But with ERG smoothing on, it’s just the number it tells you right? I take it the underlying power is still accurate to the second?
From that second video it seems not. Look at the comparison between the trainer and the Vector pedals for the same intervals.
Was more thinking about what TR registers. Like he says, what you’re doing on the bike doesn’t really change.
- Yes, absolutely this is the case. Factors like the type of resistance unit in use, the flywheel weight and rpm, and internal electronics that control the whole system all play a part. Generally speaking, the more expensive trainers respond faster than lower priced one, but that is not the whole picture.
-
I have the H2 and know it responds reasonably quickly. That is especially true when using the small ring on a road bike, because it keeps the flywheel slower and more easily controlled by the resistance unit vs a higher flywheel speed as in the big ring on a road bike.
-
Can you be more specific about exactly when you are feeling it takes 10 seconds to respond and adjust?
- I ask because the entry to the work intervals on your workout above looks basically correct and quick.
-
The only place I see a notable lag is at the end of a work interval. This is a problem for nearly every trainer, because the situation creates it as a direct matter of the circumstances. You are pushing at 90 rpm which drives the flywheel and you are providing the proper force to hit the power target… then that the trainer drops that force in a moment (1-2 seconds in many cases) which means the flywheel now has very low resistance, which allows it to spin almost completely free. That is the “negative” moment of having a large and effective flywheel.
-
It is physics at play, in conjunction with the simple “drop” in resistance as the trainer and app are following the prescribed workout. That inevitably results in a “power drop” which some see as a problem. I don’t love it, but also don’t see it as very detrimental to the workout or intended training effect in most cases.
-
I have a rather convoluted way to handle it in the cases where I want to minimize that delta, and can share it if you are interested. But it is largely a pain, even when done in the best way possible via shortcut keys on a keyboard (Win/Mac only).
-
I am not sure exactly how the Wahoo app adjustment or smoothing compares to the TR app specifically. I’ve never used the Wahoo one for ERG control, so don’t know how it compares.
-
Generally speaking, that spin up that Shane does is reasonably “gradual” to me in that he is picking up around 3rpm per second (total swag) and the trainer is adjusting. Without seeing it graphed on TR, I can’t say what the Wahoo app is doing. Considering their penchant for overly smoothed data application, I expect there is a notable amount of smoothing applied.
-
Also, if my poor memory serves correctly, your power data is actually coming from a power meter (Assioma pedals?). If that is right, you will see WAY more power data fluctuation from that fact alone, regardless of the trainer in use. Measuring data from the pedals is the most volatile location that is subject to the largest and fastest swings of any power measurement option.
-
I am betting that in Shane’s video above, he is pulling power directly from the trainer. As such, the data will be “smoothed” purely from the location the power data is captured (after the wheel drive hub specifically for a wheel-off trainer) AND the fact that the power data measurement method used in most trainers has some inherent lag in registering actual power input swings.
-
Essentially, you are likely comparing apples and oranges here with respect to what you see vs what you see in Shane’s video. Does that make sense?
Lots of responses! I’ll cover both groups…
-
I’m using the small ring with a cog in the middle of the cassette.
-
speed of response: this is at the start of intervals, it might be the “road feel” but I’ve probably turned the pedals five or six revolutions before target power feels like it’s reached its intended value.
-
I understand the physics behind the “overrun” at the end of each interval - basically angular momentum and the heavier the flywheel the longer it takes the system to spin down.
-
I’ve a left side only 4iiii crank based PM. I assume TR take a power meter’s data if it’s available in preference to that reported by the trainer. I had PowerMatch turned on.
For my next workout, my wife’s turn today, I’ll turn PowerMatch off and see how that affects things. It’s a hard start, steady state workout so will give me a chance to keep my cadence in check (!) I suspect this is just a learning curve combined with a misunderstanding of how Erg mode behaves.
1 Like
Do keep in mind that switching to the trainer power data may yield different results. Despite claims from makers, we frequently see wide differences in data values between meters and trainers. So, it may be fine and close in your case, but be aware that if it feels off, it may well be due to deltas between devices.
Testing without PowerMatch is worthwhile, because it simplifies the process and can reduce adjustment lag to a notable degree. So it is surely worth a test, just keep power feel in mind.
In my experience both inside and outside it takes a lot of control to keep pressure on and hit targets. It is possible but takes control.
Here are some 30/30-sec intervals after 2 hours of riding outside:
so easy to take too much pressure off the pedals and lose control when going from on to off. Also had trouble controlling the switch from off to on.
Inside here is one of the best example of control on some 30/15-sec intervals (Brasted) on the Kickr with PowerMatch:
I’m actively pedaling and mentally engaged - that took a lot of concentration and control. Definitely not “I’ll just pedaling along at constant cadence and let erg take control” in fact the opposite.
I have examples of the opposite where I wasn’t in control.
Outside you can work on control, at one extreme are sprints with controlled recovery like this hit and miss effort:
7-sec all-out sprints with 1100+ watt peak power and I fought hard to keep spinning with some pressure at the end of the first sprint, but clearly gave up on the second one.
My point is - controlling power transition from to off-to-on and on-to-off is a skill. It can be done in Erg but its a bit weird and different than outside. IMHO you must practice outside, and inside using Resistance or Level modes. Over-unders are a good fit for working on this outside, or inside in resistance/level mode (or sim in Zwift).
Of course if you have a Kickr, turn on Power Smoothing and fool yourself into thinking you are a hero, the king 
of power control.
OK, take 2…
https://www.trainerroad.com/app/career/bobw/rides/121717504-warscheneck
Much better cadence control now I’ve a better idea of what needs to be done plus an understanding of how the trainer’s “road feel” affects things.
4 Likes
Great looking ride, you nailed it! 
Looks like any good controlled effort (inside or outside) although I thought your Bluebell workout looked fine.
click for more blah blah blah variable power is normal & stop focusing on it blah blah blah
Using Erg on a Kickr, the conditions are so ideal that it is easy to see power graph differences between these scenarios:
- not paying attention to pedaling and sloppily pushing over the top
- paying some attention to pedaling and not being sloppy
- actively focusing on a smooth and controlled pedal stroke
- actively modulating power during on/off transitions
- big vs little chainring, with little chainring for me being unnaturally smooth, something that can’t be observed outside even in the best of conditions
Although the same can be said of outside under good conditions (flat road, consistent wind speed/direction).
It is simply easier to see “erratic vs smooth” inside, and then start focusing on how to make it smoother and less erratic. Rather than questioning if erratic is normal and not a concern.
There is no evidence I’ve been able to find that using Erg to more precisely hit target wattages is superior to either well executed outside intervals, or well executed indoor intervals using resistance/level modes. Power variations while racing or riding or training are normal and expected. Zones are approximate as is the estimate (FTP) you use to set the zone targets. And then there are minor variations in fitness.
My wife and I are sharing the trainer so using Erg mode and being able to use a different cog from my wife means we don’t get misaligned chain and cog wear. Erg mode also means I don’t have to “search” for the right gear combination at a reasonable cadence for a particular power output, always a problem on wheel-on trainers.
With a Saris H3 you should be getting all those benefits in either erg or level/resistance modes. I’m assuming H3 offers a level (or fluid trainer) mode. This is what I have on the Kickr, in addition to sim, erg, and resistance modes:
and level = 2 at ~250W FTP only requires 2 gears for most workouts - smaller gear for recovery interval, and larger gear for work interval. Its surprisingly similar to outside and I can sprint and hit 1000+ watts without shifting
I’ve recently started using an H3, and while I love it, I found some interesting quirks on today’s VO2 Max effort. I can appreciate that there’s going to be a bit of a lag between spin up and spin down, but there were a few efforts where the trainer took close to +/- 10 seconds to adapt (usually following the hard effort). This didn’t happen every time (just enough to make things interesting), and I couldn’t figure out any specific pattern.
My approach is to ramp up my cadence over 100 about 5 seconds before the interval starts, and this generally worked well.
The other thing I’ve found is that, regardless of ERG mode, the large flywheel on the H3 really makes riding at low power (60-90 watts) incredibly difficult… I feel like I’m free spinning and actually find I engage muscles trying to actively control the spin (when I should be recovering) - anyone else find this to be the case when riding on a wheel-off trainer?
-
60ish watts is quite low. I end up with many of my recovery’s around the 90w level, and the flywheel effect can seem a bit out of place. One consideration would be the gearing in use. If you are using a medium to higher gear, that will make the effect even more prominent. It may be appropriate to have that gearing for the work intervals, but it may have an undesired impact in recovery as you seem to be having.
-
Some gearing experimentation may be worthwhile. Possible swap for all use, or even consider using lower gear for recoveries vs higher for work. This adds complexity and will lead to more deviation depending on how you apply it, so perhaps not ideal.
- I do the opposite. If I alter my cadence pre-interval at all, I slow my cadence 5rpm about 10secs early… and then I jump hard the second before the “interval” since TR sends the trainer resistance change instruction early. Doing this kicks up the power level quicker because I am effectively entering the interval with more “brake effect” from the resistance unit due to the slower cadence.
1 Like