Hi guys,
New rider here My name’s Diego and i’m a 34 years old engineer guy which has a passion for creating himself things. As many of people around the world, i’m in quarantine forced to home, and i’ve never experienced the need to use trainers before, because i loved to run my MTB outside in all conditions and all seasons. Actually, since i’m forced to stay at home, and went too late for buying a trainer (and i’m a bit short of money as well), since i had to cover some freetime and doing some sport, i’ve decided to mix all passions and make my trainers myself. Steel frame and two steel pipes behind, being welded with some SKF spare bearings i had in my garage. Actually it works like a charm, i’m so happy about that and the price of the entire work was about zero because i’ve used only spare parts i’ve found in my garage.
After that, i was looking for something fun to watch/do during my workouts, and then i find this very nice software.
I’ve bought a set of Garmin sensors 2 (both cadence and speed).
Ingame, both sensors are being readed, and they work, but obviously my trainer is not supported (so i had to choose those not listed, and the zPower option is unavailable, so had to click on the “classic” way).
So what’s the problem? i think using “only” speed and cadence sensors, the watt power showed is way too much lower than my actual one i’m developing, since the trainer resistance is (i think) much more stronger than a ready to buy trainer (is all done by steel).
My sensations are that the resistance developed from my own built trainer is comparable to be always on slope, not that high but definitely not on straight.
I’ve a Garmin Forerunner 735xt which i use during my workouts, and i’ve already bought (waiting for the package right now) an ant+ dongle, so in that way i will add the heart rate sensor from my watch.
Do you think, adding the heart rate, the power calculation will be more accurate? is zwift using the heart rate as variable for the output power?
Actually i’m only doing 40w and 5mph on a 4% slope…ok, i’m not a pro, but i’m not that slow
Zwift takes the wheel speed broadcast by your speed sensor and the power curve of the trainer you select on the Pairing Screen and converts that to virtual watts. With those watts, the weight entered, height entered, in-game bike used, in-game drafting (Tri and TT bikes get no drafting boost), in-game wheel set, in-game road surface, and in-game virtual elevation changes your in-game speed is determined.
This is not a real trainer curve, only to demonstrate the concept.
Example: Your wheel turn at 40 km/h then Zwift will use the curve and determine your Power as 200w. Using a set of complex equations taking your weight, height, road incline , rolling resistance and other factors in to account Zwift will calculate your virtual Speed.
So that 200w can be 45km/h on a flat road or 12km/h on a hill.
Hi Gerrie! first of all thanks for the reply, very appreciated.
Yeah, i’m happy to share my work with you guys, maybe it can give you some ispiration and making your own version
(dunno if you have a DIY section over here, if yes then feel free to move this thread there)
About my original question, about power calculation, i see that Zwift doesn’t use my heart rate as variable so, basically, i’m not be able to calculate the resistance being produced by my own trainer, so isn’t possible to use it in Zwift by using only cadence and speed sensors. Can you confirm this?
As i said i feel that my resistance is a lot, compared to the number showed in the program (expected more power output from my performance).
I think i would need a power meter, is it right?
Thank you again,
have a nice day,
D.
Nice setup, that trainer is not going to move when you sprint.
It using a speed senso Zwift need to know your power curve. But since this is a DIY trainer there is no power curve so you need to try different curves (different trainer options on Zwift) to see if you can find something that get you in the ballpark.
A power meter will make it more accurate, but they are not cheap.
Thanks Gerrie!
yeah is very sturdy, the weight is almost the same as mine.
About power meter, yeah, i will try to take a look at some linear curve based trainer, and start from here.
Otherwise i will take a look on the power meter thing, but the prices are quite high. I think, in the same spirit of the trainer, i will choose the DIY route in that part, using Arduino/Raspberry stuff plus some strain/gyroscope elements.
Thank you for the support btw very appreciated.
Have a nice day!
Diego
ey @Gerrie_Delport , after two runs i think i’ve found a “good enough” trainer which mimic my average power curve: the CycleOps Magneto. Still thinking i’m missing few watts on slopes over 4%, but i think i’m happy enough without a power meter. At least it’s something coherent without being too fast. I prefer to be in dept of power instead of being the PRO (which i’m not at all ).
For the power meter i think i will wait this one, which seems to be the best option for me. https://limitspowermeters.com/
Volcano run is awesome! did few mins ago, and is very fun.
I’m still very happy about my DIY solution, it works well, and actually the most expensive buy were the cadence and the speed sensors mod. 2 from Garmin (basically because i’m a Garmin fan).
Cya on road! thanks for the support.
Have a nice day,
Diego
The only issue is that it does not provide any resistance so Zwift can’t calculate speed from a power curve. This will have you doing 400w while drinking coffee.
My name’s Diego and i’m a 34 years old engineer guy which has a passion for creating himself things. As many of people around the world, i’m in quarantine forced to home, and i’ve never experienced the need to use trainers before, because i loved to run my MTB outside in all conditions and all seasons. Actually, since i’m forced to stay at home, and went too late for buying a trainer (and i’m a bit short of money as well), since i had to cover some freetime and doing some sport, i’ve decided to mix all passions and make my trainers myself. Steel frame and two steel pipes behind, being welded with some SKF spare bearings i had in my garage. Actually it works like a charm, i’m so happy about that and the price of the entire work was about zero because i’ve used only spare parts i’ve found in my garage.
