Turbo Vado 2 4.0 or 5.0 or 6.0?

just speculating. i bet it's more complicated than we think and also has something to do with how the motor and pedals are "connected"
No question it's more complicated than we think — especially around accelerations and other special situations like starting out and passing through max assisted speed. Maybe even around gear or assist mode changes.

Cadence by itself would likely come into play in these cases, but not necessarily in steady-state riding above 60 rpm.

My goal all along was just to model Pm as a function of Pr in steady-state riding. But even that's looking way more complicated now than Spec documentation ever let on.

Need to step back and rethink the whole reverse engineering project. The way forward — if there is one — probably lies in the .FIT files. But their low signal-to-noise is gonna be a big challenge.
 
Very familiar with that. But @mschwett is talking about something online.
Yes, his link leads to Specialized Ride online service.
Yes, the site collects your rides that used either Mission Control or the Specialized app.

I'm surprised though. I had a ride in Ireland on a Bosch E-Bike but that ride is among my activities on the Ride! A mystery.

1769840951976.png

How come?! Was that ride somehow imported from my older Wahoo?!
 
Last edited:
Yes, his link leads to Specialized Ride online service.
Yes, the site collects your rides that used either Mission Control or the Specialized app.
The only link I saw just took me to the regular Specialized site. Couldn't find a path from there to what you're describing. Did you try the link yourself?

ADDENDUM: Correct link to separate Ride site below. I think forum software added the link we were discussing.


Not seeing anything that the current Specialized app doesn't do better.
 
Last edited:
The only link I saw just took me to the regular Specialized site. Couldn't find a path from there to what you're describing. Did you try the link yourself?

ADDENDUM: Correct link to separate Ride site below. I think forum software added the link we were discussing.


Not seeing anything that the current Specialized app doesn't do better.

i haven’t tried exporting a .fit file from the ride app, but the ones made by the website have all the data you need in them! click on the ride, click the three dots at the top left i think?

the chart can also display whatever combo of data you want, at a decent scale, unlike the phone app!
 
I just found how to use a FIT file viewer download .fit data. I had recorded a few rides on my Como last year so I had something to play with.

The .fit file records all the data we care about (and more) once each second. The 45 minute ride I downloaded had 2600 lines!

I’m envisioning many very short rides with a real effort to keep rider power constant. The resulting .fit files might look less noisy. Also with the data in a spreadsheet you can eliminate all the zero rider power troughs. I am itching to get my new bike and do some experimenting.
 
I Looked up a ride I did on my Como 3.0 last year, one of the few that had much time in Turbo mode. That's the only mode where I know the E/A was set at 100/100 .
The 50 minute ride had (after I deleted the zero cadence rows) about 12 minutes in Turbo, 706 data points (706 seconds of records). I tried several ways to show Pm as a function of Pr and Pm/Pr (boost). Even after I eliminated all the zero cadence rows there was still crazy zig zag lines on the chart. I finally averaged Pm and Pm/Pr over 5 rpm groups of Pr. For example for all the seconds where Pr was between 100 and 105 watts Pm average was 410 watts. This smoothed out the curves a good bit.

Here's the chart.

Image 2-1-26 at 12.07 PM.jpeg


I asked Google AI what the max motor power and assist level was for my 2.0E motor and it quoted directly from an EBR forum post by Stefan from last July. Pm max = 430 watts, boost = 280%. I couldn't find any info on the Specialized website and his Specialized link is now broken.

Note the lines on the chart. Note the actual data!

Some observations:

The raw data shown Pm max = 563 watts. The boost looks to be almost a linear function of Pr above Pr = 70 watts with a clear negative slope. Certainly not constant until saturation. Saturation looks to be Pm = 525 watts, Pr = 150 watts. 525/150 =3.5
 
Now, multiply the Pm electrical by 0.79 then the true mechanical power would be 414 W and Bmax equal to 2.76.

If we take 430/525= 0.82 (efficiency of your motor) then 430 / 150 = 2.86x

The specifications have been proven.
 
Last edited:
i haven’t tried exporting a .fit file from the ride app, but the ones made by the website have all the data you need in them! click on the ride, click the three dots at the top left i think?

the chart can also display whatever combo of data you want, at a decent scale, unlike the phone app!
Yes, the Ride website seems to be the only way to get raw FIT files from recorded rides. Thanks for the tip!

Tried again yesterday but still can't find a way to get a FIT file straight from the app.

Now I just need a palatable way to get the FIT files into Excel.
 
I Looked up a ride I did on my Como 3.0 last year, one of the few that had much time in Turbo mode. That's the only mode where I know the E/A was set at 100/100 .
Good idea to focus on 100/100 assist. I believe all Spec mid-drives use the same Turbo operating system, but they might tune it differently for different models.

The 50 minute ride had (after I deleted the zero cadence rows) about 12 minutes in Turbo, 706 data points (706 seconds of records). I tried several ways to show Pm as a function of Pr and Pm/Pr (boost). Even after I eliminated all the zero cadence rows there was still crazy zig zag lines on the chart. I finally averaged Pm and Pm/Pr over 5 rpm groups of Pr. For example for all the seconds where Pr was between 100 and 105 watts Pm average was 410 watts. This smoothed out the curves a good bit.
Reasonable processing method.

Here's the chart.

View attachment 205405

Note the lines on the chart. Note the actual data!

Some observations:

The raw data shown Pm max = 563 watts. The boost looks to be almost a linear function of Pr above Pr = 70 watts with a clear negative slope. Certainly not constant until saturation. Saturation looks to be Pm = 525 watts, Pr = 150 watts. 525/150 =3.5
Well, that's pretty much the final nail in the coffin of my naive Pm model.

The power ratio observations I reported a few posts back made it hard to believe that the Pm vs. Pr ramp up to saturation is a straight line, as my naive model assumed. And here we see that it's clearly NOT a straight line on the Como. Nor is there a clear saturation point where Pm goes and stays flat with Pr.

Dang! No idea how to model Pm now.
 
Yes, the Ride website seems to be the only way to get raw FIT files from recorded rides. Thanks for the tip!

Tried again yesterday but still can't find a way to get a FIT file straight from the app.

Now I just need a palatable way to get the FIT files into Excel.
https://www.fitfileviewer.com/

Download the fit file, open it with the file viewer, export as csv file, open the csv with excel.
 
Now, multiply the Pm electrical by 0.79 then the true mechanical power would be 414 W and Bmax equal to 2.76.

If we take 430/525= 0.82 (efficiency of your motor) then 430 / 150 = 2.86x

The specifications have been proven.
Good catch and reassuring that the Specialized published motor specs are accurate.
 
Had one other thought - that maybe the drop off in motor power at the high end of rider power was due to the Class one cut off of 20 mph. Checked the source data and the column labeled "enhanced speed" max value is 6.86. Assuming that's speed in meter/second that translates to 15.3 mph. So no problem.

By eliminating all the zero cadence data I also eliminated all the fast downhill coasting.
 
Yes, the Ride website seems to be the only way to get raw FIT files from recorded rides. Thanks for the tip!

Tried again yesterday but still can't find a way to get a FIT file straight from the app.

Now I just need a palatable way to get the FIT files into Excel.

The free gotoes website i posted above is super easy, drop the fit file and get a csv back, ready for excel.

i have many thousands of miles of rides on the ride website, but unfortunately (hah) most of it is with the motor off and when it’s on, i don’t think I’d have any way to recreate what the ease and max settings were. If that data was also in the files, i bet a dump of a few dozen of them into gpt or Claude or gemini would result in a very good explanation of what’s happening!

what i am pretty sure of is that over time, at a steady state, between 60 and 100rpm, the final power ratio DOES average to 2.0 x ease as long as that value is under max.
 
The free gotoes website i posted above is super easy, drop the fit file and get a csv back, ready for excel.

i have many thousands of miles of rides on the ride website, but unfortunately (hah) most of it is with the motor off and when it’s on, i don’t think I’d have any way to recreate what the ease and max settings were. If that data was also in the files, i bet a dump of a few dozen of them into gpt or Claude or gemini would result in a very good explanation of what’s happening!

what i am pretty sure of is that over time, at a steady state, between 60 and 100rpm, the final power ratio DOES average to 2.0 x ease as long as that value is under max.
Fair enough. I'll resist the urge to throw up my hands till I have a few FIT files under my belt.

But first, I should design some proper test rides to analyze. First thoughts:

1. Definitely maximize time at 60-100 rpm cadence to keep peak mechanical motor power and motor efficiency reasonably flat at test cadences.

2. Definitely minimize accelerations, elevation changes, and assist changes to minimize motor power transients and torque involvement.

Any other test ride ideas, anyone?
 
Last edited:
it's interesting that the .fit file contains headers for almost every cycling parameter you could imagine, including "ebike_assist_mode" and "ebike_assist_level_percent" !

unfortunately it doesn't look like specialized is using them. i don't think these were added by specialized since there are quite a few other unused ones which relate to sensors no specialized bike has, like pco and hemoglobin lol.

i also think "enhanced speed" and "enhanced altitude" are some kind of interpolation with the phone sensors, vs the sensors on the bike. i'll find a ride through a tunnel and confirm that. phone speed drops to zero in most tunnels around here - no gps - but bike sensor works well.

i grabbed an old ride that i remember - from my early days with the creo. it was up in the mountains and i was not really well prepared - no hydration or fuel, high altitude. climbed from about 6000 to 9000 feet, descended, and then bonked hard on a shorter ±1000' climb to get home. it's very clear from the ratio (gold, right axis) of rider power (blue) to motor power (red) when i'm in eco, sport, and turbo. the data is smoothed to 5 seconds here, and still pretty spiky, but over long steady state periods like the middle of the big climb in eco, it's really very flat at .3x support. this tells me i had eco at either 15% ease or something higher with max at 30% or so. over this fairly long slow steady period rider power averages 185w and motor power averages 54w. average support factor ... 29.4%. you can then see a similar relationship when i bonk at the end of the last climb and go into turbo. i'm toast and rider power averages only 159w, motor power is at 293w, and support ratio is at 1.87. at that point we're dealing with the max power more than anything else.

one very simple thought is that the final support ratio will vary if you're ever hitting the max for any given setting. this is why the normal microtune is an interesting way to explore this, since it has max at 100 the whole time. what we're really looking for is "what is the support factor when you aren't hitting the max."

specTahoe.jpg


edit : since sport is less likely constrained by max, i looked at the segment of sport at the beginning. rider power 209w, motor power 166w, average support factor 79.3%. probably had ease at 40%, it seems ;) my recollection is that i had eco at 30/30 and sport at 40/70 or something like that, so that sport would reward pedaling hard with lots of power but eco would stay steady, preserving battery regardless of how hard i pedaled. and turbo was 100/100 for sure.
 
Fair enough. I'll resist the urge to throw up my hands till I have a few FIT files under my belt.

But first, I should design some proper test rides to analyze. First thoughts:

1. Definitely maximize time at 60-100 rpm cadence to keep peak mechanical motor power and motor efficiency reasonably flat at test cadences.

2. Definitely minimize accelerations, elevation changes, and assist changes to minimize motor power transients and torque involvement.

Any other test ride ideas, anyone?

i'd find a hill of around 5%, ideally at least 300' of climb, no stops, keep cadence around 80, heart rate at a steady sustainable value for rider watts in the 100-150 range, do it once at 20/100, once at 40/100, 60/100, 80/100, and 100/100. you won't be in danger of hitting 28mph (heh) and the gearing on your vado should allow 80rpm for all of those scenarios. 20/100 may be a little slow, i'm guessing that'll be around 8mph, more like 65rpm. you could do 25/100, 50/100, 75/100, and 100/100 if you want to keep the cadence and speed ranges a little tighter. of course the results will be more interesting the less time you max out max, so you want to stay below 150 rider watts. a 4% grade would be better but harder to find.

for better or worse i have lots of hills like that around here o_O
 
i also think "enhanced speed" and "enhanced altitude" are some kind of interpolation with the phone sensors, vs the sensors on the bike. i'll find a ride through a tunnel and confirm that. phone speed drops to zero in most tunnels around here - no gps - but bike sensor works well.



regardless of how hard i pedaled. and turbo was 100/100 for sure.
i also think "enhanced speed" and "enhanced altitude" are some kind of interpolation with the phone sensors, vs the sensors on the bike. i'll find a ride through a tunnel and confirm that. phone speed drops to zero in most tunnels around here - no gps - but bike sensor works well.
Do you agree that "enhanced speed" is probably meters/second?
i'm toast and rider power averages only 159w,
If your "toast" level is 150 watts I'm impressed!
 
Do you agree that "enhanced speed" is probably meters/second?

If your "toast" level is 150 watts I'm impressed!

yes - it averages 3.6 times the other speed value which is kph.

before my latest health troubles i could maintain 200w for a very long time (4+ hours) if appropriately fueled! not anymore lol.
 
i'd find a hill of around 5%, ideally at least 300' of climb, no stops, keep cadence around 80, heart rate at a steady sustainable value for rider watts in the 100-150 range, do it once at 20/100, once at 40/100, 60/100, 80/100, and 100/100. you won't be in danger of hitting 28mph (heh) and the gearing on your vado should allow 80rpm for all of those scenarios. 20/100 may be a little slow, i'm guessing that'll be around 8mph, more like 65rpm. you could do 25/100, 50/100, 75/100, and 100/100 if you want to keep the cadence and speed ranges a little tighter. of course the results will be more interesting the less time you max out max, so you want to stay below 150 rider watts. a 4% grade would be better but harder to find.

for better or worse i have lots of hills like that around here o_O
All good suggestions! Will see what I can cook up. Some thoughts...

1. You're a tough taskmaster, but pretty sure I could do 3-4 runs like that in one outing at 80-90 rpm.

2. Used the RideWithGPS route planner to scout local climbs I know of that might be suitable. Found none of at least 300 ft gain with anything resembling a steady gradient, but several with somewhat steady stretches of 4-8%. Probably the best I'll do from home.

3. Writing an assist mode as E/M, my SL 1 experience indicates that motor power falls off sharply at E < 30. So probably more meaningful to test E points in the 30-100 range.

3. Probably need to keep rider power under 120W to reduce the chance of saturating.
 
Back