Air speed
 

Is there such a thing as a cycle computer that measures air speed?

Presumably airspeed minus wheel speed would give a record of wind conditions.

I've seen an anemometer that uses a bicycle computer as the wind speed readout. Mount that on tire bike or beenie.

I have visions of an anemometer mounted high atop a helmet.

Beanie cap, anyone?

Ibike.

Thanks for the link to that article, Looigi, very interesting read

You need to build a pitot tube, which is basically a very thin, rigid L-shaped tube one end of which points straight forward. The other end should contain a relative pressure sensor, (or manometer)

https://secure.wikimedia.org/wikiped...iki/Pitot_tube

The most accurate manometers in that kind of application that don't break the bank probably work by measuring the capacitance of a conductive diaphragm to a stationary conductive sheet. That way you can get an idea of the relative distance between them, which would be proportionate to the pressure.

Anerometers have problems - at a low speed it might not spin at all, also the speed of rotation is not directly proportionate to your forward speed.

As far as I know, a pitot tube is the only way to accurately measure the airspeed forward.

Watch out for icing!


https://secure.wikimedia.org/wikiped...nce_Flight_447

You should turn on the pitot heat prior to entering a cloud in below freezing temps... :D

;)

marc

You can easily get your bearing and heading from a GPS module. Just a few minutes ago I was fooling around with a fairly new GPS module that can update its data at 20 Hz, thats 20 times a second. The data maxes out a 115200 baud serial port. It also can output raw data, so it can be used in RTK, real time kinematics.

If you were willing to set up two GPSs. one at home in a stationary location, the other on your bike, with a good antenna, you could get extremely precise measurement in all three dimensions, plus time.

Is that something you might be interested in? A friend and I are experimenting with low-cost modules and various COTS and homemade antennas and the free RTKlib library ( http://www.rtklib.com ) trying to do a how to, a proof of concept showing that now DGPS/PPP can be done for very little, to an accuracy of well under a meter, at a high update rate for under $250 or so, plus the computers. (almost any computer will do, thin clients which can cost under $50 on ebay, are fine.)

I think its possible that we may be able to get 10 cm accuracy if there are no large buildings or trees overshadowing you.

Commercial equipment that does the same thing costs tens of thousands of dollars. There is a new generation of chips arriving now that output the raw satellite data and allow you to do the calculations yourself, which allows you to tweak the sources of various kinds of data. For example, you can get better ephemris data from NASA than you can from the satellite broadcasts.. you can also exclude satellites with marginal or reflected signals from messing up your 3d fix. I would imagine it would be really cool for cyclists. I know tha when its a little further along I am going to try logging rides. Basically, they should look like a GPS track you see now, but hopefully the tracks should be more accurate, by a very good margin.

The antenna has been the thing that is hard to get right. But we're making a lot of progress.

??? Bearing is a compass direction (true or magnetic) to or from something. You need the other something to define bearing: like the bearing from my current position to New York City is 102 degrees...or from NYC to my current positions is 282 degrees. GPS can not giving you heading. It can give you course. If you assume you're facing in the same direction your traveling, then you might assume heading is the same as course.

Give up any ideas of GPS based reporting. Air along a roadway is never the same as a "prevailing wind."

I bet that it would not be super difficult to measure actual air speed. A pitot tube is just a tube, and pressure sensors are not that expensive.. There are absolutely tiny microprocessor kits ( like http://arduino.cc ) that would easily interface to a pressure sensor (while you are at it, why not throw some other sensors in there, like an accelerometer, a GPS module (even if its just for the accurate time and a log you can review later) and a digital compass..

You could do some neat stuff with all that.. You'd have everything you needed and then some for so called "Dead Reckoning"

Thanks... that's very interesting. Probably have to wait a year or two to see it on a $29.95 computer.

It depends.. There's always Moore's Law, which says that the amount of computing power available for a given dollar amount doubles every 18 months. That has pretty much held true for decades now. The same economics apply to many kinds of embedded devices.

You would be surprised at how little this cutting edge technology can end up costing someone. There is a huge creative movement to build "open source" hardware devices that started maybe five or six years ago and is now expanding fairly quickly. Also, there are a great many inexpensive, very low cost devices for sale for various purposes that are basically easy to modify into other things. Wireless routers are a case in point. Basically, they are just small, simple computers. Even designing your own dedicated device is now so easy that high school kids are doing it. There's a wealth of info on the net on how to do this.

Also, people who are building a lot of devices are publishing their experiences and even schematics and plans. Engineers and pro-level amateurs can and often do devote a great deal of time to designing products as parts of communities JUST like this one, making designs public so that anybody who wants to can build them or add functionality to them for the group. So you pay the cost of the parts only. Often one of the people involved will have a bunch of PC boards custom made (the more that get made, the cheaper it is)

Its now not uncommon to see quite sophisticated technology showing up in projects that are community built and developed. Not many people realize how technologies that used to be out of reach are now in reach. For example, designing PCBs, creating a parts list and PCB fabrication instructions, and soldering surface mount components to complicated multilayer PC boards can be done with an evolving set of tools, at home, things like a toaster oven connected to a digital thermostat and timer, an electric iron, heat guns, syringes, heat resistant polyamide tape, and so on.

So, there is no reason you can't build this dream device you want, quite possibly for not much more than $50. Some people in the UK are targeting $25 as the price for an innovative ARM-based pocket sized PC, see http://raspberrypi.org/
. Use that as the starting point. Its got a full complement of input output ports, graphics, etc. Its a real computer.

A good accessible example of what I'm talking about is the Arduino family of microprocessor boards.

http://Arduino.cc

You know, I am surprised that this board, which is clearly a very active, vibrant community, doesn't have one of its boards dedicated to its own open source electronics and mechanics hardware projects.

Things like your sophisticated bike computer, cutting edge bike-to-bike communications, data capture, maybe jigs to help analyze the mechanics of bikes and how it can be improved...

Aim high! Bicycle mechanics Orville and Wilbur Wright built a flying machine. It worked!

A few weeks ao I stumbled across a car data logging device for automobile fine tuning enthusiasts that bike people might be interested in looking at because it shows how these kinds of devices work in general. Plus, its a professional quality device whose plans are freely available to examine and learn from. Bicycles are cutting edge technology, a mechanical extension of the human body that offers immediate feedback in a way thats unique and totally unlike motor vehicles.

There is a lot going on that I'm sure its worthwhile to study in detail.

The car data logger is called the DAQPac, here is a description of it.
http://www.parallax.com/DAQPacAutomo...9/Default.aspx

There is enough info on the web (the schematic and functional description) to build a bike-specific version yourself.. without spending a huge amount of money.

It uses a Parallax Propeller CPU..

Just found what looks like a much more specific (cheaper) data recorder design. This contains a pitot tube part and might be useful for somebody.

http://www.scribd.com/doc/74006835/E...ronics-2011-01

>> page 28 -

you will also need to measure wind direction, so you can do the vector math to figure out the total strength of the wind, and not just the speed relative to your forward-pointing pitot tube.

Super nerd that I am, I'm building an airspeed (and altitude and vertical speed) device based on an Arduino, but it's not really made with bicycles in mind. It could be built with 3xAA for power if you wanted, but it'd be comparatively heavy and huge for bicycle use. See http://dangerpants.com/labs/adi for more information.

I hope you don't mind my asking - IMO, pressure sensors are really cool, if they are sensitive..

What are the lowest speeds that you get readings at? Also, do you think the one you use or do you know of any that could sense very subtle differences in pressure.. Maybe the equivalent of say, 20-30 feet of elevation difference?

Basically, I have always wanted to be able to stack effect and measure wind direction by difference in vectors. Most of the wind direction sensors I know of have some issue.

Also, why do you need the more expensive arduino, is the app particularly CPU intensive?

Actually, I just thought of another use for it, monitoring the flow velocity going both into and out of my HRV. I could integrate two matching pitot tubes as well as a DHT11 (temp humidity sensor) into the ducting (for the intake, so I can avoid sucking in super humid air when its raining or right afterward in the summer)

Having a matched pitot tube setup would be excellent for balancing the HRV.

Bicycle airspeed is extremely easy to measure. I used to fly an ultralight aircraft whose airspeed was measured by a cheap but accurate thirty dollar device. The device is pointed forward into the wind and the air pressure lifts a graduated marker a specified amount which can be calibrated to speed. Just google HALL WIND METER. Once you know your bicycle airspeed, you can calculate the amount of power you exert. You could use one of many impellar type wind speed measuring devices; just hold it into the wind while riding and get an accurate speed reading, or attach it to your bike in some way.

Not that I know of, but make one.

All you have to do is duct tape one of these to your handle bars and you're set. (The further out front, the more accurate it will be, being further from your turbulence-creating body.) And if these are too expensive, there's variants that are just a piece of clear plastic with a tube inside and a ball that is pushed up by the wind.

I toyed with the idea of building my own bike computer using an Arduino. I found an pitot tube airspeed kit, http://store.diydrones.com/Kit_MPXV7...v7002dp-01.htm, so maybe I'll renew my interest in that project. It would be a great way to rationalize a slow average speed on a ride.

Would a mass airflow sensor work? I know they tend to be analog and use voltage levels that the (automotive) computer coverts to digital data, so I don't know if you could actually jerry-rig something like that up.

Chris

I'm not sure what the 20 wildly speculative and inaccurate posts subsequent to Merlinextralight's post #4 are about, but his "Ibike" statement is the simple answer. Their Newton power meter not only outputs rider power, but wind speed, hill gradient, temperature, and a lot more in a small, bicycle specific unit.

Maybe you should have read them. You'd have found that many were not wildly speculative or inaccurate at all. Some veered somewhat off topic, but that's the nature of discussions.

(And if you want to get really picky, the correct answer was simply "Yes", and yet nobody said that, including yourself.)

Also note that one word -- Ibike -- while being a useful answer, doesn't actually look like a useful answer. It looks like somebody started typing and accidentally hit "Post" before they even got started. Your post was better -- at least now we know that Ibike is the name of a company that makes a product that does solve the problem, for the low low price of $499!

I'm pretty good at spotting a headwind on my Bicycle, without instrumentation.

Well now that we've staked out our roles here-- me as the illuminator, and you as the apologist-- let me continue by pointing out that one can buy a wind speed measuring iBike Pro on eBay right now for $100, and that there are a few other options as well with BINs of $200 or less, and a couple more yet that will certainly auction off at very well under $499, and I'd wager, well under $200.

Yes, I realize that if one is poor or cheap, that may be still be out of reach, and that they'd probably be happier speculating online about how to duct tape a McGuyvered pitot tube to a handlebar, but I don't presume that's the case, regardless of how temptingly logical an explanation it would seem.

Will I be an arse-hole my whole life, or just today? I'm hoping just today. :rolleyes:

Pointing out an error that one has made is not "apologizing".

As for our roles, we both seem to be rather ... paternalistic at this point. (Though your giving more details on what the word "IBike" actually meant was indeed illuminating.)

Those kinds of wind speed measurements each have an advantage and a disadvantage. The anerometer style would be able to measure total wind speeds in a 2D plane, so they would pick up the total effect that a side wind that was adding to the front wind might have. The propeller styled one is going to measure the effect in one single direction better. A propeller with a vane behind it to steer it into the wind would be good for gauging the effect of a side wind. Then you could also measure that angle..I've been wondering a lot about how to measure wind speed at low speeds. its easy at high speeds but almost all of the available methods have very big problems at low speeds. To get decent accuracy both for speed and direction at low speed (single digits MPH) it seems to me like you need a very light sensor but that same sensor is probably not going to be so great at higher speeds.. because the superior wind catching ability that serves it well at low speed is going to be a disadvantage at higher speeds.