"Averaging" fuel level sender??

[captainboo]

They work like this:-

http://www.digikey.com/Web Export/S...meder-auto-fuel-float-sensor.pdf?redirected=1

 

[nigelmercier]

captainboo's post above suggests that there are a series of switches at points in the shaft which "switch" the resistance in "steps" over the length of the shaft ...

Seems a crazy way to do it, but I guess it is OK if you have enough switches.

 

[wipe_out]

Seems a crazy way to do it, but I guess it is OK if you have enough switches.

Yes, I think that's the issue here.. My tank is pretty shallow (~11") which probably also explains why I only have about 5 switch points.. I will be down at the boat again tomorrow so might pull the sender again and measure it on a digital multimeter which will hopefully clearly show the exact number of "switches" on my sender..

I can also get some measurements and check how easy it will be to make a tube type baffle to go around the sender and reduce the "slosh" effect..

 

[William_H]

Not much help to the OP but what he really needs is a capacitive fuel probe. Here 2 tubes one inside the other are insulated from one another. As the fuel rises inside and outside the tubes the capacitance between the tubes increases. The capacitance is converted to a voltage for the meter in electronics. The beauty of this method is that you can add as many probes as needed so that as the fuel moves from one end of the tank to the other the fuel level rises in one probe and drops in the other so always giving a correct result. They are used in aircraft with wing tanks having low height but huge area so very susceptible to fuel movement. Along with baffles of course. They have an interesting other effect with temperature change ie colder the fuel becomes more dense so capacitance effect is greater. (huge temp changes in long range jets) This means the fuel gauge is calibrated in pounds or kgs of fuel which relates then to actual energy and mass of the fuel rather than strictly quantity.(or level)
I had an old Oz built Ford car of 1982 vintage that had a capacitive fuel gauge. It was superb. Later 1998 model I got had gone back to float and resistor type. I wonder why?
Anyway if OP sees a fuel gauge system described as capacitive this might be an answer. Unlikely I suppose. good luck olewill

 

[lw395]

Not much help to the OP but what he really needs is a capacitive fuel probe. Here 2 tubes one inside the other are insulated from one another. As the fuel rises inside and outside the tubes the capacitance between the tubes increases. The capacitance is converted to a voltage for the meter in electronics. The beauty of this method is that you can add as many probes as needed so that as the fuel moves from one end of the tank to the other the fuel level rises in one probe and drops in the other so always giving a correct result. They are used in aircraft with wing tanks having low height but huge area so very susceptible to fuel movement. Along with baffles of course. They have an interesting other effect with temperature change ie colder the fuel becomes more dense so capacitance effect is greater. (huge temp changes in long range jets) This means the fuel gauge is calibrated in pounds or kgs of fuel which relates then to actual energy and mass of the fuel rather than strictly quantity.(or level)
I had an old Oz built Ford car of 1982 vintage that had a capacitive fuel gauge. It was superb. Later 1998 model I got had gone back to float and resistor type. I wonder why?
Anyway if OP sees a fuel gauge system described as capacitive this might be an answer. Unlikely I suppose. good luck olewill

What he needs is a stick with some lines on it and a grasp of how much fuel he uses per hour.....

 

[nigelmercier]

My tank is pretty shallow (~11")

From the data sheet in #41:

MEDER has developed a new float series having up to nine sensing points over a 300 mm sensing distance.

Even looking at the circuit diagram, I still don't see how it works. How could they arrange for one switch to close at exactly the same position that the previous one opened? There are bound to be spots where there are either two switches on, or none. The result of this imbalance would be wild flickering ...

 

[srp]

The result of this imbalance would be wild flickering ...

 

[JohnGC]

From the data sheet in #41:


Even looking at the circuit diagram, I still don't see how it works. How could they arrange for one switch to close at exactly the same position that the previous one opened? There are bound to be spots where there are either two switches on, or none. The result of this imbalance would be wild flickering ...

Easy to arrange for at least one switch to be always on except at the bottom end (max resistance). Like a rotary make before break switch. That way the flicker at the change-over points will be just one step or 10% in the example shown. Potentially this is much more reliable than anything with a track which will wear.

 

[nigelmercier]

Easy to arrange for at least one switch to be always on except at the bottom end (max resistance)...

Or minimum in this case, I think.

 

[Daydreamer]

Easy to arrange for at least one switch to be always on except at the bottom end (max resistance). Like a rotary make before break switch. That way the flicker at the change-over points will be just one step or 10% in the example shown. Potentially this is much more reliable than anything with a track which will wear.

Yes it will be something like this. Thanks to Nigel I am sat here with a tank sender stuck in my ear listening to the reed switches clicking . I get a range of 0-180 ohms in 20ohm steps every 2 cm. There is absolutely no sign of any flickering of resistance reading.

 

[JohnGC]

Or minimum in this case, I think.

Minimum fuel, maximum resistance assuming the mechanical drawing and schematic are drawn aligned. The same may be true at the other end as well.

 

[JohnGC]

Yes it will be something like this. Thanks to Nigel I am sat here with a tank sender stuck in my ear listening to the reed switches clicking . I get a range of 0-180 ohms in 20ohm steps every 2 cm. There is absolutely no sign of any flickering of resistance reading.

It's telling you that you're beer level is too low for safe navigation of this forum and you should immediately go to the pub.

 

[Daydreamer]

Minimum fuel, maximum resistance assuming the mechanical drawing and schematic are drawn aligned. The same may be true at the other end as well.

0ohms empty 180ohms full. Euro system. Trust me on this, remember where my sender gauge is.

 

[JohnGC]

0ohms empty 180ohms full. Euro system. Trust me on this, remember where my sender gauge is.

I don't doubt you, just commenting on the MEDER drawings.

 

[rogerthebodger]

When I was designing the fuel and water tanks on my current boat I did not like the step switch type you are describing. I found these which are capacitance.

They can be used for any type of current fuel/water gauge. I went for the 4 - 20 mA current loop which is what VDO use for their capacitance type.

http://www.centroidproducts.com/

 

[nigelmercier]

Thanks to Nigel I am sat here with a tank sender stuck in my ear listening to the reed switches clicking . I get a range of 0-180 ohms in 20ohm steps every 2 cm. There is absolutely no sign of any flickering of resistance reading.

That's good news, I think.

 

[wipe_out]

When I was designing the fuel and water tanks on my current boat I did not like the step switch type you are describing. I found these which are capacitance.

They can be used for any type of current fuel/water gauge. I went for the 4 - 20 mA current loop which is what VDO use for their capacitance type.

http://www.centroidproducts.com/

Looks like an interesting alternative.. How much do they sell for?

 

[nigelmercier]

Looks like an interesting alternative...

In particular:

Eliminates excessive needle bounce on the meter while underway

 

[wipe_out]

In particular:Eliminates excessive needle bounce on the meter while underway

Absolutely!!

In fact the word I saw was "averaging" on the home page..

 

[nigelmercier]

Absolutely!! In fact the word I saw was "averaging" on the home page..

The reason I'm interested in this thread is that I have designed a built a digital fuel gauge. I was not aware that there are stepped resistor sensors, which obviously reduce the accuracy of a gauge. My gauge has a calibration mode for linear sensors, allowing the tank to be monitored to the nearest 5 litres. It also has an averaging feature, which averages over the last 8 or 16 readings (can't remember which). I'm about to build the second prototype, so this has been interesting.