If you are @ 1,000 RPM’s in 5th gear and your speedometer says 33 mph, then at 2,000 RPM’s in 5th gear, your speedo should say 66mph. Gears are linear. Correct?
Now, to the subject of speedo’s that are off. Tell me a little about wrong speedos. I have always thought that speedos are wrong at very low & very high speeds, but correct in the middle of the spectrum. Is this correct? How often are speedos off a lot at high speeds and perfect at low speeds and vice versa?
The reason I ask is that my Formula should be going 189.75 mph in 5th gear, according to the 33mph reading at 1,000 RPM’s in 5th gear. ( This can’t be right for lack of horsepower) However, the speedo reads 66mph at 2k RPM’s, and 99mph @ 3,000 RPM’s, and so on.
Doesn’t this congruency prove the speedo is correct or could the speedo be off the whole way down the spectrum by 5 mph or so? Does a bad speedo get progressively worse, or can it simply be off by 5 mph the whole way up the spectrum? Another way to ask: Normally, Is the speedo off by a certain percentage or by a certain speed, additively or multiplicatively?
Car & Driver did a piece on this a few months back, but I don’t have the issue handy to give you the full details, so I’ll wing it as best I can.
First of all, speedos are off at all speeds. The reason being is that various laws require the speedo manufacturers to have the speedo read a few miles higher than the car is actually going. (How off it is, has to do with the car’s country of origin.)
I’m not sure how more modern speedos work, I suspect it has to do with a magnet and an induction coil mounted on the output shaft, but I do know that at one time, speedos had a gear which meshed with another gear on the output shaft which was mated to the universal joints that connected to the driveshaft. So it wouldn’t matter what gear the car was in, the speedo’s getting it’s reading from (essentially) the driveshaft, and not the gears inside the transmission.
I had one car that used a different setup entirely, and I don’t know how it worked, but unless the car was moving, the speedo never registered a speed. (Of course, considering some of the cars I’ve owned, it could have just been the speedo was acting wonky.)
I don’t know if it was US law (the article as I recall claimed that US speedos were the most accurate) which required them to be off, but the article did claim it was law. Reason being, it’s to prevent some halfwit from driving his car in a reenactment of Death Race 2000, then claiming that he wasn’t at fault because his speedo registered a lower speed than he was actually going. So speedos are built to show the car going a few miles faster than it actually is.
That still doesn’t sound right to me. A speedometer is a measurement device. Higher or lower (doesn’t really matter), what governing body would allow a measurement device to be made purposely inaccurate? And how would a few miles help anyway? If a guy is going, say, 100, what would his inaccurate speedo say? 105? Big deal. I think who ever wrote that article needs to check their sources. The very idea that a speedometer be purposely manufactured to show an incorrect speed does not sound kosher to me.
I ran a couple of cars on a sppedo shop’s dyno for speedo calibration (we we going to race a car and we were calibrating an aux. speedo that we had installed.)
Anyway, the factory speedo was pretty much within 1 mph on all the speeds that we measured it all the way up to 100.
Considering parallax, I consider 1 mph to be pretty much dead nuts.
The speedo system on this car is of the magnetic pickup type.
Back in 1970 I owned an ex LA County Sheriff’s car. It still had the calibration sticker the windshield pillar. That speedo was between 1-3 mph off at various speeds IIRC. This unit was a cable type driven off the transmision.
In general error will change as the speed of the car increases, because as the tire rotates faster, it grows in diameter. Different brand tires, will grow by different amounts. Makes keeping a speedo 100% accurate a tough deal
I’d like to see that cite too. Correct me if I’m wrong, but aren’t the odometer and the speedometers linked somehow? If the speedometer was purposely made to say you’re going 105 when you’re actually going 100 wouldn’t that also mean my odometer is off by 5%? Which would mean my 36,000 mile warranty should actually last me until 37,800.
I think what Tuckerfan meant is that it’s illegal to have a speedometer that reads lower than the true value. No measuring device is 100% accurate and stable, so the common practice is to set it slightly high.
Though you’re right, a few mph shouldn’t matter for anyone other than people seeking official records, and those people won’t rely on a speedometer anyway.
I’ve heard that some motorcycle riders install Sigma bicycle speedometers. Those are based on magnets and reed switches, and can be calibrated by the user.
Well, all I have ever heard is that speedos are wrong at high speeds for performance vehicles, for whatever reason. When I got to 145 on my gsx-r (on the speedo), everyone told me it was 130, and now that my bro got to 190 (calculated), everyone is saying it’s 165-175. I am just trying to determine how fast we were going and why the speedos always seem to err on the fast side on fast vehicles???
Think about it: If indeed my brother was only going 165-175mph instead of the implied 190, then wouldn’t it be off by 15-25 mph when reading 33mph (which is ludicrous)? There was no indication of a multiplicative-type error because the speedo was congruent with the linear gears, like I showed in the OP. Can there be random errors with no simple pattern?
Can someone please tell me what’s going on and how exactly speedos work!? Howstuffworks does not help…
You seem to be saying that your tachometer reading is not proportional to the speedometer reading even in the same gear. What makes you think it’s the speedometer that’s inaccurate? I doubt there is any law regulating the accuracy of the tachometer.
Modern speedometers are electronic devices, as stated in Tuckerfan’s article. A magnet is attached to some part of the drivetrain that turns with the wheel (or one part of the component is magnetized). The magnet passes in front of a nearby sensor once every rotation, and a digital circuit counts the number of pulses per second. So these devices should be perfectly linear. If they are off they are off by a fixed ratio, e.g. always reading 2% higher than the true value. So if the tach readout is perfectly proportional to the speedo readout, that doesn’t prove that the speedo is accurate.
If you’re so concerned about speedometer accuracy, invest in a GPS receiver. They are as accurate as anything you can find. The disadvantage is that the speed readout is averaged over some time range - about one second in the units I’ve used.
uH FUEL. I don’t see how gears are exactly quite linear. The higher the rpm’s the more loss through friction theres going to be. I suspect that your 5k rpm x 33 = 165 won’t hold water.
That doesn’t sound right. Engine RPM should be proportional to speed unless something in the drivetrain is slipping. Gears don’t slip. Automatic transmissions do slip, but I don’t think clutches do. Actually, even automatic transmissions have lock-up torque converters nowadays.
Fuel, if your speedo is 15% low (for instance), what that would mean is 33mph on speedo = 28mph real. 66mph speedo = 56mph real, and 190 speed = 161 real. It maintains the same ratio throughout, that would explain your situation, only 15% is a big error for a speedo. Because of how they work, it is unlikely that you will be off by XXmph, but you will be off by XX% instead.
One possibility is that your tires are not the correct size, if they are too small, each turn of the driveshaft will push you a smaller distance, but the speedo is calibrated for a larger distance per turn.
Yes some do as motorcycle speedo’s tend to be off. I also know a number of people who have one for distance and average time. I use my GPS which does the same thing except that sometimes it can give really funky readings, like the time I was doing 150 in downtown NYC.
The speedo on my bike, and most that I have seen is driven off the front wheel. I think my Escort is driven off the transmission somehow but I’m not 100% sure.
Sorry I can’t remember the specifics (or find the article) but I do remember reading in either * Sport Rider* or * Superbike * that the speedo on the newer Gsx-r’s were reading higher than actual (more than other bikes). I think the article was specific to the 02’s but as I said I don’t remember all the details.