Yes it’s a good test. I do it myself every now and again just to check if things are consistent. I also take a note of the calibration offsets. If they suddenly changed dramatically it could be a sign of a problem. Luckily both my crank PM and trainer have been very consistent over the last year. The only thing I noticed was that my trainer appeared to read very slightly on the high side compared to my crank PM when it was brand new and took a couple of months to settle down to a very consistent level. I do also consistently see slightly more % difference in power at the lower power levels.
Edit: When I ride outside, I just mentally increase my power targets slightly compared to riding on the trainer. It would be nice if I could apply a correction factor to make them identical, but my PM software doesn’t allow for that. But I’m talking about 5-10W difference across the board, so not a big deal.
My crank sensor software doesn’t show the offset number, but my Garmin head unit does if I use that to calibrate it. It has always given the exact same number.
Worth a try for sure. My crank PM is Specialized, but I know they use a slightly tweaked 4iiii sensor and so should be very similar to yours. Calibrating it on my Garmin didn’t make any difference to the readings in my case, but it is much more convenient than opening the PM App. I can set my Head unit to prompt me to zero the PM at the start of each ride, which is handy.
I used Open 40 with the trainer in ERG mode and smoothing disabled through the Android app. I used the in workout adjustment to increase and decrease the power it was expecting. Each block was about 2 minutes with a break between to give a noticeable gap on the graphs.
I did a ten minute warm up and then did a spin down from the Wahoo app and a calibrate on the crank from the Element head unit. The spin down numbers were about the same as normal, I’ve not done a crank calibration from the head unit before and that was 10, I don’t know if that is good or bad.
My watch recorded data from the Kickr, Sufferfest recorded it from the 4iiii left crank.
The trainer reads consistently lower than the crank, the average difference being about 15 w or 12.5%.
The difference seems to be getting smaller as the wattage increases with 100 w being the odd outlier where they both matched.
As all my riding, indoors and out, will be based off the crank, I guess it doesn’t really matter whether it is accurate or not as it at least seems consistent. I’m not vain and boasting about my FTP so it doesn’t really matter if it is the 171 I got from the crank on the FF or 150 as the trainer may think it is.
Both crank and trainer were refurbished, the trainer direct from Wahoo, the crank from a store so I guess there could be something faulty somewhere.
Okay first obvious thing I note is that in your earlier posts your trainer was apparently reading much higher than your crank (KickR = 173W vs Crank = 152W) and now it is completely the other way round with the trainer reading lower (KickR = 152W vs Crank = 171W). So something major going on there!
Your latest results are at least more in line with what you would expect i.e. crank should read higher as it is measured upstream of your drivetrain, which will typically lose 4% of your power. So if we allow for that difference, then your crank is reading on average approx 8.5% high relative to your trainer.
At low power levels there is more chance that you could be pedalling with a significant bias toward one leg, which would mess up your crank power reading as it’s a single-sided PM. At higher power levels often these leg discrepancies tend to balance out as you are concentrating more on producing power as best you can. This could possibly explain why you got an odd result at 100W - maybe you were just subconsciously resting your left leg slightly causing the crank PM to under-read total power. I’m assuming here that you don’t have a known individual leg weakness or injury that would seriously throw off your left/right power balance across the power range.
My Garmin head unit always shows a calibration value of 10 for my crank PM too (also 4iiii), so probably okay or a meaningless figure! But at least it’s the same number I get.
As for the remaining net 8.5% difference, it is what it is. I would say it’s outside of what you would reasonably expect given the accuracy of both sensors IF they were both working within their manufacturing tolerances (typically ±2% ish and often better in practice). Impossible to say which of your power sources (or both) is off.
I would repeat this test every month or so until you are confident it’s not a moving target.
I think I just got the figures the wrong way round earlier. I tied the Sufferfest app with the trainer as it controls it, but it actually gets its figures from the crank. My mistake there.
I’m about to start the 12 week FTP program so I might retry this every 4 weeks during that. Now I know it is fairly consistent and the doesn’t look like something is broken I’m not worried about it. It is more just curiosity now.
And now I’m also curious to know if I’ve got mismatched legs!
Sounds like a good plan. You can always just do a quick power check while you are warming up before a workout. I always display my crank power on my head unit while using trainer power on SUF, so I would notice pretty quickly if something was wrong.
I have to say your trainer power now looks pretty noisy compared to the crank PM. I know you’ve removed the smoothing filter, but since it looks like you got your figures the wrong way round in your earlier posts, it’s the trainer that looks noisy rather than the crank PM. It’s usually the other way round, so that might indicate a fault with your trainer. But I have no experience with the KickR myself, so don’t know if this is normal or not.
I’ll have a word with Wahoo and see what they say as I bought it direct from them. Can guarantee that it will be just out of warranty if there is a problem.
I was curious about how my Assioma Duo pedals and Kickr Core would compare after a spindown and zeroing “calibration”.
Yesterday, after Butter, I performed a spindown registering 14.4 sec and approximately 58 Celsius and also “calibrated” my pedals.
Today I did The Shovel registering the Kickr Core using The Suf (via BLE in Win 10) and the Assioma pedals using a Garmin Edge (via ANT+). HR and cadence are the same.
My main observation was overall power values measured with Kickr being about 97% of values measured by the pedals. Also, the peak values for the short sprints were considerably higher with the pedals, see the screen captures of my Garmin Connect.
3% loss from the drivetrain is what people told me I should expect to lose so that seems about right.
I don’t know how much it matters, but I though the spindown was supposed to be done after about ten minutes warm up rather than after a full session. I tried it after once and the numbers were very different to what they normally were.
That’s is outstandingly close given the different equipment - amazing that they’ve got the tech working that consistently. I’d be even more amazed if any of us could get numbers any closer (given the 3% drivetain loss often quoted between pedal/crank and rearhub/trainer). Amazing
I was happy with that power comparison. Another comment here mentioned I had spun down after a regular workout, not the 10-min warm-up recommended.
I will perform a new test with pedal and Kickr Core “calibrations” as recommended, let’s see.
Probably I will try it today before AVDP. I don’t know if it will be a good comparison, since there is no sub-threshold steps for this workout. On the other hand, I will see how both respond to short term variations at my MAP efforts.
I “calibrated” the Assioma pedals as I connected them to Garmin Edge.
Then I warmed-up Kickr Core for 10 min (started at about 60 W, ramped to 120 W in 5 min and held for 5 min). Performed spindown, which showed 12.2 s (or so) at 37 Celsius.
During warmup, I could see slightly higher values in Garmin Edge for the pedals (something as 126 W average for 120 W in Kickr Core at Wahoo app).
After the spindown, the numbers got pretty close. I just disconnected everything and started the AVDP workout using Win 10 app for The SUF, for which I showed the image comparing the 20-min max power.
It seems to be very consistent with what I saw before. I don’t remember exactly when I did a similar comparison with proper “calibration” and spindown last year, to dig into Garmin Connect and find the comparison; but I remember the numbers used to be very close.
Just for those interested or in need, I serviced my Wahoo Kickr Core unit by myself, about 2.5 months ago. It seems I did a great job on it.
I got a bad bearing making noise and Wahoo suggested the replacement of my unit by a refurbished one. Since it was not covered by warranty and I moved from the US to Brazil, the shipping costs were prohibitive. Then I decided to service it by myself, managing to replace both bearings that go on the shaft of the flywheel.
After properly adjusting the (terrible) bolt/device that tensions the belt, the spindown was showing consistent results. Anyway, I found on the internet how to perform a “factory” spindown (don’t remember if they name it that way) and did it.
I see exactly the same thing between my Assiomas and first-gen Tacx Neo. I’ll go with the 3% drive train loss and try not to worry about it. I think the main thing stick to one power source for consistency’s sake.
Everyone here is very switched on and has clearly been doing this for a very long time. I’m an absolute noob, second week on the trainer.
My question is slightly different to the above discussion, but hoping you can provide me some insight. I’m tracking my workouts both through the SUF iPad app, and my Garmin Edge 530. They are both connecting to the Kickr Core for power, and to an ANT+ cadence and HR sensor. Everything works flawlessly, and the 3-sec average power between the SUF app and Garmin unit are consistent. However, at the end of the workout, the avg power data as well as the normalised power data differ between Garmin and SUF by about 5-8W. Surely this discrepency shouldn’t exist if they are using the same power measurement input? Which would you consider more robust to align your performance and progress to?
As you say - there’s only once device transmitting data - so the data is the data, and in theory and average of a simple power number should be calculated the same way you’d think.
As you’re training on SUF I’d use SUF. Though one thing to bear in mind - it’s just a number. That avg power number is just a ‘stat’. It’s not used for anything
If you use erg, SUF will ask the trainer to hold at certain power levels in intervals and so on.
So to a certain extent it doesn’t matter (other than just being totally weird and woke probably bug me until I understood why) and doesn’t affect your training as you’re trying using a smart trainer with SUF.
I take it the Garmin gets started at exactly the same time as you start pedalling and recording and same with stopping
No stops in the workout that could cause issues ? (Is Garmin set to include zeros in its average?)
