23 replies to this topic

[alpder]

If anyone has a working 90mph 1408tpm centre dial going spare, please get in touch...

 

My NOS 18T "red" cog has arrived. Curiously it's in a Rover packet marked DAM8079 and listed for a Metro (would be "22G2024" for a Mini one) but apart from a small "circlip" type groove on the lower end of the shaft it measures-up the same as a regular Mini cog and the tooth spiral looks like the same angle. So looks compatible. Until I try to fit it, naturally.

 

However, before I have a stab at fitting it, I realise it makes more sense just to change my speedo for an identical 1408tpm model but one which works properly. Then I'll - in theory - have both a speedo and an odo registering correctly without changing anything else. Which would save me a day or more of tearing my hair out trying to get at the bottom of the gearbox through the driveshaft hole.

Edited by alpder, 12 June 2025 - 06:15 PM.

[alpder]

Acquired a cheap/faulty 1408tpm speedo. Dismantled and washed all the little plastic cogs in its jammed odometer without disturbing the speedo parts.

 

On-the-road testing shows it matches the satnav speed almost exactly at every speed from 20mph to 60mph. That's that problem solved. Whether this faulty odo will last the course is another matter - it has at least got into double-digits without jamming. Yet.

 

Sadly the fuel gauge reads 'pessimistic' but I guess that's easily fixed using the gauge and/or voltage regulator from the old one.

 

Edit (see also edited post below): it now doesn't look like my car is geared the way I thought I'd built it to be. So the following data are probably garbage. I'll check the car's gearing later: Anyway, for others' reference, the important numbers are:

 

An old, believed never re-calibrated, 1408tpm 90mph Smiths speedo reads pretty much bang-on with brand new 145/80R10 - 3.105FD - 7/16 speedo gears. (It should, in theory, read 2% over, plus whatever margin of over-read that Smiths built into it in the first place. The reduced actual reading may be due to slight fade of the magnet.)

Edited by alpder, 19 June 2025 - 10:44 AM.

[Ethel]

The milometer  should settle that.

[alpder]

The milometer  should settle that.

This odo, compared to Back Country Navigator, also reads 12% low just like the old one did. I guess BCN is wrong, not the odo's. I'll check against satnav some other way, to be sure.

 

Edit: I've now checked against BCN, DigiHUD, and a TomTom satnav. The speedo is perfect. The odo reads 11.5% under. Next recourse is to count the teeth on the odo's gears and work out what Smiths actually built.

 

Edit2: Turns out I've been talking cobblers (surprise!): I've counted the teeth on the odo, and - at first and second counts with my old eyes - it's geared for 1408tpm. Assuming a recount with a bigger magnifying glass doesn't throw up a completely different number, there are now only three realistic possibilities:

• I opted to fit a 2.76 diff instead of the 3.1 I originally decided to fit, and forgot to tell myself about it. Oh, I do hope not.
• I didn't fit 7/16 speedo gears. Hopefully this is it.
• Both odo's are skipping 12% of the time. Hardly likely.

Sorry everyone!

Edited by alpder, 19 June 2025 - 10:50 AM.

[timmy850]

It is possible to calibrate the speed and odometer separately

I sent my lovely 140kph speedo away for a repair and it came back with an incorrect calibration of speed in mph and odometer in km! That was a frustrating mistake

[Ethel]

Odometers are direct drives. I can't say for certain, but it would make sense to gear the speedo itself for km/miles & adapt for final drive ratios & tyre circumference with the drive gears. There's obviously going to be a limited number of gear ratios available for both ends of the cable, so an inevitable amount of inaccuracy for at least some applications.

 

It'd be interesting to know how satnav does distance & speed - if it's from the coordinates alone. Your position could be sampled anything from every 1 to 15 seconds, so it could cut out a lot curve in point to point calculations - more the faster you go.                                                                                                                                    

[Spider]

It'd be interesting to know how satnav does distance & speed - if it's from the coordinates alone. Your position could be sampled anything from every 1 to 15 seconds, so it could cut out a lot curve in point to point calculations - more the faster you go.                                                                                                                                    

 

I don't know the intimacy of how SatNav works, but for a location it needs to 'see' at least 3 Satellites and for speed measurement, 4. That seems to give a refresh rate of around 1 second, however the more Satellites the unit 'sees' the more accurate and quicker it becomes. The system that the surveyors use, that gets to +/- 30 mm also uses the Cell Phone Towers as well as Satellites.

 

When I say 'see' the satellites, that's pretty much exactly what the GPS unit needs. Clouds, Trees, Buildings,etc do interfere with it and at the very least, slow it down substantially, I've seen 5 - 10 second refreshes. Certainly Tunnels, Underground Carparks etc wipe it off all together.
 

[Ethel]

I'd think it's more down to the hardware in your car that the satellites & the potential accuracy. If its main purpose is to position you on a 2D map   it wouldn't have to bother with the extra complexity of an extra dimension. You might get a clue by disobeying its instructions and seeing how quickly it cottons on

[Spider]

My off road GPS is '3D' as in it will not only give a location to within 3 metres, but also altitude. Some of the street GPS's too (and el-cheapo ones from the supermarkets) also give altitude.

 

 

 

I think we're getting a bit off topic and sorry for that,,,