Fuel gauge goes up with no fuel added - Page 4

Mike Noc · 28th August 2011, 22:14

Quote:

Originally Posted by HarryM1BYT

Yes, but as said - with a pump moving it from one side to the other, I don't understand why there a need to measure the level in both sides?

Harry although the pump moves fuel from one side to the other there is also a scavenge return pipe going back to the pump side as well as unused fuel returning from the HP fuel pump and leak back on the diesel models.

So that, along with the fuel sloshing over the saddle when going round bends, make it impossible to accurately measure the total fuel in the tank from just one side.

Mike

T-Cut · 28th August 2011, 22:32

I'd ignore the OBD stuff regarding the fuel tank. I think the MGR tank system is different from the BMW system from which we got the original diagnostics information.

There is a float-driven rheostat in each side of the tank. The resistance of each is integrated by the IPK, digitised and presented as the gauge reading. You can check the current resistance of the floats by putting an ohm meter across the relevant terminals on the under-seat connector. Here's a diagram:

The fuel level in each side of the tank is more or less equilibrated by the action of fuel being pumped to the engine (petrol) and by backflow from the engine (diesel). This is done by venturi type static pumps, which only function when fuel is flowing through the jet. Neither the primary diesel pump nor the petrol pump equalises levels directly.

Wayward readings on the fuel gauge must be a reflection of the manner in which one or both floats are moving. It's not uncommon to read about sticking/binding floats, especially if there's been some tank work done that involved removing or even just adjusting one of the two modules (pump/filter/filter-adapter). The clearance for the floats is apparently very limited and the smallest shift in mounting position can cause strange effects at the gauge. I suppose floats may also develop leaks, so they may become partially weighted with fuel. This will have odd results as the tank level varies. When the level gets very low, the float may partially empty itself, so when fuel is added, it's initially very bouyant and then starts getting heavier and sits lower in the fuel. This would lead to a wandering gauge.

The OBD figures for mpg, range, etc. are only as good as the data it receives and the calibration accuracy. The data displayed are running averages, so if you frequently reset the computer, the running average starts again with what's available. That's typically not enough data points to make the calculations realistic. To get anything like meaningful figures the computer should not be reset without good reason. The calibration for mpg is described in the latest revision of the Diagnostic Routines. This calculation is the basis for the range and any other display involving fuel so there's a knock-on effect if it's not accurate enough. It also uses distance travelled in miles/Km and that's taken from the RH front ABS sensor as you know. However, the accuracy here will also vary over thousands of miles due to tyre wear and particularly from the use of non-standard wheel/tyre combinations. Getting correct calibration could take 1000's miles of brim to brim readings until the data smooths out and it can be relied on. The problem is, nothing it uses is an absolute value. They all have tolerances so even a brim to brim calculation is only as good as the displayed mileage.

TC

HarryM1BYT · 29th August 2011, 00:22

Thanks for sorting out the confusion T-Cut.

28th August 2011, 22:32	#32
T-Cut This is my second home Rover75 and Mreg Corsa. Join Date: Nov 2006 Location: Sumweer onat mote o'dust (Sagin) Posts: 21,754 Thanks: 341 Thanked 3,660 Times in 2,924 Posts	I'd ignore the OBD stuff regarding the fuel tank. I think the MGR tank system is different from the BMW system from which we got the original diagnostics information. There is a float-driven rheostat in each side of the tank. The resistance of each is integrated by the IPK, digitised and presented as the gauge reading. You can check the current resistance of the floats by putting an ohm meter across the relevant terminals on the under-seat connector. Here's a diagram: The fuel level in each side of the tank is more or less equilibrated by the action of fuel being pumped to the engine (petrol) and by backflow from the engine (diesel). This is done by venturi type static pumps, which only function when fuel is flowing through the jet. Neither the primary diesel pump nor the petrol pump equalises levels directly. Wayward readings on the fuel gauge must be a reflection of the manner in which one or both floats are moving. It's not uncommon to read about sticking/binding floats, especially if there's been some tank work done that involved removing or even just adjusting one of the two modules (pump/filter/filter-adapter). The clearance for the floats is apparently very limited and the smallest shift in mounting position can cause strange effects at the gauge. I suppose floats may also develop leaks, so they may become partially weighted with fuel. This will have odd results as the tank level varies. When the level gets very low, the float may partially empty itself, so when fuel is added, it's initially very bouyant and then starts getting heavier and sits lower in the fuel. This would lead to a wandering gauge. The OBD figures for mpg, range, etc. are only as good as the data it receives and the calibration accuracy. The data displayed are running averages, so if you frequently reset the computer, the running average starts again with what's available. That's typically not enough data points to make the calculations realistic. To get anything like meaningful figures the computer should not be reset without good reason. The calibration for mpg is described in the latest revision of the Diagnostic Routines. This calculation is the basis for the range and any other display involving fuel so there's a knock-on effect if it's not accurate enough. It also uses distance travelled in miles/Km and that's taken from the RH front ABS sensor as you know. However, the accuracy here will also vary over thousands of miles due to tyre wear and particularly from the use of non-standard wheel/tyre combinations. Getting correct calibration could take 1000's miles of brim to brim readings until the data smooths out and it can be relied on. The problem is, nothing it uses is an absolute value. They all have tolerances so even a brim to brim calculation is only as good as the displayed mileage. TC Last edited by T-Cut; 28th August 2011 at 22:42..

29th August 2011, 00:22	#33
HarryM1BYT This is my second home 75 Contemporary SE Mk II 2004 Man. Sal. CDTi 135ps, FBH on red diesel, WinCE6 DD Join Date: May 2010 Location: Leeds Posts: 17,273 Thanks: 2,160 Thanked 2,061 Times in 1,586 Posts	Thanks for sorting out the confusion T-Cut. __________________ Harry How To's and items I offer for free, or just to cover the cost of my expenses... http://www.the75andztclub.co.uk/foru...40#post1764540 Fix a poor handbrake; DIY ABS diagnostic unit; Loan of the spanner needed to change the CDT belts; free OBD diagnostics +MAF; Correct Bosch MAF cheap; DVB-T install in an ex-hi-line system; DD install with a HK amp; FBH servicing. I've taken a vow of poverty. To annoy me, send money.