Yes, you can build a DIY power meter at home … but please don’t
Sensitivus, the Danish company that brought you the “$6 power meter” is at it again, this time with detailed online instructions for how you can build your own bottom bracket spindle-based single-sided power meter at home for less than €300 (AU$500 / US$330) and a fair bit of elbow grease.
The concept that’s presented seems sound, with dual strain gauges, a compact “brain” and wireless antenna, and user-selectable rechargeable battery life. In addition, the fact that the instructions can be applied to virtually any Shimano crankset with a 24 mm-diameter steel spindle means the system will supposedly work on a huge variety of crank models.
Got a drill, some cyanoacrylate glue, maybe a small 3D printer, a fair degree of mechanical savviness, and some time on your hands? Then, my friend, you’re in business.
Except for the fact that this is an absolutely terrible and wholeheartedly irresponsible thing to suggest people do. If all goes according to plan, then yes, it sounds like you might end up with a power meter that produces some good data. If it doesn’t — and there are myriad ways for things to go south — your numbers won’t be worth much, or you’ll have spent a bunch of time and money to get nothing more than a now-ruined crankset.
But at worst, you’ll end up with a catastrophic drivetrain failure with potentially little warning, and possibly a sizable hospital bill (or worse). And let me tell you from personal experience: dental implants are no fun.
Allow me to explain.
“A simple way to build a power meter in your kitchen”
The draw of an inexpensive power meter is a powerful one, and the thought of being able to build one yourself using your existing crankset is more tantalizing still. The instructions posted by Sensitivus are fairly detailed, and the process doesn’t seem terribly daunting for someone with reasonably decent DIY skills. There are plenty of specifics listed, Sensitivus makes the required electronic parts available for purchase so you don’t have to fabricate a printed circuit board with on-board components de novo, and there’s even a calibration tool to dial in your creation once you’re done.
“Performance is every bit as good as a commercial power meter,” said Sensitivus CEO Rolf Ostergaard. “The performance is better than some commercially available power meters, similar to others, and much better than any other DIY project we have seen. We have done many more tests on this including a full calibration test on our professional calibration machine. If you do a good job on the mechanical build, you will actually end up with a very good power meter.”
However, it’s when you dig even a little bit deeper into Sensitivus’s tutorial that some issues arise.
First and foremost, the process involves drilling a 2 mm hole into the spindle of your 24 mm-diameter Shimano Hollowtech bottom bracket spindle. This provides access for the strain gauge wires to connect to the electronics module (which includes the system “brain”, the wireless antenna, and rechargeable battery) that resides inside the spindle.
Now, 2 mm doesn’t sound like a lot, right? Especially with all of that reassuring steel that still surrounds that itty bitty hole. It’s perfectly harmless, surely? Except that it’s not.
Spindle failures are hardly commonplace with Shimano’s Hollowtech crankset, but they’re not unheard of, either. There’s this concept in the field of fracture mechanics called a stress concentrator. Basically, they’re geometric features in loaded parts that — as the name suggests — concentrate applied stresses relative to the surrounding material. Ever notice how toilet paper usually tears at the perforations? Those perforations are stress concentrators. The paper is supposed to tear there by design.
When you drill a hole in something — even if it seems harmlessly small — you’re creating a stress concentrator. For a theoretical plate loaded in shear, the resultant stress concentration factor is somewhere in the neighborhood of four, meaning the edges of the hole are subjected to four times the stress of the surrounding material. Of course, this isn’t a theoretical flat plate of infinite size, so the numbers don’t translate directly, but suffice to say, the likelihood of a crack initiating there is substantially greater than it would be without the hole.
Keep in mind, too, that people aren’t exactly known for regularly removing their crankset spindles for visual inspections, meaning there’s a more-than-zero likelihood that such a crack would propagate without the user having any idea it’s there at all — that is, until it fails. That failure could well be catastrophic in nature given the high applied loads, and very possibly result in a nasty wreck.
As I mentioned already, spindle failures aren’t exactly a dime a dozen with Shimano cranksets. But they do happen, and the thought of a company suggesting to people that they should modify those cranksets in a way that would make those failures even more likely seems grossly irresponsible at best.
For its part, Sensitivus seems to be washing its hands of any responsibility, and is arguably being a bit cavalier about the potential risks.
“If you don’t know that drilling a hole in a tube makes it weaker, you should obviously not be building your own power meter,” Ostergaard said. “We have added a safety note for this now – just in case. The small hole will likely be where fatigue sets in – if you ever get to that point — something I doubt will happen for the average person. With this power meter installed, you will very likely get strange readings before anything gets dangerous. If you are seriously worried about breaking spindles, get the power meter installed to monitor it – or take up another hobby.”
There are other red flags.
While Sensitivus provides plenty of guidance on how to get seemingly accurate data from your new DIY power meter, the company has left the builder to figure out several key details on their own.
Take this line, for example: “The looks, waterproofing, and robustness is 100% up to you.”
In other words, the company will sell you a circuit board, but you need to figure out how you’re going to mount it inside the spindle.
And then there’s this: “In no way is this a finished product and you are not allowed to make and sell these. This is Do-It-Yourself only. For those that find that fascinating, challenging, or whatever drives you. As with any recipe, make sure you read the whole thing before you start and make sure you feel comfortable doing each of the steps. We provide only paid support on this and encourage user-to-user support in a Facebook group set up just for this.”
As of the time I wrote this article, that Facebook page had a total of two members (with the company itself included as the administrator). And there’s yet another disclaimer once you get there: “We can not offer support on this, as this is really just a ‘for fun’ project for us.”
I don’t know about you, but I don’t find much of this particularly comforting. Has this DIY concept been tested long-term (or even short-term) for durability? How about ultimate strength or fatigue strength? Is addition to creating a stress concentrator, will that hole also introduce the potential for stress corrosion cracking?
There are yet more nuggets on the DIY page.
“Carefully inspect under a microscope…”
I have one of these at home, don’t you?
“You would think the exact location and alignment of the strain gauges are extremely important. And it is important, but you can actually get pretty good results just eyeballing it. The calibration step will compensate for a lot of the “error” you make here.”
What’s “pretty good”? And what exactly is “a lot”?
“Then hang a known weight around 10 kg on the pedal (crank arm horizontal) …”
Needless to say, a “known weight around 10 kg” is more than a bit vague, and you’d absolutely need to verify the actual weight with an accurate scale. Thankfully, Sensitivus’ calibration tool at least allows you to input the actual weight, assuming you’re able to obtain it. Either way, how do you know the crankarms are exactly horizontal? And even if you’re able to calibrate your new DIY power meter at horizontal, what will the readings look like at 30°, or 45°, or 17° or 71°?
Just because you can doesn’t mean you should
Sensitivus stated back in 2016 that it wanted to “democratize power meters”, and without debate, it’s an admirable goal. The long-term vision that power meters might eventually become as commonplace and inexpensive as heart rate monitors is one that I still hold as well. Sensitivus is also correct in promoting the idea that the concept of bicycle power meters isn’t terribly complicated, nor are the required components all that expensive to produce.
Lots of things sound easy and/or inexpensive in theory; however, the theory or concept often isn’t the hard part. When it comes to power meters, it’s the calibration and programming algorithms that make proven power meters worthwhile. It’s the regular firmware updates that ensure long-term compatibility with other electronics. It’s the confidence of a factory warranty, or the reassurance that water contamination isn’t going to short out your DIY rechargeable battery and cause a fire. It’s the testing that promises buyers that the product is safe to use. And on, and on, and on.
Now, in fairness to Sensitivus, the concept of a spindle-based power meter where strain gage wires pass through a hole in the spindle isn’t entirely outrageous. Rotor and Easton power meters both use 30 mm-diameter aluminum spindles with strain gages on the outer surface and electronics embedded inside, and Sensitivus also sells its own ready-to-run power meter under the ZWATT brand name. Both Rotor and Easton insist their designs are proprietary and aren’t licensed from anyone else, though, and although their spindles have holes in them as well, both companies are also emphatic that they’re rigorously tested for long-term durability and safety.
“We do not pay any licensing fees for any of the technology in the Cinch power meter,” said Easton brand manager Matt Hornland. “The exact design details of our power meter are not something I can share. However, the spindle component of the Cinch power meter was specifically designed both to measure torque accurately and provide the high performance and durability that customers expect from the EC90 SL crank set, and any Easton product. We do not recommend anyone doing this to a spindle that was never designed for modifications. It does not seem reasonable to take an existing crank spindle and drill a hole in it and expect a good outcome.”
Meanwhile, Rotor managing director Lori Barrett points out that while the standard non-power meter Aldhu bottom bracket spindle measures about 3.5 mm in wall thickness, the power meter version is roughly twice as thick to provide additional reinforcement.
“You have to do the ISO4210 testing for fatigue on a commercial crankset,” Ostergaard said. “We do not have that testing capability in-house, but have outsourced those types of tests. For high-end products, some manufacturers are wisely using even tougher tests, as some bigger/stronger riders actually break cranks even when they are 100% okay by ISO4210. This is all about what return rates you want to achieve. When we work with manufacturers to design power meters, durability is a key point and there is a lot you can do both on the inside and the outside of a spindle when you design it specifically for a spindle-based power meter.”
According to Ostergaard, though, Sensitivus hasn’t done that testing for this particular DIY project.
Even if you ignore everything I’ve said up to here, or if you think I’m being needlessly alarmist, or pandering to established brands, consider this: Sensitivus is currently asking up to €300 for each one of the electronics modules that you’d need to pull the DIY solution off (along with a compelling written argument as to why you should be at the head of the line for one of the “limited number” of available electronics modules). Keep in mind, too, that other necessary costs aren’t included, either.
For the sake of comparison, a brand-new Stages single-sided power meter retails for US$299 / €299, ready to run. One from 4iiii is about the same amount of money.
Oh, and there’s also the minor detail that the module Sensitivus is supplying might be defective.
“These modules are sitting in a box here because we suspect some of them may have a problem with microcracks in a specific capacitor,” reads Sensitivus’s DIY page. “We know this problem in all its details and the fix is quite simple: replace the capacitor (we will include instructions with the module). Most modules are probably all good, and those that are not will over time start to drain the battery faster than intended if exposed to enough humidity and vibrations. No big deal for DIY, but we can not use them for our ‘real’ customers.”
Potential headaches (and physical injuries) aside, Sensitivus insists this isn’t just about saving a few bucks — it’s just “for fun”.
“For DIY to be fun, we figured you needed to do something creative yourself, right?” Ostergaard mused. “This is not intended as a power meter ‘kit’. This is something we know is more challenging and (nerd-) fun than baking another sourdough bread in these crazy times. If we can inspire [people to have] to some fun and creativity with this, we have achieved everything we wanted. DIY has ‘you build it – you break it – your problem’ written all over it. By definition, DIY is something you have to take responsibility for.”
What’s the point?
At the very bottom of Sensitivus’s instructions for the DIY power meter is this:
In other words, the company’s main business is selling its technology for other companies to implement. Sensitivus isn’t actually interested in manufacturing anything; they’re the idea people. That’s all well and good, and this concept still seems to hold some merit. Maybe someone will take Sensitivus up on its offer and be able to add power meter technology at the consumer level for far less than what we’ve seen to date, which would be fine with me, assuming that company does the hard work of sweating the details in creating an accurate and reliable product so that you or I don’t have to.
But why exactly is Sensitivus encouraging people to build a power meter at home? Or, for that matter, why did they bother to detail to the public several months ago the OEM material costs of a power meter when you’re trying to sell that technology to other brands? Maybe this is all just an unusual way of building brand recognition, or some clever method of advertising.
Or maybe Sensitivus is just trolling all of us.
Either way, I’m all for DIY projects, but this one gets a hard pass from me. Even if it works — and given that Sensitivus has provided itself with more than a few outs here — why take the risk?