Fuel Level Sensing

[IPF Racing]

Currently I only have a low pressure warning on the surge-tank for fuel level sensing (I'm not paying for an ATL fuel level sensor to place in the fuel cell). How is everyone monitoring fuel level? Stock sensor, pump outs and injector pulse-width, flow meters? I'm considering building a flow meter setup for the supply and return fuel lines with a quick math display that would show the most accurate remaining capacity, but want to see what everyone else is doing before I dive down that rabbit hole.....

[DRVOLKS]

search in the old posts on this, I got one for a airplane  mounted in the fuel cell with a light comes on at 3 gals. works fine its mounted in a pick up box in the cell.

https://www.ebay.com/itm/Float-Switch-Fuel-Aircraft-Airplane-USAF-Surplus-P-74-607-NOS-FREE-SHIPPING/221986297852?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

Edited November 12, 2019 by DRVOLKS

[Indysupra]

We always just figure fuel burn and use that. 

[Ian]

Old method:  Drive till it starts shutting off in RH corners.

 

New method:  Watch for stock fuel light.

[IPF Racing]

3 minutes ago, SonsOfIrony said:

Old method:  Drive till it starts shutting off in RH corners.

 

New method:  Watch for stock fuel light.

 

Thats basically what happens now, I just noticed at Pitt that fuel strategy is crazy crucial at some parts of the race and I want to be able to know EXACTLY how much fuel we have left and are currently using....

[Andrew D Johnson]

When it stumbles, pit next lap!

[hotchkis23]

We have the airplane style unit that has a rod that measures resistance as the level drops to move the gauge.  It kinda works....

[turbogrill]

9 hours ago, IPF Racing said:

Currently I only have a low pressure warning on the surge-tank for fuel level sensing (I'm not paying for an ATL fuel level sensor to place in the fuel cell). How is everyone monitoring fuel level? Stock sensor, pump outs and injector pulse-width, flow meters? I'm considering building a flow meter setup for the supply and return fuel lines with a quick math display that would show the most accurate remaining capacity, but want to see what everyone else is doing before I dive down that rabbit hole.....

 

A level switch on surge tank must be the best warning. If surge is half, you are in trouble.

 

The idea of integrate the injector pulse width is interesting. Is the flow from an injector linear with the pulse width?

[TimS]

If I had the money and a high end standalone ECU I would use a flow meter.  ECU would alert driver when fuel was used.  I believe top level endurance teams use this type of setup.  

[Black Magic]

10 hours ago, turbogrill said:

 

A level switch on surge tank must be the best warning. If surge is half, you are in trouble.

 

The idea of integrate the injector pulse width is interesting. Is the flow from an injector linear with the pulse width?

 

If that was radically off mapping and engine would be even harder. 

 

In new ecu's you can see lbs\hr as an output, integrate that to get fuel burn....

[mender]

17 hours ago, turbogrill said:

 

A level switch on surge tank must be the best warning. If surge is half, you are in trouble.

 

The idea of integrate the injector pulse width is interesting. Is the flow from an injector linear with the pulse width?

There is a bit of hysteresis at short pulse widths (idle) but for the most part it's pretty linear. One of the bigger variables is fuel pressure as some index to manifold pressure but again pretty linear for mostly full throttle as seen in racing. 

[Hi_Im_Will]

Did I show you the reserve system on Friday night at PIRC?  Got one in my car, there was one in another E30 at PIRC, and two more are about to get delivered.  Two level warnings in the surge side, plus a reserve to get you back to the pit when you ignore them and run out on track:

 

 

You know you want one.

[erich]

I gave up trying to get a good sensor reading from the tank while the car is moving.

As the fuel sloshes around while driving a level sensor is useless. 

A low pressure sensor in the surge tank is probably the best bet if you want to know if you're just about to run out of fuel.

 

We use historical fuel burn rate at each track and plan accordingly.

[Ws6 Mike]

I got one of these mounted in my surge tank. Float Switch

When the cell is no longer filling the surge tank the float switch drops triggering a light on the dash letting the driver know he has 1/2 gallon left and to pit immediately.  Has worked great for me so far. 

[Originalsterm]

52 minutes ago, Hi_Im_Will said:

Did I show you the reserve system on Friday night at PIRC?  Got one in my car, there was one in another E30 at PIRC, and two more are about to get delivered.  Two level warnings in the surge side, plus a reserve to get you back to the pit when you ignore them and run out on track:

 

 

You know you want one.

 

We're able to go 2+ hours with this beauty and the stock 318 tank.

 

 

[mender]

https://www.futurlec.com/sensor/flowfuel30l0.shtml

[enginerd]

34 minutes ago, mender said:

https://www.futurlec.com/sensor/flowfuel30l0.shtml

Know of a digital integrator and display which would work with this??

[TimS]

Will

Is that surge tank something you manufacturer?  

[Slugworks Paul]

On 10/12/2019 at 9:59 AM, mender said:

https://www.futurlec.com/sensor/flowfuel30l0.shtml

 30 liters per hour? Are you racing a 50cc dirtbike or something?

[Slugworks Paul]

But seriously though, get a plate for your fuel cell (I think you were saying you have a fuel cell?) with a 5-bolt SAE flange, and buy a resistive or capacitive sensor. In addition you could add a float switch for a light at the very bottom. You can also do level sensing in a surge tank various ways (I've tried just about all of them). You could sell a kidney and buy Will's surge tank, or you could do like I did and make something just as effective for around $50.

Edited October 14, 2019 by Slugworks Paul

[IPF Racing]

On 10/11/2019 at 3:26 PM, Hi_Im_Will said:

Did I show you the reserve system on Friday night at PIRC?  Got one in my car, there was one in another E30 at PIRC, and two more are about to get delivered.  Two level warnings in the surge side, plus a reserve to get you back to the pit when you ignore them and run out on track:

 

 

You know you want one.

 

Will I think you were busy at Pitt with an install but I had looked at your previous posts about this. My concern with any of these solutions, even the fancy reserve system, is that they are all what I'd call "last minute warning systems" where you have to pit directly or close to directly after the warning light. Using historical data works well too, but only if you've been to that track before. I'd like to have a quick display that shows fuel usage per lap so that on a practice day you can gauge max stint duration (if less than 2 hours) with a real number, and that shows a real number amount of fuel left in the tank. Its most likely not all that necessary, but the engineering side of me wants to know the numbers..... Also knowing the difference of fuel usage between drivers seems useful. 

[Slugworks Paul]

56 minutes ago, IPF Racing said:

 

Will I think you were busy at Pitt with an install but I had looked at your previous posts about this. My concern with any of these solutions, even the fancy reserve system, is that they are all what I'd call "last minute warning systems" where you have to pit directly or close to directly after the warning light. Using historical data works well too, but only if you've been to that track before. I'd like to have a quick display that shows fuel usage per lap so that on a practice day you can gauge max stint duration (if less than 2 hours) with a real number, and that shows a real number amount of fuel left in the tank. Its most likely not all that necessary, but the engineering side of me wants to know the numbers..... Also knowing the difference of fuel usage between drivers seems useful. 

 

Then your choice is:
-2 flow sensors
-if data is available calculate based on injector pulse width, fuel pressure and injector design parameters

Edited October 14, 2019 by Slugworks Paul

[IPF Racing]

4 minutes ago, Slugworks Paul said:

 

Then your choice is:
-2 flow sensors
-if data is available calculate based on injector pulse width, fuel pressure and injector design parameters

 

Agreed, I guess I'll see if I can make a cheap solution and report back!

[Hi_Im_Will]

23 minutes ago, IPF Racing said:

 

Will I think you were busy at Pitt with an install but I had looked at your previous posts about this. My concern with any of these solutions, even the fancy reserve system, is that they are all what I'd call "last minute warning systems" where you have to pit directly or close to directly after the warning light. Using historical data works well too, but only if you've been to that track before. I'd like to have a quick display that shows fuel usage per lap so that on a practice day you can gauge max stint duration (if less than 2 hours) with a real number, and that shows a real number amount of fuel left in the tank. Its most likely not all that necessary, but the engineering side of me wants to know the numbers..... Also knowing the difference of fuel usage between drivers seems useful. 

 

A gallon of fuel is ~.5" in the bottom of a standard 18 gallon fuel cell.  When that's sloshing around at well over a g, changing direction every few seconds, how the hell are you going to measure depth with that level of precision? And that .5" is precision, if you get it, is only going to tell you your range within ~7 minutes.  That's a big error.

 

I think the closest you're going to get to comparative fuel consumption data is integrating pulse width over the lap.  It will tell you a lot about how hard people are driving relative to each other, how effective your conservation is, and even effects of traffic. However, you will still struggle to get within the ~2% accuracy you'll need to know which lap is your last. 

 

Further, you'll have a hard time measuring loss out of the vent system, and doing a good job compensating for fuel temp, air temp, etc.  It can all be done, but it's not easy.  

 

So my solution was to forget all that noise, and since you're either going to run to a time target or run out, you're fuel system should allow you to suck out every last drop without fear of getting towed in.  

 

Now, if you want to do some driver-only craziness on rain races, or heavy conservation to get to 2hr for strategy reasons, you can integrate FPW to estimate remaining range, attempt to correct it with MAF/lambda to help accuracy, then use reserve to stretch out the stints and save your bacon if you get 2 cups less fuel in the tank than you expected.  

 

Hope you like programming, and it would be awesome to see that done at the Champcar level.  That strategy is some pro level goodness.

[IPF Racing]

43 minutes ago, Hi_Im_Will said:

 

A gallon of fuel is ~.5" in the bottom of a standard 18 gallon fuel cell.  When that's sloshing around at well over a g, changing direction every few seconds, how the hell are you going to measure depth with that level of precision? And that .5" is precision, if you get it, is only going to tell you your range within ~7 minutes.  That's a big error.

 

I think the closest you're going to get to comparative fuel consumption data is integrating pulse width over the lap.  It will tell you a lot about how hard people are driving relative to each other, how effective your conservation is, and even effects of traffic. However, you will still struggle to get within the ~2% accuracy you'll need to know which lap is your last. 

 

Further, you'll have a hard time measuring loss out of the vent system, and doing a good job compensating for fuel temp, air temp, etc.  It can all be done, but it's not easy.  

 

So my solution was to forget all that noise, and since you're either going to run to a time target or run out, you're fuel system should allow you to suck out every last drop without fear of getting towed in.  

 

Now, if you want to do some driver-only craziness on rain races, or heavy conservation to get to 2hr for strategy reasons, you can integrate FPW to estimate remaining range, attempt to correct it with MAF/lambda to help accuracy, then use reserve to stretch out the stints and save your bacon if you get 2 cups less fuel in the tank than you expected.  

 

Hope you like programming, and it would be awesome to see that done at the Champcar level.  That strategy is some pro level goodness.

 

Totally agree with what you're saying and why we currently only have a low pressure sensor on the surge tank, its hard to use a traditional sensing for level in the tank. Also my tank has too much crap in it to fit a level sensor. PW is an option and easily doable on the CAN gauge with Megasquirt. Your solution is definitely solid when ignoring all that noise and most teams do run to a certain time or just the 2 hour mark so my thoughts may be overkill. 

 

It should be a fun challenge, I've certainly got a few ideas....