Power discrepancy between smart trainer and Garmin Vector
 

First time I'm using my Vector 2s power meter on my Elite Rampa smart trainer. I'm using TrainerRoad and measuring 175w, but my pedals are only registering around 125w. Is it normal for this much discrepancy between a smart trainer and an actual power meter? Or is my power meter not reading properly?

It shouldn't be that big of a difference.

Did you make sure both were calibrated correctly? My only other thought would be that the 2s is single sided, perhaps it's measuring your weaker leg and averaging that, giving the lower reading.

I have seen differences even larger than that (Tacx Genius compared to Powertap C1.) I completely ignore whatever numbers the trainer spits out.

Oh dear. That's a large difference. At least one, and possibly both, of your devices is providing misleading information. There may be ways to determine which one(s) but it can be a fair amount of trouble. Those are expensive devices to have this problem. My condolences.

Some people claim that power measuring devices don't need to be accurate, they only need to be consistent. This is one of the reasons why they're wrong.

I don't bother using the reported power off of either of my trainers (Kickr Snap and Kickr). Both consistently read 5-10% higher than my P1 or G3 (which I suppose is within the spec range). It's almost comical. On the Snap in erg mode, I can be doing 100w, then the next interval is at 250w and the reported power instantly jumps to a perfect 250w before the resistance even kicks in. Which makes me think that the Snap just reports target power.

I had the same issue recently as well. Vector 2 user, and I was testing a CycleOps Magnus smart trainer at a shop where I work. The trainer consistently spit out numbers ~40w higher than my Vector. Now, since the Magnus is a wheel-on smart trainer, tire rolling resistance could be a variable.

Is it therefore correct to say that the power numbers from smart trainers, even $1200 smart trainers, are mostly inaccurate?

Anyone know if the Cycleops Hammer is as far off as the Kickr and others?

Cycle ops is made by the power tap folks so I'm hopeful it will be reasonably accurate but for my needs it does indeed only need to consistent. When I use "x" force I don't care if the reported number is correct as long as it's repeatable each time I use x force. Others may need to know that number matches a known true output... I have two Hammers on order but no way to calibrate them anyway.

Where did you order from? ETA as far as shipping?

I'm about to move to a much colder area and won't be able to ride year round like I do now, let us know how the Hammer serves.

Hmmm. I don't think it's correct to say that. There is a chance (low probability, but non-zero) that both of the OP's devices are reading as intended.

What if both the OP's devices were reading consistently (as is possible)? What would you advise?

Lbs
 

I ordered from my local bike store. They have just become available to order

I would advise simply not worrying about it. When you are off the pedals however you will have the pedals for reference so when you want to do a FTP type ride outside it is your pedals that you will be listening to so remembering that number will be more important than remembering the numbers from a smart trainer . It would be interesting to note power ratings when riding with somebody that is the same weight as you. If they read the same power as you do with your Garman power meters then you would know that they were reasonably accurate . That assumes that you were on at least similar bikes .

I have no advice, but I'm curious - how does the discrepancy change with output? Is there a fairly consistent 50w difference or is there a consistent % difference or is it completely willy-nilly? It'd be interesting to see a graph comparing the readings as you exert different amounts of effort.

I don't think that is necesarily the case. What seems to be correct to say is that different power measuring devices (of any kind) tend to result in different power numbers, which makes using several of them interchangeably a bit of a headache.

That is all well and good, until you decide to use another device for measuring power, then everything goes out of the window. That is why I suggested to the OP to use his pedals and ignore the trainer, because he can use the pedals both indoors and outdoors. If you get a smart trainer and it is consistent, then it will serve you well. Until you decide to buy a power meter to use outside, then if the numbers don't match you will have some decisions to make. It is just the unfortunate nature of the current power meter market.

I am not the OP but in my case the numbers were all over the place. If it was a consistent difference then it would be really easy to account for it, but I don't think that is ever the case.

To follow the device that he can use on every ride, and that is the pedals.

The load on "smart" trainers typically isn't terribly smart. On most consumer-priced smart trainers the load is varied by an electronic load generator that moves magnets closer or farther away from a rotor to change the level of resistance for the same roller speed. However, these units do have limitations on their operating range, and how quickly they can vary the load. Most trainers will tell you the maximum power they can dissipate; what they don't often tell you is the maximal load they can generate at very low roller speed. You may know that on the road you can generate equal amounts of power either with high cadence and low pedal force, or low cadence and high pedal force. A similar thing happens on trainers -- you can generate the same power with low roller speed and high load or high roller speed and low load. So different trainers can have different operating ranges for roller speed and load -- and that's a long explanation why the error can vary with power level.

Here's a plot of some measurements I made comparing my electronically-controlled ergometer and my (accuracy validated) PT hub. As you can see, the error varied with roller speed. The black dots show the "default" settings, the red and green dots show how the readings varied with a different ergometer scaling constant. Subsequent to this, I ignored the ergometer mode on that trainer.

http://anonymous.coward.free.fr/rbr/tacx_correction.png

Measuring what's through a trainer is somewhat akin to measuring calories. Both our interpretations of data and not a direct measurement. When you have a power tap or other system installed on your bike you were getting a direct measurement of forces like strain gages and therefore it can be reasonably accurate. I would always put my money on an actual power meter versus a smart trainer

I wouldn't necesarily. A direct drive trainer could measure power as well as a hub power meter. The problem is, you can't take your direct drive trainer outdoors. I'd recommend trying to use the same device for all your measurements as much as possible.

Unless you're willing to invest on the really good stuff of course, then you can have different PMs in different bikes and be failry confident of both accuracy and consistency. Need deep pockets though.

A general question on pedal power meters: is the direction of the force on the pedal taken into account? Let's say the rider is pushing vertically at 100 pounds on the right pedal. At the 3 o'clock pedal position, all of that 100 pounds (100 X sin (90 degrees) = 100%) is in line with the crank circle, hence used to power the bike. But if that rider keeps pushing straight down at the 4:30 position of the pedal, now only 70.7% (100 X sin (135 degrees) is in line with the crank position. If he is still pushing straight down at the bottom of the stroke, he is creating no useful power at all. 0%. But a dump load cell on the pedal itself would just say the rider is putting out 100% power at all three of those pedal locations.

Depending on what one is looking for, this could make a big difference. If you just want to be able to boast of your 400W output, the dumb unit works just fine. Your body will feel like it it putting out real power (and benefit from it). But if you are trying to go fast on the bike, it matters a lot. Only the useful power is felt by the drivechain and rear tire.

Seems to me that unless the unit has a way of filtering the pedal force to only read the force in the direction of the crank circle (tangential force vs radial force), it is being deceptively optimistic. Anyone know if the force direction is taken into account and if so, how?

Ben

Well after a long road ride yesterday and another trainer ride today, I think it is the Elite MyEtraining software that's been causing the problem. I'm supposed to use it to calibrate the trainer, and a few weeks they did an app update and it now doesn't appear to function properly.

I kept the same power zones from my ftp test on the trainer for my 3hr road ride yesterday and the numbers felt appropriate. Trainer ride today was utilizing my pedals instead of the trainer power and everything was spot on. I'm going to put in a ticket with Elite support to see if there's a way to fix the app/trainer calibration.

The Vectors are instrumented with a lot (I don't quite remember the number) of piezoresistive load sensors arrayed around the pedal spindle. The rider "zeroes" the pedal when it's vertical so the sensors can tell whether the pedal is vertical or horizontal or anywhere in between. In theory, they could use this information to refine their power calculation (they have the force but they assume constant rotational velocity -- if they processed the position data, they could improve on the assumption of constant rotational velocity).

Yes the direction of force is taken into account. That's why Vectors require you to "set the install angles" before you can use them. You get on the bike and pedal at 80 to 90 rpm for a moment and the system uses its accelerometers to figure out how the strain gauges are oriented so it can do exactly what @79pmooney asks about.

That undermines the idea that measured kJs = kCals, slightly. If your pedaling efficiency is terrible and you're pushing straight down at 4:30, then the power meter can only see 70 % of the energy you're putting out, and a good HRM system is actually more accurate. Guess that's a big if.

Well sure you could burn way more calories than you produce in work if you insist on maintaining terrible form.

Which is why good coaches talk a fair bit about efficiency, form, technique, etc.

Loose shoulders, bent elbows, still upper body, pedal full circles, focus on smooth high cadence, etc.