BIKEPGH MESSAGE BOARD ARCHIVE

grade is distance of rise over distance of run expressed as a percentage.. so if a specific length of hill is 100 ft long, and you go up 10 ft, then it is a 10 percent grade.

I would say find out what the minimum and maximum elevation of the hill in question are, take the difference and divide by the length of the hill. Just make sure you use the same unit for both numbers (feet or meters)

Length of rise divided by length of run. Move the decimal point two spots or multiply by 100 to get it in a percentage.

A hill in the middle of a ride might be really steep and have a 10% grade or something, but lengthen the ride and include some long flat sections and the percent grade over the length of the ride will lessen.

A complex and perplexing question as it turns out. Not very easy to do, especially if the road curves(as opposed to going straight ahead and up). Garmin Edge bike computers (and probably others) have a grade feature built in. I haven't ridden up that stretch since 2008 but I can see in an old ride that it topped out at 16% That seems about right from memory.

(five minutes later) Now that I look at that again, it might be that this was going the other way from Saxonburg Blvd. toward Gibsonia and up the hill on the other side. So I'm not sure. It's a good climb though, either way.

Mapmyride doesn't estimate grades very well over short distances (<1 mi). The best way I've found is to get the elevation data and distance off gmap-pedometer.com and then do the math.

@johnwheffner ... it doesn't do that great job on long distances either. Before leaving for the trip I mapped out my Wales tour on it (all 300 miles), it said the steepest grade was 7%.

The actual steepest grade that we know of was 20% (there was a sign), and we hit multiple hills that were that steep or even steeper.

That being said, the new beta.mapmyride.com seems to do a little better, and it lets you zoom in on sections of your ride for a closer look.

By the way, for those of you who have a Garmin Edge and are wondering where I got the % grade from 2 years later--you can only see this in the Garmin Training Center software. Back when I had a 205 you had to upload to GTC, but with the 705 not. But you can still download GTC and then export rides from connect. It's not perfect.

There's a new site out there, sadly fee based after you upload 10 rides, but which seems to address a number of shortcomings in Garmin Connect called Strava.com. Check it out if you have an Edge.

And now back to your regularly scheduled thread.

The z-coordinate (elevation) is the least accurate part of a GPS reading. Most of the mapping your ride websites will just toss this out and snap your route to the ground, based on some usgs or other topo data for the elevation bit of it. Your grade is determined by this data, which comes in various flavors of accuracy and resolution (not real great).

What I find most annoying is if you go through a tunnel, or over a bridge, it ignores that, and pretends you went over the mountain, or swam the river. I have yet to find one that knows better.

You can do much better if you have a GPS with a barometric altimeter to determine your elevation. Most of the mid-range and higher models these days do. I would make sure your mapping software isn't tossing out your elevation data, if thats the case.

I even had a speedometer for my bike with an altimeter that did a decent job and gave me a steepest grade data point for the ride.

Here is a link to a gradiometer (i think that is how it is spelled) that you make on your own. This will give you an idea of how steep a section of the climb is.

http://graphics.stanford.edu/~lucasp/bike-grade.html#Build

This tool is also called an inclinometer

http://en.wikipedia.org/wiki/Inclinometer

Here is a how to make one

http://www.exploratorium.edu/math_explorer/howHigh_makeInclino.html

this is done in degrees but it would be simple to make one the measures in % just remember that 45 degrees is 100% so 22.5 deg would be 50% and so on.

EDIT: you can even purchase one for your handlebars

http://circlecitybicycles.com/inclin.htm

Google Earth's database includes elevation, and allows you to measure the distance of a curve (or rather, several straight lines that together, approximate a curve). I've tried measuring the big Negley Avenue hill this way, for instance, but don't remember what answer I got. I wouldn't be surprised if their altitude database wasn't good enough to do these calculations in any case.

Using the beta map my ride with the rise/run strategy given here I am guessing between 15-16% for that hill. But it's hard to say.

I've also rode that hill the other way, and it's a beast in the opposite direction too (just past the turnpike).

Also, ironically, this was my very first ride with my new Garmin Edge 250 that I picked up at Performance Bike. I have no idea how to use it yet but plan to try to figure out how to get it to cooperate with my Mac soon.

So, if its the hill I think it is, http://maps.google.com/?ie=UTF8&hq=&hnear=3610+Wickshire+St,+Pittsburgh,+Allegheny,+Pennsylvania+15212&ll=40.585637,-79.883459&spn=0.00396,0.010815&t=h&z=17

then it looks like its about a 200ft rise over about 4/10ths a mile. Thats about a 10% grade over the whole hill. Looks like on the steepest part it hits about 12-13%

Yep, that's it.

I had [(1100-840)/1660]*100 since I had the total climb at about ~ .3 mile and the map thing showed a starting elevation of 840 feet and the top at 1100 feet.

I believe your calculation more than mine, but can you tell me how you got the values so I can learn how to do this for myself going forward? I don't see where to even get elevation in Google. Thanks so much!

For mac, check out Ascent. Gives much more info than Garmin Training Center - rate of climb, power output estimation, nicer user interface as well. As long as you limit the number of imported rides to 10, it's free.

google earth has elevations, google maps doesnt seem to. Also, try http://ridewithgps.com It lets you import gps data from a whole lot of different file types and sources. Also lets you click/plan routes and stuff.

Anybody know of a GPS tracking app for android? I know there's one for the iphone.

runstar works pretty well. my wife has another one that she uses as well, but i cant think of the name off the top of my head.

My Tracks from Google. Small footprint, sits in the background logging away or you can view your position on a google map, your elevation plot, or a stats page. Upload the route right from your phone to your google my maps and view them in google earth/maps.

You can get Google Maps to show the grade for a route, like this:

1. If you've hidden the panel on the left of the map, display it again by clicking the arrow on the light blue bar.

2. Click "My Maps".

3. Click "Browse the directory".

4. Find Path Profiler, and click its "Add it to Maps" button.

5. Click "Back to Google Maps" (upper left).

6. Check the boxes "follow roads" and "show grade". Don't check "curved Earth" because it seems to be broken.

7. Click each point on the route. You'll see the grade as a percentage in red on the left.

8. If you like, click the Popup link to see a graph of the elevation in a new tab or window. Example:



Unfortunately, it just shows the overall grade for each segment, not the grade of the steepest part.

Wow, Steven! Very Cool! Great info.

According to the data from my Garmin, and it's probably not accurate, but whatev, over 0.2 miles you go from 871 feet to 1025 feet on those switchbacks which would make it a 14.5% grade.

one of the things I learned about from Geocaching is that most Gps have a accuracy down to 10ft if they have and good signal and higher if they dont. This variable in accuracy can be in all directions including elevation. take a Gps out and once it is running stand still and watch the readout as it changes. That can make the readouts from GPS ok in general but hard to get really exact. the chance in distance over miles is minimal while if you go up 200 feet and your accuracy is + or minus 10 feet your overall change in elevation could read between 180 and 220 feet and that is if the signal is good.

just my thoughts on it.

DB

BTW, I don't know why this page disappeared from the WPW site but luckily you can still get it from archive, I always thought it was interesting...

Negley is 15.81% according to that.

BTW, Steven - that path profiler thing is pretty cool, I didn't know that existed.

Yup, it's kinda neat. One trick: If you click your start and end points, it'll show you the grade for the whole route. But if you click a bunch of intermediate points, the page you get by clicking Popup will show the grade for each segment, color-coded to the graph. (But see their FAQ, linked below, about limits on the number of segments you can use. And of course you run into measurement accuracy limits if your segments get small enough.)

The Elevation Contours mapplet, from the same content page, can also be useful. It draws contour lines, and you can tell it to show the elevation every time you click someplace.

These are both by the same folks, and their FAQ page includes lots of good info, including where their altitude data comes from and a list of similar sites.

The WPW page on Pittsburgh hills tells me S Negley is 15.6% but the Google Path Profiler tells me that Negely between 5th and Fair Oaks is 9% and even less if I map it all the way to Wilkins.

Oh well. I guess I'll never know for sure unless I quit my job and become a land surveyor. I really appreciate all of the great info given here.

Fair Oaks and Wilkins are on the other side of the crest, so what you saw makes sense. I measured it from just south of 5th to Dunmoyle and got something pretty close, in the 16% range.

Hey yeah, I just got it to work. There is hope for me yet!

Couldn't you simply attach a compass and a string with a rock from your frame after zeroing it out on a flat surface?

spak - you need to click on dbacklover's links. although being able to measure grade from the comfort of home is a good thing too.

a while ago I put up some tracks taken from my garmin gps (with altimeter) Looks like I never tagged them nicely, but there are a few here

link to flickr pages

ah yes salty, whoops.

top. Steven, the Google path profiler seems to have moved.... do you know how to get this now??? thx

Looks like Google discontinued the facility the old page was using (see the FAQ), so it's dead.

This page is by the same folks, and might be useful. But I've been using ridewithgps.com myself. It shows some kind of local grade info.

If you read the forums on mapmyride, there are quite a few comments on how mapmyride calculates grade. The longer the ride, the less the grades become. I got tired of waiting for them to fix it and switched to ride with gps. It seems to be closer on grade calculations to what I am seeing on my cycling computer.

thank you Steven!

Is there an affordable GPS that incorporates elevation info?

@dmtroyer: It's not GPS but is a simple computer. I wouldn't call it cheap at $139 on their website but it is the one that I know so I thought I'd throw it out there. It does Altitude and Grade.

http://bontrager.com/model/08205

in terms of precision and accuracy of determining the percent grade of a road, physically measuring the road is best. From there the quality of the information recorded of few items discussed on this thread in descending order is: cycle computer with a barometric altimeter and variable time-step for recording, google maps or other mapping service that references USGS mapping, then GPS without a barometric altimeter.

recording elevations based purely on the location, signal strength and number of GPS satellites is highly imprecise and wildly inaccurate.

cycle computers are very accurate, as long as you correct your altimeter to a known elevation regularly, as it can drift with weather and the seasons. the precision of cycle computers can be adjusted by the frequency of recording - more frequent, more precise.

online mapping services like google maps, reference USGS 7-1/2 minute quadrangle topographic maps that publish changes in elevation which can be off by at least +/- 5 ft. So there will be a wide range of values returned if seeking elevation changes over short distances.

In regards of a slope grade. http://www.mathopenref.com/images/trigsolving/triiggeneral.png Bicyclist is riding on H. Grade is just a tangent multiply by 100 -- http://www.mathopenref.com/images/trigsolving/eq3.jpg. Now if we mesure H and O (height) and divide I by H then we get a regular sine -- http://www.mathopenref.com/images/trigsine/Eqn1.gif. So to get a correct persantage we nee to do the following Pgrade=tag(arcsin(O/H))*100.

this is done in degrees but it would be simple to make one the measures in % just remember that 45 degrees is 100% so 22.5 deg would be 50% and so on.

Not quite correct.

Allow me to be pedantic.

90 degrees (perfectly vertical) give 100%.

one mile of road = 1 mile increase in elevation.

45 degrees gives you 1 mile rise for every 1.414 miles of road, about 71%.

22.5 degrees is about 38%. 10 degrees is 17%.

no, no, you're both wrong. the percent grade is just the coefficient of the slope, as in rise/run. so, 100% = 1.0x = 45°, but, in fact, 50% = 0.5x, which is actually a slope of 30°. it's just a bit of trigonometry.

@Mick, your mistake is to consider the distance of the tangent, whereas percent grade measures the run as the amount of horizontal distance along the plane that is normal to the direction of gravity, i.e. the flat theoretical ground beneath the hill.

Only in Pittsburgh: " flat theoretical ground"

This is why they put sin, cos & tan buttons on calculators.

HiddenVariable has it right, but Mikhail is right too.

Mikhail is saying, suppose you only know the length of the slope (not the horizontal run, as HiddenVariable uses) and the rise. Using Wikipedia's diagram from their article on grades, you know the slope length l but not the run d (as well as the rise h).

To precisely compute the grade from that, it's not as simple as

grade = 100 * rise / run

You need a little trig.

Mikhail made a few typos that make things a bit confusing, but I think his formula

grade = 100 * tan( arcsin( rise / slopelength ) )

is right. Using the notation from the diagram, the grade g is

g = 100 tan( arcsin( h / l ) ) = 100 h / d

Of course, the slope length l (as measured by counting the revolutions of a bike wheel) will usually be close to the run d (as measured on a map), except on very steep terrain. And how often do you know the slope length but not the run?

Since everyone seems to be an expert on slope % now, anybody knows what is the slope percent on Penn Avenue @ Bloomfield?

Penn between Butler & Main varies around 5 - 7%.

NNNNEEERRRDDDSSS

Just kidding.

I prefer qualitative measures of slope. "Flattish", "A bit of a climb", "Steep", and "F'in steep" cover the bases pretty well.

HiddenVariable has it right, but Mikhail is right too.

indeed he does. i should clarify, then, that when i said "you both", i meant mick and the poster he was quoting. mostly, i was just pointing out that 22.5° is not 50%.

@reeddan Flattish", "A bit of a climb", "Steep", and "F'in steep" cover the bases pretty well.

For most places, yeah. Pittsburgh has the additional "OMFG! They have to be kidding!"

Like various south side slopes, and Canton Ave. Basically any dirty dozen hill and then some other hills, too.

The kind of hill that when I walk my bike up it, I put on the brakes, take two steps, plant my feet, push the bike up a few feet, brakes again, etc.

Federal Street has all of the first four.

"Flattish" is from North Ave up to about the library.

"A bit of a climb" describes from there up to Henderson.

"Steep" describes from Henderson to the Perrysville split.

"F'n steep" from Perrysville to Lafayette.

If you want OMFGTHTBK, continue on up to Burgess (note, you're still not at the top yet), descend to East Street, then reverse. Also known as Hill #7.