There are free apps that can record power, like Wahoo I believe. I am not sure if Strava premium works with a power meter. I can try later to see though.
Mine shows weird left/right balance numbers when I brake with my feet. If the entire ride is pushing forward on the pedals, then left/right is normal.
Update:
Strava cannot connect to a powermeter.
Wahoo app can.
I overreacted a bit, a soft reset fixed my problem. My workout yesterday has no missing data.
I also imputed the missing power values from the previous 3 rides by averaging the adjacent values. I get power data consistent with my other efforts.
Anyway, it’s good to know that there is an application on smartphone that allows you to connect a power sensor.
I get some pretty funky values, potentially it is because of the amount of eccentric braking I am doing in hockey. Some sessions I will get 52-48% LR balance, others I will have 68-32%.
Are you using a power meter while playing hockey? If so, what do you want to get from the power data?
I’m largely interested in understanding our sport better. I want to know what is required for Matchplay at different levels. How many accelerations and eccentric decelerations we do in matches and how large they are.
It’s largely for my own interest of understanding my sport. I believe hockey may be quite different to a lot of other team sports as I think the eccentric braking load is comparatively pretty large.
Team sports on feet have eccentric loads while running but for stopping you can usually plant a foot at and angle, ice Hockey has a hockey stop. Bike polo use brakes. So I think we have a unique situation where we often slow ourselves down eccentrically before accelerating again.
Id also like to know how often I turn left and right in a match and see if it seems optimal, but may need an IMU for that.
Interesting. I played my first unicycle hockey match in a long while on Friday and my quads felt it the day after. I think you have a point about the eccentric braking.
I used the same power meter for the 10K, marathon and hill race.
First, I had no missing data on the 10k and marathon. But I had more than 10% on the hill race. I think I need to at least do soft resets from time to time… before I find another solution.
There was maybe a bit of unfavorable wind on the 10km, but the passage of the bridge and the numerous cyclists, forced me to burn some W in the brake. On the other hand, after crossing the bridge, I held 3.3km at 35.5 kph (264W). I probably spent more energy (for a lower average speed) because the effort was more irregular.
It is also possible that a larger development ratio would improve performance on the flat. Being able to run at 35 kph with a 120 rpm cadence would certainly be more comfortable, which would be a ratio between 1.7 and 1.8. Of course, in the more technical turns and climbs, it would be more difficult to keep up…
It was partly to have the power data from my races at UNICON that I invested in this power sensor. It’s interesting 
Interesting data Simon, what about using shorter cranks instead of longer ratio (or bigger wheel but 36 is already the maximum available) ?
On the 3 races and on my 2 unicycles, I used 145 mm cranks. This is the shortest crank length I’ve ridden in the last few months.
With shorter cranks, the motion is not the same, and it would be harder for me to produce as much power.
I already tried 137 mm cranks a few years ago and what is sure is that the gap is too big between 145 and 137 mm, and I think that in absolute terms 137 mm is too short for me. Maybe I should try 142 or 140 mm, but now I’ve been riding 145 mm for years. I don’t even want to try.
Wow, it is so great to see power numbers to compare with! If you are interested, here is some rough analysis I did with your numbers.
Road Climb estimates
I estimated your CRR 0.005, and CDA 0.42, to match up your climbing road race power and times.
For me on a 29er, I’m estimateing 0.004 CRR and CDA 0.40 to match my power and time.
Theoretical 36" Tire Performance
For fun, assuming the CRR/CDA are correct, and ther ewas skinny and thin 36 tire with similar CRR and CDA as my 29er, you would be more than 1 minute faster!
Of course, with even slight weight inaccuracies for each of us, such as +/-0.5kg for water, sweat, clothing, etc, that could be +/- 20 seconds difference and affect the CRR/CDA estimates…
I think the 36er is the best choice at your riding level though. Going to a 29er for the better tire and weight probably isn’t worth it since 51:00 would be about 130rpm and likely impact your power output. Though, you spin that fast for the 10k/marathon so maybe…??
Marathon
Using the same numbers for your marathon results, would give a schlumpf efficiency of 87% which seems way too low 
So maybe CDA is actually a bit higher (or other errors)
With my tested 95% schlumpf efficiency, CDA would be closer to 0.46
Conclusion
All of these are pretty rough guesses since we don’t know the exact weight, speed variations throughout the races, etc. But, CDA around 0.40-0.46 seems normal, so I think the power data is quite accurate and a good reference. Really glad to see your power meter working, and the numbers seem good. I always wonder if mine is working normally or if something might be off, but this is reassuring!
I found out my 3d printed platform adaptors had cracked after getting home.
They were making noises especially after riding the cross country course, but I thought it was just normal play in the fit.
The small rectangular blue/black pieces are pushed into the main plastic piece to keep it from sliding out.
However, around the rectangular hole, there are cracks on both pedals.
The left side (with blue part) has slid out a few mm.
The right side (with black part) has a thin visible crack, but the main piece has not slid out significantly yet.
Luckily, this should be an easy fix… I just need to make it a bit thicker there
Thank you for your analyses!
I’m not familiar enough with all these numbers, but I’m amazed at the total power difference between our 2 performances.
That’s a lot! I think that to improve the performance on the road, you would need a 36" tire with a width of 2" (I don’t think it should be narrower) and really slick.
This is off topic, but what rim and tire do you use? I’m planning to build a new G29 to travel in the Pyrenees with a hooked 25mm carbon gravel rim and a schwalbe g one speed 28*2" tire mounted in tubeless.
I don’t know if we can use my 10k and marathon rpm as a reference. I develop much less power than on the hill race.
In training, I could see that in my favorite segment (2.5km with a 1% slope), the days I had a more favorable wind (and was going faster), my power on the segment was lower. I would have to make a graph with the power and the cadence on this segment.
I would have imagined that at this speed, there was more W for rolling resistance and less for aeropower…
In your calculations, the weight has no influence on the 10km and the marathon? Maybe it does a little? My total weight on the marathon is at least 4kg heavier than during the climbing race, and a little less than 3kg for the 10km. I had a waterbag of about 1kg for the marathon and the hill race, I had planned 1L and 0.75L respectively, and I didn’t even drink half of it on the 2 races. And the weight difference between my 2 unicycles is 4kg (9.5 for the G36 & 5.5 for the 36").
Otherwise the fact that I didn’t have a brake for the hill race must have saved some W.
Loving your nerd discussions guys! Keep it up, really fascinating stuff.
My replacement SRM X Power Pedals have been sent to me, but not riding until October I don’t think due to school commitments.
After that I will have time to do whatever I want. I very much want to record a hockey game in real time which means I will get both positive and negative torque and can calculate a positive and negative(?) power for eccentric work. I will need to do it in R though as the sampling rate is extremely high on real time data collection. I think the csv files are around 18mb per 10 minute match.
True, I don’t think “skinny” is really necessary. Just lighter with good rubber!
Skinny just makes it more aerodynamic, and a bit lighter though. Rolling resistance is theoretically better with wider tires even.
This website has a nice graphical interface to see the different drag contributions. It should be exactly the same as my spreadsheet, just no cadence/wheel size comparisons.
My rolling resistance estimate could be low; it might be more like 50 W instead of 30 W, given how knobby and thick 36" tires are… However, aero drag will still be the main componen at 33kpht. The weight will increase rolling resistance, but it is linear. For your example, 70kg total vs 74g total is 6% difference. So rolling resistance at the same speed would increase by 50 W * 6% = 3W.
Definitely something that matters, and is included in the equations. However, in real life using the right tire pressure, right tire, and body position/aerodynamics will be more important for speed on flat ground; I would almost completely ignore weight.
Weight on a climb though, is of course a big deal. It is proportional to the power to overcome gravity…And gravity is taking around 200W. Even 0.2 kg of brake is worth 0.3% * 200W = 0.6 W. About 7 seconds for your road climb race time! I took my brake caliper and lever off too to save those few seconds, no reason to not use a lighter unicycle and parts… However, rolling resistance is probably just as important if not even more… the 1 minute saving I mentioned above is probably almost all in the rolling resistance. Of course, finding the “fastest” tire is a lot harder than finding a lighter component.
Now for the off topic portion!
29 Tires
I loved my 29x2.0 Marathon Supreme. It scored well for rolling resistance. However, it is a bit heavy for uphill races. It is also not tubeless compatible. It is great for turning on all my rim widths though. I like to use this for my G29 riding in general since it turns well. However, for a big trip, I would definitely want something tubeless.
Marath Supreme rolling resistance
For my ungeared uphill races, I have been using the Big One Liteskin 29x2.35. Great for most uphill races ungeared, since it is a bigger diameter. It is super light and scores extremely well in rolling resistance. I set it up tubeless, so I think this might be the lightest and lowest rolling resistance tire out there. However, it doesn’t seem to be available anymore.
Big One rolling resistance
The new G-One Speeds, which are what replaced the Big One, do not seem to score well for rolling resistance.
G-One Speed rolling resistance
The newest version (G-One R and G-One RS) use a different rubber compound, and seem faster, so maybe you would choose this instead of the G-One Speed. Maybe, they will come out with a slick version soon too…
G-One R Rolling Resistance
G-One R
G-One RS
I recently bought and used Rene Herse Antelope Hill 29x2.2" Ultralight Casing with a latex tube. Rene Herse has narrower widths, and the model name for each width is different. It should be comparable weight and rolling resistance to my older Big One. However, I have read people have problems with tire blowouts when using the ultralight casing, high pressure, and tubeless. I have not tried it tubeless yet.
However, if you use a narrower version, like 48mm, that should help with tire blowouts.
The 2.2" version does not turn too well though for me. I had it at 45psi on a 25mm internall width rim.
29x2.2 Antelope Hill
700x48mm Hatcher Pass
Rolling resistance for Rene Herse Slick Tires
If I had time and money to test for a long Guni trip in the mountain, I think the Rene Herse 48mm might be something to look into.
Rims:
I used the narrowest MTB rim from light bicycle they had at the time for my ungeared 29.
The used to advertise at 70psi max, but changed it to 40psi. I usually run my Schwalbe Big One Liteskin 29x2.35" at 45 psi tubless. There was no flyweight version when I bought mine. At 45 psi, the spokes lose a lot of tension (rim compresses). I would not go to the flyweight version on an asymmetrical disc brake hub.
Their newer 29" mountain bike rims look very nice though, asymmetric to help with spoke tension, and a little bulge so it is not a straight hookless profile. I just worry it might be too weak. They are very flexible with customizations though if you email them, so maybe if they could do a reinforced version, this would be a very good choice. 250grams, with around 100 grams reinforcement would still be 350 grams!
I use this gravel rim on my guni. Mine is hookless, and I regret not getting hooked! (On my bike, I’m happy with hookless though, since I do pinch the tire and hit the rim when I hit holes very fast) I run exactly at their max pressure rating of 55psi with a 32mm Continental Grand Prix 5000 for uphill races. A bit higher pressure would be better for sure with such a thin tire. Even if you use 2.0" tires, I think the extra reassurance will be good with a hooked version.
My G29 uses the standard weight version. My gravel bike has the flyweight versions. Again though, there is considerable spoke tension loss when you inflate the tire. I would not feel safe using the flyweight version on a unicycle, even at my body weight. It will be a tiny bit heavier than the mountain bike rims since it is a deeper cross section, but I think it looks cool
I have emailed them about drilling the holes offset, and they said they can do it. 1mm offset is no problem. 2mm offset, and they would add 30 grams to the spoke hole reinforcement layers.
I can’t wait to see this data!
I am a bit disatisfied with my flat pedal setup… It forces my feet to be further apart, which I don’t like. (At least 16 mm per side)
Few things I will try
If these don’t work, then I’ll have to look at the SRM X Power Pedals again! I think the high sampling rate data would be cool for road riding too.
Decided to try some modified VCS cranks with 5.5mm offset each.
This decreases the Q-factor from
100mm hub +
12mm bearing +
6x2mm spacer +
27x2mm crank width -
16x2 mm powermeter pod
= 210mm, down to 199mm (-11 mm)
This is still a bit wide, but now within what fat bikes use.
road/gravel bike = ~150mm
mtb = ~170mm
fat bike = ~200mm
Another interesting thing I learned from this exercise is that standard “0-Q factor” unicycles with 100mm hubs have a Q-factor of 178mm. This is wider than a mountain bike, and much wider than road bikes. I almost never ride my mountain bike because the extra 20mm Q-factor feels very uncomfortable to me…
Unfortunately, the Favero Assiomas do not give “platform center offset” data. Garmin Vectors give this data. It seems it might be interesting to compare to bikes given the different Q-factor.
Looks good, that 3d printed filing guide is a neat idea. That powermeter pod really does look big, but with the modified cranks it looks like a similar offset to many normal pedals now.
That’s why I’m always confused that some people like “high Q-factor cranks” on unicycles.
regarding Q - perhaps sometimes (eg in muni) feet catch on cranks unexpectedly (mitigated by outward curved cranks), or fat tires rub against legs (or maybe some people walk like ducks?).
I really like a high q-factor crank because my heels will always hit the crank otherwise. On my square taper cranks with just a slight q-factor I rubbed most of the paint off and they rust if I don’t use them enough to wear off the rust. I suppose I do walk like a duck also. The outer edge of the heal of my shoe soles gets worn down first. I only have 100mm hubs though, so perhaps I wouldn’t like high q-factor cranks on a 125mm hub, but I haven’t tried that yet.
Made some 3d printed platforms that are thick enough so that I stand on top of the pod without touching it. No I don’t have to modify my cranks (which was not possible with hollow bicycle cranks)
I am not sure how robust I can make them. One good thing is it can be super custom… but with a new challenge of figuring out what type of pedal shape and pin placements are best. Did my best for now to just look similar to other pedal designs.
If anyone is interested, I could share the other files/parts.
Tried the first version (larger, clear/white colored) one indoors on my bike, and it worked fine. Hopefully I can find time to ride them more soon
