In this review of the ROTOR INpower 3D+ I am going to cover the following:
- The general issues of installing a crank-axle power meter on YOUR bike – is it easy? or not?
- A look at OCP/OCA fine tuning for Q-Rings using ROTOR’s software and MAS spider
- A look at Torque Effectiveness and Pedal Smoothness across two power meters
The industry is aware that MANY cyclists are investing in power meters for the first time. It’s an expensive but important investment. There’s a LOT of cycling-specific terminology that scares many people so hopefully I will present this review in an accessible way for all to read and understand.
For those of you who know your UBBs and BB30s then hopefully there is some value here for you too.
What Kind of Power Meter Is It?
If you like, the INPOWER 3D+ is a ‘proper’ power meter using strain gauges. These are located inside the axle/spindle that supports the two crank arms – one for each pedal.
The INpower 2D+ is left-side, ANT+ power only. It automatically provides cadence via ANT+ too. The 2INpower dual-sided versions are out in May 2016.
The INpower 3D+ comes as two aluminium crank arms with the left one pre-attached to the axle. With my model here, the drive-side/chain-side also has the integrated MAS spider onto which attach the two front chain rings/cogs.
The earlier “ROTOR power” series are crank-based power meters. Don’t get them confused.
What are the pro’s and cons of this location?
Along with any power meter solution located ‘somewhere on the bits you pedal’, the system has the potential to be able to support all the fancy Garmin Pedal Stroke metrics Pedal Smoothness (PS) and Torque Effectiveness (TE) as well as provide cadence. That’s great.
Unlike most other locations the crank axle is a safe place for the location of your expensive power meter; it won’t easily get damaged like a pedal, or stolen like a rear wheel. It pretty much can only be installed in one precise way. There are no fiddly bits to fasten onto your crank arm and no specific need for frequent recalibrations.
You even have sufficient space inside the crank axle to house a nice big AA battery – powerful, long-lasting and easy to change.
The only real downside is interchangeability between bikes. BUT it really IS simple to change the whole thing from one bike to another in less than 5 minutes IF both are fully compatible in terms of the Bottom Bracket – that’s the hole in the frame where it fits into, complete with the bits such as bearings. More on that later.
Specifications & Compatibility
The INpower is UBB compatible – Hint: the ‘U’ is for universal. ie potentially very easy to fit and change between many bikes.
These two diagrams might help you with your compatibility and, as you can hopefully see, the UBB means compatibility with a wide variety of BB factors namely: frame shell width (68-86/92mm); inner diameter (37-46mm); threaded vs non-threaded; and 30/24mm spindle.
The INpower 3d+ Power Meter is compatible with ROTOR’s elliptical Qarbon Q-Rings which attach to the MAS spider. The MAS spider can allow even more precise Qring positioning than the spider you most likely currently use. This setup has the potential to make you EVEN faster than training by power alone. Potentially!
When I say ‘compatible’ I mean that the elliptical chainrings should not mess up the power data. Very few power meters (non-hub based) support elliptical chainrings.
The ROTOR Qarbon Q-Rings themselves will most likely fit your current bike and are designed to work with all 9, 10 and 11-speed drivetrains. It is available in 110 BCD and with 3 teeth versions: 53, 52 and 50. Qarbon can be combined with any 110 BCD inner ROTOR 5 arm chainring – NOT Shimano 4 bolt 110 BCD.
Larger chainrings are more prone to flexing and the Qarbon construction is supposed to reduce that.
The ROTOR INpower 3D+ specification document is (here).
Other Techie Details
Power measurements are taken 200 times per second ie 200Hz. Data is transmitted at the end of each complete revolution. Although from the specs it further says that complete-revolution power is sent 4 times/second.
This is frequent data transmission (others are higher) but note that your display device may choose to display and/or save power data over totally different timeframes eg per second or SMART recording.
Now you know. For most of us I don’t think it matters too much.
Contents & Unboxing
Contents and unboxing are covered separately in this post (here).
I like painting. I like watching it dry. Reading about me unboxing a product can be as exciting as you watching me watching paint dry. So you CAN skip that link. You won’t be sorry if you do!
Generic Installation Instructions
The installation instructions will vary slightly from bike-to-bike. You can get a flavour (here) of how up-to-the-task you are to do it yourself by following my efforts. So that link could be worth a glance if you are contemplating buying these and installing yourself.
RESOURCE: The manual containing detailed instructions is <here>.
RESOURCE: The technical specifications are <here>.
Optimising the Qarbon- Q-Ring Position
As part of the installation you’ve just completed you would hopefully have downloaded new firmware using the INpower software. You’ll need the same INpower software (V1.4 and above) for the optimisation as well as for that firmware upgrade.
RESOURCE: <Here> is the manual for the INpower software.
RESOURCE: This is the official ROTOR OCA Profile Test Protocol (summarised!)
“3x 90 secs @ FTP, 90 secs active recovery”
The INpower software disables the communication between the power meter and my head unit. Data will only display on the INpower software once the two are properly connected. Well, that’s what happened to me.
Once you have connected the INpower device with the software, you skip straight to the Advanced Training Mode. If you are familiar with the WATTBIKE Polar View then this is a similar pedal stroke display concept to that.
So you start pedalling and perhaps try a few different ride-positions. The green line in the image below shows the Optimum Chainring Angle (OCA) and in the top right hand corner it shows me that the corresponding Optimum Chainring Position (OCP) is 4 [more likely 4.5]. Actually mine was 5 most of the time but pressing prtscrn whilst pedalling was tricky – you can optionally save data to a CSV/FIT file when testing. As it turned out I had never used OCP=5 before, implying that all my previous use of elliptical chain rings was sub-optimal. Now I know.
I found that the recommended OCA was consistent so long as I was seated. When standing the OCA was higher by 5 or 6 degrees. I don’t stand much. Now I know why 🙁
You can see my poorer pedalling technique, above, when standing. The larger red circle is the positive power created. The inner red circle represents negative power or failing to de-weight the rising foot. This is a great way of telling me to sit down when going uphill. When I do that the smaller circle gets much smaller. ie a more efficient technique, wasting less torque.
Power Meter Accuracy Using Elliptical Chainrings
I usually have 1 or 2 other power meters to beg or borrow but not at the moment, just one other to compare to ROTOR’s INpower 3D+. So I thought I would do something a little different on the testing.
You will no doubt have seen other reviews that show the INpower to be consistent in its readings. That seems to be true from most vendors’ power meters these days.
My take is going to be to check the power variances between bePro pedals and the INpower 3D+ using Qrings. It’s specifically the Qrings that I anticipate will throw out BePro (because Favero acknowledge an inaccuracy there). I’m just doing this out of curiosity and to be different. I appreciate most of you won’t use oval chainrings.
- Lezyne Super GPS (nice MTB bike computer) + Favero BePro (forced to left-side only)
- Garmin 920XT + ROTOR INpower 3D+ (naturally left-side only)
Test Conditions: Varying road surfaces from perfect and flat to fairly bad; indoor trainer and the Surrey Hills. No cobbles, no rain. About 100km.
I kept getting charts that were seemingly so similar that I thought I was recording the same power meter on each watch. The chart above shows the red line disappearing as I back pedal to recalibrate the BePro and its power goes to zero. So that sorted out my concerns there! ie the ROTOR was still producing power at that time.
These results were not what I expected. Remember I was using the Q-rings. So I expected the BePro to be 2-3% higher, as advised by Favero. You can’t really tell much looking at the chart. Looking at tables and averages the INpower is generally 1.0-1.5% higher than the BePro ie in the 200w to 400w power range that’s 2-4w/3-6w.
Power discrepancies caused by elliptical chainrings may well be effected by different cadence levels. my understanding is that any quoted variation (eg the 2-3%, above) is based on normal cadence in the 90s rpm. You could perhaps draw the inference that if you have a higher cadence then any single error per revolution will be magnified by the higher cadence.
Both have a sampling frequency of (up to) 200Hz ie they should be broadcasting at pretty much the same number of points per revolution.
So other than being able to conclude that they are both consistent with each other, I’m not quite sure what else I can say. I can’t draw the conclusion that they are both accurate. They could, of course, both be equally right/wrong or both equally affected by the Qrings. Given that Qrings are one of ROTOR’s differentiators in their product offerings, it’s very unlikely their product performs badly in that area.
Pedal Smoothness (PS)
PS is Pavg/Pmax as a %age, where P=Power – ie evenness of distribution throughout the stroke. Commonly in the 10-40% range. I had to double the INpower data.
Looking at the PS data, the INpower is clearly quite a bit higher (several percentage points). In terms of what we should make of that I’m not so sure. The bePro recorded slightly lower averages but higher maxes compared to the INpower. But, again, the cause of that is only speculation.
Either way it’s probably irrelevant. PS is the sort of measure that you might be looking at to track technique improvements over time or across different positional tweaks. IF you want to compare this data with your mate’s then you can now rest assured that it’s not worth trying to bother and that may well apply to all power meters not just these 2!
Torque Efficiency (TE)
TE is the percentage of total power that you deliver which is positive. eg if you fail to fully de-weight the rising foot then you introduce negative power. I had to double the INpower data.
As with PS, the TE for the Inpower is higher than the BePro.
Price and Availability
The full crankset shown here, excluding rings, will be in the region of: £700. As of October/November 2016 you can use the 15% discount code at PMC, below.
The Qarbon Qring, at £170, is not cheap. You should be able to get them from any ROTOR dealer or PMC.
A Rotor Q Ring Inner Chainring 110BCD 4 Bolt Shimano is about £49 (not the one used in this review) and you should be able to get them from all ROTOR official dealers such as PMC, below.
Disclaimer: I have no affiliation to Sigma Sport. Your nearest ROTOR dealer will be fine. The power meter is on loan from the UK distributor. I make nothing from you buying one of these power meters. If you want to say thank you for the review please make your next Amazon purchase from the icons on the right hand menu.
Technical Service is here: firstname.lastname@example.org
Front derailleur info
The pedal axle location was much easier to work with than I had anticipated. On my test bike it was as easy to change the entire cranks/pedal axle as it was to change 2x pedals. In fact it was actually EASIER to change the INpower 3D+ than the two BePro pedals as only one allen key was needed and the subsequent precise positioning and calibration are both slightly more time consuming on the BePro.
If you have a standard/fairly modern bottom bracket you will almost certainly find the ROTORs to be compatible.
Assuming compatibility, the only major issue you are going to have changing between bikes will be when the axle gets fused to the bearings and if you also need to change chainrings. Changing chainrings IS super easy it just takes a wee bit longer and will require derailleur adjustment in the case of Qrings.
The point of me saying this is that previously I would have thought it might be a bit too involved to install/move the INpower 3D+ – it really isn’t hard, anyone can do it. Even me.
Having noted that, if you have an older bike then your best option may well be to get the dealer to install it as special tools may be required.
The highlight of the test experience was the Qrings. I got some extra free power. Quite a bit. The ROTOR INpower software is near idiot-proof in the speedy determination of the OCA/OCP needed to optimise the Qring’s position based on power output from the INpower 3D+. It was certainly me-proof.
ROTOR’s INpower 3D+ power meter itself is a tricky one to sum up. You’ll have seen from me and lots of other places that it seems accurate and consistent – as do many other power meters. The 300 hour battery life and the ‘hidden away’ axle-location are great positives, as is the lack of a need to continually recalibrate. I’m not really sure what to say about the PS and TE data – probably fine to use by most of us to guide our improvement.
FWIW: I liked it
Futures: A 2INpower version is released in May 2016 with dual-sided power. Left-side only power is most definitely an option as we’ve seen here and with other products in ROTOR’s range.