Dylan & Peak Torque explore cycling suspension & cognitive dissonance

I linked the video. Give it a watch it’s well worth it. Even just to pop over there and give it a thumbs up. Peak Torque did some good work, here, let’s give him a little support.

3 Likes

What about his cadence seated compared to standing? Does he do each run at the same cadence? An interesting phenomena coming from the dark world of indoor racing is that many are turning to low cadence standing efforts to maximise power and minimise work.

If you want to measure drive chain efficiency, you’d need a pedal based and crank based power meter on the same bike.

Too many variables to come to any conclusion.

There is no free lunch. Lower cadence requires higher force to get the same power. And standing requires more overall muscle contributing to the support of the rider (as seen in the HR aspect of the PT vid).

All that to say, low cadence and standing are not some hidden cheat.

3 Likes

That’s exactly the way Frank explained it to me. :smiley: I did not want to hear it. He was correct, though.

Standing is less efficient than seated. I think that’s been well accepted for years now. Although I’ve heard mike woods say on Zwift it’s an advantage for him to stand. There’s no aerodynamic penalty for doing it on Zwift.

2 Likes

There may be some potential advantage. I’d like to hear specifically what that may be. I agree it drops an areo disadvantage that happens outside.

But I take issue with the specific aspect mentioned above, “maximise power and minimise work” because it is essentially impossible with respect to what we know about what power is.

1 Like

I’m guessing he was referring to the Zwift racing cheat due to their algorithms that kept power up even after the initial surge. That’s a topic for a different post though.

1 Like

It actually does maximise power and minimise work. There’s loads of evidence about low cadence leading to less metabolic demand at a specific power. And there’s no aero penalty for standing indoors. As I understand, the core is engaged in a different way too.

I’ve experimented with these techniques, and they have been widely used in the Zwift Premier League. In race 4, one of the riders had an average cadence of 56 for the entire race, standing most of the time. During an A cat Zwift race, I can now get recovery by dropping my cadence. I’ve managed to get my HR down to 108 in a race with NP of 311W. On the Bologna climb, I’ve had an indoor power PB for the 7 minutes on the hill of almost 6w/kg standing the whole way and low cadence. The couple of weeks before at a regular cadence, for a similar RPE I was only able to achieve 5.3w/kg at a similar HR and RPE. It’s absolutely a different technique that does work indoors.

1 Like

No, that’s something completely different. And very much cheating.

2 Likes

I stand corrected. And I agree!

1 Like

Maybe! But if you watch the virtual tour de france you’ll see both images of Woods riding out of the saddle for extended periods of time & you’ll be able to watch his cadence oscillate between the 70’s and low 80’s for the duration of a climb. So I took that to mean he’s just out of the saddle pretty much all the time on those ‘climbs’. I’m thinking stage 2 and stage 5 of the vTdF.

Later on he talked about this on a podcast & said that 1.) He felt like he was able to stand and ride better than most because of his training as a world elite runner 2.) He felt like on Zwift that was an even greater advantage because there was no aero penalty for doing it.

Watch stage 5 starting at about the 1:34 minute mark where Woods drops Pozzo. He’s out of the saddle the whole time. 5 minutes later they flash back to Woods and he’d doing 65rpm & still riding out of the saddle. Crazy. 400+ watts.

1 Like

Yeah, definitely, those are 1st order effects. Shock locked out or not? Tertiary.

:rofl: @iamholland Easy, now! I kind of sympathize with PT on this one…it’s a lot of work to take data like this. I was entertained by the video presentation of the data & it’s fun to talk about it on forums like this. So I don’t want to pick apart methodology TOO much.

I’ve spent a lot of time taking aero data. I spent hours & hours characterizing my orthogonal strain gauge array & optical torque sensor. Things like that are never done. There is always one more little thing you could have done…and when people talk to you about your results everybody thinks you should have done this or that. :slight_smile: Pfft.

If you’ve got a bike with a shock and a temp gun, share your results! I think it’s a neat idea. But I don’t want to be down on PT because he didn’t do it. Heck, I think he should get a VO2master and retake all the data while wearing it. Now that would be a video worth watching!

2 Likes

Any change in drivechain efficiency (pedals to wheels) due to the suspension moving is going to be minimal. The real issue is how much energy it takes to put the same amount of power into the pedals when your shocks are turning your effort into heat.

2 Likes

The way that shocks are turning your energy into heat is through the drivetrain.

I think dylan misses the point and peak torque understands it but doesn’t explain it too well.

Power is a measurement that is measured at the crank so it is only measuring mechanical power at this point in the system. So the only things that will have an effect on the timed runs will be downstream of this point and external factors: drivetrain efficiency, rolling resistance, aerodynamic drag (including wind, atmospheric pressure, humidity), distance travelled (line choice) and weight. If you keep all these varibles the same the times will be the same. Things that won’t have an effect will be human movement efficiency and frame efficency (where the spring dampers are placed).

Peak torque refers to metbolic cost but we can also consider this to be power production of the human body from stored chemical energy, this is probably only about 25% max efficiency to human mechanical power since the majority is lost as heat.

But this is human mechanical power (overall movement) without movement efficiency and frame efficiency so if we assume this to be at 95% efficiency then you are producing 316W of human mechanical power to provide 300W to the crank.

So if you are producing 300W at the crank then you are actually probably actually producing 1264W with 948W lost as heat from the skin breath etc.

Now add a spring and damper into the system and say you can feel the heat in the damper this could be like 6W mechanical power lost prior to crank so like a 2% increase in the overall power production. So now you are still producing 300W at the power meter but your body is producing 322W human mechanical power and 1288W overall. A 2% increase might feel like a lot if you are at the limit in a race.

Sorry for the long answer, I kinda got carried away. Also if there are any other engineers or physicists or nerds out there I’m sure they can correct me if I’m wrong or refine any of the numbers I’ve made assumptions on.

4 Likes

Overall, I agree with what you are saying, however you are missing a key point here and that is effect of the drivetrain on the suspension system. That is a key component of the equation and one which can be measured…if the drivetrain is causing loss of energy through suspension activation, then the times for a given run should be slower. In theory, locking out a system would eliminate those forces….but the data seems to be indicating that it is not a factor.

Also worth noting that using HR as a proxy for metabolic costs, best option for real world testing, is probably not terribly accurate. After doing 12 runs, one would expect HR to rise with each subsequent effort. Think about when you do numerous intervals on a TR workout…by the time you get to the last one, your average HR is higher than it was for the first one, even though the power for each interval remained the same.

Net-net: the data here is interesting, but there is a LOT more testing required to back up the hypothesis.

2 Likes

Agreed….or it simply isn’t enough consumption to make a meaningful difference in time, similar to how little difference weight makes in climbing times.

2 Likes

I don’t see how the drivetrain activation a horizonal force could compress the suspension system in any non negligible way. There is effectively a very small lever when it is loaded with a rider. The movement you see of the suspension compressing in the peak torque video is from his weight shifting up and down vertically when pedaling but this is wasted prior to the power meter.

I have never designed a bike but I imagine you would want to minimise this lever to almost a ridgid bar perpendicular to the radius of the contact point with the ground to the rear hub. Hence minimal wasted power. However there are probably other points in suspension geometry that it is probably a compromise with.

You could get a PHD on it but it would be surprised if they found out anything not already known about by the engineers who design these bikes.

Perhaps the lack of lockout, when the spring unloads, is actually propelling the rider up the hill? This sort of makes sense, and why lockouts (at least for me) only seem to help on the flat smooth areas, where you don’t want to go up, just forward.

1 Like

Designing suspension systems that are not negatively impacted by pedaling forces is a complicated task. There are a multitude of forces that need to be factored in, and ultimately balanced against each other. Anti-squat, system responsiveness, small bump compliance, brake dive….addressing or minimizing one can exacerbate another.

The challenge was further complicated due to patent issues as some of the most efficient systems were aggressively protected back in the day (improperly so in many cases).

But suspension designs have significantly improved over the years in terms of efficiency….it is nearly impossible to buy a bad suspension design these days. IMO, some of this data re: locked vs. unlocked suspension is the result of those improvements.

But the data we are talking about indicates that lockouts don’t help on flat, smooth areas.