Accuracy of fuel range indication

Scorpio / Scorpio Cosworth Discussions - Questions, problem resolution, general talk, technical tips and modifications.
User avatar
DPDISXR4Ti
Site Admin
Posts: 13526
Joined: Wed Jan 08, 2003 11:40 pm
Location: New York

Accuracy of fuel range indication

Post by DPDISXR4Ti » Sun Jan 12, 2020 1:18 pm

On my Audi when the fuel message indicator tells me I have 0 (zero) miles until empty, I actually have about 2 gallons left and so could travel another 50+ miles if I had to. I'm wondering what the similar real story is with the Scorpio fuel range indication. Anyone have any real-world data-points here?

Taking the fuel computer out of the picture for a moment, IIRC the Scorpio fuel tank is 17 gallons, so even if I was getting a lousy 20MPG, range should be 340 miles.

Any yes, I know it's not a great idea to run the tank all the way down until empty.
Brad

User avatar
andyofcolumbusmerkur
Level 4
Level 4
Posts: 387
Joined: Mon Jul 15, 2013 7:58 pm

Re: Accuracy of fuel range indication

Post by andyofcolumbusmerkur » Mon Jan 13, 2020 10:19 am

On my Toyota truck I have about 6 gallons left when the needle barely touches the E, and the low fuel light illuminates. I thought this was weird and read on the forums about it. I guess Toyota took the version with the lowest mpg (v6/4x4/extended cab) and designed it to get you 100 miles to a gas station after you notice the light. Remember the vehicle is sold all around the world and some places gas station are not a prevalent as here in the US. Can't add any info on the scorpio but I thought this was interesting.
You're only as good as the last car you build.

User avatar
DPDISXR4Ti
Site Admin
Posts: 13526
Joined: Wed Jan 08, 2003 11:40 pm
Location: New York

Re: Accuracy of fuel range indication

Post by DPDISXR4Ti » Mon Jan 13, 2020 12:58 pm

I can provide a little bit of an update... The fuel range indication is NOT very accurate. My wife went out yesterday for what would amount to about a 25 mile drive. When she left, the range was indicated 36 miles to empty. When she returned, it indicated 31 miles to empty. She was very proud of how efficiently she drove. :notworthy :beats head
Brad

Ed Lijewski
Level 8
Posts: 7193
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: Accuracy of fuel range indication

Post by Ed Lijewski » Mon Jan 13, 2020 3:28 pm

Meaningless. Unless the car was completely level when parked then later started and driven, the computer's calculations will be based on skewed fuel tank level readings which during the drive become more realistic.

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

User avatar
andyofcolumbusmerkur
Level 4
Level 4
Posts: 387
Joined: Mon Jul 15, 2013 7:58 pm

Re: Accuracy of fuel range indication

Post by andyofcolumbusmerkur » Mon Jan 13, 2020 7:39 pm

You could pump out all the fuel, even replace pump afterwards with a new one for reliability. Then fill it to the top at your favorite gas station (even the same pump number) as the reference point. See how much you are actually using versus what the computer says. I found you need to average it a few times since sometimes you get a little more or less gas, or the ground is not 100% flat.
You're only as good as the last car you build.

brokencase
Level 4
Level 4
Posts: 414
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: Accuracy of fuel range indication

Post by brokencase » Thu Jan 16, 2020 8:17 pm

I've never ran it all the way out. But I seem to recall reading somewhere that it was about a 20-25 mile reserve.

The Scorpio fuel computer is pretty accurate. The tank level has some influence, mainly for resetting the range but the majority of the calculation is based off of the fuel injection pulse width (fuel used) and the speed sensor (distance traveled). Unlevel terrain does not introduce any error.

If the majority of the tank was consumed by a heavy footed driver and then your wife gets in and drives conservatively, then yes, the computer's estimate of miles to empty will have to drastically change. Nothing wrong with that. My new truck does the same thing.

There is just one issue. The fuel computer in the Scorpio was made in the time when there was no ethanol in the gas. E10 has about 5% less energy content. So today the MPG reading will be a little lower. So if you see 26mpg on a long trip you actually did a little better.

Ed Lijewski
Level 8
Posts: 7193
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: Accuracy of fuel range indication

Post by Ed Lijewski » Fri Jan 17, 2020 7:37 am

I assume the fuel computer reads TPS positions against speed sensor input.

A common issue on most cars that don't get driven regularly or get fuel system additives is that the fuel level sensor sticks producing erroneous readings. Tom Taylor at Rick Auto reported that Valvoline fuel system cleaner worked for his car. I found that it freed up a sticky float on my car as well.

YMMV

Edit: See last article here:

https://www.rockauto.com/Newsletter/arc ... -3-15.html
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

User avatar
andyofcolumbusmerkur
Level 4
Level 4
Posts: 387
Joined: Mon Jul 15, 2013 7:58 pm

Re: Accuracy of fuel range indication

Post by andyofcolumbusmerkur » Fri Jan 17, 2020 8:10 pm

On my Moms Impreza I can get up to 45mph then reset the trip odometer and basically coast home. Going downhill a little and just barely touching the throttle. The display will read something like 80mpg when she gets in the car and pulls away. Pretty funny.
You're only as good as the last car you build.

User avatar
DPDISXR4Ti
Site Admin
Posts: 13526
Joined: Wed Jan 08, 2003 11:40 pm
Location: New York

Re: Accuracy of fuel range indication

Post by DPDISXR4Ti » Fri Jan 17, 2020 8:46 pm

brokencase wrote:
Thu Jan 16, 2020 8:17 pm
But I seem to recall reading somewhere that it was about a 20-25 mile reserve.
Thanks for actually answering my question. :mrgreen:

Apparently I'll be testing this soon as a ~12 mile trip was taken the other day and I am told the range is now indicated at 10 miles. I'll be getting it filled up this week at the gas station that is about 10 miles away. Do I feel lucky, or should I bring a gas can with me?

I'll take note of the gallons required to fill it up. Anyone know the exact fuel tank capacity? I know it's 17-something.

OBTW, entire fuel pump assembly is new. Thanks Darren!
Brad

brokencase
Level 4
Level 4
Posts: 414
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: Accuracy of fuel range indication

Post by brokencase » Fri Jan 17, 2020 11:52 pm

Ed Lijewski wrote:
Fri Jan 17, 2020 7:37 am
I assume the fuel computer reads TPS positions against speed sensor input.
No, it just needs the instantaneous fuel consumption, and that is proportional to the pulse width sent to the injectors.
Airflow is not consistent and can change with temperature. In addition, the EECIV runs the car rich when the engine is cold and that eats more fuel and is not expressed by the throttle position.

The ethanol/fuel mix is interesting...

Lets say your rolling down the highway in 1988 with pure gasoline at a constant speed and the EECIV is running in closed loop mode with the O2 sensors. Your burning the fuel at a perfect stoichiometric ratio and you making a fixed amount of power to to keep the car moving along at 60 mph. Lets say that the fuel computer reads 27mpg.

Now let's say we substitute E10 that has 5 percent less energy content than regular gas. To make the same horse power we have to burn 5% more.
Well it's not directly obvious but the energy content of the fuel just so happens to be proportional to the stoichiometric ratio. And since we are running some ethanol the stoichiometric ratio has to change. It interesting because the energy produced is more about the oxygen combined with the fuel rather than what the fuel is. The EECIV sees the difference in the O2 sensor reading with the ethanol blend and automatically extends the pulse width to the injectors. However the fuel computer does not know about the stoichiometric change. It just sees that we are burning more fuel that normal. So it reads the 5% difference 27mpg*.95 = 25.65mpg.

So i was a little wrong in my earlier statement. The fuel computer is still calculating correctly with E10, but you should expect lower reading for mpg with E10.
Last edited by brokencase on Sun Jan 19, 2020 9:23 pm, edited 1 time in total.

Ed Lijewski
Level 8
Posts: 7193
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: Accuracy of fuel range indication

Post by Ed Lijewski » Sat Jan 18, 2020 7:46 am

I've observed that with speed control On, MPG skews slightly favorably even though approaching hills the system increases rpms sharply to maintain constant speed. Whereas with speed control off, on the same/similar route I can modulate the throttle deftly to maintain exact speed without that rpm burst which surely should affect/diminish omg efficiency. Yet the trip computer calculation somehow isn't affected by those rpm bursts.

(Don't suggest the speed control/trip computer is better at maintaining constant speed than my highly trained foot.🙄)

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

Ed Lijewski
Level 8
Posts: 7193
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: Accuracy of fuel range indication

Post by Ed Lijewski » Sat Jan 18, 2020 9:36 am

Re pulse width signals, please post a pic of the Scorpio EEC spinouts identifying which pin connects to the trip computer.

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

User avatar
andyofcolumbusmerkur
Level 4
Level 4
Posts: 387
Joined: Mon Jul 15, 2013 7:58 pm

Re: Accuracy of fuel range indication

Post by andyofcolumbusmerkur » Sat Jan 18, 2020 11:54 am

What about changes in air pressure, humidity and of course temperature?
You're only as good as the last car you build.

brokencase
Level 4
Level 4
Posts: 414
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: Accuracy of fuel range indication

Post by brokencase » Sat Jan 18, 2020 12:38 pm

Pin 34 of the EECIV, pages 36 and 90 of the Electrical and Vacuum Troubleshooting manual.

What about changes in air pressure, humidity and of course temperature? It does not matter, to calculate MPG you just need to measure the fuel rate and the distance traveled.

I hate to say it Ed, but I trust the cruise control to do a better job than your foot!
I know, I know, it a cold, calculating, and heartless world out there.

Ed Lijewski
Level 8
Posts: 7193
Joined: Thu Sep 04, 2003 5:53 pm
Location: The Belly of The Beast

Re: Accuracy of fuel range indication

Post by Ed Lijewski » Sun Jan 19, 2020 11:35 am

I enjoy saying it Dean: only a die-hard techie would put total trust in technology.

https://www.caranddriver.com/news/a1536 ... to-fix-it/

YMMV
Descartes: "Cogito Ergo Sum"
Lijewski: "Sum Ergo Drive-O. Mucho!

Post Reply