Carbonfiberboy 

My experience is that there will never be any definitive plot like you want. Smoothness depends more on what the rider's trying to do than on cadence. OTOH if what the rider is trying to do is to be ultimately smooth and that's all, the rider would pedal on rollers in a small gear at very high cadences, say 120+. That would be as smooth as one could get. Trying to be that smooth at say 70 cadence would be much more difficult. OTOH, pedaling seated in a very large gear against high resistance at say max effort at 40 cadence - pedaling like that would also be fairly smooth. One attempts to get smoother when one is trying to do something difficult which depends on smoothness. Last week's ride involved a short but steep climb up a gravel track heavily covered in wet leaves. I was trying to be very smooth and made it with no wheel spin. So I doubt there's any correlation between cadence and smooth, per se.

We suspect there's a correlation between smoothness and crank inertial load which causes riders to be smoother on a climb than on the flat though climbing at a lower cadence than they used on the flat.

Hermes

I have Vector 2 dual. I checked my Garmin Connect file but there is no smoothness plot generated. This may be due to having an older Garmin head unit. According to Garmin, Version 2.4 software should generate a smoothness plot. I have a new Garmin 820. The next time I use the Vector pedals, it will be with the Garmin 820. According to Garmin, I should be able to turn on the smoothness function via the head unit from the sensor details page. I assume that will generate a smoothness graph in Garmin Connect.

What I have observed with Garmin Vector 2 is that the transient response with the Garmin 800 and 810 head unit is very slow. Cadence is slow to respond for fast cadence changes at the track. However, I will post up a file if I can make my stuff work. We can then revisit the validity of the graph.

Carbonfiberboy 

I did an experiment leading a group of the same riders every week on SS bikes, some fixed mixed in. We rode hilly rides of about 60 miles, pedaling maybe 40 on 10% climbs, pacelining at 110-130. When we switched back to doing similar rides on geared bikes we found it hadn't made the slightest difference. Everything was just as hard at it'd always been and we were no faster. I don't know about smoother. Maybe. Those group rides were very difficult and we all thought we were killing it. Not so. Principle of specificity. We did get better at riding fixed and SS but it didn't translate. It took a long time for the pros to figure that out, too. When training, the hardest thing to figure out is what did what because one can't do 2 things at once. Even alternating them on alternating years doesn't prove anything because we're all different from year to year anyway.

The other thing that's a bit difficult for a freewheel-only rider is to start riding fixed and not get hurt by simply doing what's instinctive for a freewheel rider. So what you say about not switching between bikes is a good idea. The fixed bike should have normal brakes too, so as not to wind up under a cement truck as has happened here. You maybe should post a little how-to-do-it along with the suggestion.

Hermes

As promised, here is data from last night's track workout at Velo Sports Center in Carson, CA on the 250 meter indoor wooden track. I used my Garmin Vector 2 power pedals and my Garmin 820 head unit with 1 second recording mounted on my track bike - fixed gear no brakes. When I set up the pedals on my track bike the Garmin asked to set angles, set crank length and calibrate which I did. I clicked on power sensor details and turned on smoothing.

The session included a 60 lap warmup with the last 20 laps a motor burnout and then 2x20 minute motor paced efforts in a 15 rider pace line with each rider taking 2 laps on the motor before pulling up and going to the end. However, in set one, when a rider got to the end of the line, we were supposed to launch an attack from the back and gain 1/2 lap on the motor before swinging up. We each did that twice. Speed was approximately 25 mph. The next set was the same without the attacks but the speed was 26.5 mph.

HR is blue, cadence is light blue and power is brown on the first two graphs.

Overview of the workout from Cycling Analytics


I added the second interval to show how variant power is on the track.



Left right balance



Torque Effectiveness.



Pedal Smoothness



Platform Centre Offset



Power Phase



Peak Power Phase

srode1

Is smoothing an option on the app or from your Garmin? I haven't seen that option yet for the Vector 3s, maybe looking in the wrong spot?

Hermes

Smoothing option is selected from the Garmin head unit by selecting power sensor details.

Carbonfiberboy 

@Hermes - that's all very cool. Since you were on the track, we know that power and cadence are directly related, so more cadence is smoother, right? Thus the OP is answered, at least in the case of one rider. That was my conjecture from my experiences with very fast pedaling, but of course no data.

Hermes

The track is more complicated due to geometry. Velo Sports Center is 250 meter but it has a longer straight and sharper turns than most 250 meter tracks. And the shape of the track is a bowl. So if I ride in the pole lane at the bottom of the track, the surface is banked but flat. It is true that there is a slight difference between turns one and two and turns 3 and 4. Even in the pole lane when going through a turn, one will experience centrifugal force which will increase ones weight and frictional forces between the tire and the surface increases. At 35 mph, we pull 2 Gs for example.

As one rides higher on the track, there is an increase in elevation change into the turns that increases progressively the higher one rides such that riding at the top in the turns is very hard. When we did the drills, our coach sets the motor at 26.5 mph but half way up the track. So in the pace line, I rode constant cadence and speed but changes in elevation and increased frictional forces plus surges in the pace line after an exchange causes power fluctuations. For me, there is a 125 watt power difference between the straight and the turns at the same speed and cadence. And there are the attacks and exchanges which require more power.

Our lap times were 21 seconds, so every lap, I would get two power spikes from the turns and many times had to float on the straights. My average cadence was 101 rpm for the interval and constant.

I do not know how to interpret the data other than to say gee whiz that is pretty cool. So I do not know if OP got the answer to his question. Let's see what he thinks.

redlude97

interesting to see how biased your L/R balance is, have you ever noticed this in practice and have you every tried to adjust for it? I'm left leg dominant from an old knee injury on my right side. I can ride closer to 50:50 if I really focus on it but it seems harder to produce a given power especially above Z4.

Hermes

The data may be misleading. This is the 250 track where we are constantly adjusting pedal force on the banking and to react to surges in the pace line. Also, Garmin accelerometers are slow - at least by my observation so fast transient changes in force may be lost. So the imbalance may be worse or better.

Here is an expanded graph of my first acceleration that was intended to be a breakaway. The goal was to accelerate hard so as not to allow other racers to jump on my wheel. Inspecting the L/R balance, it is 49/51 at 550 watts and 112 rpm at 92.6 gear inches. BTW, the speed on the graph is wrong. It is more like 30 mph. If you look at the other new metrics, I blew the doors off of torque effectiveness and etc.



So high power seems okay but I still am suspicious of the data. I do not have a left right power issue.

Let's say I did. Right before our session, the para olympic guys were on the track. Did you ever see a one leg olympian cycle? He is effing fast. All the O2 in his blood is available to his remaining leg. And we know that it does not take a lot of strength to make a lot of power. Of course steady state and accelerations are two different matters.

Thunder Horse

It can take a long time to pedal smoothly, as in equal force over the whole revolution of the crank. May even take years.
I found several distinct stages of development. Each one seemed to be what I thought was 'smooth'. But later I experience an improvement, which then becomes my new understanding of smooth pedaling. In any case, each leg constantly has equal forces applying throughout the whole revolution of the crank arm, like stirring a large bowl of cream with a wooden spoon. Once the cadence, of equal forces, increases, one finds themselves floating on their saddle (not bouncing).

smackpotato

Thanks for the information. I found it very interesting