Adjusting Displayed Fuel Economy on Mk6 TDi

Since I bought my car new, I’ve been disappointed with how inaccurate the mpg numbers that are shown in the MFD. When I recently bought a VCDS, I found out that I could adjust the calculations used by the MFD to make the mpg information more accurate.

https://idpartsblog.com/2018/03/26/adjusting-displayed-fuel-economy-on-mk7-tdi/

I hadn’t been tracking the MFD indicated mpg per tank, but I did feel that the numbers seemed to be about 8-10% high. So I made my initial adjustment to 1.05 and started tracking the indicated mpg against my externally calculated mpg for the same tank.

The next three tanks looked like this:
tank 1 - 478 mi., mfd mpg = 36.6, calc mpg = 35.9, error 0.7 mpg or 2.0%
tank 2 - 508 mi., mfd mpg = 43.6, calc mpg = 41.8, error 1.8 mpg or 4.3%
tank 3 - 527 mi., mfd mpg = 42.3, calc mpg = 40.6, error 1.7 mpg or 4.1%

That led to an averaged 3.5% error overall with the adjustment set to 1.05. So I changed the adjustment value to 1.09 using my VCDS, while sitting at the gas station of that last fill up. I am just about ready to fill my first tank with the new setting of 1.09 and will be interested to see how close it is now. Of course, one tank is just one data point and not enough to declare victory.

My goal is to get the mfd to match my calcs +/- 0.2 mpg. That would put my mfd at 99.5% accurate, which would be good enough for me.

So the point of this thread is to ask if anyone has done the adjustments and what adjustment value did you decide was good enough?

BTW, tank 1 mpg looks low because it included 250-300 miles towing a 1400# trailer. Given that, it looks better now, doesn’t it?

On a semi related note, I'm fairly certain that the MFD uses indicated speed / mileage to calculate the mileage. I've found that with stock sized tires and stock VCDS settings, my speedo (both digital and needle) were out by 8-10% anyway. I'm sure that didn't help your economy calculations.

Concerning indicated versus true speed - I have confirmed that my speedometer indicates 2 mph fast at speeds ranging from 25 to 70 mph as confirmed by multiple external radar speed signs. I have also checked my elapsed time over measured miles to check true speed against an indicated 60mph (it was 62 seconds, which calcs to 58mph, but my elapsed time measurement probably has a greater inherent error than all the other ways that I’ve checked my car’s instruments. In this case, it looks like I got lucky). I thought that a consistent 2mph error was weird because I expected a consistent percentage of error, not a fixed value. I have also checked indicated odometer readings over long trips and have never observed an error of more than +/- 0.5 miles compared to freeway signs on non-stop runs of more than 300 miles, recognizing that freeway signs may be off by +/- 0.5 miles too. Worst case scenario of both errors compounding in the same direction, the error calculates to be less than 0.35%. So the dash indicated speed appears to be off by 2mph and the dash indicated odometer appears to be dead on accurate.

Recently, I installed a P3cars ODB2 multi-gauge which gets it’s data from the OBD port. One function is speed display from the ECU via the data port. Surprisingly, the ECU speed data is consistently 2 mph slower than the dash indicated speed (including the digital speed for the cruise control), at the same observed speeds. Therefore, running the stock tire size, my ECU very precisely and accurately displays my known speed as observed via external measurement.

So based the analysis of available measurements, I believe that the ECU does know my actual speed and the actual distance travelled to an observed accuracy greater than 99.6%. Therefore, I’m going to stick with the assertion that the inaccuracies of the MFD mpg data are coming from the ECU fuel consumed measurement, which can be adjusted with the procedure referenced above.

Thanks for taking the time to share your findings. Somewhere it was posted that the mpg inaccuracies where similar % differances to possible innacurate calibration using liters.

No worries. I think that everybody’s car may end up with a different adjustment factor to dial in their displayed mpg and range. In the next month or two, I should have mine dialed in fairly well.

If we get a few more examples of what adjustment settings have worked for others, we’ll be able to see if there is a “sweet spot” of good adjustment values for our cars and if so, owners will be able to start their adjustments there and maybe finish the project in just one change. If there is no typical adjustment factor that works for most of us, then each person will have to go though their own trial and error to dial in their displays.

We already know that virtually all of cars are reading off in the same direction, now I want to figure out if they are off by similar amounts.

Update: I did my first tank fill up with the display correction value of 1.09. Mind you this is only one measurement, but the results do look promising.

tank 1 - 520 mi., mfd mpg = 38.6, calc mpg = 38.5, error 0.1 mpg or just 0.26%

So another few tanks to confirm the results, I’ll post data as it becomes available.

This is super helpful, thanks for doing the leg work on this! Might save me from constantly updating my spread sheet after each fill up.

Update: another tank with the display correction set to 1.09.

tank 2 - 490 mi., mfd mpg = 39.2, calc mpg = 38.6, error 0.6 mpg or 1.56%.

I’m going to keep this setting to see if the data smooths out the measurement errors over time, probably another 3 tanks. I’ll post info here as it becomes available.

FWIW, I believe all VWs (if not all German autos) have a built-in speedo "nanny" which says you're going just a wee bit faster than you actually are. The ECU knows the true speed, but the speedo itself tells you that you're going 2-3 MPH faster. This is a safety requirement according to some German law, IIRC.

Anyhow, all German cars I've owned and driven have always displayed my speed as just a bit faster than I was actually going. In general, 2-3 MPH difference between real and indicated.

Hope this helps.

I believe you are correct...

I had my cruise control set (on the digital display) at 78 mph. My GPS said I was doing 79 and my speedo said I was doing 80

update: third tank with the display correction set to 1.09. here’s all the data for this setting so far:

tank 1 - 520 mi., mfd mpg = 38.6, calc mpg = 38.5, error 0.1 mpg or 0.26%
tank 2 - 490 mi., mfd mpg = 39.2, calc mpg = 38.6, error 0.6 mpg or 1.56%
tank 3 - 505 mi., mfd mpg = 38.6, calc mpg = 39.0, error -0.4 mpg or -1.03%
average 505 mi., mfd mpg = 38.8, calc mpg = 38.7, error 0.1 mpg or 0.26%

So the accuracy is better than my target rate of 99.5% accurate, although, I am surprised at the variance in precision with single tanks measuring 0.6 mpg over and 0.4 mpg under the calculated numbers. This variance might be primarily from differences in how full each tank was filled. I always fill until clear fuel is visible at the fill fitting, but temp and ground slope may effect how much fuel is on-boarded each time.

So at this time, I am sticking with the 1.09 display correction value and will continue to track the accuracy over time. I will probably post long term updates of just the average data every few tanks to let you guys know if things are staying within spec.

I have noticed that with the MFD showing more accurate (lower) mpg data, I am having to reset my expectations of what the mileage should look like under various conditions and on my regular commute. I know that the display was only 10% off (high), give or take, but it does take some getting used to not seeing 50 mpg very often. It used to come up for my commute to work, and going home was often above 40 mpg. Now, 50+ mpg only comes up on long fwy speed drives without much climbing involved. Oh well, I guess that I would rather see real data, instead of the “optimistic” factory VW data, even if it doesn’t make me feel as efficiently awesome behind the wheel...

Since we are geeking out over this (and we do appriciate the effort). What do you think about the fuelly mpg difference? Odometer inaccurate? GPS vrs read out?

I wish I could figure out how to do this on my 2015 Passat.

But the Adaptation button is grayed out.

JELLOWSUBMARINE said:

Since we are geeking out over this (and we do appriciate the effort). What do you think about the fuelly mpg difference? Odometer inaccurate? GPS vrs read out?

Click to expand...

At this time, I think that my fuelly numbers are good. I’ve verified that my odo is accurate and precise via gps and other methods. The only other factor would be variances in fuel measurement, but I believe that gas pumps are accurate and my fill method is pretty consistent.

My recent 3 tank average of 38.7 mpg is pretty close to my 154 tank average of 39.6 mpg in my fuelly signature, which also includes about 60k miles of pre-fix data where my calculated mpg was 40 mpg. Doing the math by projecting my current trends, my average at 120k should be right around 39, which would mean that the difference between pre-fix and post-fix would be just about 2 mpg (60k @ 40 mpg and 60k @ 38 mpg = 39 mpg overall), matching the fix info provided by VW. Given that calculation, my recent 38.7 mpg looks slightly higher than my expected post-fix 38 mpg. But minor variances like that are likely due to different driving conditions, regular commute versus longer trip, etc.

On 3 separate MKVI Jettas, a correction of 1.07 has worked nicely for me. I used the same method as you to arrive at that number. I think I may have even adjusted one a bit after having the fix done about a year ago. In any case, the display now gives a much more accurate estimate for instantaneous and tank MPG average.

Since then, I haven't given much thought to it's exact accuracy and haven't bothered checking the precision of the correction. I still make a quick mental comparison of the measured MPG to the display for each tank. Just knowing that it's doing a much better job of telling the truth makes me happy.

It's good to see someone being proactive and making a simple correction with VCDS instead of whining incessantly about how inaccurate the MPG lie-o-meter is each time they fill up.