It’s been just shy of 6 years since the original Kinetic inRide sensor was announced (Fall 2012), and today they’ve updated it with the guts it always wanted. The inRide sensor allowed you to take virtually any existing Kinetic fluid trainer (which is pretty much all of them) and get proper power broadcasting from it, so that apps could consume that data. But far more importantly, it allowed you to calibrate the sensor on your trainer, accounting for differences in how tight you twisted the roller wheel or your tire pressure.
That calibration often meant the difference of 20-40w in accuracy between the inRide method and just a generic power curve that trainer apps often utilize. In other words, it was a big deal.
And to that end the original inRide sensor was solid. It bought power to probably a decade’s worth of older trainers that weren’t equipped for the world of TrainerRoad, and years later, Zwift. And did so in an accurate way. But there were some issues. First, it never support ANT+, so you couldn’t connect it to your Garmin or numerous other devices or apps that used ANT+ (even an older computer). Second, it didn’t actually support the Bluetooth Smart power standards. Instead, they cooked their own (which wasn’t actually a huge deal 6 years ago when there was no power standards). But that meant that over time the inRide sensor got more and more isolated as the rest of the world moved to standards.
Today though, that changes. The new V2 sensor not only supports ANT+ and Bluetooth Smart, but it does it four different ways. Atop that, they sliced the price considerably from when it first launched. Now it’s a mere $49. Oh, and it’s still compatible with a decade or so’s worth of trainers.
Now, in addition to the updated Kinetic inRide sensor, they’ve updated their ‘Smart’ trainer lineup. These are essentially just their Road Machine and Rock and Roll Trainers that ship with an inRide sensor already slapped on. They are very different from their ‘Smart Control’ trainers, which use an entirely different resistance unit. Though, that resistance unit can replace the flywheel on an existing trainer frame, reducing your costs somewhat.
What’s in the box – inRide V2:
First up on the unboxing docket is the updated inRide unit itself. You’ll see that it proudly displays the ANT+ & Bluetooth Smart logos on the front, while doubling down on that on the back with more text related to it and a bunch of apps that are compatible listed. Of course, it’s far more than these, but these are the bigger ones out there.
Once we shake all the parts out of the box, here’s what’s left over:
Inside you’ll find the pod, CR2032 battery, magnet/magnet plug, an installation guide, and some product safety stuff. A nice tidy package so to speak.
Here’s the pod itself, which from the outside is identical to the existing pod, except not green instead of black.
Then we’ve got the battery and magnet/holder. The battery should last about a year or so, and then it’s just a simple replacement with another coin cell battery.
Oh, and here’s the all important product safety guide:
And that’s all there is in the box. The pod itself has simple double-sided tape on the back of it, so it sticks to the trainer as you’ll see in the next sections.
What’s in the box – Road Machine Smart 2:
In addition to the updated Kinetic inRide pod, they also updated their fluid based trainer bundles, which include the inRide sensor built in. Specifically, the Road Machine Smart 2, and Rock & Roll Smart 2 trainers. Those take the Road Machine and Rock & Roll trainers/frames and plop an inRide pod on it and call it a bundle. While I have an older Kinetic trainer myself from yesteryear, here’s a look at what the latest bundle for the Road Machine looks like:
Note on the box you’ve now got the ANT+ & Bluetooth Smart glory listed:
Inside you’ll find the trainer frame, the flywheel/roller unit (detached), a few screws and press-on-screw/knob, a trainer skewer, and the manual.
The inRide sensor is already attached to the back of the trainer, and the magnet already placed inside the roller itself.
‘Building’ the entire thing only takes a second. You simply attach the resistance unit with a single bolt and then twist in the press-on knob. Done:
At this point, there’s no difference between a standalone inRide sensor on a Kinetic fluid trainer and the pre-stickered variant I just unboxed. So the rest of the post will just focus on the inRide pieces, so those are identical here. Finally, for lack of anywhere else to stick it – the Smart Fluid bundled trainers will start shipping in August (whereas the new inRide pods ship immediately).
Assuming you’re starting with inRide sensor alone, you’ll need to get the magnet installed into your trainer. this simply pops into the small hole on the rear of the trainer near the flywheel. This magnet comes in the box and sits inside a little rubber stopper:
The purpose of the magnet is to measure wheel speed. From there it can then align that to a known power curve to determine power. With a fluid trainer, the specific speed corresponds to a specific power level (at least after the fluid has warmed up a bit). Because fluid trainers generally have no resistance knobs, it’s a relatively well understood curve. Here’s an example of power curve chart:
But the most important piece is what comes next – calibration. See, while the power curve is accurate, what’s not accounted for is how tight you’ve attached the wheel to the trainer. If you overtighten it, that isn’t accounted for in the resistance curves, and thus you’ll actually be working harder than the math assumes. The trick is that there’s no perfect methodology for tightening, since everyone’s wheel size is different. Kinetic in their instructions says ‘tighten the knob will it touches, and then rotate the knob fully 2-4 more times’.
As you might presume, depending on your wheel there’s a vast difference between 2 times and 4 times in terms of resistance applied.
And that’s specifically what calibration seeks to resolve.
So to calibrate you’ll open up either any app that supports ANT+ FE-C calibration or Bluetooth FTMS calibration, and then trigger calibration from within the app. Alternatively, if your app doesn’t support that (cough, Zwift over Bluetooth Smart), you can do it using Kinetic’s free Kinetic Fit app (that app also has paid training aspects, but that’s not in play here). The process merely has you speed up to 21MPH, and then stop pedaling to let it coast down.
As you coast it measures the duration it takes for the wheel to reach a certain speed, which allows them to figure out an offset value to use. And as a result, transmit accurate power values. Note that generally you want to wait a little bit for the fluid to warm-up, about 10 minutes usually does the trick. Or, if you’re in a hot garage in Florida with the temp is already 92*F…then I found it was nicely warmed up already.
Also, if you don’t change the roller knob in between trainer rides, and then you also ensure your tire pressure (PSI) is the same each ride, then you can get away without calibration each time since nothing has ‘changed’ in the system. You’ll likely see some slight variances during the warm-up period, but that’s pretty much normal for most fluid (and even non-fluid) trainers anyway.
While calibration is ‘the feature’ of inRide, it’s of course not exactly why you bought it. You bought it because you want to see your power numbers. But inRide broadcasts more than that, it’ll give you speed and cadence too. In fact, it’ll do that more ways than a Swiss Army knife. Here’s all the different methods it broadcasts your data:
ANT+ Power Meter device profile (Power + Speed + Cadence)
ANT+ FE-C Smart Trainer device profile (Power + Speed + Cadence)
Bluetooth Smart Power Meter device profile (Power + Speed + Cadence)
Bluetooth Smart FTMS device profile (Power + Speed + Cadence)
Why use FE-C and FTMS if you can’t control the resistance levels on the trainer you ask? Simple: It’s how they can allow 3rd party apps to trigger calibrate commands, which are normally reserved for high end resistance controllable trainers. It’s kinda an ingenious idea when you think about it.
If you’re in an app like TrainerRoad, you’ll see the inRide sensor listed a number of ways, depending on whether you’ve got both ANT+ or Bluetooth Smart (or both) enabled/connected. The key thing in TrainerRoad you’re looking for is to use the ‘Electronic Trainer’ option, versus the power meter option (since that supports calibration). TrainerRoad started supported Bluetooth Smart FTMS last summer when it was first introduced.
And you can calibrate it directly from there too:
Meanwhile, in Zwift, you’ll see it come up in both the power meter section and the resistance controllable trainer section. You’ll note that in this section only the ANT+ trainer-specific variant (‘ANT+ Trainer 108’) is showing (the icon is of a trainer, as opposed to a power meter). That’s because Zwift doesn’t actually support Bluetooth Smart FTMS yet properly. They support a variant of Bluetooth Smart support, which companies tend to code to. But not yet FTMS. So if you want calibration inside of Zwift you’ll need to use ANT+ FE-C instead.
The second piece Zwift supports (just like TrainerRoad) is just showing it as a regular power meter (sans-calibration). You can see that above as well ‘Powermeter 108’. Notice you’ll see both the Bluetooth Smart (inRide 00108) and ANT+ variants (Powermeter 108 108) of the same sensor (and 98 other power meters on my bike). It doesn’t really matter whether you pick ANT+ or Bluetooth Smart in this context.
In the case of Zwift you’ll also want to ensure the cadence pairing selector automatically picks the inRide sensor as well (it should), unless you have a separate cadence sensor, in which case I’d use that instead.
Once all that is up and cooking – you’ll just Zwift or TrainerRoad (or whatever your app is) like normal. You’ll see your power, speed, and cadence data shown on the screen just like a higher end fancy trainer. And most importantly, that data will actually be accurate too since it’s calibrated.
And, just to round things out, here’s the inRide sensor showing up via ANT+ as a regular ANT+ power meter on a Garmin Forerunner 935 watch:
The world is now complete, everything displays everywhere.
It’s long been established that the inRide sensor is actually very solid once calibrated (logical, given the entire point of the inRide sensor is calibration). In terms of the device itself, nothing has changed in that respect – all that code is the same. What changed was the broadcasting pieces over ANT+ and Bluetooth Smart. Still, I never trust any time. Or even if I do trust companies, I always verify.
Thus, this section is about verifying. I’ve been playing with variants of the inRide V2 pod for a few months now, but for accuracy testing we’ll look at data from last week on the final production unit. For this test I looked at Zwift, and compared it to a handful of other power meters concurrently. On this data set we’ve got the trusty PowerTap G3 hub, alongside newcomer Avio and their Avio power meter. I had two other units on the bike at the time, but they both had issues – so we’ll remove them from this data set to reduce WTF factor.
Here’s the overall ride at a high level (and the data set is here):
As you can see, things are very close between the three devices. Some minor nuances here and there, but overall all very close. These were captured on pretty different devices, so that contributes a bit. The Avio was captured on a Garmin Edge, the PowerTap G3 on a Wahoo BOLT, and the inRide data is from Zwift itself.
Let’s zoom in on the first 6 minutes to start. What you see here is some definite stabilization of all the units. Somewhat normal actually in the first few minutes of a ride. That’s usually when power meter temperature sensors start to stabilize for offsets, and when the fluid on the trainer is also stabilizing:
After that I do two sprints, at which point all the units seem to ‘lock’ quite nicely into place:
I’m seeing some slight variances/wobbling within the Avio unit, but on the whole these three were tracking quite nicely through here.
And here’s the next little while as well. Note that the change in scale here means that the wobbles are slightly more pronounced. These charts at smoothed at 5-seconds. Since this is a non-interactive trainer, it means that you won’t see quite as perfect stability as you’d see in a typical smart trainer since the wattage is purely variant on my cadence/speed/gearing (versus something like ERG mode).
One thing to notice is that even on both the sprint and the recovery (post-sprint) are proper. It’s actually the recovery off the sprint that is most challenging for trainers, since you can sometimes get into a spot with a trainer where it needs the speed to ‘catch-up’ a bit. Even the KICKR SNAP can show some issues when you sprint and then instantly back-off the wattage/speed a bit, causing a bit of a gap. Not so here though.
Overall though – the power accuracy is just as it was more than half a decade ago here: Perfectly fine.
But, one last question – what about cadence?
While something I consider a ‘bonus’ feature, it’s not quite perfect. It seems a little bit low:
In fact, if I pull in all five units onto that chart from that test, you’ll see that the odd-man out was the inRide 2, lower than the rest by about 8RPM. Though, there were also some spikes and dropouts on cadence that you see there. These corresponded to the sprints a bit, which is somewhat normal for cadence coming from a trainer (when it goes awry, it’s usually due to a big sprint and the pedaling stopping, as the estimation gets confused).
As I noted a moment ago, I generally don’t consider cadence a huge deal on trainers, since so many people have it on their bike anyway. It’s nice when it’s there, but it’s hardly a deal killer. And given how consistently this looks to be offset, it might just be an easy fix for them (you know, like ‘add 8RPM and call it done’). But I’m not a programmer, nor do I play one on TV.
Instead, I focus on power for trainers. And in this case, the power is exactly where I want it to be. So that makes me happy inside.
(Note: All of the charts in these accuracy portions were created using the DCR Analyzer tool. It allows you to compare power meters/trainers, heart rate, cadence, speed/pace, GPS tracks and plenty more. You can use it as well for your own gadget comparisons, more details here.)
Simply put, if you have a Kinetic fluid trainer of any vintage, the updated Kinetic inRide sensor is the best deal in the trainer/power meter realm today. There’s nothing anywhere near $50 that can provide accurate data and do so properly across all the major protocols. They nailed it.
More importantly though, I’m optimistic that we’ve seen Kinetic turn a page in their support of industry standards. With the original inRide they were hesitant to support ANT+ (or even Bluetooth Smart standards when it finally emerged). They of course doubled-down on that with the Smart Control, costing them significant market share that should have been easy pickins for them. But in my discussions with the company they seem to now understand how critical it is to support these standards, and over the last few months I’ve been playing with an inRide V2 variant have gone out of their way to ensure the pods are nailing the nuances of broadcasting different power metrics in the most open ways possible.
It’d say the only area that gives me pause is the pricing on the overall Kinetic Smart Fluid trainer packages (non-Smart Control variants). In other words, the ones where you buy a (non-interactive) Road Machine or Rock & Roll Trainer and get the updated inRide pod attached. I’m not sure in this day and age that those prices are as competitive as they used to be. For the money, I’d spend slightly more (or even slightly differently) and get the interactive smart control aspects from other options on the market. Road-like feel on non-fluid trainers has improved dramatically, and the gap isn’t as wide as it used to be (if at all).
But for the inRide pod, yeah, that’s brilliant now. Especially if you’ve got an older Kinetic trainer sitting in your garage somewhere, this just breathed a gigantic breath of new life into it in terms of compatibility between apps and devices. While Zwift and TrainerRoad did support the older inRide sensor, that support didn’t extend to other devices from Suunto, Polar, Garmin, Wahoo, and countless other apps. With the new pod though – all that changes.
With that – thanks for reading!