September 25, 2000, at the NSX Modified Shop

Dialing in the Stupid Fuel Gauge
"The Driver vs. The Mechanic"

The Fuel cell is triangular I don't think the fuel gauge will work.

Okay, so after our last embarrassment about the stupid fuel gauge at the last TCRA race, Larry says that he is ready to fix this once and for all.  He orders a Stewart Warner fuel gauge for the fuel cell, and he tells me you can calibrate the gauge to work with the fuel cell.  There is a an adjustment screw on the fuel cell that measures the resistance from the floater in the fuel cell.   I look at this with a very skeptical eye.......and I tell him that it ain't gonna work.  He thinks about it, and says it will.  I absolutely positively say it ain't gonna work.  He says of course it will work.  He will set the Fuel gauge to 0 for empty via the adjustment on the fuel cell, and then when we fill it up to the top with fuel, he will make the adjustment on the fuel cell to read full on the gauge.  

I tell him it ain't gonna work, because:
1.  The fuel cell at empty will measure 0 ohms, or whatever.  I agree.  But we don't know how low the fuel pickup goes.  
2.  One you get to the top and set the top, that is fine.  But I maintain that when the tank is half full (11 gallons), it won't read half on the gauge, as the floater will be at half height on the triangular fuel cell, which means that there will be more like 3/4 of a tank, due to the irregular shap.  In other words, the height goes down SLOWER as the closer you get to the bottom, as the width of the cell is bigger.  I maintain that most gas tanks are somewhat boxy, they aren't triangular, and regular gas tanks have floaters that measure linear height, just like the Stewart Warner gauge will.

Larry responds with:
1.  No, you bonehead driver.  He will set a pump to pump gas out of the fuel cell until we know where EMPTY actually is.  So he can check the fuel pickup level.
2.  So if you want to half to read 11 gallons, he will just set the gauge to half, using the adjustment screw.

I respond with:
1.  Okay, good point here, that is a good idea on checking the fuel pickup.
2.  No you bonehead mechanic, if you adjust the meter for half, it will throw off your "full" setting, and then it will throw off the 3/4 and 1/4 settings.

Larry responds with:
1.   Of course it is a good idea, I think of everything.
2.  No, you bonehead driver, the gauge measures volume, not height.

I respond with:
1.  Okay, you get a point there.
2.  No, you bonehead mechanic, the floater has to measure height, as it by definition "floats" on top of the height of the liquid.  Comprendo Senor Bonehead???

Larry responds with:
1.  He is always right when it comes to mechanical stuff.
2.  If you think the 3/4 setting and 1/4 setting will be off, I can calibrate the gauge for those also when we put in gas.

I respond with:
1.  Every bet I ever made with you I won.  And I will win this one also, as it is based on theory that we can prove.
2.  You can't just arbitrarily set the stupid adjustment to all theses settings, it is proportional you idiot.  Setting it at 0 and full will measure the liquid level proportionally.  You can't just set it at 11 gallons to be half, as 11 gallons will probably be at the 1/4 level of the fuel cell, and not the half height of the fuel cell, and expect it to measure 1/4 and 3/4 and full correctly.  3/4 of the height of the fuel cell will really be 7/8 of a tank due to the triangle shape, so the gauge will be confused as you go back down.

So we argue for a couple of days on this.  Frank also comes by, and he says that it ain't gonna work either, and it will also be off because we are on a bench operating a gauge with a standalone 12 volt car battery.  Frank says that when I am driving the car, voltage will vary from 10-14 volts, and thus throw off how the meter reads ohms off of the fuel cell, as it is electrical.  Larry says that the 12 volt battery on the bench is NOT for the fuel gauge to test that is for something else.  And then he looks at both me and Frank like we are amateurs.  Me and Frank think we need to send Larry to a math class.  A remedial math class for 7 year olds.

So the big day finally comes where Larry has time to do the fuel cell.  He wants me there while he is doing this, because he doesn't want to hear any crap that it is his fault that I ran out of gas because he installed a fuel gauge that could never work, in my opinion.

I go over to his shop, and he has rigged up what looks like a mad scientist experiment.

There is a running car at the front of his shop with the hood up.  He takes out a bunch of stuff so  he can get to the voltage regular of the running car, and hooks my fuel gauge up to the voltage regular, to simulate what it will be like when I am driving in the car and I look at the fuel gauge.

Okay, so the fuel cell is on a dolly, and is completely empty.  We have 22 gallons of gas that Mark has on the table.  We have a small electrical pump that is powered by the 12 volt standalone car battery that will pump gas out of the the fuel cell using the two fuel lines (for two fuel pumps in the car) that have a "T" connector that has a single line into the fuel pump.  You follow me so far??  I grab a BC 10 Fire Extinguisher, since we have 22 gallons of gas, a running car, and an electric pump that sparks when you start if off the 12 volt battery.  Larry says the fire extinguisher ain't gonna do shit, because if this stuff blows up, it will blow the fire extinguisher halfway across the parking lot, and our body parts along with it........

Okay, so we pour two gallons of gas into the fuel cell.  Larry fires up the electric pump, and the fuel pickup point pumps out 1 gallon of gas, then stops.  Okay, so now we know that "EMPTY" on the gauge is when there is one gallon left in the cell, as the pickup point can no longer retrieve gas.  Larry sets the analog gauge to "E", using the adjustment for the floater on the fuel cell, which sends a signal via a wire to the fuel gauge to signal "E".

I keep telling Larry he is going to owe me a steak dinner, because even if he calibrates the gauge as we fill the cell (he wants to calibrate at 1/4, 1/2, 3/4, and full), when we empty the fuel cell, it ain't gonna read the same, as the proportions are all screwed up by him screwing with the adjustment so much.  He insists it will work......and bets me on it.   I tell him I am going to put on my web page what an knucklehead he is about his "calibration theory".  

This is how they blew up the FBI Building in Oklahoma

Okay, so now we put in 5 gallons,  which is 1/4 full.  Larry sets the fuel gauge to 1/4 based on where the floater is.  We now put another 6 gallons, to make 11, and which is now half a tank.  Larry sets gauge to half.  And now I am jumping up and down saying I am right, I know I am right.  Because.....the fuel cell is only filling up on ONE SIDE of the fuel cell, the driver's side of the fuel cell.  So I maintain the floater is all screwed up, because the floater is now at the top, and we still have to fill the OTHER side of the fuel cell.   The floater can't go any higher to measure the rest of the gas we have to put in.  Larry smiles, and says, "No, you are an idiot driver".  And we find out why the gas had to be "burped" in.  The breather tube gets pinched from the top of the fuel cell pressing into the top of the gas tank compartment, and that is why gas did not flow smoothly into the cell.

But, now I have a minor brainstorm.  Remember at the last race, and Larry did a visual check on the fuel cell, and he said it was almost full, and I took his word for it, because we were "burping" the fuel cell a lot, and it seemed full?  Well....I think Larry did see the fuel cell filled at the top, the problem is that he was looking at a 3 inch section of the fuel cell at the top that happened to be the side that FILLED FIRST.  So by looking at that section, the tank was full according to Larry's eyes.  When in fact it was half full, as the other side of the cell did not fill up yet.  Let's see, that would mean for half a fuel cell, 11 gallons = 44 miles, and I did 10 practice laps and 8 laps in the race before I ran out of gas prematurely (I needed to go 15 laps in the race), 18 laps around Willow Springs = about 47 miles.  I'll be damned.

But now, I am jumping up and down, tell Larry what a moron he is, because the gauge will be all screwed up, as when we keep filling the tank, up, the floater ain't gonna move anymore, because the right side of the fuel cell is at the top, so it can't move anymore, so now the gauge will be off due to it being a triangular fuel cell AND because it only fills the left side of the fuel cell first.  Damn, I am a theoretical genius.  

Larry ignores me.  We put in 5.5 gallons to get us to 16.5 gallons, or 3/4 of a tank.  Larry again calibrates his gauge to 3/4.  I am way is this going to work.  We put in 5.5 more gallons to get us to 22 gallons, and Larry makes his final setting to "FULL" on the gauge.  I start thinking about using some equation to represent the futility of this exercise...

Fuel cell has two lines to battery powered fuel pump that empties into a bucket.

I am jumping up and down.  I tell Larry, "You gauge is going to be so far off as we empty the gas out, it will be ridiculous.  3/4 on the gauge will not be correct, 1/2 won't be correct, and 1/4 won't be close to being correct, because you kept recalibrating the gauge as you filled it up.  The floater measure LINEAR height, not volume, you bonehead.  You can't change adjustments on the way up, and then expect it to read the lower levels correctly when you pour gas out.  The gauge measures linear, and you kept adjusting the gauge as you went up.  Therefore when it goes down, since you changed what the gauge read for 3/4 and full, it will be all screwed up at 1/4 and 1/2."

Larry just shakes his head and smiles, and turns on the pump.  We pump out 5.5 gallons of gas, which means we  have 16.5 gallons left, or 3/4 of a tank.  We look at the gauge.   It measure 3/4. reads correctly.  But I am not worried about that reading, it is the lower part of the gauge that will be incorrect.   We start to pump out another 5.5 gallons to get us to 11 gallons, or half a tank, and I predict the gauge will still measure something like 5/8ths, as the two fuel lines are pumping out gas out of both sides of the fuel cell evenly, so now height will be ALL screwed up.  The height is now much lower than it was before.  Because when we filled it up, it filled up one side to the top first, and then the other side to the top.  But now, as we are simulating what the fuel pumps in car the would do, it is emptying both sides of the fuel cell at the same time, thus the level of gas is going down faster.  I am cracking up, you mechanics are idiots.   We have now pumped out 11 gallons, and I smugly look at the gauge to see how far it is off.  I look at it.....and it says, exactly 1/2.   Just like it should be, the reading is accurate.  Uh I have a puzzled look on my face, and Larry is jumping up and down laughing his ass off.  "You drivers don't know shit" he says in my face, with a big smile on his face, "Man you drivers are dumbasses."    

I am in shock.  I can't believe I am wrong.  I tell Larry, "Wait.  You still have to check the gauge at 1/4 and empty.  It will be wrong at those levels."  He shakes his head again......we pump out another 5.5 gallons to leave us with 5.5 gallons left or a quarter tank left, and the gauge reads 1/4.  It is accurate.  DAMN.   He pumps out another 4.5 to get to 21 gallons, checks the gauge, and it is on "E" exactly.   Unbelievable, I can't believe I was wrong on this......Larry just keeps laughing......and reminds my dumbass that I owe him a steak.  I guess I have to concede victory on this to him........this one time......