small and not particularly obvious unless you go looking for it. Oh, and there's a small magnet stuck to the driveside chainstay with either a black or a white cover over the top. This is required not only for cadence but to get any power information from the system.
The strain gauges on either side of the chainset measure independently so you can get data on each of your legs – you can see the balance between them, for instance. You can also get smoothness and efficiency measurements, but that's the limit of the pedal stroke analysis – for now, at least.
The separate parts of the system are joined together by an 'LR connector' wire that threads from one side to the other, with just a single signal being sent to the head unit (as opposed to data being sent separately from each side). A rechargeable, non-replaceable battery lives in the axle – just one battery for both sides.
You recharge via the control unit, the cover flipping down to reveal a magnetic charge point for a USB cable. This looks the least robust part of the system by a long way because the hinges are tiny-going-on-minute. That said, Shimano claims that a charge, which takes about 2.5 hours from empty, lasts up to 300 hours so you'll need to open the cover only rarely and it has never come open unexpectedly during six months of testing. With that in mind, you'd have to say it works fine, but I'd still be happier if it was meatier. Call me highly strung but it makes me nervous.
An LED on the control unit shines when you turn it on and during calibration and charging. Press a little button and the LED will glow green if you have more than 20% battery remaining or red if you have less.
The pods containing the strain gauges are sealed on to the cranks – they can't be opened. Shimano says that this design means the system is completely waterproof although it doesn't give an IP rating. As far as I know, no water has got in during wet winter and spring testing or, if it has, the results have been unaffected.
The Shimano Dura-Ace R9100-P power meter adds about 70g over a non-power meter chainset, which is more than a dual-sided Stages system (a claimed 35g), for example... but it's still only 70g. You get all of the other Dura-Ace chainset features including HollowTech II cold forged hollow cranks, a hollow outer chainring to reduce flex, and a four arm spider design.
Shimano boasts that its system features Bluetooth and ANT+ technology, and that's true, but it currently sends out your performance data over ANT+ only. That means you can display it on a Garmin Edge bike computer, for example, a Wahoo Elemnt, or a lot of other devices, but not all. The Bluetooth-ness refers to communication with Shimano's E-Tube app, which allows for wireless firmware updates, although Shimano intends to add Bluetooth Smart power broadcasting later in the year.
Speaking of data, the Shimano Dura-Ace R9100-P power meter transmits power, power balance (left leg/right leg), cadence, pedal smoothness and torque effectiveness. Other than left/right, the R9100-P doesn't show where in the pedal stroke the power is being produced like Shimano's Bikefitting.com system does for further analysis.
You should perform a zero offset before every ride. You simply unclip from the pedals, position the cranks vertically and hit the relevant button on your head unit or Shimano's app on your phone. Alternatively, you can press and hold the button on the control unit until the LED glows blue.
Reviewing a power meter takes an age. I've been using this one since before Christmas in order to get a full understanding of its capabilities. What we do at road.cc is fit two or more power meters to the same bike and compare the results over several weeks' use (of course, using more power meters doesn't always tell you anything extra in that an outlier isn't necessarily wrong; it's not a vote!).
I've chosen the PowerTap G3 Hub as my reference power meter because it has a good reputation for accuracy and it can be used alongside most other power meters out there. If I chose a crank-based power meter as my reference, for example, I wouldn't be able to compare the results to those of other crank-based power meters because I couldn't fit both to the same bike at the same time. That's the rationale. I've also compared data from Shimano's power meter with that from Garmin Vector 3 pedals.
When looking at data from different power meters we can't ever say that one is right and the other is wrong, we can only explain how they compare.
Shimano claims an accuracy level of 2% for the R9100-P, which is similar to that of most other brands (their claims vary slightly but they're usually 1.5-2%).
I've found the overall power data provided by the Shimano R9100-P to be remarkably similar to that of our PowerTap G3 Hub over many, many rides.
Check out figure 1 below, for example, which is a chart showing data from a 54-minute ride. The pink line shows Shimano data, the green line shows PowerTap data.
What I did in that ride was:
5 mins warm up
5 mins at about 300 watts
20 x 15 secs on, 15 secs easy
5 mins at about 300 watts
I then had 5 mins easy before repeating the intervals.
You'll notice that the pink Shimano peaks tend to be a little higher than the green PowerTap peaks, and the PowerTap troughs are generally a little lower, but there's really not a lot in it. They pick up jumps in effort equally well with no real lags or anomalies.
If we go in close, with no smoothing of the chart, we see that the Shimano data is very slightly delayed compared to the PowerTap data, but they essentially show the same thing.
I stress that this is just an example. We have loads of other charts that show the same thing. Other reviewers have found the PowerTap hub data to be a little higher in sprints, but I've found the results to be very similar throughout, no matter what I've tried. Steady state, sprints, hill reps, through and off, you name it, I've not been able to get the Shimano R9100-P to do anything weird compared to the PowerTap hub. Differences between the data have virtually always remained within the margin for error claimed for each device so I've little hesitation in saying that the Shimano's overall power data is reliable.
Figure 3, below, shows data from the Shimano R9100-P (pink) and data from Garmin Vector 3 pedals. The Shimano system measures a little higher, particularly at a few peaks, but the overall results are remarkably similar.
Figure 4, below, shows a series of five-second bursts with five seconds riding easy in between. Both Shimano (pink) and Garmin Vector (blue) pick these up. In general, Shimano measures them a little higher, but usually only by a few watts.
Things get more interesting when it comes to left/right balance: the percentage of the total power that's coming from each of your legs. I compared data from the Shimano R9100-P and Garmin Vector 3 pedals here.
Other people have found that the Shimano R9100-P tends to give a high left leg figure compared to other systems, and I initially found the same.
Figure 5, below, shows data from a 25-minute sector of a ride. The pink line shows Shimano data, the green line shows Garmin Vector 3 data, and the blue horizontal line shows a 50:50 split.
You can't tell a lot from that jumble so below is the same data smoothed out (figure 6). As you can see, both power meters show a dominant left leg, but the difference is far more pronounced according to Shimano. Over the entire ride, Shimano measured my left leg doing 55.5% of the work while Garmin Vector 3 had it at 53.8%, and that discrepancy was fairly typical of what I found over many weeks of testing.
However, Shimano has offered a firmware update since then that seems to have brought its data more into line with that of Garmin.
Figure 7, below, shows a section of a recent ride (Shimano data is pink, Garmin data is blue). At the start of the ride, I unclipped one foot from the pedal and rode one-legged for 30 seconds, reattached and rode with both feet for 30 seconds, then unclipped the other foot, and so on. The Shimano power meter has picked this up whereas the Garmin pedals haven't.
There are a few Shimano peaks later in the ride that I can't explain, but in general the data is fairly similar to that from the Garmin pedals.
Figure 8, below, shows the same ride but with the data smoothed. I couldn't tell you which device is more accurate in its left/right balance figures but I can tell you that the Shimano data is now much closer to the Garmin data than it was prior to release of the latest firmware. On more recent rides, sometimes Shimano gives a slightly higher left leg figure than Garmin, sometimes it gives a slightly lower left leg figure, and sometimes they're exactly the same. Previously, the Shimano figure was always higher.
On the whole, then, the Shimano R9100-P appears to provide reliable overall power data and earlier issues that I had with left/right balance have been resolved.
There are still a couple of negatives, though. First, swapping this system between bikes isn't easy compared to moving a pair of power measuring pedals, for example.
Second, there's the price. Shimano's R9100-P power meter is available in 170, 172.5 and 175mm crank lengths, and with 53/39, 52/36 and 50/34-tooth chainrings, all of them for £1,499.99. A standard Shimano Dura-Ace R9100 chainset has an RRP of £529.99, so you're paying just under a grand more for the addition of the power measuring capability. You can also buy the system without chainrings in 165mm and 180mm crank lengths, the retail price being £1,299.99.
For comparison, a Stages Cycling Shimano Dura-Ace R9100 LR dual-sided power meter has a retail price of £1,199 – £300 cheaper than the equivalent from Shimano – and it already has Bluetooth Smart compatibility. On the flip side, there are more expensive options out there from the likes of SRM and Quarq.
Overall, the Shimano R9100-P offers reliable data, it's easily recharged and it integrates well into a complete Dura-Ace system if that's what you have on the rest of your bike. It's a solid start in this market now that firmware updates have sorted out some teething problems, but there are cheaper options out there.
Consistent, reliable and well-integrated power meter although there are cheaper options on the market
road.cc test report
Make and model: Shimano FC-R9100-P Dura-Ace power meter chainset
Size tested: 52/36t, 172.5mm
Tell us what the product is for
It's a crank-based power meter for monitoring your performance.
Tell us some more about the technical aspects of the product?
Shimano lists these features:
Reliable electric system
Own developed strain gauge
Software compensate automatically
Possible to start quickly
Waterproof in all conditions
BluetoothLE and ANT+
Wireless firmware update system
Integrated rechargeable battery
Easily change chainrings without affecting power
Enough clearance with our direct mount brake caliper
Close integrated design
Active temperature compensation
Battery: rechargeable Li-ion battery
300+ hours of ride time
Quick response and operation without stress
Clearance with frame
Accurate data collection
It's integrated well into an excellent chainset.
Like many other people, I found that the left/right balance data was skewed to begin with, but a firmware update seems to have sorted that. Performance has been consistent and reliable.
The tiny hinges on the control unit worry me, but nothing has gone wrong in six months of testing so my fears appear to be unfounded. That aside, you'd have to be pretty clumsy or unlucky for any real damage to occur.
It isn't as light as a Stages system, for example, but the addition of 70g to your bike isn't noticeable.
It's £300 more than a Stages Cycling Shimano Dura-Ace R9100 LR dual-sided power meter at retail price, but there are certainly more expensive options out there too. It makes much more financial sense if you need a new chainset than if you don't.
Tell us how the product performed overall when used for its designed purpose
It performs very well (after a firmware update addressed left/right balance issues), with our testing showing very similar data to that of the PowerTap G3 Hub.
Tell us what you particularly liked about the product
Neat integration with chainset, infrequent need for recharging.
Tell us what you particularly disliked about the product
Not easy to swap between bikes.
Did you enjoy using the product? Yes
Would you consider buying the product? I'd consider it, but there are cheaper options these days.
Would you recommend the product to a friend? As above.
Use this box to explain your overall score
Now that the left/right balance issue has been sorted, it's an 8 all day long.
About the tester
I usually ride: My best bike is:
I've been riding for: Over 20 years I ride: Most days I would class myself as: Expert
I regularly do the following types of riding: commuting, club rides, sportives, general fitness riding
Mat has worked for loads of bike magazines over 20+ years, and been editor of 220 Triathlon and Cycling Plus. He's been road.cc technical editor for eight years, testing bikes, fettling the latest kit, and trying out the most up-to-the-minute clothing. We send him off around the world to get all the news from launches and shows too. He has won his category in Ironman UK 70.3 and finished on the podium in both marathons he has run. Mat is a Cambridge graduate who did a post-grad in magazine journalism, and he is a past winner of the Cycling Media Award for Specialist Online Writer.