Re: [MV] M series speedo calibration

Richard Notton (Richard@fv623.demon.co.uk)
Mon, 5 Apr 1999 09:27:33 +0100

-----Original Message-----
From: Mark C. Johnson <mcj115@psu.edu>
To: mil-veh@skylee.com <mil-veh@skylee.com>
Date: 05 April 1999 01:45
Subject: [MV] M series speedo calibration

Hi,

>
> I am using as a temporary fix in one of my vehicles a M series
>speedometer. The problem is that it registers an estimated 7 or 8%
higher
>than what the speed is really.
>
You cannot really begin to consider accuracy without a calibration
reference, how is exact road speed to be measured ? Using another
vehicle as a pace car is not a true reference.

Why is the over-reading estimated ? Indicated speed minus true speed,
divided by true speed x 100 = percentage high.

>My question, Is there a way to calibrate the speedometer?
>
The reading is dependant on the drive ratio of the speedo drive gear
arrangement designed by the manufacturer and the rolling circumference
of the tyres, the difference between new and worn tyres could account
for several percent.

Apart from some very old devices and modern digital things all
speedometers operate on electrical principles (surprisingly) using eddy
currents. The drive cable rotates a bar magnet inside a metallic U
shaped cup (usually alloy) concentric with the magnet but not touching
it and connected to the pointer on a short shaft with a spiral return
spring.

The rotating magnet creates current in the cup like a small generator
but as the cup represents a short circuit this makes a magnetic field
opposing the magnet and the cup is dragged around, with the pointer,
against the force of the return spring until the opposing forces are
equal. A mechanically driven rev counter is exactly the same.

> And if so how is it simple enough that a person could do it them self?
And lastly >how does one go about calibrating it?
>
The insides of a speedometer are much like a watch so if you feel happy
taking a watch apart go at it. Should the error be _proven_ to be + x%
throughout the whole speed range then the spiral hair return spring can
be adjusted to apply a little more force, usually the fixed anchorage is
a concentric friction locked arm. The pointer is commonly only a push
fit on the fine drag cup shaft and could be re-positioned.

Without some sort of jig or fixture providing an accurate and calibrated
drive on the bench though I'd suggest it will be going it in and out of
the vehicle like a fiddlers elbow whilst you iteratively adjust it.

All these internal items are very delicate however and you stand a good
chance of wrecking it.

It would be surprising if the error is linear, the principle used is not
noted for inherent accuracy and manufacturers often build in a small
amount of speed flattery to enhance their vehicle product and alleviate
some speeding tickets, 10% at the top end isn't unusual with just a few
percent at 30mph usually, its largely down to where they print the
numbers on the dial.

In the UK, traffic laws allow +10% on speed to account for instrument
inaccuracy and surprisingly you are required only to have ".....a device
to show when a speed of 30 mph is exceeded." There is no requirement to
be able to measure any other speed or that the "device" shall be visible
to the driver.

Regards,

Richard
(Southampton UK)

===
To unsubscribe from the mil-veh mailing list, send the single word
UNSUBSCRIBE in the body of a message to <mil-veh-request@skylee.com>.