Which way to turn 3.2 air flow disk?

[mysocal911]

Originally Posted by Discseven

I'll have a go at that after completing the city MPG test with the Ox sensor disconnected.

To "probe" the circuit, will wrap the scope wire around the male connector on the car side and pushing the female side of the connector onto it.

With no idea what your background in Porsche electronics is Gilles, I must ask if you're absolutely certain this test will not fry my ECU/DME or the Ox sensor?

This is laughable, as he knows more than most who have claimed knowledge and posted about the 911 3.2 DME ECM in this thread!

[scarceller]

If you want a very quick test to verify the O2 signal line (DME Pin #24) is functional you can do this easily with a simple AA 1.5vdc battery.

- Engine running at idle
- Stock O2 disconnected
- In the engine bay locate the signal line in the harness for the stock O2
- First, test to confirm that line is at 0.5vdc when it's unplugged from the O2 sensor.
- Next, take a jumper and jumper ground to the O2 signal line and leave it jumpered for few seconds. This will set 0vdc on on the line and will tell the DME we have a false lean condition.
- When DME sees this false lean condition it will start to richen the fuel mixture and you will hear a noticeable change in the exhaust note.
- If you have a WBO2 gauge monitoring the mixture you will instantly see the mixture start going way rich. If you had been at 14.0AFR with sensor unplugged the mixture will easily go into the low 13s or 12s. You can see this really well on the WBO2 gauge.
- then remove the ground jumper and you'll be back at the default 14.0AFR
- Then, we can now take a 1.5v AA battery and ground the '-' of battery to a ground screw and then jumper the '+' of battery to the O2 signal line. This puts 1.5vdc on the signal line. A normal O2 sensor outputs no more than 1vdc but the 1.5vdc won't damage anything.
- The 1.5vdc is seen by the DME as a false rich condition and the DME will start leaning the mixture.
- The engine will not like the super lean mixture and the idle quality will instantly suffer. Engine will not like idling at AFR greater than 16.0
- Again the WBO2 will instantly reflect the super lean mixture.

WBO2 gauges are wonderful for doing testing like this. Super simple method to prove that the DME closed-loop ability is in proper working order.

Then you can also just plug in the stock O2 sensor and monitor the same signal line as mentioned above. But what if you have a faulty stock O2 sensor you won't be able to easily prove that the DME and the wiring is in working order. But with the 1.5vdc AA battery and ground method it's easy.

[mysocal911]

The best test is still by actually monitoring the O2 signal with a properly setup o-scope.

[scarceller]

Also, the DME has a self test pin to test O2 circuit. See bottom of this page scan.
Pin A of test connector goes to DME pin #12 and that's a special output just for testing that you are indeed in closed-loop O2 mode.
And for the rest of the testing on the page if you don't have a CO meter you can easily use a WBO2 gauge instead

[mysocal911]

Here's where the O2 signal (rich/lean) goes to in the DME ECM;



Simple, right?

Got this info from this company; https://www.systemsc.com/technical.htm

They reversed engineered the 911 3.2 DME ECM in 1984 to develop schematics for repairing them.

[Discseven]

Thank you for all that info Sal. Am adding the 1.5v battery test to the menu.

City MPG results were just calculated. Is with the Ox sensor disconnected...

Highway = 27 mpg (From prior run to the Keys.)
City = 10 mpg

City mpg is with exact same fill up procedure at exactly the same pump as was done for the highway test. For city... throttle was not heavy footed and I made a point of coasting to red lights. 10 is not good news.
.

[mysocal911]

Originally Posted by Discseven

Thank you for all that info Sal. Am adding the 1.5v battery test to the menu.

City MPG results were just calculated. Is with the Ox sensor disconnected...

Highway = 27 mpg (From prior run to the Keys.)
City = 10 mpg

City mpg is with exact same fill up procedure at exactly the same pump as was done for the highway test. For city... throttle was not heavy footed and I made a point of coasting to red lights. 10 is not good news.
.

Hopefully, you don't have a CAT! If you do, remove it or please carry a fire extinguisher.
My '88 911 3.2 gets 18-21 MPG in the city, with an O2 sensor always connected.

[Discseven]

Originally Posted by mysocal911

Hopefully, you don't have a CAT! If you do, remove it or please carry a fire extinguisher.
My '88 911 3.2 gets 18-21 MPG in the city, with an O2 sensor always connected.

No CAT Dave. I do carry 2 Elements. Used to get 18 city. Goal is to get the engine back to that.
.

[Discseven]

Originally Posted by scarceller

If you want a very quick test to verify the O2 signal line (DME Pin #24) is functional you can do this easily with a simple AA 1.5vdc battery.

- Engine running at idle
- Stock O2 disconnected
- In the engine bay locate the signal line in the harness for the stock O2
- First, test to confirm that line is at 0.5vdc when it's unplugged from the O2 sensor.
- Next, take a jumper and jumper ground to the O2 signal line and leave it jumpered for few seconds. This will set 0vdc on on the line and will tell the DME we have a false lean condition.
- When DME sees this false lean condition it will start to richen the fuel mixture and you will hear a noticeable change in the exhaust note.
- If you have a WBO2 gauge monitoring the mixture you will instantly see the mixture start going way rich. If you had been at 14.0AFR with sensor unplugged the mixture will easily go into the low 13s or 12s. You can see this really well on the WBO2 gauge.
- then remove the ground jumper and you'll be back at the default 14.0AFR
- Then, we can now take a 1.5v AA battery and ground the '-' of battery to a ground screw and then jumper the '+' of battery to the O2 signal line. This puts 1.5vdc on the signal line. A normal O2 sensor outputs no more than 1vdc but the 1.5vdc won't damage anything.
- The 1.5vdc is seen by the DME as a false rich condition and the DME will start leaning the mixture.
- The engine will not like the super lean mixture and the idle quality will instantly suffer. Engine will not like idling at AFR greater than 16.0
- Again the WBO2 will instantly reflect the super lean mixture.

WBO2 gauges are wonderful for doing testing like this. Super simple method to prove that the DME closed-loop ability is in proper working order.

Then you can also just plug in the stock O2 sensor and monitor the same signal line as mentioned above. But what if you have a faulty stock O2 sensor you won't be able to easily prove that the DME and the wiring is in working order. But with the 1.5vdc AA battery and ground method it's easy.

Thought I had a 1.5v batt on hand but don't. To progress today without chasing a battery, I just tested for the .2 - .8 voltage oscillation with a 3mm Allen in hand. Turned the air bypass screw different ways and it did indeed alter the voltage by obvious fractions. iPhone overheated by the engine so no vid. Set bypass screw so oscillation is generally between 0 - .9v. There are spikes higher and lower on occasion. Couldn't tighten the resulting range better than as noted. Am thinking if the average should be at .5v, the setting as it is is good and the DME is doing its job.

Why the highway mpg would be reasonable and the city mpg terrible... that's out of my scope of understanding (for now.) I'll keep at this until it all makes sense and the MPG is in order. I understand you recommending a wideband O2 gauge. I'd like to avoid that cost and adding hardware to the car (or shelf) if possible so am going at this with a scope for the time being.

Plan to do 1.5v test tomorrow. Thank you Sal.

Monday am driving to Orlando so another highway mpg test is in order. Will leave the Ox sensor disconnected given having adjusted the air bypass.
.

[mysocal911]

Originally Posted by Discseven

Thought I had a 1.5v batt on hand but don't. To progress today without chasing a battery, I just tested for the .2 - .8 voltage oscillation with a 3mm Allen in hand. Turned the air bypass screw different ways and it did indeed alter the voltage by obvious fractions. iPhone overheated by the engine so no vid. Set bypass screw so oscillation is generally between 0 - .9v. There are spikes higher and lower on occasion. Couldn't tighten the resulting range better than as noted. Am thinking if the average should be at .5v, the setting as it is is good and the DME is doing its job.

Why the highway mpg would be reasonable and the city mpg terrible... that's out of my scope of understanding (for now.) I'll keep at this until it all makes sense and the MPG is in order. I understand you recommending a wideband O2 gauge. I'd like to avoid that cost and adding hardware to the car (or shelf) if possible so am going at this with a scope for the time being.

Plan to do 1.5v test tomorrow. Thank you Sal.

Monday am driving to Orlando so another highway mpg test is in order. Will leave the Ox sensor disconnected given having adjusted the air bypass.
.

Hopefully, you are aware that the mixture screw ONLY affects the idle mixture?
Without testing the O2 system with the engine under under a load, it's basically a waste of time!

[Discseven]

Originally Posted by mysocal911

Hopefully, you are aware that the mixture screw ONLY affects the idle mixture?
Without testing the O2 system with the engine under under a load, it's basically a waste of time!

What Steve Wong has to say about adjusting the mixture screw...

"...you could use a DVM and adjust the screw measuring the voltage fluctuation of the O2 sensor with it unhooked. Properly adjusted at idle, it should fluctuate between 0.2 and 0.7 volts. This will set it at Lambda 1.0, where the O2 sensor will try to modulate the mixture to when connected. Clockwise on the screw reduces the opening of the air bypass, thus richening it, and vice versa."

Post #5 https://forums.pelicanparts.com/porsche-911-technical-forum/170123-air-fuel-mixture-adjustment.html

What I gather while wasting my time is the idle adjustment in the throttle body controls idle speed. What the mixture adjustment in the AFM does is set the AFR at idle. Lambda is then set at idle by adjusting the AFM mixture screw.
.

[Discseven]

Originally Posted by scarceller

If you want a very quick test to verify the O2 signal line (DME Pin #24) is functional you can do this easily with a simple AA 1.5vdc battery.

- Engine running at idle
- Stock O2 disconnected
- In the engine bay locate the signal line in the harness for the stock O2
- First, test to confirm that line is at 0.5vdc when it's unplugged from the O2 sensor.
- Next, take a jumper and jumper ground to the O2 signal line and leave it jumpered for few seconds. This will set 0vdc on on the line and will tell the DME we have a false lean condition.
- When DME sees this false lean condition it will start to richen the fuel mixture and you will hear a noticeable change in the exhaust note.
- If you have a WBO2 gauge monitoring the mixture you will instantly see the mixture start going way rich. If you had been at 14.0AFR with sensor unplugged the mixture will easily go into the low 13s or 12s. You can see this really well on the WBO2 gauge.
- then remove the ground jumper and you'll be back at the default 14.0AFR
- Then, we can now take a 1.5v AA battery and ground the '-' of battery to a ground screw and then jumper the '+' of battery to the O2 signal line. This puts 1.5vdc on the signal line. A normal O2 sensor outputs no more than 1vdc but the 1.5vdc won't damage anything.
- The 1.5vdc is seen by the DME as a false rich condition and the DME will start leaning the mixture.
- The engine will not like the super lean mixture and the idle quality will instantly suffer. Engine will not like idling at AFR greater than 16.0
- Again the WBO2 will instantly reflect the super lean mixture.

WBO2 gauges are wonderful for doing testing like this. Super simple method to prove that the DME closed-loop ability is in proper working order.

Then you can also just plug in the stock O2 sensor and monitor the same signal line as mentioned above. But what if you have a faulty stock O2 sensor you won't be able to easily prove that the DME and the wiring is in working order. But with the 1.5vdc AA battery and ground method it's easy.

Above testing is done:

O.5vdc at O2 signal line leading to DME = Yes
Jumper between engine ground and signal line= rpm increases very slightly almost immediately... after 27 seconds, rpm falls off a bit and is then slightly rough.
1.5vdc battery neg side to engine ground and pos side to signal line= immediate, weak and very rough idle. This idle is measurably rougher than 'false lean' test idle.

.

[icarp]

Thank you Sal for sharing this quick and effective testing system .
It works , and it is an elegant solution

Ian

[Discseven]

Drive to Orlando was done with O2 sensor CONNECTED. 27 mpg result is same as with O2 disconnected (Key Largo test drive done earlier.) Did check my odo was correct using the Turnpike's posted mile markers.

Am continuing to fiddle with mixture screw and scope. First scope, Etepon, failed to work properly not long after first use. A bit confusing until it was apparent the scope was f'd. Zeewweii 154PRO has stepped in. Am now looking at 0.2 to 1.0v oscillation (at idle, engine warmed) for the O2 sensor. Gunning for better. Slowly this 3.2 fuel control system, diagnostics, and adjustments are making sense to me. Appreciate all your input.

Doing city milage test again. Should the city milage persist at 9-10 mpg, and given the highway mpg being reasonable, the difference that stands out is the AFM's flap movement. For highway, it's relatively stable. For city driving, the flap moves back and fourth. Am thinking there may be too much tension on the spring. ??? I'll not touch the disk until the mixture at idle is as good as I can get it.
.

[scarceller]

The stock O2 sensor is designed to oscillate in voltage when connected to the DME. It simply swings between 0.2 and 0.8vdc when working properly. The rate of the swing, how long it takes to swing from 0.2 -> 0.8 -> 0.2 is directly related to the age of the sensor. As the sensor ages the swing rate of time becomes longer.

The O2 sensor closed-loop setup was not designed to fix fuel issues that are way off base. It simply will not do that. This is why you must be certain that the fuel mixture is working properly under all conditions. Idle, Part Throttle and moderate loads.

The very best tool to verify fuel ratio across all driving conditions is a WideBandO2 gauge. Trust me and invest the $200 for a decent gauge otherwise you are wasting a lot of time. You don't need a scope to setup mixture you need a CO Meter (the old way) or today we use much better device, WBO2 gauge.

For the money the AEM 31-4110 is a really good gauge and uses the LSU4.9 modern sensors that do not need free air calibration.

I'd be happy to get you my person test sheet for gathering 7 important testcases for AFR.

Enjoy.

[scarceller]

Your hi-way MPG at 27MPG is great, nothing wrong with that.

But city depending on driving habit should be in the 16-22MPG range. 10MPG or lower is not correct, it indicates an issue. Again get a WBO2 gauge and we will know exactly what's going on with Air Fuel Ratios, no need to check MPG. AFR numbers is what you need to dial this in quickly and diagnose any possible issues.

[Discseven]

On menu to check:

• With engine at operating temp, temp of each exhaust down pipe.
• After city drive, plugs pulled and compared (photo.)
• After engine at operating temp, disconnect idle switch by throttle body---note engine's tone delta.

.

[ant7]

Slightly off topic, good info from Sal by the way;
I am sure I read somewhere that there is a way of testing the AFR barn door spring resistance using a weight gauge, anyone here have any more info on that ?
It would make sense' and provide peace of mind to at least know that the tension on the door is correct to start with, surely this would then mean that the toothed wheel in the AFR is set correctly for fuel and air ratio as the barn door opens and closes with fluctuating driving conditions ???

[Discseven]

Originally Posted by Discseven

On menu to check:

• With engine at operating temp, temp of each exhaust down pipe.
• After city drive, plugs pulled and compared (photo.)
• After engine at operating temp, disconnect idle switch by throttle body---note engine's tone delta.

.

Exhaust down pipe temps were checked under 3 different scenarios:


1.) Engine warmed up while at standing idle. 210 dF dash temp is "warm."


2.) Engine driven to Whole Foods. Parked a short while. Started then temps checked.

3.) Last sequence was shot without changing gun from C to F so have converted those temps to F below.

1 - 469
2 - 477
3 - 541
4 - 440
5 - 496
6 - 542

Engine driven from Whole Foods home and shut down when just parked. Temps immediately taken.



After city drive of aprox 20 miles, engine is shut down, allowed to cool off and plugs pulled. Bosch platinum W6DPO. Roughly 1,600 miles on these since rebuild.




Rather than stuff my opinion in here, am holding off. Am also not doing anything put cleaning the plugs and reinstalling them. Once this fuel puzzle is solved, I'll likely check them again.




Made error with this idle check. Disconnected with engine running but was not warmed up. Will edit this after redo. As it was, idle immediately increased slightly upon the disconnect. Immediately drops when reconnect is made. EDIT to come.

Credit to Sal for suggesting these item checks. TY Sal.
.

[Discseven]

Fuel Pressures...



Pump ON / Engine OFF




Pump ON / Engine ON, at idle.

New pump installed 600 miles ago.
.