Speedometer calibration

[lambojohn]

Does anyone know what RPM of the speedometer cable equates to speedo mph. This would enable calibration of the speedo out of the scooter.

[Warkton Tornado No.1]

Presumably you mean standard Lambretta type, not, Smiths, SIP etc. If so, I don't recall it ever being specified anywhere, but I 'get' why anybody would want to know.

The information you seek will probably be closely guarded by those in the know

Having got this far in asking the question, don't be thwarted by a lack of response, as it can be calculated using the rolling circumference of a wheel & the worm wheel : worm ratio

This might help on the circumference:

http://www.csgnetwork.com/tiresizescalc.html

If you don't have access to all of the 'numbers' let me know & I'll count the teeth on the speedo drive.

[dickie]

I stumbled across the answer within the last week. Either on here or the other unmentionable one. It was specified in revolutions per kilometre and I'm pretty confident that it was 1000 cable revs per kilometre. I'm pissing around with my speedo at the moment so I'll measure it on mine either tonight or tomorrow.

[dickie]

I stumbled across the answer within the last week. Either on here or the other unmentionable one. It was specified in revolutions per kilometre and I'm pretty confident that it was 1000 cable revs per kilometre. I'm pissing around with my speedo at the moment so I'll measure it on mine either tonight or tomorrow.

I've just remembered I found it in an old 70s Lambretta book I have. I'll check tonight.

[drivera]

There is a simple way of doing it , get someone to drive a car in front of you at a set speed ( say 40 ) , look at what the speedo is actually reading and remember it , then after taking the speedo unit out take the glass out and mark on the speedo housing where 40 is and what the speedo actually read , take out the speedo face and turn the pointer needle to what speed you read , using fine needle pliers hold the spindle under the pointer tightly then carefully move the pointer to 40 , reassemble and you should have a fairly accurate speedo .
I saw this in practical classics car mag and have uses it on mine , it works

[dickie]

1610 revolutions per mile. Which is 1000 revolutions per kilometre.

I'm sure you can do whatever maths you like based on that.

[Fast n Furious]

I concur......... 1610 revs per mile.

[lambojohn]

So 1610 revs per mile
1 mile per minute=60mph
60mph =1610rpm
30mph = 805rpm
All agreed?

[Warkton Tornado No.1]

Post by lambojohn » Tue Jan 24, 2017 6:39 pm

So 1610 revs per mile
1 mile per minute=60mph
60mph =1610rpm
30mph = 805rpm
All agreed?

Not quite. Based on one mile divided by Rolling Circumference (RC) & a worm wheel/worm ratio of 13/10, my calculations result in something different....

One mile = 1760 yards
Multiply by 36 to obtain inches = 63,360”

The theoretical* RC of a 3.50 x 10” tyre is 53.407”

Therefore, the number of revolutions the wheel will make in one mile is 1,186

So, if the machine travels @ 30 mph for one hour, there will be 30 x 1,186 wheel revolutions in one hour = 35,580

However, the speedo cable rotates @ 13/10 times faster. 2967 x 1.3 = 46,254 RPH

Divide that by 60 to obtain RPM.

771 RPM @ 30 mph

Or 1,000 RPM @ 39 mph

* The only assumptions I’ve made is whether my maths is right & the RC (Rolling Circumference) which should be measured...

I hope this helps.

[Fast n Furious]

Post by lambojohn » Tue Jan 24, 2017 6:39 pm

So 1610 revs per mile
1 mile per minute=60mph
60mph =1610rpm
30mph = 805rpm
All agreed?

Not quite. Based on one mile divided by Rolling Circumference (RC) & a worm wheel/worm ratio of 13/10, my calculations result in something different....

One mile = 1760 yards
Multiply by 36 to obtain inches = 63,360”

The theoretical* RC of a 3.50 x 10” tyre is 53.407”

Therefore, the number of revolutions the wheel will make in one mile is 1,186

So, if the machine travels @ 30 mph for one hour, there will be 30 x 1,186 wheel revolutions in one hour = 35,580

However, the speedo cable rotates @ 13/10 times faster. 2967 x 1.3 = 46,254 RPH

Divide that by 60 to obtain RPM.

771 RPM @ 30 mph

Or 1,000 RPM @ 39 mph

* The only assumptions I’ve made is whether my maths is right & the RC (Rolling Circumference) which should be measured...

I hope this helps.

Your calcs work out the Revs per mile for the speedo cable to be 1542 r/mile.
The true rolling circumference would of course be less due to tyre / road compression.
So the calculated rotation of 53.4 inches / rotation, would only need to be reduced to just over 51 inches to give the Innocenti produced figure of 1610 revs per mile.
Measuring the true RC is a conundrum.

[Warkton Tornado No.1]

Post by Fast n Furious » Wed Jan 25, 2017 3:48 am

Your calcs work out the Revs per mile for the speedo cable to be 1542 r/mile.
The true rolling circumference would of course be less due to tyre / road compression.
So the calculated rotation of 53.4 inches / rotation, would only need to be reduced to just over 51 inches to give the Innocenti produced figure of 1610 revs per mile.
Measuring the true RC is a conundrum.

Somehow, I cannot imagine there being as much as 2.4" variation between the theoretical & measured (actual) RC.

The other thing to bear in mind is the UK law concerning speedometer accuracy, although I don't know how long it has been in force. Without checking, I believe that current legislation dictates that for (@ least) current vehicles the speedometer MUST not read slow but may read up to 10% fast.

If my calculations are anything like correct, the implication is that Lambretta based scooter speedometers were set up to read SLOWER than the actual speed, but we are discussing an object last produced in Europe as OEM fitting some 40 odd years ago.....

[CHRIS in MARGATE]

No don't... You'll have everybody saying 60 is the new 70

[Raveydavey]

There is a simple way of doing it , get someone to drive a car in front of you at a set speed ( say 40 ) , look at what the speedo is actually reading and remember it , then after taking the speedo unit out take the glass out and mark on the speedo housing where 40 is and what the speedo actually read , take out the speedo face and turn the pointer needle to what speed you read , using fine needle pliers hold the spindle under the pointer tightly then carefully move the pointer to 40 , reassemble and you should have a fairly accurate speedo .
I saw this in practical classics car mag and have uses it on mine , it works

Brilliantly simple. At least you can be fairly certain at least one speed reading is correct.

[Fast n Furious]

We should bear in mind here that the issue is about the revs per mile of the odometer input drive. Not revs per mile of the speedo drive output connection at the wheel.
Some compensation will have been considered to ensure that the speedo reads higher than true to meet construction and use rules as pointed out.

[Warkton Tornado No.1]

It would be interesting to see where this concept goes.

By now, most of us will be aware that car speedometers are optimistic & that GPS speed readings are the only ones to be relied upon (on the flat, that is...)

A calibrated speed of a particular known RPM to the speedo on the bench should establish accuracy @ one speed.

I'd probably opt for 45 mph as travelling most of the time will aim to be in a range of 15 mph either side of that.

[roli150lam]

There is a simple way of doing it , get someone to drive a car in front of you at a set speed ( say 40 ) , look at what the speedo is actually reading and remember it , then after taking the speedo unit out take the glass out and mark on the speedo housing where 40 is and what the speedo actually read , take out the speedo face and turn the pointer needle to what speed you read , using fine needle pliers hold the spindle under the pointer tightly then carefully move the pointer to 40 , reassemble and you should have a fairly accurate speedo .
I saw this in practical classics car mag and have uses it on mine , it works

There are very few if any family cars that have a speedo that reads true.
A better idea would be to have a truck in front at an agreed mph.Do this immediately after said truck has had the bi-annual tacho calibration.there is then a small chance for an accurate reading.

[rossclark]

From my observation my car speedo reads about 3mph over the flashing signs with 'your speed' and the same for GPS - I assume this is to keep you under posted limits.

I've also seen arguments by senior police officers that's suggest the 10% discretionary allowance on measured speed, i.e. beyond the posted limit, where there should be a prosecution, should be scrapped, as the indicated speed on the dash is higher than the limit and the driver should see this and therefore are knowingly exceeding the limit.

[solostax]

Admire the maths from WTno1 and F&Fall , dickie for coming up with the Lambretta data ,driveria for a practical method ,Roli150lam for a finer definition and all contributors .But In considering the variables that creep into and cause differences to the theoretical , it’s a wonder the old speedos get anywhere near to the truth.

Being prompted to investigate why there was so much indicated speed variation between all my different speedos that on test, were being fed the same constant source of cable rpm,I discovered that within the actual mechanism there are so many possible factors.

Because there is no direct mechanical link from the spinning speedo cable to the actual pointer,subtle pointer needle bearing wear ,.electro-mechanical alignment of the spinning magnetic components and resistance spring properties can make a significant change from the theoretical to the actual.This then is part 1 of the problem.
Then as F&F has pointed out the centrifugal force exerted on the tyres at differing speeds will alter the circumference and wouldn't be surprised to learn ambient temperature might also be a factor.This part 2 combined might well justify the instrument to be renamed a “Guessometer”

Considering the problem and solution from a pragmatic angle ,surely an individual calibration using two solid references, that is time and distance could be used .This is the method I have used in the past to form a deviation chart( much like a boats compass ) spanning the range 30 to 50 MPH.

I am lucky to have in my vicinity a 1st class unbroken 8 mile dual carriageway. Maintaining a constant speed I measured the time between two known points ,noting the indicated speed .Furthermore,using stepped up velocities I track more runs giving me a fair idea of correction necessary to apply to any actual speedo display.
These days the obvious alternative to stopwatch and travelling distances at constant speed calibration is GPS . GPSversus Veglia ? ! I did enjoy the challenge of maintaining constant speed exercises.So gone now is a valid reason ,that had a practical purpose, to go for a ride.
Cheers