I haven’t the faintest idea how strong you are. You could be very strong but that kind of power is still unlikely. It could be hundreds of watts less and still be a great number that you should be proud of. The problem is simply not knowing.
1 Like
Why some of the more virtual elite races require two power sources and many race series now require a power meter.
This has been an issue for years…spin bikes way over estimate power to make you feel good. Spin bike owners trashed races or hung out with the lead group in A races (highest level). Some know and do not care and others refuse to believe their $300 spin bike is that inaccurate. Until they go ride IRL (in real life) and cannot climb their driveway.
Even Peloton bikes overly optimistic ($2K spin bike)…my wife cranked it up to 350 watts but can barely do 50 watts when I had a Kickr Snap.
1 Like
Cheapest way is to get a wheel on trainer used - Snap, Saris M2/3, Tacx, Elite - that has a power meter and some cheap bike. Throw on a slick tire and give it a go. $150 + $100 + $50…cheaper than power pedals.
Ok points taken, thanks Chris and everyone
3 Likes
Well it’s a different set of buyers to those who are going for a $5000 Kickr Bike.
The folks buying an entry level spin bike may not be aware of power meters and training with power, they are just looking for a bit of activity. They may not be aware that 2000w is track-sprinter levels of power, same as 500w for 30 minutes is dubious for most people - they probably don’t understand power at all.
Where the folks on the high end trainers with power meters are serious because they are athletes or racing. They need accurate power that matches up with their real bike which normally has a power meter on it.
Where the smart trainers are great is that you get the variable resistance when the route changes so it adds a lot more realism and enjoyment.
I see. But isn’t it a straightforward enough calculation it has to do for the power where it looks at the resistance and the rpms? Surely from those 2 variables something can be ascertained?
It’s not really the maths that’s the issue.
If you’re seeing really high power numbers then it’s likely to be wrong. Some would say obviously wrong. Chris Hoy power numbers are not obtained in a ‘oh, I never realised how strong my legs were’ fashion - unless you’ve been sleep walking to the gym and squatting 3x your bodyweight you most likely didn’t do 2kw of power.
As for the rest of the power numbers you get the only way you can really tell if they are good or not is by comparing them with something else that you either know to be good or believe to be good.
And the only reason we believe our trainers, perhaps, is because people like DC rainmaker do a lot of testing and comparing and wahoo et al have done a lot to work on the accuracy of their data because otherwise people notice and complain.
The reebok fr30 is not (as far as I’m aware) reviewed or tested by them, nor is it widely used - although they advertise it that you can connect to zwift - and looking at a picture it seems like a relatively cheap exercise bike. That doesn’t immediately suggest that they’ve spent a lot of time making sure any power data is accurate because I doubt 90% of the customers know or care that much about power.
In theory you know the resistance blah blah you can calculate power, but in practise it needs calibration and so on - it’s simple to measure the power incorrectly and increasingly more difficult to get closer to the right numbers. The latter usually increases the cost. Although these days power meters are much cheaper than back in the day.
If a Wahoo trainer reads 20 watts low DC rainmaker is writing an article about it and customers would be saying “it doesn’t match my power meter” - so they need to fret about it and they have to design something that’s as accurate as they claim - because 90% of their customer base is using something like zwift looking at that power data. I doubt that is the same for your exercise bike - but the bottom line is, aside from the really high figures I don’t know if your trainer gives accurate numbers or not, but I’d suspect more likely not.
Unless you want to race I wouldn’t let it spoil your enjoyment. If you get more into zwift and end up with a smart bike or direct drive trainer undoubtedly you’ll end up with something that you can be more confident gives accurate power figures.
2 Likes
I think it’s a glitch over the few seconds yes but then I’m wondering why they can’t produce a bike that has accurate calculations over the slower speeds/longer time periods for the work involved in doing 1 revolution in 1 second at a particular resistance, surely it’s a relatively simple equation that only degrades at very short time intervals/bursts? I guess now I’m wondering about the rest of the data and how inaccurate it is. I can imagine also that after a while these bikes degrade and so the numbers are even further off. Thankyou for your considered reply Michael.
It sorta would be that straightforward, but the problem is that Zwift can’t “see” or measure the resistance of your trainer.
The trainer measures the power and sends it to Zwift.
To do the calibration you just measure the force required at a particular resistance to do 1 revolution in a second or such and then work out the force at other resistances to get an equation that allows you to set it up to calculate the power at any resistance for 1 revolution in x seconds. It would be non-linear and at very small time sequences / high resistance it might not be accurate if there was significant variability in how the parts were manufactured and assembled or if they didnt spend longer than a few hours on the calculations. But they’d have to be clowns to screw it up badly. What I suspect is that after a while the bikes degrade which then results in substantial inaccuracy at any resistance which then people notice.
The upshot is there’s a bunch of different ways power is calculated.
Some of these methods are flawed to begin with - a good example here is something that claims to measure your body fat %age. It’s not a question of whether they measure accurately whatever they do but that their whole premise for measuring is flawed to begin with.
Some power measurement is like this because it’s trying to work out the power backwards so to speak and it either ignores or fudges many of the variables.
Other methods appear simple in concept. e.g most of the cheaper power meters people attach to their bikes have strain gauges so the simple idea is, you push on pedal, that force deforms the strain gauge that’s either in the pedal or on the crank. They measure that voltage change, they measure your cadence, they do some simple math and have your power figure.
And it’s all very easy in concept. But when you actually try to do (and not only have several large brands done it now, quite a lot of amateurs have attached strain gauges to bike parts with arduinos and so on and created power meters)
When you actually do it you realise that your data is wrong. Perhaps for multiple reasons. The most obvious for a strain gauge is that temperature affects the bike part and thus affects the strain gauge reading. So you either have to calibrate at the start of a ride and assume the temperature is constant, or you have to start measuring the temp and accounting for it.
And so on - and that’s why whatever method your trainer uses to measure power however simple that appears to be it’s only simple to get an inaccurate measure of power.
At that point they’ll test and test and change firmware and so on - and they’ll arrive at some figure for how accurate it is. 1%, 10% or whatever. Now I’ve searched and I can’t find a figure for your exercise bike in manual or in the specs on the sites selling it. To me that suggests they don’t think it’s very accurate. Unless it does come included with a different manual that claims a figure?
In fact there’s barely any references to power at all. Which makes me think they’ve stuck a screen on that just has one of the table lookups or formulas that people used to use for the ‘wheel on a roller’ dumb trainers to tell you what power you were supposedly doing at different speeds. It was basically a manufacturer supplied curve and they were not accurate.
Is it a simple process to get accurate power readings? Well yeah at this point you can just click ‘add to cart’ on a range of different products. But I’ll reiterate I don’t know how accurate your exercise bike is, but neither do you, the manufacturer doesn’t appear to tell you and we have some obviously inaccurate data. So I might be confidently incorrect but I think it’s not very accurate at all.
1 Like
I could be wrong, but I’m pretty sure the Reebok doesn’t have an actual power meter built into it, so it can’t measure the power. It probably reports and tells you watts, but this is probably a simple table lookup value that is built in, based on revolutions per minute and which resistance level you’ve put the trainer in.
1 Like
Yes interesting about other variables in power meters but in the case of this bike surely the force required to do 1 revolution per unit time does not change hardly even after using for half an hour, 1 hour etc. by being affected by temperature or deformation of any parts? It just needs to record the time for the revolution, it doesnt need to record anything else?
So then it is just a simple equation to show the watts that is shown on the bike? It would use a simple formula, quadratic or cubic rather than a lookup table as that would require storing all the values.
I’m trying to understand really how inaccurate the power figures are for when I’m doing 100 or 200 watts because if its significant then I’ll have to buy another bike but I don’t think it is. For the 2000W/5 seconds it doesnt matter to me and I can see how it can be a glitch.
Thanks.
You make it sound like creating an accurate power meter is simple but it really isn’t. Many companies have struggled with it. Even an industry giant like Shimano has been trying to produce one for years and they keep putting out ones with inaccurate data.
2 Likes
I think he’s suggesting that using an accurate power meter, why can’t a company build a spin bike and map the power to cadence and resistance levels settings and it then would be transmitting these tabled values accurately forever.
1 Like
indoor smart trainers measure power through a few different methods, and as far as i know only a 1st gen kickr (the 2016 version) used anything resembling a strain gauge in it. no comment on the accuracy since i dunno if they were actually good trainers. wahoo are still in business so it can’t have been that bad i guess.
the accuracy of an indoor trainer, and most of the higher end ones are pretty much good enough despite this honestly, depends wholly on how much money, time and attention was paid to that during the design process and the quality control… all of which drives the price up for the customer. though you can get a “good enough” direct drive for probably £400 now
1 Like
Yes I’m trying to understand in more detail because you can really easily in a lab with, even high school physics students, measure the force and therefore effort/power that it takes to push a wheel or disc round in unit time. You can then repeat that a few minutes later and get exactly the same result. You can set the thing running and unless temperature or pressure doesnt affect the experiment find that in an hour say its still taking the same force to push that wheel round. You then do that at a different intervals of resistance, plot the results and get the equation. That’s not difficult surely?
I think the issue will be that over time, and that could be as little as a few months of all the resistance changes and use that it’s not able to set the resistance accurately?
this is where it gets extremely difficult, basically. of course, these variables (among others in both an economic and an engineering sense) can be accounted and calibrated for and many established trainer and power meter producers do a very respectable job of it, but it’s easy to get it wrong and cut corners here
and depressingly common. power meters aren’t my field of expertise but i’ve seen enough companies do a bad job of it over the years to know it’s easy enough to do a terrible job of it
1 Like
Yes so after 10 minutes cycling on it it gets easier? How much difference do you think it would be once the thing has warmed up because I can’t tell easily. Thanks.
i don’t have any insight into how your equipment works, but a lot of purpose built smart trainers will automatically self calibrate at regular intervals while riding now… i think the kickr core is the cheapest trainer available currently that has this feature. not my personal trainer of choice but people seem to like it
i used to have a saris H3 and through dual recording i can tell you that thing wouldn’t read correctly for at least the first 30 minutes of every ride. that’s the only relevant anecdote i have to offer you
1 Like