My TJ1 speedo vs GPS indicated speed fluctuates wildly.
When speedo reads 80kmh, GPS says 73.
When speedo says 100kmh, GPS says 96.
When speedo reads 110kmh, gps says 103.

Thought Id throw it to the crowd, see if that's normal..Ive chosen 100kmh as the benchmark for consistency.

Format=
Speedo kmh= 100
GPS says=
RPM=
Gearbox=

Mine would be
Speedo=100
GPS says 96
RPM= 2000
Gearbox = 4SP non tippy auto

Sorry, haven't done the GPS vs speedo check.

Not uncommon for the Magna speedometer to be generous in it's readout. 10% over is within specs, I believe.
I don't mind if it reads 110 kph at an actual 103 kph, gives room for accidentally going over the speed limit and still avoid a ticket.

Most rural highways have km markers on them, maybe check with those?

All vehicle manufacturers fit speedos that indicate a higher speed than what you are actually doing. As you say, indicating 80 kmh when your true road speed is 73 kmh. This potentially stops any claims against car makers when you get booked for speeding. If I remember correctly the speedo will always show up to 8% higher than road speed. If anyone has a more recent or accurate number please let us know. While GPS's are far more accurate they are not 100% either. Depends on how many satellites are being accessed at the time. My 01 TJ Advance 4 speed auto is doing 95 kmh when the speedo is indicating 100.

When my speedo reads out 105km. My gps reads out between 101km and 100km fluctuating

Do what I did on the 380... Find a tyre size calculator and see if you can legally go up a size on your current rims, and fix your problem.

Success will depend on your current rim size and suspension setup.

I took my stock 215/55/17 Dunlops which were dead at 43,000km and replaced them with 225/55/17 Bridgestones.

Heaps more grip (more width AND length of tyre on the ground, plus better rubber) and speedo went from indicating 60 at a true 55, to indicating exactly 60 at true 60. All speeds up to 100 it is within 1.

Not so many people driving past me anymore. Plus better for the environment.

Although this may be comparing Pink Ladies to Granny Smiths, and I haven't used a GPS in a long time (don't own one that functions apart from my phone), but every time I drive past those "Speed Advisory" checkpoints, their readout is exactly the same as the speed indicated on my speedo.

There are two of these near Ballarat on the Western Freeway, one a couple minutes west of Ballan (between Ballan and Gordon), and another at the Wendouree exit (4th Westbound exit to Ballarat out of 5) on the Gillies Street overpass. I've tested both checkpoints (when the-powers-that-be decide to turn them on) on both lanes in both Eastbound and Westbound directions, and all of them always read exactly what my speedo reads - usually 112 km/h as that seems to be what my cruise sets itself to, but I have tested them manually at 110 km/h and 115 km/h just to check my speedo, no hooning here! (I've never pushed it to the maximum 117 km/h readout.) Maybe I should try them at 100 km/h and 105 km/h, too.

Speedo: 112 km/h
Checkpoint: 112 km/h
Engine speed: 2800 RPM
Gearbox: 5sp manual (3.5 L)
Tire size: 215/60R16 (factory size) (Thought I should add this too, as it plays a part also)

TW, std 205/65/15 rims. 96 is what I get on the GPS for indicated 100.

While GPS's are far more accurate they are not 100% either. Depends on how many satellites are being accessed at the time.

Hmmm, disagree.

While the position solution may not be super accurate if only limited satellites are available, provided your GPS is actually providing a position, the speed is very accurate. GPS may not know where you are with millimetre accuracy, but it knows the exact distance between fixes and uses the time interval to calculate speed.

Even if the US DoD were to turn on Selective Availability to deliberately put the position GPS solutions out by up to 100m for security measures, the computed speed would still be absolutely perfect.

Any cheap, crappy hand held GPS (yes, even IPhone) will give precision speeds. By many magnitudes, the rounding up or down to the nearest km/hr will be the biggest error.

Actual, rather than nominal tyre diameters can have a significant effect on speedo readings for a given true speed. Two different make/model tyres with the same 215/60*16 spec, for example can have different diameters. So too can under-inflation and wear affect speedo accuracy.

FWIW, my TW VR-X AWD with 215/60*16 Yokohama Blue Earth tyres reads about 2kph higher road speed than my GPS indicates. The difference is more or less constant (50-110kph), rather than proportional to speed. IOW, I could probably get a spot-on reading by repositioning the needle or increasing the spring pre-load inside the speedo. Not worth the effort IMHO.

FWIW, compared to a 215/60*16 tyre's nominal diameter (664mm), a 205/65*15 tyre (647mm) is 2.5% smaller, so you'd expect a corresponding difference in speedo and RPM readings on that basis alone.

Speed/RPM correlations are also affected by whether you're driving a manual or auto - Hyphen's car does ~2800 RPM @ 112kph, but my auto does less than 2500 at the same speed.

I run I think 240/40/17? (gotta love wide tyres) that may be wrong could be 235/45 but anyway different tyres. And yes the speedo is out. I set cruise control at 109kph and it is actually doing about 103kph

Mitsi don't have a great reputation with speedo accuracy - as I recall the Qld Police found the 380 speedo so badly out they had to install specially calibrated 3rd party speedos even in the general duties cars (Highway patrol cars are always calibrated)

According to ADR 18/03, the difference between the true speed and the indicated speed must between 0 and 4kph + true speed *.01. So the worst you'd expect at
• 60km/h (true) is 64.6km/h (indicated)
• 80km/h (true) is 84.8km/h (indicated)
• 100km/h (true) is 105kph (indicated)
Granted, that's pretty awful, but the only time you'd be liable to see anything approaching that would be if a given model had a wide range of tyre sizes and yours happened to use the smallest. In the Magnas, for example, differences in standard tyre sizes alone meant a possible 2.6% variation. The following table shows what the worst-case effect of this would be for the same speedo on smallest (205/65*15) and largest (215/60*16) diameter stock Magna tyres:


km/h 205/65*15 215/60*16
True Ind. Ind.
40 44.4 43.3
60 64.6 63
80 84.8 82.6
100 105.0 102.3
120 125.2 122.0


For 380s the ADR speed compliance was required at were 40, 80 & 120 km/h.

As for the police, have you ever seen a current or ex-police general-duties 380 with a non-factory speedo? Has anyone?

My 380 is more then 10% out. Need 112 indicated to do 100km/hr. I sit on the 110 mark which means I'm doing 98-100km/hr

The speedos in my cars are always fairly accurate on those occasions i've cross-checked them with the GPS or with the highway checks. i.e: speedo shows 100kmh and GPS shows 99, 100 or 101.
AFAIK speedos of that era should be fairly accurate.

My 380 is more then 10% out. Need 112 indicated to do 100km/hr. I sit on the 110 mark which means I'm doing 98-100km/hr
Have you checked it with a GPS and/or timing over an extended distance (e.g. 20km) via roadside distance markers on a major highway? Although distance markers are not always spaced the exact indicated distance apart, they're usually within +/- 100m which is more than sufficient for this purpose. At 100km/h it should take exactly 12 minutes to travel that far. Unless your car takes 13:12 to do it at an indicated 100, it's not 10% out. At 110Km/h it should take 10:55. A 10% error at 110km/h indicated means it would take exactly 12 minutes...

According to ADR 18/03, the difference between the true speed and the indicated speed must between 0 and 4kph + true speed *.01. So the worst you'd expect at
• 60km/h (true) is 64.6km/h (indicated)
• 80km/h (true) is 84.8km/h (indicated)
• 100km/h (true) is 105kph (indicated)
?



Bit of a problem with your maths - should be 10% not 1%


The rule is: The speed indicated shall not be less than the true speed of the vehicle. At the test speeds specified in paragraph 5.2.5. above, there shall be the following relationship between the speed displayed (V1 ) and the true speed



0 ≤ (V1 - V2) ≤ 0.1 V2 + 4 km/h


In addition in the table below I have the MMAL specs for speedo accuracy on the 380 - a bit tighter than the ADR :_





40 KMH ADR 40 to 48, MMAL 40 t0 44


80 kmh ADR 80 to 88, MMAL 80.5 to 85.5


120 kmh ADR 120 to 136, MMAL 121.5 to 127.5


Sorry about all the edits trying to paste complete table for XL

Bit of a problem with your maths - should be 10% not 1%
Oooops!

Still, it's good to know that MMAL had much better standards than the ADR's minimum.

Thank you all for your feedback.
It appears to me there are three outstanding finds from all of this...
1) There IS an inbuilt margin of error, as per ADR's
2) Tyre size plays a part in all of this, as expected, but wasn't expecting a 3-5km/h difference, as per the previous info
3) The set up of gearboxes / final drive ratios is HUGE in magnas...I cant imagine doing 2800rpm or 2500rpm (Macropod / Hyphens results) at 110...Mine's a lazy 2000rpm at 100 indicated, or 2200 at 110...
That's probably why there is so much variance in fuel economy for the same make / model???
Again thanks to all.:neutral:

Madmax: You said that most rural highways have markers to calibrate your speedo. Correct. You will find that the odometer checks almost spot on, although your speedo will be up to 10% high. Damned nuisance because these days it is so easy for a manufacturer to give good accuracy. Hell - you buy a $5 watch which keeps time to within seconds over a year!

I read a lot about how people fit oversized tyres to their vehicles to correct the speedo and then complain about increased fuel consumption. In reality, the consumption is still probably the same, it is only the odo which is giving wrong readings.