I hardly look at the speedo...get a calculator for rpm speed that uses your gearing and tire size...I just use the tach.
Mines about 6 mph off and everyone with an SC that I know says about the same.

I had mine rebuilt, the gear broke inside. They also 'calibrated' it.
I replaced the funky reed switch in the tranny at the same time.
It works but it's off by about 5 mph. Who knows why.

If you drive with the traffic your odds of getting a ticket are low.
If you get something like a 9500ix and radar is in the area, it will
show you your speed. The 9500ix is an expensive investment
but with its built in memory, update capabilities, accessory"live" mapping on
your droid or iphone... we'll its a pisser!
Can't beat it and less false alarms than the Valentine.
One of the best investments I've made besides wide band AFR display.

an optimistic speedo is the cheapest way to get performance out of the car. :D

When I fixed my odometer gear, I used a gps to calibrate the needle, making small adjustments until they agreed.

Smart idea!
What did you adjust on the board?

Not sure what your question is.

I read an article (not Porsche specific) that claimed this was because of variations in actual tire dimensions from different manufacturers and because of variations due to under/over-inflating the tires.

Mine reads way fast and gets near accurate at high speed.........Like someone said...they are all off........maybe just live with it......

If I did this math correctly, then this example of using a 16-in tire on a speedo calibrated for a 15-in illustrates how significant the difference can be.

A 15-in tire has a theoretical circumference of 47.1 inches.
A 16-in tire has a theoretical circumference of 50.24 inches.

If the speedo is calibrated to a certain number of revolutions, then:

800 revolutions
15 in: 47.1 * 800 = 37,680 in (total distance traveled)
16 in: 50.24 * 800 = 40,192 in
delta = 2512 in (6.7%)

1200 revolutions
15 in: 47.1 * 1200 = 56,520 in (total distance traveled)
16 in: 50.24 * 1200 = 60,288 in
delta = 3768 in (6.7%)

1600 revolutions
15 in: 47.1 * 1600 = 75,360 in (total distance traveled)
16 in: 50.24 * 1600 = 80,384 in
delta = 5024 in (6.7%)

Oh and a 6.7% calibration difference at 20mph would register a difference of about 1.34mph and at 60mph would be a difference of about 4mph. I think this is consistent with what most people see as the difference between real and actual.

15 inch tire? 16 inch tire? Those are rim diameters, not tire diameters The rim diameter has nothing to do with the outside diameter of the tire, and the circumference thereof. The only thing that matters is the rolling diameter of the tire, which is 25.0 " for the most popular tire sizes for SC or Carrera, 205/55/16 and 225/50/16.

From all my experience and what I've seen reported on this board for the last 9 years, the typical difference is twice what you suggested.

How did you "adjust the speedo".
What procedure?

My SC is off by almost 9%. At 80MPH indicated my GPS will read 73MPH. Must by a German thing. I also have a BMW R1200RT and an R60/6. Both of those speedos are also optimistic.

I believe the German government has VERY BIG FINES on a manufacturer if any of his cars are found to UNDER report speed. Thus they design the stock speedo to OVER report speed by 5% or so. Not sure when this practice started.

You are correct. However, since I didn't have the actual outside measurement handy and I was too lazy to look it up, I just used those simple numbers. The actual numbers don't matter for the sake of illustrating the point that a larger tire circumference makes a difference in distance traveled for the same number of revolutions of two different tire diameters and mechanical speedos are calibrated to a number of revolutions/unit time.

Seems like I saw a post on here once (or maybe I read it somewhere else, I don't recall) on how to calibrate your speedo using an electric drill and a 12v power source.

Thinking out loud here, but what happens if someone changes the gear ratios in the transmission. Does the final gear ratio change at all?

I thought about the wheel size as well. Although my 81SC came from the factory with 16-inch wheels, the PO or dealer changed my speedo - I think my car came with the idiotic 85 MPH speedo. What I have often thought about was when the speedo was replaced, were there two speedos available? A 15-inch version and a 16-inch version, each calibrated for the appropriate wheel size? It just seems like my speedo is calibrated to 15-inch wheels, not the 16's I have.

I wonder how NHS "calibrate" their speedos? Is it done on a bench? If it is, how do they know if they are calibrating for a 15-inch wheel or a 16-inch wheel, because when they "calibrated" mine, they did not ask, although they had my car at their shop. I simply assumed they used the dyno rollers because I saw them there. That is why I was quite surprised that after my speedo was calibrated, it still registered higher speeds than I know I am going. And for me, I just didn't like the fact that my mileage is probably higher as a result of the incorrect speeds.

I looked for it a little while ago and couldn't find it. I think it was a guy sitting at his kitchen table with a cordless drill and a car battery. He had the drill hooked up to the speedo cable somehow and I can't remember what he used to measure the RPMs.

According to one source I found says the US requires something like no more than 1.5% variation. About 5% seems pretty common so it is kinda interesting that so many people are seeing greater variation than that.

Not too much you can do about it though. Send it off to a shop that knows what they're doing or live with it. If I ever send off my gauges I would have it done, until then I'll live with it. I keep the Escort app running on my smartphone when it counts the most anyway. :)