Power measurement discrepancies?

[jmschuur]

I’m new to the world of indoor training, so maybe it’s just a case of misunderstanding how things work. Hope you can help.

I have a brand new Wahoo Kickr v5 with firmware version 4.2.5. I’ve updated my FTP in Systm (didn’t do the 4DP test yet). When doing a workout I noticed I have to push the pedals significantly harder than expected. I did a quick check with my Quarq AXS power meter on my bike. I don’t have the exact numbers, but it looks like when Systm says I’m pushing 180w, Quarq says I’m pushing over 200w. If anyone can explain to me how to dual record (I use Systm on a Macbook and a Garmin Edge 530 bike computer) I could upload a comparison. Or even a triple record (Kickr on Systm, Kickr on Garmin and Quarq on Garmin).

As I said, maybe the discrepancy can be explained because of misunderstanding or a misconfiguration. My ultimate goal would be to be able to train indoors as close to as how I train outdoors. Not sure how to setup things for that either, but let’s first focus on the wattage differences I see and feel.

Thanks!

[CPT_A]

Welcome to the club!

First - mandatory - response is to put in a ticket with the Minions? When I first did my setup last year they helped walk me through setup and calibration with the app.

[jmschuur]

Thanks! I was thinking about that, but wasn’t sure where to start; forum or the Minions. I’ll give it a try.

[AkaPete]

I think I read somewhere that power measures depend in part on where power is measured. KICKR measures power at the back (I don’t know if they have strain sensors or calculate it from flywheel acceleration and braking force). The Quarq AXS power meter measures strain the spider—very close to your feet which are applying the power. So it seems plausible that friction and slight elasticity in the drivetrain causes power at the flywheel to be less than power at the spider.

Or—the two meters aren’t calibrated the same.

I’m curious what the minions have to say if you go that route.

[JSampson]

@jmschuur It is normal to have different inside and outside power metrics. Further, note that you are measuring power in two different places in your example - the trainer and the spider and so they are not equivalent. I would suggest just focusing on using the power measured from your trainer.

The easiest way to get your inside numbers is via the Half Monty test which will give you your FTP and MAP numbers and can be done quickly without a lot of rest beforehand.

The Full Frontal test requires about a week of rest and preparation and there is a week long training plan that can be added to the calendar. It is a tough test so be sure to read about it beforehand.

Good luck and be sure to keep us advised of your progress!

[jmschuur]

Thank you all for your quick response. For clarity, I’m not looking for numbers that are exactly the same on my Quark PM and my Kickr. I know there will always be differences because of different factors, like the examples you’ve mentioned. My problem is that the power that I have to put on the pedals also feels very different. So, 180w on the Kickr feels significantly harder that 180w on the bike outside. Although I do know inside and outside cannot be compared, I’d expect the numbers to be closer to each other.

I was able to compare the two powermeters during the exact same ride. I took the log from Systm and compared it to the log of Garmin (which logs the Quarq data). Take a look at the graph below for example. Red is Kickr, blue is Quarq.

The numbers aside, I’m particularly worried about overtraining in this situation. I have a counselor that 'll create a new training plan for me, based on a baseline (from a test) and based on my power. If power measurement on the Kickr is not accurate… The Kickr specs says the Kickr has an accuracy of +/- 1%. Quarq AXS has an accuracy of +/- 1.5%. That is a total of +/- 3%. I think the difference at this time is around +10%.

[AkaPete]

The discrepancy is very consistent. That’s a good thing—it means both power meters are generally working and detecting changes in power. My suggestion (of course, I’m not a coach) is to use the power numbers you get from the trainer when you do Full Frontal or Half Monty for your indoor training. Then, if you do structured training outdoors (I’ve never been able to really pull that off with the complications of highly varied terrain and traffic), jack the numbers up by 10%. In fact, at the end of this old Sufferfest article it says

For the outdoor ride, the goal is a higher average power (90% FTP) but if you are riding indoors it should be adjusted down to around 80% as seen in the NoVid workout. This is because most people will have a naturally higher power outdoors than indoors. The effort will be similar.

I don’t know if the “naturally higher power” referred to is because of different power meter placement or other differences when outdoors. But either way, you need to adjust: different FTP measures for different situations and configurations.

[jmschuur]

Thanks again. I’m trying to wrap my head around it. The fact that the discrepancy is consistent is a good thing indeed. I just read a more recent article from Wahoo about the difference in power output indoors vs outdoors.

https://eu.wahoofitness.com/blog/ask-experts-indoor-cycling-power-vs-outdoor-power/

It’s interesting, because I would not expect such a big difference in power inside vs outside. Nevertheless I do know my core is far from strong enough, although I’m not even a bit of a ‘swayer’ and I have no back pain issues. I’m quite stable and pain free on my BMC TeamMachine SLR01. Well, maybe this is an eye opener with a positive side (there is room for improvement, like there wasn’t any… ).

I guess best next step is to do the Full Frontal and go further from there. Though I still like to get an expert view on this from the Minions, as well as some good help and advice on setting up the indoor trainer for my needs. If anything worth mentioning comes out of that route, I’ll make sure to update this thread.

Thank you so far!

[JSampson]

@jmschuur Best of luck on Full Frontal. Search in the forum and you will find lots of threads about the test that are worth reading and will help you prepare.

Also check out the Elements of Style video. It is a drill video that will help you identify your potential posture issues and provide you a list of items to keep thinking about.

You are definitely correct that core is key and really helps to effectively drive power into the pedals. Check out the Wahoo Knowledge Podcast: What Core Training Can Do For You for some additional tips - lots of low planks, bridges and bicycle crunches will go a long way.

The Knowledge Podcasts

[jmschuur]

This morning I played around with my Kickr setup, checked everything and decided to do the “Recharger” workout in Level Mode @ 0.5% (feels much better and more realistic then ERG mode I must say).

Below you’ll find a comparison chart of power and cadence; Kickr values logged by Systm and Quarq values logged by Garmin. I thought I understood why I was seeing what I was seeing, feeling what I was feeling, but I don’t as it turns out. I do understand the difference between indoor and outdoor cycling (I think) and I do understand that these two power meters (Kickr and Quarq) will never show the exact same numbers. But shouldn’t both power meters measure more or less the same? I mean, they are both inside, operating under the samen conditions. Only difference is how these devices measure power (spider vs rear wheel). In other words, I don’t expect my Quarq power meter to measure the same power inside as it does outside (it’s clear it doesn’t), but I do expect the power measured by both Quarq and Kickr to be more or less the same. I do know my Quarq power meter is pretty accurate (I can never be really sure, but I’ve tested it more than once against other power meters). So, what am I missing? Or should I do that “factory spin down” and see if that changes anything?

PS. When I’m doing the FF on this Kickr and do the training based on the result of that, the training will be ok. But I didn’t pay so much money for a trainer that doesn’t show the correct numbers. although the discrepancy is very consistent between Kickr and Quarq. Does that make any sence?

[bukharin]

I think this discrepancy is more than it should be. I have a similar setup (Quarq power meter sending data to Garmin, and Kickr Core sending data to SUF/SYSTM), and these are my power curves for a recent Recharger session:

I don’t know if this will help but you might want to consider trying to set up the Kickr to use the Quarq as its power source. You can do this in the Wahoo app (on your phone). The instructions seem to imply this would only work in the Wahoo app (not other apps like SYSTM), but a number of people use this setting and it seems to work (?).
Using Your KICKR and Power Meter With the iOS Wahoo Fitness App

[jmschuur]

Thanks! That looks a lot more as expected. I just added Quarq PM as a device in Systm (great tip btw) and did a quick test. The values on the Garmin and on screen in Systm are pretty much the same. This is expected of course, but it’s good to know that it rules out any connection influences, software faults, etc. I cannot be sure yet, but it looks like the values measured and/or presented by the Kickr are not correct.

Level Mode and ERG Mode work as expected when using the Quarq PM as power meter. Of course this is not how it should work and not how I want it to work.

I also wanted to test how it works when riding an On Location workout. I hope to see that when there is a climb, Kickr will constantly adjust the resistance based on how steep the climb is and my physical values (weight, bike, etc.). Do you know how that works?

[bukharin]

Glad you got that working at least! Personally I’m fine with using the Quarq as the power meter - it means the indoor and outdoor rides are directly comparable. I also use the Quarq as a cadence sensor because it goes higher than the Core (the Core was maxing out in Cadence Builds). I can’t think of any real disadvantage apart from reducing the Quarq’s battery life, but it still lasts a long time and is very easy to replace.

As for the On Location climbs, the power requirement goes up, but it’s set as a fixed % of your FTP/MAP/etc, based on what the rider was doing, and doesn’t take weight into account. It is not trying to simulate the physics of riding up a hill, but instead is trying to replicate the effort (relative to ability) that the rider put in. So you will go up the hill at the same speed as the rider did, no matter what your power profile looks like.

This is how pretty much all of the rides in SYSTM work - the power targets are based on your 4DP results, and your efforts don’t affect what’s going on on the screen.

[jmschuur]

I understand your point of view on using the Quarq as the power meter. Nevertheless, I’d like to get what I’ve paid for and I’m thinking about creating a fixed setup with a cheaper bike/frame on the Kickr for indoor training.

As for the climbs; I feel a feature request coming up . Shouldn’t be too hard to add this functionality into Systm. I’d love to follow a structured training plan on the Kickr and ride epic climbs on it as well.

For now I’ll try to contact the Minions again to discuss the power meter issue. I’ll update this thread with the outcome. Thank you all so far!

[carytb]

A power match feature would sort this problem out however this from the Sufferfest guys
"No plans at all for Power Match. While there are some other reasons for wanting it, the accuracy of smart trainers reaching pretty much the same levels of power meters, the requests for power match has declined quite a bit.

[bbulkow]

I have the same issue as jmschuur, and it messes up my training in a number of ways.

1. TSS targets. TSS is by watts, and a 20% difference means my indoor TSS are really low and don’t accurately reflect my training.

2. Watt targets while outdoors. While doing a new SYSTM plan (Gran Fondo), they wanted me to do a certain workout (4 8-4 intervales) at a certain power level. Those were all calculated based on my indoor training - and when I was out riding, the numbers were so low as to be not effective training.

I can’t buy the people saying “your outdoor watts are higher” and “just train to a 10% difference if you’re outdoors.” If my training plan has me hitting certain watt targets, and certain training amounts, and we actually believe in these Watts things, it would seem a 20% difference is enough to be worried. Wahoo says the trainers are within a few percent, and I believe them - and I can’t believe it’s really happening.

If I had only one watt meter, I wouldn’t be annoyed. But I got an on-bike meter, the Garmin RS200 pedals, and the watts I’m getting on my outside rides are in a different category. When I ride inside, I can’t do anything like what I can do inside (even with a fan).

Eventually, and this is why I’m writing jmschurr, there’s something I just don’t see mentioned. Reflash, and factory calibration. I’ve re-flashed my firmware and done a “factory recalibration”. It’s a hard to find setting in the app (at least in android, you have to click in a dead zone 15 times for the dialog to come up) - so I mean really hidden. That’s also where I found the ability to reflash my firmware.

Good news is my bluetooth connection is now stable - it wasn’t before. I had to re-power my tablet (the one I run SYSTM on) every time to get it to connect, after doing the calibration spindown. This is very encouraging. I don’t know why the systm app hadn’t recommended a reflash.

Will my power numbers be better? I don’t know yet! The “factory calibration” does a regular spindown, then does it again a second time, but clearly adds a few watts of resistance, and based on the difference in spindown time, figures what resistance “means” for your exact device.

I don’t know why Wahoo has this kind of calibration and doesn’t tell people. There’s some mention that this “causes wear” in your trainer. I can’t see why this would cause more wear than running normally applying a little resistance, but maybe it does. People suggest doing it no more than a few times a year.

You can find other information online about the “advanced spindown”.

Good luck —

[Coach.Andy.T]

So those accuracy figures are within themselves. So When the KICKR reads 200 watts, it’s anywhere between 198 and 202 watts and will be consistent and reliable within itself from session to session (an old problem with power meters was that they could swing from day to day by 20+ watts due to various environment factors). The difference between the two is common due to where the power is being measured from. The close the power meter is to your foot, the higher the reading will be, as when you travel further away from there, power is lost through various factors. If you have a spider based power meter (Quarq) it will read higher than a hub based power meter (more often than not). This is because some energy/power is lost through flex in the chainrings, then more through the chain, and finally some through the cassette as well. Using a very clean chain (potentially waxed due too contaminants meaning it should last longer) is a way to help ensure the recorded power on the KICKR is closer to the Quarq. But you may have to accept that there is a difference between the two, and use both indoor and outdoor training profiles. Fortunately this shouldn’t be too hard as you can use the 4DP testing for indoor sessions, and then go off your outdoor power levels for those sessions. Or just add 10-20% on for your outdoor efforts. Although worth checking that the percentage difference is consistent across different wattage levels, i.e. 200 compared to 180 should be 400 compared to 360, but if the range shrinks or grows then that needs to be factored in as well, some power meters have slightly different slope

[jmschuur]

There will always be differences between readings of power meters. The different methods (e.g. spider vs hub) and different circumstances (e.g. inside vs outside) all have impact on the power readings. That’s a fact. My problem is that the discrepancy between two professional, high ranking power meters is more than what ‘feels’ right. I mean, I have a very well maintained top level bike (BMW Teammachine SLR01 with SRAM Force AXS, everything as good as new), a Quarq AXS power meter (also as good as new) and a brand new Wahoo Kickr v5. So, the big difference between the power readings of both should be explained by the difference in accuracy (Quarq 1.5%, Kickr 1%, makes 2.5% total) and the rest in loss because of the drivetrain and frame stiffness. That feels like an insult to my bike So, I’m very much ok with a discrepancy in power readings between the Quarq and Kickr, but my guess is that this is to much of a difference. And don’t we all spend a lot of money on power meters and trainers for their accuracy.

I’m still busy with Wahoo on solving the issue with the power readings. Hope to give you all an update on that soon.

[jmschuur]

I’ve tried different firmware versions and did a factory spindown after each firmware upgrade/downgrade. Although there are some differences in readings, I think / not sure, but I haven’t notice any real differences.

The factory spindown option is suggested by Wahoo when you use their self help articles. Just keep answering ‘no, this didn’t solve my problem’ and you’ll be suggested to do the factory spindown. The article will guide you through it.

[CPT_A]

@jmschuur @bbulkow, Referring you back to @Coach.Andy.T’s explanation of power differentials, at some point simple physics would like to introduce itself.

The only way programmers can overcome the basic physics of waste heat, friction, and environmental factors would be to build in compensation into the algorithm that generates your numbers. But that doesn’t make it “better”, it just hides the issue from you.

So you’re a happy customer with “acceptable” indoor/outdoor wattage output, but the data are no more valid than if you just accept that there’s only so much engineering can overcome.

Two cents, FWIW.