Speedo Inaccurate

[SteveK]

So, think of it this way: if you have standard wheels and tyres on your car, then, all else being equal, your car knows how fast it's travelling, exactly. That accurate perception of the speed is used to calculate the distance travelled and everything else based on that - so, if you drive at precisely 100km/h for 1 hour, your trip meter will read 100km.

However, your car lies to you about how fast you're travelling. It knows you're doing 100km/h. But it tells you (via speedo) that you're doing 105km/h (or 107, or even 110), so you'll slow down, and stay under the limit. The only error in this whole equation (and it's an intentional one) is in the instrument you're reading. Australian law requires that speedos must read 100% accurate or overestimate up to 10%, but must not underestimate.

What you want is to have an instrument that shows you a (more) accurate representation of how fast you're actually going. So you use VCDS to alter the car's settings to achieve this. Trouble is, rather than having reduced the error margin in the instrument itself (the amount by which the car is lying to you), you're actually making the car think it's going slower than it really is. The nett result is that, yes, the instrument will read a lower speed, but the odometer/etc will underestimate how far you've driven - throwing out its calculations.

Ok, so you've confirmed that the setting in VCDS is not adjusting the correction, it's setting the BASE default speed reading for the car, i.e. a base for a car that is either a different model or different stock tyre ratings. All clear now.

So is there an actual way to correct this outside of adjusting this value? I have a feeling the 19s on the car are not part of the stock setting for the Golf R and hence throwing out this value but alas there is no other default value that it can be set to that has been stored for Golf + 19s, if you get my drift. I wonder if there is a noticable difference with a Golf R riding on 18s, both set to cruise on 100 dead. Me thinks rolling on 18s would be slightly faster.

Thanks again Adam.

Cheers,
Steve

[AdamD]

So is there an actual way to correct this outside of adjusting this value?

Have a read through this thread; loads of good info there. In particular, take a look at logger's comments - he's done a lot of playing around with VCDS and vehicle speeds. I suggest that any further discussion on this topic be moved to that thread, thanks!

[Corey_R]

Wow! Lots of conversation overnight. I took AdamD's lead and moved all this conversation to the speedo thread he's linked above.

Are you sure about that Corey? Not having a go, just trying to understand your thinking. I've worked with guys to do tunes on various local cars and adjusted speedos via electronics to correct deficiencies and my understand is that you are just adjusting the reading from the rolling tyre to the speedo.

Yeah - it's not adjusting the speedo. I can confirm the conclusion that you've managed to come to with AdamD that you're adjusting the "BASE default speed reading" so effectively what you've effectively changed is the distance the car measures (odometer) whilst still leaving the 7% speedo error there.

AdamD mentioned that you can do a test with an external GPS. If you reversed your VCDS changes to get it back to stock and then did the following test you can confirm all of this:

- Set cruise control to 100km/h
- When you reach 100km/h as indicated via your analog speedo your GPS unit should say that you're doing 93 or 94km/h (if you have stock wheels and tyres at "correct" inflation).
- If you then switch the MDI to the "average speed" display, and then hold down the OK button to reset the trip meter, the first figure it will display for your "average speed" is 93/94km, the same as what your GPS will be displaying.

It used to be that speedo's could be +/-10% accurate. Now with digital speed cameras booking you for only 113 in a 110 zone, the law was changed so it can only read slower than your actual speed. Additionally, as consumers are allowed to change their wheels and tyres within certain restrictions, it means that the manufacturers can't even make the speedo PERFECTLY accurate, as a legal wheel change could then make your car illegal. (Wheel changes can change the speedo by 3%+)

Japan and Australian cars have lagged behind European cars on this topic, because I can recall Wheels or Motor magazine reporting on "actual speed when 100km/h reported" as far back as the mid 90's and particularly the Mercs, Audis, BMWs of the time were starting to report 100km/h when they were only doing 93 to 95km/h. Like it or hate it, it's just the way it is now.

Logger and Maverick probably did the deepest investigation in this, in conjunction with Ross-Tech, the makers of VCDS, and whilst I can't remember where that conversation took place, the outcome was that there was no way to correct the 7% built in error of the speedo without negatively affecting other measurements.

[SteveK]

Wow! Lots of conversation overnight. I took AdamD's lead and moved all this conversation to the speedo thread he's linked above.



Yeah - it's not adjusting the speedo. I can confirm the conclusion that you've managed to come to with AdamD that you're adjusting the "BASE default speed reading" so effectively what you've effectively changed is the distance the car measures (odometer) whilst still leaving the 7% speedo error there.

AdamD mentioned that you can do a test with an external GPS. If you reversed your VCDS changes to get it back to stock and then did the following test you can confirm all of this:

- Set cruise control to 100km/h
- When you reach 100km/h as indicated via your analog speedo your GPS unit should say that you're doing 93 or 94km/h (if you have stock wheels and tyres at "correct" inflation).
- If you then switch the MDI to the "average speed" display, and then hold down the OK button to reset the trip meter, the first figure it will display for your "average speed" is 93/94km, the same as what your GPS will be displaying.

It used to be that speedo's could be +/-10% accurate. Now with digital speed cameras booking you for only 113 in a 110 zone, the law was changed so it can only read slower than your actual speed. Additionally, as consumers are allowed to change their wheels and tyres within certain restrictions, it means that the manufacturers can't even make the speedo PERFECTLY accurate, as a legal wheel change could then make your car illegal. (Wheel changes can change the speedo by 3%+)

Japan and Australian cars have lagged behind European cars on this topic, because I can recall Wheels or Motor magazine reporting on "actual speed when 100km/h reported" as far back as the mid 90's and particularly the Mercs, Audis, BMWs of the time were starting to report 100km/h when they were only doing 93 to 95km/h. Like it or hate it, it's just the way it is now.

Logger and Maverick probably did the deepest investigation in this, in conjunction with Ross-Tech, the makers of VCDS, and whilst I can't remember where that conversation took place, the outcome was that there was no way to correct the 7% built in error of the speedo without negatively affecting other measurements.

Thanks Corey, I read through this thread overnight on Adams recommendation and totally understand the workings now, whilst not ideal.

I'm a stickler for knowing what speed I am doing. Personally, I think this discrepancy encourages people to try and find their accurate speed and we'll see more people driving at 105kmh+ on the speedo to 'assume' they are doing 100kmh, etc. Different for each car. I'd prefer it to be accurate but I can live with it.

Cheers,
Steve

[Corey_R]

Yeah - I understand completely!

Your comments about people driving faster than their speedo says is one of the side effects of this situation. But then, that fact is also part of the reason that these laws started coming into effect in Europe etc. Because you set the limit at 60, and people drive at 65 anyway (unless they're coming up on a speed camera).

Once you've done the "trim meter reset test" or used a GPS device to find what the "real speed" is at 100km/h, then you can just make yourself a little label if you're really obsessed

[Ryan_R]

At least you guys don't have 16" wheels.

Mine is out by 8%, as confirmed by my phone's GPS and overhead checkers. When I first drove the car on the freeway at an indicated 102kph I was wondering why on earth everyone was overtaking me.

Rather than worry about maths while driving, I've just remembered the following numbers:
43, 54, 65, 76, 86, 97, 108, 119 (which are the correct indicated speeds for 40, 50, 60,... 110). Use a calculator or Excel to work them out for yourself.

Maybe we should calculate the speed offsets for each wheel/tyre size and list them in the first post, and call it a day?

[SteveK]

Rather than worry about maths while driving, I've just remembered the following numbers:
43, 54, 65, 76, 86, 97, 108, 119 (which are the correct indicated speeds for 40, 50, 60,... 110). Use a calculator or Excel to work them out for yourself.

That is another argument for having an accurate speedo. We shouldn't have to remember certain numbers, we should just go off the dash.

What happens when you jump in the wifes car and you're still in that mentality? Flashing blue and reds.

Ok, enough from me, too busy waiting impatiently for my RNS510 to arrive. I think I might go for a lunch time spin to just enjoy the car.

*sigh*

[dave_r]

That is another argument for having an accurate speedo. We shouldn't have to remember certain numbers, we should just go off the dash.

You're not wrong, its a pain to remember. Not sure why it is so hard to be accurate. Doing GPS based comparison between my R and a mates RS5. Mine out by about 4km/h, his near perfect match to speedo!

[Ideo]

I like having an innacurate speedo. When I do a true 110 on the F3 it feels like I'm the fastest car on the road.

[Ryan_R]

Yeah, I like doing an indicated 120kph on freeways that say 110 too

@Stevie, Well, no. I've driven a few company cars and had no problem reverting to driving at the indicated speed rather than compensating. It's not like I'm in a hurry to get back to work or anything Don't have any other drivers of the car so no worries there either.

I don't mind remembering, since I almost-always have the digital speedo showing (never got used to the other one in this car for some reason). The only thing I don't like is being aware of passengers looking at it.

I made a quick spreadsheet here: https://spreadsheets.google.com/ccc?...gE&pli=1#gid=0

If anyone with larger tyres wants to fill in some of the details then let me know. Otherwise hopefully you know what your offset %age is by now (just drive at 100kph with a reliable portable GPS unit on the FWY).

[AdamD]

I made a quick spreadsheet here: https://spreadsheets.google.com/ccc?...gE&pli=1#gid=0

If anyone with larger tyres wants to fill in some of the details then let me know. Otherwise hopefully you know what your offset %age is by now (just drive at 100kph with a reliable portable GPS unit on the FWY).

Just bear in mind that the error margin does vary from car to car, not just between models and tyre configurations. There's the ever-present tyre inflation issue (and pressures vary depending on if the tyre, road and air temps are warm or cold), tyre age (treadwear affects rolling diameter)... and that's before considering the possibility of an actual speedometer error.

Caveat emptor! And don't get booked guys!

[Corey_R]

Yeah - those are my feelings too AdamD. Best to just give the knowledge to the forum members on how to determine the inaccuracy % themselves, and then let them decide what to do with it. Providing a spreadsheet with a "model / wheel / tyre size" lookup seems to be just asking for trouble :/

[Ryan_R]

I agree and will remove the mentionings of that. Although for someone to come along and just assume what they read on the Internet is true in realation to something like that without checking a few things out for themselves is also asking for trouble

[Corey_R]

pfft!!!! assumption is the mother of invention!

No wait... something's wrong there

[Ryan_R]

Give me a few hours to work that one out