Fuel consumption sensors MB

[BackintheFold]

My average fuel consumption indicator is playing up.
The most it will show is 18.8mpg (oddly specific) and when I reset it it can come back up with 11mpg and drop rapidly down to 0.0!

The average speed is correct. If I reset it at 60 and drive at 60 it shows…60
The time elapsed is correct.
The actual speed indicated is correct.

I’m thinking I may have disturbed the fuel sensor while I’ve been working on it recently, but I don’t know where that is!

Thanks

EDIT: side of the metering head of course! The cable is covered by yellow plastic.
My earth wire looks a little dirty. I’ll clean it and see what happens-UPDATE-Nothing.

Update: remove the plug from the sensor and drive and you get 255 mpg showing.

Brown wire is -ve, not designated as earth, and heads back to the dash.
Yellow/black wire should give +5v.
Green/white wire is the sensor.

18.8 seems to be the reset value. Sounds like a reasonable starting point when working out the average.
Next step will be to measure the 5v to a good earth, and see if the -ve is an earth.
The good news is it doesn’t affect the car and is of little real use. Also it is more likely to be the sensor than the instruments as everything else works there.

[Colin Aitchison]

Can only comment about WR engine, but from memory, 3 pin plug on side of metering head base, ground, 12v supply and signal are the 3 wires, can't remember the order, with ohms meter connected across 2 of the 3 pins on side of metering head you should see 1000 ohms with air flow plate at rest, when air flow plate lifts the ohms reading should change, rise or fall can't remember, you can adjust but tricky, pick out the sealant from the head of the 4 fixing screws in the corner of the plug housing, loosen screws a turn or so and rotate plug housing to adjust base ohms reading, correct air flow plate level to be confirmed first before adjusting, reseal screw heads with black silicon.

[BackintheFold]

Ah thats very helpful-thanks.
So it works on airflow plate position-great-that makes it easy to test at rest.

[BackintheFold]

Just tested.
At rest I get 845 ohms between lowest two
Terminals of Plug on sensor.
Between lowest and highest plugs on the lead, I get 4.95v with ignition on but engine off.

I’m assuming that the wr and mb would both be 1000 at rest, so mine needs adjustment but I’d be grateful if someone with an MB could measure their ohms before I start fiddling with adjusting!

Out of internet’s, I got an actual 27mpg (indicated 11.6 mpg)on a mixed motorway and pootling around town test, and 40.4 mpg (indicated 21.8 mpg) on a steady A3 and clear M25 run this afternoon.

So I’m not overly worried about the discrepancy!

[tractor Dave]

Wow! That's amazing.
A clear M25 on a Sunday afternoon

[BackintheFold]

Originally Posted by tractor Dave

Wow! That's amazing.
A clear M25 on a Sunday afternoon

Haha! Yes. Reckon it’ll be solid tomorrow.

[BackintheFold]

Cleaned the connections again, and now working fine.
Have considered removing to check readings again…but am not going to mess with it.

[Hanuman]

This for the WR, so I’m assuming the same for the MB.

There are a number of adjustments on the electronic display for which there was a calibration procedure.
For instance, the fuel tank sender being matched to the display cluster would start with an empty fuel tank, and set zero. Then a measured quantity of fuel- 20litres exactly, and the dash would be adjusted to correspond. Then a half tank, then full. Or something like that. There was also (I’m pretty sure) adjustments for the fuel consumption.

A lot of folk just assume the consumption shown is accurate, when in fact it hasn’t been calibrated in 20, 30, 40 years. My WR, for a very long time was pretty accurate when compared against a miles driven against fuel consumed calculation. But I know of people proudly proclaiming how they routinely got 36mpg from a WR, while driving with a lead foot, yet wondering why they had to fill up so often.

In our old machines, it was a couple of potentiometers, one for offset, one for %trim, to get a reasonable best fit.
(In my V8s, the same is achieved using simple dip-switches to trim the reading.)

The consumption sender, as found on the side of the metering unit, is simply a potentiometer, and like everything is prone to issues due things like age, hardening of wires, damaged shield, tarnished contacts. Anything that affects the impedance in the system.
Often, this manifests as wildly erratic readings going full scale in both directions.

The other sensors that have an annoying tendency to give wild readings that are an issue-
Temp sensor, front of the block, due to its exposed location.
Gearbox sensor, giving a bad output, often caused by an accumulation of oily water residue tracking in and shorting across the pins. Or at least giving a high impedance in parallel which is enough to screw with the readings.

[BackintheFold]

Originally Posted by Hanuman

This for the WR, so I’m assuming the same for the MB.

There are a number of adjustments on the electronic display for which there was a calibration procedure.
For instance, the fuel tank sender being matched to the display cluster would start with an empty fuel tank, and set zero. Then a measured quantity of fuel- 20litres exactly, and the dash would be adjusted to correspond. Then a half tank, then full. Or something like that. There was also (I’m pretty sure) adjustments for the fuel consumption.

A lot of folk just assume the consumption shown is accurate, when in fact it hasn’t been calibrated in 20, 30, 40 years. My WR, for a very long time was pretty accurate when compared against a miles driven against fuel consumed calculation. But I know of people proudly proclaiming how they routinely got 36mpg from a WR, while driving with a lead foot, yet wondering why they had to fill up so often.

In our old machines, it was a couple of potentiometers, one for offset, one for %trim, to get a reasonable best fit.
(In my V8s, the same is achieved using simple dip-switches to trim the reading.)

The consumption sender, as found on the side of the metering unit, is simply a potentiometer, and like everything is prone to issues due things like age, hardening of wires, damaged shield, tarnished contacts. Anything that affects the impedance in the system.
Often, this manifests as wildly erratic readings going full scale in both directions.

The other sensors that have an annoying tendency to give wild readings that are an issue-
Temp sensor, front of the block, due to its exposed location.
Gearbox sensor, giving a bad output, often caused by an accumulation of oily water residue tracking in and shorting across the pins. Or at least giving a high impedance in parallel which is enough to screw with the readings.

Thanks for that-but I am sure the MB dash is a different setup to the WR.

Now that the on-board computer is giving realistic figures, I can check it against the real world next time I fill up.

The 40.4 mpg above was based on steady driving and petrol used (full-full) vs mileage covered (from dash and Goole maps), currently it is shiwing 28 ish for local traffic...