Converting GPS data to Power?

Here's a power equation that I use. It has 2 scenarios where you can compare power required under 2 sets of conditions, like checking a difference in weight.

As far as estimating power from a whole ride, that's tough. You may want to get a fix on your FTP using a long climb and this equation, then using PE to guage each ride.

Try Paul at Intellicoach.ca

Thanks for that file. This is a good start. It really shows why my friend (who is 40lbs lighter than me ) is leaving me behind on the steep stuff. He has to produce 60 Watts less then I do!!!

I need to reformulate the equation so I can input hill "total elevation gained" , "distance" , and "time for climb".

Tony

tbowren said:

Thanks for that file. This is a good start. It really shows why my friend (who is 40lbs lighter than me ) is leaving me behind on the steep stuff. He has to produce 60 Watts less then I do!!!

I need to reformulate the equation so I can input hill "total elevation gained" , "distance" , and "time for climb".

Tony

Ah, I have a version that has another section for that; I'll look for it, but it should be easy to add a few cells for that. Just make sure to "Unprotect" the sheet first (on the Tools/Protection toolbar).

tbowren said:

...
I need to reformulate the equation so I can input hill "total elevation gained" , "distance" , and "time for climb".

Tony

With that data you can try this link: rst.mp-all.debergauf.htmOpen ↗

Is not in english but is pretty easy to use, hope it helps!

Ale Martinez.

tbowren said:

I mostly want a way to quantify the TSS for a given ride and get some CP datapoints for a Power Curve.

For point in time "snapshots" of your power curve, you could do that with a trainer. You don't need a PM, but it would be helpful to use a trainer that someone with a PM has "calibrated" for a reasonably accurate speed/power curve (e.g., the Kurt Kinetic -- see frenchyge's data). For a road or MB ride, I think you'd want to define a relationship between grade, speed and power (given total weight of rider + bike). Then you could take a GPS data file (I assume you can get a raw data file) and convert it into a power file by using this relationship. But, to track your progress over time, I think the trainer protocol will be more accurate. I think a trainer is the next best thing to a PM for measuring performance over time.

tbowren said:

RapDaddyo's MakeWKO utility.

A new and improved version will be posted within the next 24 hrs. I have to update the readme document (obviously, the thing that I never actually use).😄

Amartinez: Thanks for that link. Seeing the "pros" power compared to mine is pretty interesting.

RapDaddyo:
Hey! I was hoping I'd see you reply. I've become quite a fan of your posts and approach to training. Your idea of converting a GPS file into something I can put directly into Cyclingpeaks via a "relationship" is worth looking into. I could throw away data that corresponds to a negative grade because I don't pedal that hard downhill offroad.

I actually do have and use a trainer. It is a Minoura Rim trainer that I use with both my road and mountain bikes which is hooked up to a PC running NetAthlon. This trainer's Power profile is "supposedly" calibrated in NetAthlon so I do get Wattage on this. (And I do my 3 x 20min FT intervals on there as well )
The "problem" is that in sunny SoCal I get to ride outdoors almost everyday and we mountain bike during our lunch break 4 times a week, so I want a way to get somewhat reliable TSS and power numbers from those rides.

Thanks for your help and ideas.
Tony

tbowren said:

Your idea of converting a GPS file into something I can put directly into Cyclingpeaks via a "relationship" is worth looking into. I could throw away data that corresponds to a negative grade because I don't pedal that hard downhill offroad.

It's far from perfect, but the hardest efforts should show up and that's what you really want.

tbowren said:

I actually do have and use a trainer. It is a Minoura Rim trainer that I use with both my road and mountain bikes which is hooked up to a PC running NetAthlon. This trainer's Power profile is "supposedly" calibrated in NetAthlon so I do get Wattage on this. (And I do my 3 x 20min FT intervals on there as well )

Unfortunately, I don't know anything about that particular trainer and it isn't on this list of trainers that were tested (unless it's the Nashbar Mag 2) geocities.comtrainerpowerOpen ↗. All the trainer mfgrs. are aware of the "power boom" and they all want to advertise that you can ride their trainers at power targets, but I'm afraid it just ain't so.

tbowren said:

The "problem" is that in sunny SoCal I get to ride outdoors almost everyday and we mountain bike during our lunch break 4 times a week

Bummer. You should move to Maine where you can ride on a trainer ~6 months/year and have better power data.😉

tbowren said:

so I want a way to get somewhat reliable TSS and power numbers from those rides.

It may be "in the ballpark" for Performance Manager purposes, but I would still fall back on trainer rides to track my performance over time.

Drats!!!
I just uploaded my gps data from last weekend and the data for Sunday's recovery ride is not a bit accurate.

I knew while riding as I went under some tree canopy that speed reading was off. The reading would flucuate from 12 to 20 when I knew for certain that I was keeping a relatively constant pace. If you see the link on Motionbased you can see that it has a max. 35 mph speed and I know that I never went above 24 +/- for the ride. Since it was a recovery ride I tried to keep around 17 to 20 mph on the speedo, but reading was all over the place and the uploaded data reflects the same irregularities.

I suppose one would have to be careful when analyzing the data as to how the gps was reading for the particular day. My ride on data for Saturday seems to be more accurate.

here's a power equation sheet with distance, time, elevation incorporated.

Felt_Rider: You are so right about the lack of accuracy of GPS readings. I was thinking of all they ways I could simplify the conversion process by omitting downhill grades and disregarding wind calculations, but I still have to accept errors coming from things like carrying momentum into hills from preceding descents and GPS screwups. Pretty soon you start to realize you are getting a very accurate measurement of a wildly INACCURATE reading. Not very useful at that point.

It seems the most useful thing to do would be treat the hills as intervals and the descents as rest and look at the entire ride from the "big picture". The latest .xls file from jws is perfect for getting that info since I know quite accurately how high and long the hills are and how long it took to climb them. I may still try to create my own .csv file but I will have to be suspicious of some of the readings.

jws: FYI thanks for that latest .xls file, just wanted to give you a heads up that the grade needs to be multiplied by 100 for the watts to calculate correctly. I compared the results of the same ride (13.8% grade for the first file shows an i value of .1383 and for second file it shows .0014). I figured you just whipped those extra boxes up for me, so it probably doesn't matter but others might be surprised to see their climb only require 7 Watts.

RapDaddyo: I checked out the graphs from the power tests on the various trainers. Wow they are all over the place. I really cant trust my NetAthlon readings either. Time to start saving for a real powermeter, luckily I have alot of time to figure out if I want it for my road or mountain bike. Probably the road bike because thats the one I try to regulate my power on, on the MTB I just ride as hard as I can without falling of the bike.

Tony

tbowren said:

on the MTB I just ride as hard as I can without falling of the bike.

That is my riding approach for MTB. 😄
I tried the 305 on the trail and had problem with dropping signal in dense canopy so I think I will not risk damaging the 305 with a potential crash on the trail and just go for the adrenaline rush instead. No numbers to look at just let it rip.

Thanks guys for the spread sheets

Felt_Rider, If you mount the GPS on the stem it seems very safe from a crash. I have gone down a number of time and it's never been scratched. The handlebar mount seems very sketchy to me.

I use it to check out my heart rate on climbs and recovery and see how many feet I have climbed. Once I even got lost on a trail and used the map mode to see where I had ridden to help get my bearings and pick the right way home.

Some jerseys mess up the heart rate data, and some locations mess up the speeds but those are small prices to pay for all the other times it works great. If tree issuse are a big problem, the cadence/speed adapter will override the GPS signal for much more accurate speed info in those situations. Plus, I love the graphs and charts, that why I want to use Performance Manager so badly.

Tony

tbowren said:

jws: FYI thanks for that latest .xls file, just wanted to give you a heads up that the grade needs to be multiplied by 100 for the watts to calculate correctly. I compared the results of the same ride (13.8% grade for the first file shows an i value of .1383 and for second file it shows .0014). I figured you just whipped those extra boxes up for me, so it probably doesn't matter but others might be surprised to see their climb only require 7 Watts.

Tony

thanks, tony....i did just add those changes this morn and didn't check it. OOPs