A Power Meter Deep Dive with Special Guest 4iiii's Scott Cooper – Ask a Cycling Coach 229

Special Guest 4iiii’s Scott Cooper joins the team to discuss all things power meters, including how power meters are made, calibration, and what the future of power meters may look like. Tune in to hear Scott’s insights in episode 229 of the Ask a Cycling Coach Podcast!

Youtube Live Video:

Topics covered in this episode:

Subscribe to the Podcast

Listen to previous episodes on SoundCloud

Episode Notes:


Great episode.

Power meters should just include a pamphlet with a link to this video in the box to explain this to people.

1 Like

I can definitely sink my teeth into more fitness tech deep dives.

I will admit I haven’t listened yet, so don’t know if this was talked about or not: any comment from 4iiii on the purported underlying power data unreliability for the right side given the current Shimano asymmetric crank arm design?


Ive been thinking of getting a 4iii for a little bit and I think this video just sealed the deal!
Great video TR

1 Like

This is also worth a look on the topic,

Love my 4iiii (single sided 105 crank), never missed a beat.


Excellent episode! Love to see more of the tech/equipment deep dives on the podcast, thanks!

I skipped though this looking for them addressing this issue… I couldn’t find it. (Someone correct me if they did and I missed it?)

A few comments on this specific topic…

  • It’s not an issue with the technology (gauges, radios, battery, reliability, etc).

  • My testing indicates the same tech on other non Shimano cranks (right side specifically) works brilliantly.

  • It’s a problem on Shimano’s very own power meter they themselves have not solved.

  • I was going to mention something along the lines of “it doesn’t take a degree in mechanical engineering to…” but it kind of does. This is why I referenced Keith Wakeham, a lot, in my report. (A thorn in the side of 4iiii there with him previously being their Director of Power Meter Tech)… so he KNOWS this issue well.

  • Cutting to the chase - Shimano are producing a potato of a crankset when used in this way (right side, specifically, left is fine). Everyone is just caught up in the “Oh, we better make a L/R power crankset version” and are holding a left crank and a potato to work with.

  • It’s a PR/media problem if any of them admitted the issue. So they won’t. Which shits me to tears the amount of hours/days/weeks I put into trying to get to the bottom of wtf was happening as a consumer, and now I’m just the prick who’s talking about it.

  • I’ve had denial from Shimano the issue exists… and zero follow up from the rest. Seven manufacturers I believe the count was.

  • I’m sitting around waiting to see what Shimano do next. They’ll have to pivot somehow to make a better power meter which I suspect will shut out the rest who are piggybacking off their cranks at this point. An internal gauge design would do it. So would making a set of SPD-SL power meter pedals. I’d be ok either. I just want things to work as printed on the side of the box…


My 4iiii single side shimano XT on my hardtail is the best power meter I have owned. Fitted it and forgot it. calibrate every ride and change the battery as required

Previous PMs:
Stages gen 1 (x2) - battery killers
Powertap P1 - good but died, and would occasionally wander off
Vector 3 (still in use on road bike) - least said the better

1 Like

My journey with my 4iiii is mixed: after (as best as I can tell correct) installation with bolts torqued to specs, power readings were high. After a few weeks, readings were normal, though. However, the battery unexpectedly failed before one race (going from 70 % to 0 % over night) and the Bluetooth connection crapped out in the beginning of a hill climb race (although perhaps iOS 13 is to blame here). Oh, and the app drives me nuts: they want to force me to create an account just so that I can update the firmware over my phone.

Nevertheless, given the price, I am mostly a happy customer.

Being an out-and-out geek, this was a great episode :nerd_face:

Interesting thought. By that I guess you mean that everything is housed within the crank arm on the drive side prior to bonding the two halves of the arm together. This will allow the strain gauges to be moved much further away from the spider, which is what is causing the problems in the first place. Currently the sensor is very close to the spider because of the chainrings.


1 Like

Following on the topic of zero offset, I’m wondering if anyone can help me. I have an older PowerTap G3 on a track bike. Worked great for several years then started reading too high. Only briefly did I entertain the idea that my FTP jumped to 400W. If I ride it on the trainer and do the zero offset “calibration” in the TR app, it’s all good. But as soon as I ride on the track, it’s too high. I can’t see any way to do a zero offset through my Wahoo head unit or the PowerTap app on my phone. Customer support at Quarq says there is no way to do zero offset and I need to send it to factory for calibration. I’m sceptical about this because the TR zero offset works. Any suggestions?

Can you just use the TR app at the track, via your phone, to do the offset?

  • Maybe you tried this and I missed it above?

The way I understand Wakeham’s analysis, this would not eliminate the problem as other problems crop up once you move the strain gauges farther away from the crank.

I think what @GPLama is something like Rotor’s InPower power meter, where the strain gauges are affixed to the crank spindle.

I don’t understand why moving them further away would be worse than they already get on the non drive side where they can put them wherever they want.

In the spindle is non drive side anyway so that wouldn’t solve anything.


Have a look at Keith Wakeham’s video that @Aseldi posted in this thread. He compares four different scenarios, 2 cases with Shimano-style, asymmetric cranks and 2 cases with symmetric, SRAM-style cranks. The two variations differ by the distance to the crank axle.

Specifically, go to 14:00 of the video: while there is an improvement for the calibration curve, it is worse for the axial contribution which leads to manufacturers “fudging” with the curve. Or do you understand that differently?

1 Like