Joining VeloClub not only supports the work we do, there are some fantastic benefits:
by Matt Wikstrom
March 7, 2017
Photography by Matt Wikstrom
After creating its opulent Neo direct–drive smart trainer a couple of years ago, Tacx went to work on a lower priced alternative. The result is the company’s second direct-drive trainer called the Flux. The new trainer is built around a belt-driven 7kg flywheel and it boasts many (but not all) of the same features as the Neo.
In this review, Australian tech editor Matt Wikstrom shares his impressions on the performance of the Flux.
There has been a quiet evolution in stationary trainers in recent years as automated resistance units have become more affordable and smart technology has been embraced to create a rich (though virtual) training environment. These so-called smart trainers have done a lot to improve the appeal of indoor cycling so cyclists that are finding it difficult to get out on the road can still enjoy the activity.
Tacx has a deep history with indoor cycling that can be traced back to 1972. The company’s founder, Koos Tacx, had been operating a bike shop since 1957 and was well known by local pro riders. He wanted to help them improve, and having opened his own factory in 1969, he started manufacturing cylinders for rollers.
The market for rollers was very small at the time but Tacx remained committed to indoor cycling as the company grew. In time, Tacx would move to a larger factory and its catalogue would grow to include trainers, tools, workstands and water bottles.
Tacx now has eight different models in its collection of smart trainers. All but two are rear-wheel driven with resistance units that vary in sophistication and the amount of load on offer. The other two smart trainers — the Neo and Flux — are direct-drive units that take the place of the rear wheel.
CyclingTips’ Australian editor Matt de Neef had a close look at the Neo previously, a futuristic-looking trainer that boasted a sophisticated direct-drive mechanism with a motor brake. It was also very expensive, and while Matt found that the Neo had some nice touches, he didn’t feel that it was enough to justify the high cost of the trainer.
Eighteen months on and Tacx have answered those criticisms with the Flux, a significantly cheaper direct-dive smart trainer. The new trainer gives up some of the extravagant features that furnished the Neo but retains all of its connectivity and much of its performance and accuracy.
The greatest advantage that a direct-drive trainer has to offer is that by substituting for the rear wheel, users no longer have to worry about wearing out the rear tyre. This extra convenience comes at a significant cost though, such that all direct-drive trainers on the market are high-end products.
Tacx has set of price of AUD$1,099/US$900/€799/£700 for the Flux, which brings it into line with other direct-drive trainers such as Wahoo’s Kickr, Elite’s Drivo, and CycleOps’ Hammer. With that said, prices for these trainers can vary a lot, depending on the market and retailer in question, so there is no clear winner on the basis of price. Nevertheless, where the Neo was an outlier on the basis of price, the Flux now joins the herd, albeit at the high end of the market.
Tacx have managed to bring the price down for a direct-drive smart trainer by using a few strategies. First, the Flux is built around a belt-driven flywheel that is much simpler than the true direct-drive mechanism devised for the Neo. Second, the Flux uses cheaper hardware, such as the magnets for resistance, which aren’t as strong as those installed in the Neo. And third, the electronics have been economised, so the accuracy for power, speed and cadence measurements drops to <5% compared to <2% for the Neo.
As a result, the Flux lacks some of the grunt of the Neo since maximum resistance is capped at 1500W (the Neo offers 2200W) and slopes are limited to 10% (the Neo can replicate 25%). In addition, the Flux doesn’t attempt to replicate road feel and without a motor brake like the Neo, it won’t speed up with descents. So, the bells and whistles are gone, and there’s no light show either, since the Flux doesn’t light up like the Neo to reflect the amount of power being generated.
Connectivity remains completely unchanged. Like the Neo, the Flux offers both ANT+ FE-C and Bluetooth Smart to connect with a variety of devices including smartphones, tablets, laptops and desktops, as well as bike computers and watches. For the former, most devices (except suitably equipped bike computers and watches) will require an additional ANT+ antenna, while the latter will simply depend upon whether the device in question is equipped with Bluetooth (BLE 4.0).
The Flux is delivered in a box containing the trainer in two parts (the support legs are separated from the main unit), a power cord, quick-release skewer, customer information, and a bag of bits. This bag contains the pair of bolts and a 8mm hex key needed to assemble the trainer along with a few fittings for the axle and the freehub body.
Buyers must supply their own cassette for the Flux, and if they’re planning to install it themselves, then they’ll need the appropriate tools. They will also need a suitably equipped smart device and some downloadable software as well as a nearby powerpoint.
Strictly speaking, there is no need for power or a smart device to use the Flux, since it can be used unpowered and unconnected as a simple resistance trainer, but I fail to see why anybody would ever choose to use a smart trainer in such a manner (power outages not withstanding).
With just two bolts to install, I don’t expect many people will have trouble assembling the flux. What they will have trouble with is picking it up and carrying it around. At 21.5kg, the Flux is heavy to lift, and with all of its weight centred on the flywheel at the rear of the unit, it’s also awkward to carry. As a consequence, this is a trainer that demands a dedicated spot. For those riders that need a portable smart trainer, there are much better choices on the market, but they are all rear-wheel driven.
The Flux is 64cm at its widest and 67cm in length, which makes for a compact footprint, so once the owner finds a spot for it, it won’t take up a lot of room. The power cord is only 1.75m long, so chances are, an extension cord will be needed to get power to the unit.
While buyers must supply their own cassette for the Flux, they have the freedom to fit just about any brand and model thanks to Edco’s clever MultiSys freehub body design. The body has an arrangement of grooves and splines that can accommodate all of Shimano and SRAM’s 9-, 10- and 11-speed cassettes as well as many of Campagnolo’s 9-, 10- and 11-speed cassettes. However, the latter depends upon the number of splines, so any Campagnolo cassette that that has four splines will fit, while those that have eight (9-speed Centaur, 10-speed Veloce, older (2009-11) Centaur 10-speed, and 11-speed Centaur) will not.
The MultiSys freehub body is a thoughtful touch, however users may have to install one of the spacers (2mm and 0.5mm) that are supplied with the Flux before fitting their cassette. If it is an 11-speed Shimano/SRAM cassette or a 9- or 10-speed Campagnolo cassette, there won’t be any need for a spacer. The 2mm spacer should be used with Shimano/SRAM 9- and 10-speed cassettes, while the 0.5mm will be needed for Campagnolo’s 11-speed cassettes.
There is an added complication concerning the lockrings for the cassettes. Shimano/SRAM cassette lockrings will not fit Edco’s freehub body (too large), which is why Tacx supplies Edco lockrings with the trainer (one to suit 11T and another for 12T cogs). In contrast, Campagnolo’s 10- and 11-speed lockrings will fit the freehub body.
The Flux is supplied with a couple of spacers and lockrings to help with fitting different cassettes to the unit.
As mentioned above, those buyers that want to handle the job of fitting the cassette will need some suitable tools. This includes a compatible lockring tool (Shimano or Campagnolo) as well as a chain whip. The latter is necessary since there’s no way to lock the direct-drive mechanism of the Flux, so a chain whip must be used to stop the cassette from turning as the lockring is tightened.
The Flux is supplied with a 130mm quick release axle along with a spacer to increase it to 135mm for MTB/road disc bikes. A 5mm Allen key is all that is needed to unwind the axle cap on the non-drive side to add the spacer. For those hoping to mount a bike fitted with a thru-axle, they will need to get an additional set of adapter skewers made by Tacx.
Once the bike is fitted to the Flux, there is no need to use a riser for the front wheel. While this keeps the number of necessary accessories down, it does limit the amount of room between the axle of the unit and the top of the legs to 200mm. While this is plenty of room for short- and mid-cage rear derailleurs, long-cage derailleurs will end up rubbing on the top of the right leg when the largest cogs are in use. Thus, while it’s possible to mount a MTB on the Flux, it’s perhaps better to think of it as a trainer best suited to road bikes.
At face value, the Flux promises the simplicity of plug-and-play connectivity, and to a certain extent, this is true. But anybody that has gone through the process of integrating a smart device will understand that a measure of preparation is required before that simplicity can be realised.
Edco’s freehub body will accept all Shimano and SRAM cassettes as well as many Campagnolo cassettes.
Perhaps the best place to start is to download the Tacx Utility app. This is a free app that will run on a smartphone or tablet (iOS and Android) that allows buyers to perform diagnostics on the Flux as well as updating the firmware and calibrating the unit.
Having been supplied with an early release unit, I wasn’t really surprised that the Flux needed a firmware upgrade straight out of the box. This immediately addressed some of the difficulties I had been having with Bluetooth connectivity however connection times remained relatively slow for my iPad4. I quickly learnt to get the process going first and then get dressed in my cycling kit, so that by the time I returned to the bike, it was ready to ride.
By contrast, connecting my Mac laptop to the Flux via ANT+ was almost instantaneous. This required an extra piece of hardware in the form of a USB ANT+ antenna that was part of the Upgrade Smart kit that Tacx had supplied with the Flux, but any USB ANT+ antenna will work with the Flux. It’s also worth noting that this kit is purpose-built for the Windows environment, so Mac users won’t get any utility from the rest of the kit.
The Flux provides a little bit of feedback when in use in the form of indicator lights on the side of the unit. A red light comes on when an ANT+ connection is made; blue when Bluetooth is in use. Other than that, there is no other feedback from the trainer — except for the disturbing blinking green light indicating the Flux is operational — and no way to carry out any kind of testing or diagnostics without a smart device.
Which brings me to apps and software. This is where the heart of a smart trainer lies, directing the changes in resistance associated with a structured workout or a ride in the virtual countryside. This is not something that I will go into any detail on other than to say that there is a variety to choose from (e.g. Tacx, Zwift, TrainerRoad, and The Sufferfest, to name just a few), and while all will work with the Flux, there are differing platform and/or connection requirements that the user will have to sort out for themselves.
The Flux arrived for review in the middle of the Australian summer, so there wasn’t any great need to put it to use because of unfavourable riding conditions. Be that as it may, I tried out a few different apps on a couple of different platforms over the course of a few weeks, and through it all, the Flux was a smooth, quiet and dependable performer.
Just how quiet was the Flux? I’ve never used the Neo, so I don’t know how the Flux compares with what might be the quietest trainer on the market, but I went to the trouble of measuring the noise levels, which showed that it was only marginally louder than a bike in use.
I measured the ambient noise in my outdoor area at 60db; after mounting my bike on my workstand, I measured 70db when pedalling and 90db when freewheeling. Doing the same with the Flux, I measured 80db when pedalling and 85db when freewheeling (the Edco freehub wasn’t as loud as my DT Swiss hub). Moving the training setup into a small room (3m wide, 2m long 2.5m ceiling) had very little effect on the noise levels, which were 80db when pedalling and 90db when freewheeling.
There’s a 7kg flywheel in the Flux that’s a little difficult to get going but it makes for a quite realistic pedalling action.
As a consequence, there was no need for earphones to listen to music or follow the audio cues from any of the training apps I was using. With that said, a rider won’t go unnoticed if he or she is sharing the space with other people, though I suspect the noise they’ll be making with each effort (i.e. the gasping and groaning) will be much louder than the Flux itself.
There was some mild oscillation in the sound of the Flux when the flywheel was rotating, and early on, some light clicking that could be traced back to a loose cassette. I was lucky that there was no need to re-adjust the rear derailleur to suit the Flux, but for those that are in the habit of using different wheelsets, they will already understand that minor adjustments to the limit screws and/or cable tension are sometimes necessary for crisp and quiet shifting.
There’s also the possibility that if the bike has a worn chain there may be some skipping and/or lazy shifting when paired with a fresh cassette on a trainer.
My maximum power output is very modest, so I never tested the limits of the Flux. The unit proved to be very sturdy while I was in the saddle however it could rock with a concerted effort when I was out of the saddle. I wouldn’t count this as a shortcoming though, since it probably had more to do with my technique. Having made use of rollers extensively in the past, one of the gifts that stationary trainers have to offer is to teach the rider how to remain steady on the bike while making a big effort.
The pedalling action was smooth, consistent, and quite realistic though some effort was required to get the 7kg flywheel moving. Once in motion, every change in resistance occurred on cue without fanfare or fuss and I could immerse myself in the workout. My awareness of the Flux quickly disappeared thanks to the quiet, unobtrusive manner in which it went about its business.
This has to be counted as one of the strongest features of the Flux. If I closed my eyes — and ignored the fact that I wasn’t moving — then it was possible to imagine being out on the road. I still prefer the feeling of having the wheels rotate under me that comes with using rollers, but otherwise, the Flux is very effective at bringing the cycling experience indoors.
The only time that this spell was broken was when I was freewheeling. The flywheel was always quick to lose its momentum and the buzzing from the freehub would come to a sudden halt, yet the data display would still show me coasting along at the same speed.
I started using the Flux with the free Tacx Cycling app on an iPad, and while it delivered a threshold test with clinical efficiency, there wasn’t anything to really engage with (though I did have some fun changing my weight in the settings to see what kind of effect it had on a 1km climb; needless to say, dropping 20kg makes it much easier to rocket up a climb!) I then went on to experiment with a few of the films that Tacx has created, however I couldn’t shake how artificial the interface looked as the frame rate sped up or slowed down in an attempt to emulate my current speed.
Zwift was a different thing altogether, and I was immediately hooked. I found the semi-realistic gaming interface to be the perfect match for the semi-realistic cycling experience that the Flux was offering.
In fact, the two meshed seamlessly to the point where I was tempted to lean with each corner and I was able to dose my efforts by reading the terrain. The Flux seemed flawless in the way that it was able to recreate the subtle changes in resistance that come with riding an undulating road. The various power-ups and awards also kept me engaged to the point where I was choosing to stay on the bike and time was passing quickly, rather than slowing down as it normally does when I’m on a trainer.
Figure 1: validating power and cadence measurements using InfoCranks. A 20 minute training session was carried out using the Flux and data on power and cadence were collected from the unit. At the same time, power and cadence measurements were collected from a set of Verve’s InfoCranks that were fitted to the bike. Panel A shows power and panel B cadence over the course of the session, where the green line represents data from from the Flux and the purple line represents data from the InfoCranks.
It is inevitable that any device that promises to provide power data will be subject to close scrutiny. With a set of Verve’s InfoCranks on hand, it was a simple matter for me to compare the output of each to verify the claims that Tacx had for the Flux.
Over the course of a few sessions, I found that the numbers for power and cadence from the Flux generally agreed with those from the InfoCranks. Throughout a workout, the average values reported by the Flux for power (Figure 1A) and cadence (Figure 1B) were very close to those determined by the Infocranks. The Flux couldn’t match the InfoCranks when it came to measuring the maximum power for explosive efforts, and while there were plenty of instances when the Flux fell short (Figure 1A), there were also instances where it reported a higher value than the InfoCranks (Figure 2).
Figure 2: Further validation of power measurements using InfoCranks. A six-and-a-half minute training session was carried out using the Flux and data on power was collected from the unit. At the same time, power measurements were collected from a set of Verve’s InfoCranks that were fitted to the bike. The green line represents data from from the Flux and the purple line represents data from the InfoCranks.
The extra “noise” associated with the measurements made by the InfoCranks was a product of the meter’s very high sampling rate, so it was better able to capture momentary changes in power and cadence that were overlooked by the Flux. Clearly, the Flux will not satisfy those riders with high demands for the immediacy and accuracy of power measurements for explosive efforts (though this may change in the future as Tacx refines the firmware for the Flux). For everybody else, the Flux lives up to its claims.
Tacx promises that the Flux is maintenance-free, however I’d caution against neglecting the unit altogether. At the very least, Tacx recommends calibrating the Flux at least once a week, which involves getting it up to 30km/hr and letting it run down. It’s also a good idea to keep an eye on the axle caps, which I found could unwind over the course of a week. And while the bike is being used in a clean environment, it is important not to neglect the chain by remembering to apply some lube on a regular basis.
The Flux is a sound stationary trainer that offers buyers a suite of thoughtful features, including a clever freehub body along with a fair amount of grunt in terms of resistance and slope choices. It is expensive though, so I don’t expect it will have much appeal for novice and/or occasional users. Some experienced users won’t be satisfied with the Flux either, since it struggles to measure explosive efforts with greater than 95% accuracy.
In broad terms, the Flux is like a high-end bike because what it has to offer is more a matter of nuance than anything else when compared to cheaper trainers. This includes the convenience of a direct-drive mechanism to save wear on the rear tyre, steeper slopes for tougher workouts, and higher resistance for testing the fitness of near-elite cyclists. These are the sorts of things that are missing from significantly cheaper rear-wheel-driven smart trainers.
As an occasional indoor cyclist, I don’t see a lot in the Flux to tempt me. The low portability of the unit is as off-putting as the price, and having wrestled with Bluetooth, I can’t see why a smart trainer at this price doesn’t include an ANT+ antenna to simplify connectivity of the unit. Nevertheless, the Flux does everything it says on the box, plus, it is very quiet with a very smooth and realistic pedalling action.