Speedo error... cause?

budgetzagato

budgetzagato

Administrator
Moderator
My speedo has always been off a fair amount, didn't just how much until recently when I finally upgraded to a smartphone. Using a GPS based speedometer app, I found I am off by about 14%. At an indicated 70mph, the app displays 60.

This is on stock wheels with 205/60R13 tires. Given that the stock tires size is 185/70R13, my speedo error compared to stock should be closer to 1%.

So, what causes so much error? Could the transaxle be from something else? It sure looks stock, and had the studs on top for the smog stuff Scorpions had (until I removed them). Maybe the speedo drive gear is incorrect?

Thoughts and ideas welcome.

(cross-posted from Lancisti)
 
I would think the biggest source of error is in the speedometer head (spring constant, strength of the magnet etc.) which can be calibrated. If you compare the odometer to a GPS distance reading you'll probably find that the odometer is much more accurate than the speedometer, meaning the drive gear ratio at the differential is pretty close to what it should be.
 
Greg,

I hope you find an answer (I don't have one), but you are not alone. On my recent trip I compared my GPS speed to the speedometer and there was always a difference. At least mine matches yours with the 70/60 you describe. Mine is the same at that speed.
 
The Queen's speedo is off and indicates 65 * 70 mph. Also, Matt Brannon told me that the tack was slow.:eek:mg:
 
Bjorn,
Do you know more about calibration of the speedo? Is it something we can do on our workbenches?
 
When I switched from an '85mph speedo to a 140 it was off, had to fiddle with it a lot, changing starting needle position and such, eventually I got it pretty darn accurate but I believe I ended up just putting in another 140, they must vary quite a bit. You can send 'em off for calibration but it's pricey (to my budget) and you need to address so many factors (tire size and all that) I don't know where you'd end up, maybe be better, may be worse. I tried swapping around the guts, the cups and magnets and such. They are pretty fiddly and sensitive little devices, simple, but sensitive.
 
Italian gauges are relaxed and approximate. It's just Italian... You want 50 when it says 50? Ahhh fahgettaboutit. :italia:

That being said, since having the access to a gps speedo, I've noticed most cars run about 5% optimistic on the GPS. The gauges in the 2012 500 Pop I owned were almost laughable. With stock diameter tires, the GPS would show 65mph, the Speedo dial would show 71mph, and the EVIC digital readout would show 68mph. Funnier yet, the digital clock ran 1 minute slow per month! How in the world in this day and age do you screw up a digital clock? LOL

My 500T is better, it shows about 5% error, with the needle only being 1% faster than the digital readout, and the clock takes 6 months to lose a minute.

As for X1/9, the '78 speedo is dead on now after changing rear tires to 205/50-15. If I recall that was a 3% diameter increase.
 
funny, the clock in my X is spot on and never gets off (although it seems the battery gets disconnected for one reason or another every few weeks so it gets reset a lot!, I still mean to do the Bob Brown mod to make it read 12 hour instead of 24)
clock in my old Opel was mechanical with electric rewind, I loved repairing that! Once done, it was also spot on.
 
All new cars do this...

zonker said:
That being said, since having the access to a gps speedo, I've noticed most cars run about 5% optimistic on the GPS.
Click to expand...

And it's not an accident. The reason, you ask? All the better to expire the warranty a few percent sooner, my dearie! Believe me, the bean counters are on to that trick and have been for decades.

Pete
 
Someone (here no doubt) said it's law that speedos must not read low by only a tiny percentage, but they may read high by a bigger margin, the theory (I guess) is it's safer to be going slower than you think, rather than faster than you think, speedo and odometer aren't necessarily related though, the eddy current speedo can tell you anything while the odo could still be accurate, right?
 
The speedometer error consists of mainly of a fixed mph error and a percentage error (offset and gain). The mph error (offset) is calibrated out simply by repositioning the needle on the shaft. For the percentage error (gain) you need to change the effective spring constant of the return spring or the strength of the rotating magnet. I have never done this, and don't know what the "proper" way is, but looking at the guts of the speedometer:

DSCN1167M.jpg


If you were to un-solder the spring from the tab then pull out it out a bit and re-solder it, you would have increased the spring constant, so less movement of the needle for the same actual speed.

I would start with a test drive at, say, 20 mph (as indicated by the speedometer) and take a GPS speed reading. Then do the same at 60 mph (again as indicated by the speedometer) and take another GPS reading. That is a 40 mph difference indicated on the speedometer, and probably something less than a 40 mph difference in the GPS readings. Using the difference between two readings eliminates the fixed mph error (offset) and allows you to calibrate the percentage error by itself. Tweak the spring and do another test drive. Repeat until you are happy with the accuracy of the difference readings. Finally reposition the needle to remove and fixed mph error.

With all the hassle of disassembly, test drive, reassembly and repeat, it might be worth it so pay a speedometer shop to do it.
 
Last edited:
Yes and no...

Yes and no... the odo works off of the same cable so it will be as accurate as the rpm are, so tire size is the variable.
 
Makes sense...

Makes sense, just need to prioritize it in the big scheme of things. I can think of worse winter projects.
 
budgetzagato said:
Yes and no... the odo works off of the same cable so it will be as accurate as the rpm are, so tire size is the variable.
Click to expand...

I guess the cable input can be good but the speedo mechanism itself could develop issues: magnetism "loss"? spring lose springiness? wear on needle shaft, binding on shaft, misplaced needle, cup/shaft alignment, the thing is just fiddly,
 
You must log in or register to reply here.
Back
Top