Power measurement discrepancies?

Exactly and agreed. Question is what the power meter is worth if the measured output is always questionable. I would never spend that much money on a power meter if that’s the case. What I would accept is the difference because of indoor/outdoor/method. I will not accept a difference because of a faulty PM. That is what I’m trying to find out with Wahoo. Is it faulty or not. For now, the discrepancies between the Quarq and Kickr are to big (measured at the same time, the same bike, the same circumstances. I’m not trying to match outside Quarq with inside Kickr).

For clarity on my concern. Imagine I’m visiting my coach for an exercise test, based on which he will create a training plan for me specifically. He’ll collect the numbers with his test bike that has its own power meter. With that numbers I have to train on my bike outside, with the Quarq and inside with the Kickr. It is ok to take into account the differences within margins, knowing that the power meters are correct (within margins). But what if you have reason to doubt the power meters. This would even be true if I use the 4DP test on the Kickr, let System create a plan for me and I would only use the Kickr for my training. You need to be able to trust and rely on your power meter. If you cannot, it’s simply not worth the money.

So, all I want and need is certainty about my power meters. The current output/numbers do not give me that certainty.

3 Likes

Fair point. :+1:t2:

I can’t agree more. The point is the Watt is an international standard well defined for a long time (100+ years?).

I believe I need about 1% or 2% accuracy (bodies are too complex to need much more). 5% would show up in my training graphs, it’s too much. I believe I am buying gear promising that. I believe most gear is delivering it. If gear is faulty, if Wahoo’s Kickr Snap can’t deliver it, then that’s not OK (eg, caveat emptor). There are other brands - although I like the fact that rear-wheel trainers are easier to swap bikes, which is why I haven’t moved on. I might have to, to get accuracy.

It seems some people are getting 1% out of their snaps.

But life’s a compromise, no? I am sad my Kickr Snap doesn’t work in full sunlight because I thought it would be nice to set up outside on the deck. But the Snap uses a photo sensor and despite me taking it apart and re-taping it, it doesn’t work in bright light. So it goes, I’ve given up on riding on the deck between about 10a and noon when the sunlight is wrong. That’s a small thing :-).

If the issue is the difference between rear wheel measure and crank/pedal measure loss, then we (you, the community) need to pick one (I would pick crank/pedal) and if you have a wheel measure, you need to find a way to calibrate back to pedal measure. In my case, I now have a pedal-based measure for my road bike, so I could set up a calibration and figure out an accurate loss value (move the pedals over to my training bike, take the measurement, enter it as a tuning parameter, move the pedals back).

Basically I need to be pretty sure I’m getting to 1%ish, or I’m junking this thing.

@Coach.Andy.T I read from a pretty knowledgable source recently (GPLama) that the Kickr power model factors in drivetrain loss, which would typically be in the order of 3-4%. Maybe your Wahoo techs could confirm if this is true or not?

For me it is not clear whether the Kickr is reporting power at the hub or at the crank with it’s power modelling. If GPLama’s info is correct then it should be reporting power at the crank (using a generic drivetrain loss factor).

1 Like

Yes, they should both measure the same power within a couple of percent. Whether or not you happen to produce more power riding outdoors vs indoors is totally irrelevant to the issue of your Quarq and Kickr measuring 13% different on average. Even if the Kickr doesn’t take into account drivetrain loss (see my other post above asking for clarification on that) that would only account for 3-4% of the discrepancy you are seeing. Basically either your Quarq or Kickr is out of spec.

2 Likes

When people say “your outdoor Watts are higher” they just mean that, for various reasons, you may be able to produce more power outdoors for the same perceived effort. It doesn’t mean the value of a Watt actually changes. Only your ability to produce them.

FWIW I don’t personally see much of a difference between my indoor and outdoor sustained seated power.

2 Likes

I know what they mean, I don’t believe it. I do find I roll at a higher perceived exertion, so I buy that, but for a given PE and heart rate, I think the watts should be about the same. Every exercise day is different, fatigue, yadda yadda, but not consistently different.

1 Like

Interesting thread, I understand and really sympathise with these issues. I don’t understand why adding power matching is no longer a priority in Systm because that would surely resolve much of this type of issue.

I do hope Wahoo have something up their sleeve because otherwise I fear there will be much upset as and when they roll out the outdoor training features.

e.g. My 15 year old Tacx trainer works consistently but I am 99% sure that the power output as shown is incorrect. It’s consistent so I don’t care, I just know that when it says (say) 150Watts that is not the power I’m pushing out.

I’m hopefully moving to a crank based meter on my outdoors bike in the spring and I’m hoping the power it gives will be “true” or at least “closer to correct”, but compared to the trainer I am expecting to consistently see a different wattage for the same output from me. I am sure this will be true for many people to a greater or lesser extent.

If not by powermatch, I wonder how this will be resolved. Will we need separate 4DP numbers per “location” and Systm will automatically change the targets between inside & outside sessions, but then you would need some method (outdoor ramp test?) to calibrate the difference. Perhaps Wahoo have something else very clever lined up?

I don’t use TrainerRoad but I think they have a “move this workout outside” feature so I assume they have found a way of asking for consistent effort levels from the rider across multiple power meters / bike setups?

Anyway, apologies for a bit of an off-topic ramble but I am interested to see what happens in this area when outdoor training is added to Systm.

1 Like

It would still be recording at the hub, I’m not sure there is a compensatory factor in there for drivetrain losses as that’s variable and dependant on several factors such as chain/cassette wear as well as cleanliness of it

1 Like

I’m not sure how power matching would/should work. But most important is an accurate power meter that can be trusted. If power matching could be used to make wrong measurements right, this would only make things more complex and less trusty I guess. I’d be happy with an accurate hub PM that has a discrepancy with a spider PM because of things like the drivetrain and would therefore just add a percentage myself to match my real power.

1 Like

An other thing to consider maybe is what PM method would be most valuable for a cyclist. A pedal PM might be most accurate, but no matter how much power I have in my legs, if this power is not transferred to the road it’s kind of useless. This is why we want stiff bikes for climbing for example and a very clean drivetrain, etc. It’s the same for cars, you can have a very strong motor, but if that power is not transferred to the road, it’s pretty useless.
This is a bit off topic, sorry for that. My priority for now is my confidence in the Kickr PM. :slightly_smiling_face:

1 Like

Yes, it would have to be a generic estimated loss (typical drivetrain efficiency is around 96-97%), but GPLama was led to believe by the Wahoo guys that “some” drivetrain loss is built into the Kickr power model. It would be interesting to know if this is true or not. Looking at DC Rainmaker’s comprehensive review of the Kickr V5 it would appear that there probably is some attempt at drivetrain compensation because he only saw a couple of Watts difference between his Quarq DZero and the Kickr, although the Kickr was slightly on the lower side as you might expect. But the difference was only around 1.5%. Certainly nothing like the difference the OP is seeing here.

We’re talking here about a difference in the order of 3-4%. So I don’t think it’s at all critical, but I think it is quite important to at least know where the power is being quoted (i.e. hub or crank). For physical power meters like pedals, cranks, spiders and wheel hubs, it’s self-explanatory. But for trainer power, which is often modelled and not actually measured directly by a physical power meter, then it is not quite so obvious. Hence my question about how the Kickr power model is derived.

Where this becomes more critical is in high-level virtual racing, where that 3-4% power difference between crank and hub suddenly becomes important!