10 mpg 3.2 - can’t find where the fuel is going

[PeteKz]

Re bad city, but good highway mileage: Has anyone mentioned a throttle position switch badly out of adjustment yet? If so, you might be running open loop (with full enrichment) a lot in city driving, but not at cruise on the highway.

[proporsche]

Karl..this one............
https://forums.pelicanparts.com/porsche-911-technical-forum/1046664-fuel-overflow-tank.html

[Discseven]

Originally Posted by PeteKz

Re bad city, but good highway mileage: Has anyone mentioned a throttle position switch badly out of adjustment yet? If so, you might be running open loop (with full enrichment) a lot in city driving, but not at cruise on the highway.

Pete... interesting. Am familiar with two throttle switches. Switch towards front of the car. And there's one on the passenger side of throttle body. I recall checking the front one. It clicked when the throttle is moved off it's base position. Switch on the side I do not ever recall checking.

No idea how to genuinely check either of those switches. School me on this and it'll get done.

Originally Posted by proporsche

Karl..this one............
https://forums.pelicanparts.com/porsche-911-technical-forum/1046664-fuel-overflow-tank.html

Ivan... Canister there is not the one I'm thinking of. Bear in mind, my 911 is an '80... engine is '85. That guys 911 is '86. Could be different fuel overflow configurations between '80 and '86. In my front left wheel well there is the water reservoir tank. There's also 10 gallons of fuel "missing" per tank---"missing" by comparison to the engine getting 18 mpg city. If that amount was leaking, it would/should be very noticeable. No leaks. So this "missing" fuel must be running through the engine.
.

[Discseven]

Looking for "friction"...

Raised the rear end. Checked both rear wheels for any rock (bearings.) No rock of any wheel on its spindle.

With rear in the air... checked spin of rear end with the wheels on... then off. In neutral and in gear. Both sides are even in resistance always. Smooth rotation. Some fluidity. Removed RR wheel and spun hub. Slightly more resistance now. Am calling that the difference in resistance when turning tire's OD compared to smaller hub OD. (Only checked RR wheel this way.) Can't say I have any great experience as to how the rear end should exactly feel when turning it. Over the years, I've done this turning-of-the wheels for any number of reasons. What I sense now is based on those infrequent turnings.

Pulled RR brake pads and spun hub. Slightly less resistance... as I'd expect.

Grinding when turning rear axles? None.

Hesitation moments during turns? None.

CVs were packed before rebuild. All boots are attached to shaft and in good form.

Front wheels were checked for rock-on-spindle and for spin a while back. Both sides are good. Of course front wheels are easier to spin than rear wheels.

Gear box lube? Checked level. Full. Fluid looks fresh---as it should. Was drained and refilled with new Motul during engine rebuild.

.

[proporsche]

Karl, your model has it too..just look under the left front fender for loose hoses .part number /91120107702

Ivan

[PeteKz]

Karl, re throttle switches. As I said, I'm not as familiar with the setup on the 3.2 Motronic engines. I hope someone else with more expertise in the Motronic engines will weigh in on this issue.

However, there are two switches on the throttle body. One indicates to the ECU when the engine is at idle, and the other clicks when the throttle goes past about halfway. That's what signals the ECU that you are at "full throttle" and it goes to open loop full enrichment for full throttle operation. If it was misadjusted, such that the switch was clicking at, say, 1/4 throttle, then it would go to full enrichment at 1/4 throttle.

More than that, I would have to dig into the manual.

[wazzz]

To add to what Pete says above, it is also important to check that the idle switch closes as soon as you release the throttle, because that's what tells the Motronic to cut fuel supply to injectors until the rpm has gone down to some threshold (can't remember exact threshold value, but must be somewhere around 1200 rpm). This plays an important role in fuel economy.

[Discseven]

Originally Posted by proporsche

Karl, your model has it too..just look under the left front fender for loose hoses .part number /91120107702

Ivan

Ivan... will look in there later today.

Originally Posted by PeteKz

Karl, re throttle switches. As I said, I'm not as familiar with the setup on the 3.2 Motronic engines. I hope someone else with more expertise in the Motronic engines will weigh in on this issue.

However, there are two switches on the throttle body. One indicates to the ECU when the engine is at idle, and the other clicks when the throttle goes past about halfway. That's what signals the ECU that you are at "full throttle" and it goes to open loop full enrichment for full throttle operation. If it was misadjusted, such that the switch was clicking at, say, 1/4 throttle, then it would go to full enrichment at 1/4 throttle.

More than that, I would have to dig into the manual.

Pete... Thanks! Not putting it on you to dig! I'll do it.

Originally Posted by wazzz

To add to what Pete says above, it is also important to check that the idle switch closes as soon as you release the throttle, because that's what tells the Motronic to cut fuel supply to injectors until the rpm has gone down to some threshold (can't remember exact threshold value, but must be somewhere around 1200 rpm). This plays an important role in fuel economy.

Gilles... Got it. Thank you. Get back to topic of these 2 switches after sorting out how to test them and what positions they should be in---and where mine are.
.

[Discseven]

AFM wheel adjustment - moved 3 notches clockwise

Done to see if wideband picks up any change. If wideband fails to see a change, I’d move wheel further clockwise. If wheel progressively changes position and wideband data doesn’t change… reputation of wideband system & yours truly would obviously be in the toilet. O2 levels do change. Was also done to see what comes of engine's performance. It declines…

Here’s comparison of what was and what is:



Wheel’s original position. O2 logs for this wheel position are below. 14.64 AFR / 1.0 Lambda is always at horizontal red dot-dash line.




City.




Steady 3000 rpm in 4th gear.




WOT to redline.




New wheel position. Adjusted on 5.16.25. Wheel is rotated clockwise 3 notches from prior position.



Resulting logs...



City.




Steady 3000 rpm in 4th gear.




WOT to redline.



Engine setup notes:

Installed: Bosch 158 fuel injectors*, Iridium spark plugs, Blaster coil.
ECU fuel adjustment setting = 0
ECU chip = stock
Engine when recording logs: done at op temp / 210
Wideband = dual 4.9 Bosch sensors placed symmetrically in SSIs with dual Innovate LC-2 controllers.
Logging software = LogWorks
Trace lines: green = 1-2-3 bank / blue = 4-5-6

*Tested by Bill, Mr. Injector.

Was prepared to go through any number of progressive notch adjustments. 3 notch adjustment is enough. Small adjustment to the barn door's tension... makes a clear difference in O2 and engine performance. That O2 moved in the right direction, leaner, indicates (to me) wideband gear is working. Working correctly? To repeat myself because of the oddballness of this fuel-use puzzle, wideband gear has been tested and both Bosch sensors even replaced in an attempt to find the log readings faulty. After a considerable effort to find fault in the system... that didn't happen.

Engine’s performance with 3 notch clockwise adjustment:

Lumpy cold idle.
Sputters lightly when getting off throttle.
Slightly lower warm idle rpm.
WOT to redline… Seems sluggish but not sure on this.

Absolutely no question the engine ran better with the prior (original) wheel positioning. Wheel goes back to where it was.
.

[rsscotty]

[Discseven]

RS... Nice! Thank you.

[Discseven]

Throttle position switches

Tested ohm meter to itself = OK

Idle throttle switch: Off throttle = 0 ohms. On throttle = infinite.
Tested multiple times. Same results.

Full throttle switch: Partial throttle = infinite. Full throttle = 0 ohms.
Tested multiple times. Same results.

Checked positions of throttle plate…



No play in throttle cable when plate is closed.




At full throttle, plate looks & feels 90 degrees to throat. This is when moving the throttle lever by hand. Still need to get a person flooring the gas pedal to see where this goes. EDIT: Just did this. Pedal floored = fully open throttle plate (butterfly.)

Opened full throttle switch…







Bosch part # 0 280 120 309. Looked around just in case mine was bad. Amusing price range... $60 to $677 for a "genuine" part.

When should these points close?… According to those looking at Bentley, it's at 100% throttle. I assume this being relative to the throttle's rotation from 0 to 100% and not a measure of pedal movement. Bosch Motronic book I have offers no spec. Only says “loosen the screws to adjust.” Steve Wong says "You need to see continuity on the upper two pins after 3/4 of full throttle. Without it functioning, you will stay in part throttle mode, and never go to full throttle mapping. The difference is 20 hp or more." This quote is from https://forums.pelicanparts.com/porsche-911-technical-forum/452398-checking-wot-switch.html post #4.
.

[wazzz]

Your test results show that both switches are good. The WOT switch closes at about 70% of full range.
As for the idle switch, there is a setting procedure. The drag arm should be set with a 0.2mm play. Here is a copy of the related page from the DME Test Plan. PM me your email if you want a pdf copy of the full manual.

[Discseven]

Originally Posted by wazzz

Your test results show that both switches are good. The WOT switch closes at about 70% of full range.
As for the idle switch, there is a setting procedure. The drag arm should be set with a 0.2mm play. Here is a copy of the related page from the DME Test Plan. PM me your email if you want a pdf copy of the full manual.

Gilles... Thanks for that spec image. Not destroying the switch from "impact"... nice piece of info. I set that up so switch trigger begins to move when drag arm begins to move. Interesting to see there's a spec. Will PM you.
.

[wazzz]

Karl, got your pm. PDF sent to your email. Sorry for the delay. Time shift...

[Ian Comerford]

This won't necessarily contribute to the diagnosis and apologies if I've missed it elsewhere in the thread but have you considered having an engine oil sample analysed for fuel content/saturation? DO you know whether you have a lot of unburnt fuel in there? It may tell you a couple of things and isn't expensive to do. It will also inform an early, or not, engine oil change.
Ian

[Discseven]

Originally Posted by wazzz

Karl, got your pm. PDF sent to your email. Sorry for the delay. Time shift...

Gilles... Got it. Much appreciated!

Originally Posted by Ian Comerford

This won't necessarily contribute to the diagnosis and apologies if I've missed it elsewhere in the thread but have you considered having an engine oil sample analysed for fuel content/saturation? DO you know whether you have a lot of unburnt fuel in there? It may tell you a couple of things and isn't expensive to do. It will also inform an early, or not, engine oil change.
Ian

Ian... that has not been done. Will be now. Thank you ~~~

Speculating. If fuel exists in crank oil, I suspect there'd be a problem with that engine's fuel injectors dripping fuel after shut down. Dripped fuel then makes it past what open valves there are, and poorly sealing valves if they exist, then along the bottom of cylinder, past the rings... into the case.

Just looked at Bill's report on my Bosch 158 injectors. They were checked as part of hunting down this fuel use issue. His report:




I plan to use Blackstone for the oil analysis. Will wait a while for what comments there may be on this company before getting on with the process.
.

[Discseven]

With my WOT switch opened, thought to mount it without the cover and see where the switch signals WOT. Early readings suggested moving the switch clockwise... It has ended up fully clockwise given the mounting hole slots. My estimate is this switch signals WOT at beginning at 70 percent throttle. To increase this percentage (given my switch), slot holes could be Dremel-adjusted.



Follower’s position on cam when points just make contact according to ohm meter.

Black arrows:
Left = follower’s idle position on cam
Right = follower’s WOT position on cam

White line = 50% throttle position
Red line = 75% throttle position

Prior signaling---not recorded---cannot be any later than it is now.
.

[Discseven]

Originally Posted by proporsche

i would test the consumption without filling the the whole tank-and go from there.
I suppose there no problem with your smelling, you would smell fuel right?
The reason i am asking is from my experience.Once i drove from south France to Bohemia --cc1300 km one way..So filled my tank in Saint Tropez where i lived close-by..
After filling the tank i took off and in couple of seconds i smelled fuel..it was from left side of the front fender.Where i prior remove the fuel tank breather canister-mistake.I drove gently for the rest of the trip ..never filled to the top again.
So do you have the canister there or not?In city taking off and braking the fuel moves in the tank much more then driving smoothly on the highway..
just an idea

Ivan

Ivan... There are two canisters in my front wheel well. The bottom appears to be a water reservoir. The top, smaller, I assume to be fuel related. I think if 10 gallons per tank is leaking from there---or anywhere---there would be a distinct smell of it. Never any fuel smell in the least. I looked all around both canisters as much as possible given their location, and I see no sign of fuel leaking. I'm sure there would be clean areas where leaked fuel would have removed road grime. Spotted nothing like that.

Your thinking about taking off in the city and compared to driving smoothly on the highway... credit due you. That's clever. Thank you.
.

[proporsche]

Originally Posted by Discseven

Ivan... There are two canisters in my front wheel well. The bottom appears to be a water reservoir. The top, smaller, I assume to be fuel related. I think if 10 gallons per tank is leaking from there---or anywhere---there would be a distinct smell of it. Never any fuel smell in the least. I looked all around both canisters as much as possible given their location, and I see no sign of fuel leaking. I'm sure there would be clean areas where leaked fuel would have removed road grime. Spotted nothing like that.

Your thinking about taking off in the city and compared to driving smoothly on the highway... credit due you. That's clever. Thank you.
.

sounds good without problem..
Still ,you go through so much testing to find out the fuel los --crazy...still thinking it is something missing.If i understand 10 gallon difference -between freeway and city right...?