Powered by Outside

Review: Motion Instruments System 2 Data Acquisition Kit

May 14, 2024 at 20:41
by Henry Quinney  
photo


Data acquisition is becoming more commonplace, and you'd be hard-pressed to find a serious racer who hasn't experimented with it. For your more casual rider, though, there have been barriers to overcome before benefiting from using it. Motion Instruments has tried to make acquisition more accessible to more riders by making a cheaper, user-friendly system that's easier to live with.
System 2 Details
• Waterproof tracers
• Bluetooth Operation
• Electronic position sensing fork tracer
• Rear sensor uses link rotation to plot axle position
• $499 USD
motioninstruments.com


photo

The Sensors

System 2, which uses two independent tracers that run on rechargeable batteries and are read via a Bluetooth connection to your mobile phone, put the emphasis on ease of use and simplicity. The fork tracer is a very tidy bit of kit and does without the membrane roller system that tends to attract dirt or wear. Instead, it uses electronic position sensing with no wearing of electromechanical components.

Simply put, a flat, plastic rod passes through a slot attached to the lower. This has advantages not only in longevity but also in terms of alignment and cleaning. Both tracers are 100% waterproof, which means not only you can ride in the slop but also wash your bike care-free afterwards. The top of the tracer mounts using a custom air valve cover to clamp down on the sensor on to the top of the fork. Should you be using dual-crown forks, there is another clamping style.

photo

While the front sensor is a reimagined version of the rods that you're familiar with, the rear sensor is something else altogether. It's a clever device that manages to avoid some of the pitfalls of normal systems. That said, it is also compromised in certain aspects.

The rear tracer can be fitted with several different sizes of Allen key fitments to secure to the centre of your frame's main pivot. You then clamp down the wing of the sensor on the rocker itself, set the position at 0% travel, and again at 100%. Once the system knows these two extreme positions, it can work out where you are in the stroke. There are limitations, however. Firstly, this won't work on bikes where the main rocker is concealed behind another piece of the frame, or at least it will work less effectively. For best accuracy, the link with the most rotation will be best. Secondly, because you set the end position, it's limited by the stiff elastomers that you can find in some shocks. However, every system has drawbacks, and not worrying about alignment makes for a very quick and easy install. Motion Instruments also say that this solution is stop gap, and they hope to one day have a bank of bikes with all the leverage curves saved, meaning you can just fit the part, select the bike and you're good to go.

photo

System 2 also does without external cables or a central box. Again, this makes installation far easier. The tracers can be used by themselves, but for a system that puts so much emphasis on balance, it might undermine it to an extent.

photo


The Software

The MotionIQ app gives riders plenty to go on. Used via a mobile app on iPhone, iPad, Mac or Android, it shows you all you could need to know regarding position, balance, axle speed. Whenever I use data acquisition, how much learning I have left to do always strikes me. Maybe that feeling never goes away. That said, I think the data-led approach will work better for people who already have a decent understanding of different types of graphs and technical language. While the app does a good job of breaking down the information for the more casual user, I think it's best to go into using it, anticipating that it will be a new skill to learn and not something that confirms things you already know.

System 2 also uses the GPS function of your phone. That means you can drop in as soon as the sensors are connected. The recorded data is then overlayed on a map. This is particularly useful if only to remember which run is which.

The software is free when purchasing the kit and can be upgraded to the Pro version for an additional subscription of $99 per year.
photo

The Setup

The app focuses on getting a balanced bike and is far less prescriptive than the BYB Telemetry system. The setup that you'll get to isn't so much about what's fastest but rather what balances front and rear. This isn't bad, but it feels like it's missing a certain piece. Data acquisition is inherently limited because it only goes off the data that it's been fed. What is balanced on one track might be out of balance for another. With that in mind, I think it would be helpful if that app gave some guidance regarding axle speed and how to interpret it.

photo
photo

The idea seems to be much less rigid regarding setup and takes a more pragmatic approach. Not everyone will like a certain attribute, but everyone can benefit from a better-balanced bike, even if they then bias it in a certain direction. I often rode with the system recording data, but using it to reflect upon my setup after I'd already established how I felt about it.

Almost always, when I thought the bike felt best was when it was most balanced according to the data. That said, I am very happy to not have the deepest events equal, especially on the front. Often, when using data to setup my bike, I don't pursue using full fork travel as I would rather have something spare for extreme hits. It can only read the data it's fed, after all, and I try not making a habit out of bottoming my fork out on every single run.

That said, the app does a very respectable job of giving users a very solid interface that allows them to cut sections out of the track for more accurate readings and interpret the data in a variety of graphs. It explains, expressed as a plus or minus, whether the shock is compressing faster or slower than the fork. From there, you can tune it to be as near to equal as possible. When you have a clear idea of what you want, this is easy because you can read what is closest to the speeds you want in the data. For beginners, though, it would be good to get some ratings or suggestions to guide them a little.

photo

What's it Like Compared to Other Systems?

The system takes literally minutes to put on, and is so easy to live with. You don't have to worry about crashing on it, or having to bend it back when you do. I loved how robust and clutter-free it was. In some ways, this feels like a cross between the all-out tech of BYB's system and the user-friendly and hardy nature of ShockWiz. I feel like it will bring data acquisition into the realm of more riders, which is a good thing. That said, for ultimate accuracy and guidance I think that BYB's system is going to cater to the people wanting to extract the most out of their bike. There were two occasions that the Motion Instruments linkage sensor felt loose in its Allen-key fitting and was giving a dead zone to the travel. After reclamping it, it worked well, but all in all, it just felt less precise.

BYB uses data collected from various riders to give you prescriptive instructions on what to do. While this isn't the end goal, it does help guide users to a solid base setting. From there, they can familiarise themselves with the data and inputs to add their own flair to the suggestions.



Pros

+ Waterproof
+ Ease of installation
+ Software is fast and easy to connect
+ Focusing on balance rather than suggesting a particular setup


Cons

- Elastomer bumpers shocks limit rear tracer
- Focusing on balance rather than suggesting a particular setup
- Data takes time to learn how to interpret, and what works for you



Pinkbike's Take

bigquotesMotion Instruments have done a very good job with System 2. While it isn't perhaps as high-performance as BYB's system, it also does without much complication thanks to its wireless functions and lack of data rods on the rear shock. That said, it does mean that it can never be quite as accurate, on the rear at least. If I had to throw one on a bike for an hour, it would be the System 2 that I would use. If I were to spend a day getting increasingly granular, it would probably be the BYB.

System 2 is waterproof, easy and vastly cheaper than most other systems. Bike nerds, rejoice. If you want to tinker with your suspension, there's probably never been a better time than now. 
Henry Quinney


Author Info:
henryquinney avatar

Member since Jun 3, 2014
338 articles

52 Comments
  • 32 0
 I work in pro motorcycle road racing as an engineer and dabble in MTB's for fun, so this might be my first ever comment.
Solutions like these have been popping up more and more and I am very happy to see, especially at this price point. The data acquisition partner we use for our Superbike (2D Data Recording) has been dabbling in MTB/E-MTB data acquisition as well, and they're the industry leaders in our space, I fully expect this segment of the MTB market to develop rapidly.

The Motion Instruments solution in place of a traditional front/rear linear potentiometer is an elegant solution and I hope folks recognize how hard it is to design something that has so much versatility and protection from the elements. Data sampling resolution and accuracy will also be a large consideration, I don't see any mention of how many samples per second are taken, but for a linear potentiometer for a motorcycle I'd want no less than 400 Hz to trust what I am seeing, and that's on smooth tarmac, let alone someone doing some chunky downhill.

The crux of this is going to be either tailoring the data analysis experience for the end user to draw conclusions for them (ala Shockwiz) or providing the kind of training to your end users that they can best utilize the data being gathered. All of the data in the world isn't of any use if you don't know what to do with it.
The sky is the limit here.
  • 32 0
 200 Hz is native. We have debug knobs to enable 1000 Hz. We will offer a SW release in the future that will double the default rate to 400 Hz. If you want 1000 Hz, happy to give anyone the back door to enable. Just don't do an epic ride. Nobody wants to read the manual, but we do have a user guide that doesn't totally suck. It was written describing what question we were trying to answer when coming up with the data visualizations. So you get the back story of the feature. We also go into bike balance and how we quantified that. We haven't come up with a magic algorithm to set everyone up. It's not an easy thing to achieve. But, knowing how your bike interacts with the ground and correlating the data to what you think you are feeling is an insightful exercise. After a few recordings, it starts to make sense. Despite the technical looking graphs and statistics, it is not that hard to understand what's going on.
  • 4 0
 @RobMI: Great to see a range of sample rates supported for different use cases. I assume at 1KHz you either run out of memory buffer or storage on an epic ride?


I'd like to be the dissenting opinion on not liking to read the manual. There is very little in this world better than good documentation. I skimmed the Motion IQ documentation and I am impressed at not only the verbosity, but also giving real world comparisons of what to look for in a users data. Bonus points if you allow an export of data or CAN recording so I can parse the data using 2D Analyzer on a desktop.


The "Magic algorithm" for set everyone up is a challenge for sure, the biggest variable not accounted for is an individuals frame modulus of elasticity. The frame is just a big spring at the end of the day.

You're likely past the point of needing testers, but if not, and you want one let me know. Otherwise I'll likely end up buying a kit anyways.
  • 5 0
 @mbassani: With Pro MotionIQ, you can export all the data in standard formats (GPS tracks, raw data). Our data is just a CSV with easy to read column headers. So you should be able to import this into matlab, or whatever tool of choice. Send me an email, rob@motioninstruments.com. Happy to collaborate and try to be useful getting you what you need.
  • 5 0
 @mbassani: In the weeds here, but we store data in ram, then Flash after the recording is done. I've done a couple hours forgetting I was on 1000 Hz and it miraculously saved the data. Phones can be picky on how much data they will give apps so we can be put in the penalty box hogging too much memory. If you are under an hour, should be good to go. We are also smart about not storing data that doesn't matter. So if you take a break, we stop recording. If GPS is unavailable, then we record everything regardless since we can't tell when you are not moving. We err on the conservative side.
  • 1 0
 @RobMI: are you guys affiliated with Motion Industries? I couldn’t help but notice the logos are borderline the same
  • 2 0
 @joecrosby: No but we get confused with those guys all the time. Our logo was inspired by looking at a spring compressed from the side. Once we put our logo out, then I saw the Motion Industries logo and saw the similarities. Which one looks better?
  • 23 0
 wait, are you suggesting we're supposed to fiddle with our suspension and not just ride it exactly the way the guy we bought it from had it set up?
  • 13 0
 @henryquinney Thanks for the writeup! For the bump stop, you will be off a little bit from max position. However, you'll still be able to tune your bike. Generic soft tail has always been a stop gap feature for us. There was a time industry insiders told us we'd never get curve data from bike companies, but now we have 100s of bikes in the app that came from every OEM. When Specialized Engineering gave us all of their bike data, we knew we had a workable solution that could scale. With the new system, we require angle displacement data to get a complete curve. We already support this. If you have this curve in the app, all you need to do just get the open position and then sit on the bike to compress it. Then it works normally. After the PB article this morning, we've already had a few bike companies reach out to us to ask what we need. Shout out to Trek Engineering and Yeti being the first to reach out early this morning. We've had really good luck with the VHB once it sets up. The 2.5mm screw may require Loctite if it loosened up. We recommend the blue Loctite. We'll update our FAQ to reflect this. Thanks! Rob
  • 6 0
 Thanks for this, Rob. I'm going to amend the article to reflect the idea that the generic soft tail isn't the end goal. Thanks for making a great system, it's been fun to review and will go on plenty more bikes this year.
  • 4 0
 For having a quick chat in your local coffee shop and dropping off a system with no documentation, you did a quite thorough review with really great insight. The points you brought up were valid and are things we take note of. We miss a new user perspective since we've been in the weeds on this project for too long.
  • 11 0
 Wow, getting near the price of my shock wiz and so much more data. I was genuinely shocked when I saw how much they are charging for this new system.
  • 1 0
 Yea because of the order of the photos I could’ve been convinced that price is for a fork only product
  • 2 0
 *Fork Tracer is compatible with concentric single air valve forks (Rockshox, Fox, etc.). The Fork Tracer does not support EXT air springs or traditional coil spring forks. The rear sensor is designed to work on suspension links with bearings pressed into the link, and NOT the frame, with a max diameter of 34mm.

Would you consider a Fox38 with a Smashpot "traditional"?
  • 2 0
 Not traditional. We just need the Schraeder valve to mount the Fork Tracer Flexure (90 degree bent sheet metal).
  • 2 0
 @RobMI: Correct me if I'm wrong, but you could get a spare smashpot cap, drill and tap it. And then screw in a bolt with thread locker and mount it that way, right?
  • 2 0
 @bek998: Only one way to find out. Let us know how it goes.
  • 1 0
 @bek998: I mean it's totally doable, but I'd err on the side of caution. If the drill and tap works, then it's a done deal.
  • 6 0
 If you're not measuring, you're guessing - nice work team!
  • 4 0
 I know Minnaar helped them for a long time. He still working with them?
  • 1 1
 Like some of y'all, I have access to training data from a test track, a segment about 45 seconds long set up for transponder timing (00:00.001) and I had a minute to put the data into a calculator. A/B testing two different bikes, 23 runs each (N= 46) over the last few months, A, B, A, B, etc. The bikes are super different; this isn't hypothesis testing like "Is the mean time (s) of A different from the mean time (s) of B?" It's obvious they're different. Instead this is useful to get a baseline for how much individual times on the segment vary not driven by bike choice. It also quantifies how different bike A and B are. Run the analysis, see the variance and the effect size, and see that A is about 5% faster than B.

Here's what this has to do with mountain bike DA. I'd guess that if you expect to see a 5% reduction in your segment times driven solely by a DA+tuner guided change in spring rates and damper settings, you're awfully optimistic. But let's run with it, this is for illustration. You don't need to do 46 runs, thank goodness! You do what's called a power calculation based on the effect size (the effect of A vs B) and find that to credibly see evidence that Tune A is faster than Tune B with a magnitude similar to my bike A vs B, you _only_ have to do 14 pairs of runs (N = 28.). 28 runs where you don't miss a line, don't make a major mistake, don't shut down your effort part way through. On the same segment, with real timing. That's the minimum, under the best conditions, to detect a hypothetically huge difference between A and B. But it might not be a huge difference. What if Tune A really (really) is faster, but only 3% or 1%? Smaller effect size, larger sample size required or else you won't see it. Could be real, but undetectable. And if your system can't detect it, you don't know if it's real.

I'd like to see the technology-curious amateur who is actually doing the program and not just looking at the pretty lights, the gorgeous scatterplots, the alluring histograms. A hell of a lot of work. I'd like to applaud that rider.
  • 2 0
 It would be impossible to do a ton of runs in one day and ignore the role of fatigue - so instead you'd be relying on timing results in variable trail conditions from different days, which isn't fair for each tune. It would also be impossible to hit the same lines every run - but also note that one tune could give you the confidence and control to hit different lines than you might with the other tune, which is overall a win. More sensor data would really help tell the story beyond timing results. 28 runs is a lot. If one tune is definitely better than the other, you may be able to get statistically significant results with way fewer. Either way I love your thinking here!
  • 2 0
 @mtb-phd: Totally. The N is unreasonably high because I suck. There's too much variance in my baseline lap times. Someone good would not only run faster laps, but would be more consistent. That would drop the number of laps needed to tease apart signal and noise by a long way. The reason I use myself as an example isn't just convenience -- that target audience of this product are not going to be machines who can throw down 4 3:00:000 laps in a row at their trails and be super sensitive to the direction a setup change and give the tuner useful feedback based on years of experience -- they're going to be slow, inconsistent dorks more like me, and they're in denial.
  • 1 0
 Current gen Yeti's are out of luck for the rear tracer so far as I can tell, since all bearings have been moved to the link (which is a good thing). Unless someone smarter than me sees a way this could work?
  • 1 0
 160-E is supported on the shock pivot. We have open sourced all of our mechanical for folks to make their own custom brackets to mate with our sensor. We are friends with them. If they share a mounting design to 3D print, we'll post it on our site. But for now, some bikes will be DIY if you are mechanically inclined.
  • 1 1
 It would be nice to see this integrated with a cadence sensor.
That way you could see which settings worked better for pedaling, and more importantly you could get information to balance the best performance for pedaling vs coasting/descending.
  • 3 0
 Damn, Giant must be really stoked on the wind tunnel, is this Velo?
  • 8 9
 "Focusing on balance rather than suggesting a particular setup." This deserves an entire article to unpack, which won't fit in a comment and I'm off the clock anyway, so to briefly summarize: is more balanced better than less balanced? How much more, and under what circumstances, and what's your evidence? The DA system, no matter how much data it logs, cannot on it's own validate itself. Whether your tuner's special sauce is something simple or something complex, whether you've logged 10 minutes or 10,000, the DA is modeling a proposition that can be simplified to IF A THEN B. A is the data acquisition and the tuner's algorithm, like set up the bike such that when the DA is on we see what we want: equalize such and such positions or velocities or accelerations, maximize these events, minimize those events, anything the tuner can dream up an the DA is capable of measuring, in any combination. All of the is on the "A" side of the proposition. The "B" side is the rider being faster on a segment of trail, the rider reporting being more comfortable or less fatigued or more confident in a post-ride survey. The DA system cannot does not measure "B". That's why pro teams can use these tools, they have test tracks and staff to help do timing. That's why consumer DA systems are hot b.s. unless the consumer is going to arrange (pay) for the same help that the pros get. No matter how much data and how much creativity goes into the "A" side, it's a guess. The model generates a hypothesis, if you want it in fancier terms. And it remains a fcking guess until the model is validated. That's "B". None of these products do "B," which, let's remember (faster, more comfortable, less fatigued, more confident) is _the_whole_point_of_tuning_the_bike_.
  • 2 0
 I think you are taking the opinion of the editor over what the founder of the company has suggested on the podcasts that he has done in regards to the balance in the motionIQ app. If I am not mistaken he specifically says the balance is not a be all end all and just provides a comparison. Everything is reference points, so if you find 15% higher rebound balance is a good reference point, then you can utilize that to help set up your bike further. Also the app syncs with strava and has built in timing, so I dont quite understand what you are getting at. The bike setup and the rider are a synchronous relationship, and the data acq. system does not claim to separate the two completely, it just provides data to further understand how your "feeling" on the bike is reflected in your suspension and its settings
  • 4 0
 Hi all, I did try and address this in the article.

"Everyone can benefit from a better-balanced bike, even if they then bias it in a certain direction". I think this is largely true, and where their desired balance is going to sit will be a preference. That said, being empowered with the knowledge to choose where that point is is the key bit. That said, I think that compression and rebound speed are best matched at the axle rather than not. I hope that helps and if you feel it's unclear I can amend. Thanks
  • 4 0
 I demoed the first version of MotionIQ for a few weeks in 2020 and spoke with him about this subject. I experimented a lot with my 2020 Enduro (Ohlins coil F/R) with this system.

My conclusion is 1:1 front and rear is not always 100% ideal, it depends on the geometry and kinematics of the bike, however when it comes to rebound velocities you want them closer than not. There seems to be more variability in compression and especially travel usage, which don't forget has an effect on the geometry of the bike and your balance. The system provides enough information for you to make these judgement calls by reading the data without being spoon fed instructions, but it will need to be correlated to feeling. It also gathers real data you can observe, not data gather indirectly and the way it presents the data allows you to eliminate outliers.

When I ran it, I ran the same track over and over. The GPS data allowed time measurement across segments, but for me it mostly came down to feel. Reading the balance data helped enormously because I found a ratio of balance front/rear where the bike felt the most centered, so when I made tweaks or adjustments one way or another, I could adjust the opposite end to find that balance again. So if I had to tweak the fork because I was bottoming out or it was too harsh, I could make adjustments to the rear to keep it within that balance point to compensate. This is very important and an undervalued part of suspension setup IMO, if one side is setup wrong then it can be really hard to diagnose where the problem is.

It'll always come down to how the rider feels, but most riders are so far off where they should be that this gives people a really valuable tool to demonstrate a rough starting range for setup. These data points can be really helpful especially for coaches looking to help people who have no idea what adjusters do or how suspension works. I learned how this balance works and how my suspension should feel when it is balanced properly, which helped me learn down the line how to set stuff up even without this system in place.

This is a huge improvement over the earlier version, very well done. I am sad to see it doesn't work with coil forks, though.
  • 3 3
 The point isn't whether balancing f/r wheel velocities moving thought their travel specifically is or isn't an interesting phenomenon to measure compared some other, it's that al of these keebler elves trying to sell a brake ace or a shockwiz are trying to sell you on a set of targets inside their gamified app -- so that you can adjust your settings to chase down whatever reward in on offer without leaving the product. It is saying IF A THEN A. And leaving the user to imagine that that must mean the setup is better because the targets must have been chosen for that purpose. It's deceptive. And someone brought up Strava. Jesus, just use your hat racks here. Imagine a 3 minute test track. Just use your best guess how much faster you'd expect to be from manufacturer's base settings for body weight versus God-sends-His-angel-to-you perfect settings. How many seconds faster? Now ask yourself what is the random error of fcking Strava over that 3 minute test track? Uh oh. That's before we get into the actual variance in times just due to the rider. If y'all are 'nerding out' step your nerd game up Smile
  • 4 1
 @Snfoilhat: I'm not sure what your point is.

Data acquisition like this makes people more comfortable and confident on their bikes. We're paying $1k+ for high end forks, close to that for shocks, and thousands for bikes yet I'd argue 90% of mountain bikers or more have zero clue how to set any of that stuff up properly. They are mostly over sprung, over progressive, under damped, and that's if they even bother setting those things up in the first place.

There are objectively correct setups for all of these things and putting people at least in the ballpark will make them more comfortable and confident. It's not always about speed or shaving seconds off, for the average rider it's about the realization of how these things should work together and how that can make them more confident and comfortable on the bike, along with getting the value out of their setup.
  • 4 0
 @shinook: TBH, our old system didn't really work with coil forks either. We did some things to make it work, but it was not ideal. Probably obvious, Coil DH forks are supported with the DH fork mount.
  • 2 5
 I'm surprised to see you come out so strongly for objectivity and measurement @shinook when you display such comfort straight up inventing that 90% of mtbers line from what I guess we have to recognize as a Jim Henson-like imagination Big Grin
  • 3 1
 @Snfoilhat: Go find 20 mountain bikers, sit on their bike, then tell me I'm wrong.

Regardless, it doesn't change anything I said. I obviously don't have an objective measurement, but in my experience most people have things setup wrong and not even in the ballpark. If you wanna miss the forest for the trees, that's on you.
  • 1 1
 @RobMI: Yea, I understand mounting on a coil single crown is difficult. I just got a lot of value out of it when I ran it years ago on mine and I'm gravitating more towards the two types of forks that make mounting difficult, so it'd be nice to see something that works, but I get that doublesided tape isn't ideal either.
  • 1 0
 @RobMI: To clarify, the reason you state it works with air forks is due to the need for the air valve on one leg, correct? If I'm running a hybdrid coil/air setup using the Avalanche damper with the coil over the damper, this still works because I retain the air valve on the non damper leg?
  • 1 0
 @genericmk: If you have the air valve, and it's centered, it *should* work. If I can see a pic of the air valve, I can verify.
  • 2 0
 @RobMI: The fork has the standard RockShox air valve so that works as I understand. The coil is on the damper side so it doesn't interfere with the air side. Cheers.
  • 1 0
 What do you do with the information? Does the software make recommendations on how to tune your setup in response to the data generated?
  • 3 0
 This is sweet. I want.
  • 1 0
 Would love for this to fit on the EXT Era! (We have them on two bikes, the 2.1 is the most amazingest fork evar!)
  • 2 2
 This seems very limited. Pretty sure this won’t work on an evil. Too thiccc.
  • 3 0
 Works on the Delta link opposite side of the sag indicator
  • 1 0
 Will this work on a VPP or other design that does not have a "main pivot"?
  • 2 0
 Their launch video shows it mounted to the top link of the new RM Altitude, you should be able to use the same placement on a VPP assuming the hardware fits the size requirements.

www.youtube.com/watch?v=vWIhRzrLVI8&ab_channel=MotionInstruments
  • 2 0
 If santa cruz vPP, the top tube link is the pivot of choice.
  • 1 3
 Need a rent option for regular folks. Price aint bad really, I just don't see me needing it after getting my bike dialed in for my local trails.
  • 3 0
 buy one and rent it out then
  • 1 2
 Can I borrow it for a couple weeks?
  • 1 4
 Like a smart watch, go pro and Strava, it ain't for me.







Copyright © 2000 - 2024. Pinkbike.com. All rights reserved.
dv65 0.035780
Mobile Version of Website