Calibrate that cheap speedometer

I took my last bike that had a speedo out to the local 1/4 mile track and made adjustments every 4 laps until speedo was perfect.

Ya I did the same thing with the measuring tape and contverting it to mm. It wasn't dead on after that but within .3 mph to .5 depending on speed matched to my phone gps.

I just recently matched my F-12 with the gps and I had the same findings as ToxicAZ.

they are all fairly close , so i never bothered with, it i just set it for the wheel size and went , i am more into mileage than MPH .

oh man! that's some great advice. i didn't realize that's what was being measured, i've used four different ones so far, and settled on the last setting when it seemed to feel right distance-wise around town. but now i'm going to give this a shot and see if i come up with something better.

I just got a Bell wireless speedo at wallyworld. Don't know if its the f12 or whatever, but it was the only wireless they had. I programmed it the other day, using the given 2073. I have a gps tool on my phone that allows me to see what speed I travel. Today I installed the speedo and checked it against the gps tool, and it was dead on! I know everyone won't have the same results I did, but it worked for me. So before measuring and reprogramming that speedo, find a way to check its accuracy against something you know to be accurate.

Guys, its all very well to start being more precise but before being precise you have to ask yourself if you've thought of everything.
1. One size does not fit all. A new tire has more tread than a used one. Not so important for racing tires with hardly any tread but on those thick studded moutainbike ones it makes a huge difference so even if you could blindly trust (more on that later) the original theoretical value given by the manufacturer, you would regularly have to remesure and set the value for wear.
2. Going round an athletics track in circles is by no means accurate, firstly usually it is a soft material which gives way as your tire pushes into it, this lengthens the travel as the wheel turns (very tough maths to calculate by how much). Secondly, the 400m lap is only along a specific line at a specific distance from the lane separation lines, so unless you're driving perfectly steady and straight (tough going round the bend), you're not doing 400m, it may even be safer to just do 100m straight, at least you can stay +/- straight and every line is the same distance.
3. A GPS can help but for it to be more accurate than the old mark the pavement and let the wheel turn 5-10 times trick you need to take your GPS on a nice long stretch with no forests, bends, hills or potholes and ride as long as you can counting the revolutions of the wheel (or setting your speedometer a few times until both devices correspond). Don't bother doing this unless you can do it over 10000m at least. Your GPS may be 50m out which is 0.5% on 10k, that same percentage on measuring 1 wheel revolution would be over 1cm and over 10 wheel revolutions 10cm. Any 25 or 30m tape measure will be accurate to within 1 max. 2cm over that distance.
4. DO NOT rely 100% on internet info and tables or manufacturers info, double check yourself. It seems a few people started some differing tables and since half the world follows one side and the other half are following the other. For my 700x23c tires you will find dozens of sites claiming and discussing the circumference being around 209.6 or 209.7cm whilst dozens of others claim it is 213.3, obviously something went wrong somewhere along the line a long time ago. The right distance for that tire is not 213.3cm. Anyone who goes out and test his tires and comes up with that value should be locked in prison for being stone drunk on a public road.