It's too bad you didn't get a look at what chip is in there.
No need to de-solder anything just open those black nylon standoffs and you can open up the two boards, they are hinged with a ribbon cable.
Listen to Sal, he is an expert on this.

You must confirm that the AFR is below 13 at WOT.
You made a few mods which means the stock chip may not be correct for your motor. You really need to confirm the AFR

Cold idle speed is not 800, that's why we never adjust idle speed till fully WARM.
At cold start the DME has a idle speed MAP that controls idle speed based on the CHT sensor. At initial cold start it MUST idle above 1000RPM, typically around 1100RPM.
Not until the engine is fully warm does the CHT sensor drop down to below 200ohms and then the idle speed is 800RPM in 84-85 or 880 86-89 engine.

This is why all adjustments are done fully warm.

After you figure everything out, you really should upgrade to the 87-89 chip that idles at 880RPMs. But don't worry about this yet till things are figured out.

If you plan to keep this car a while, best under $200 investment is a decent permanent WBO2 gauge in the cockpit. You can setup your mixture perfectly using it and verify everything is correct. In addition, it's amazing how many tests you can do with a WBO2 gauge. I really like the AEM 30-4110 gauge because it uses the better LSU 4.9 sensor that does not need free air calibration. But be careful as China knock-offs exist, only purchase from reputable sellers. Or better yet the AEM store directly here:

https://www.amazon.com/AEM-30-4110-UEGO-Ratio-Gauge/dp/B00N3VGPYS

Then learn all about these gauges and ideas for mounting them here:
http://forums.pelicanparts.com/porsche-911-technical-forum/895323-aem-wideband-o2-gauges.html

Again, you really need this to properly setup that Engine like the pros do.

Am at sea level... but planning drive through Rockies. Will get to 9,000 feet.

Nice vid Sal. Thank you. Am back into it.

Insides...


http://forums.pelicanparts.com/uploa...1714766068.JPG

http://forums.pelicanparts.com/uploa...1714766068.JPG

http://forums.pelicanparts.com/uploa...1714766068.JPG

Your have the '85 thru '87 4K chip (XXXXXX180).

All the mods have been returned to the original settings including the rotary fuel switch inside the ECU---so back to original status but for the disconnected Ox sensor.

Just did get the ECU open (Sal's vid for ref) and Dave posted the chip inside is '85 - 87. My engine is '85. First test runs today proofed the engine doing well without the ox connected. ECU reinstalls after I post this and tomorrow I'll test run some more.

Provided test runs tomorrow confirming today's performance, fuel tank gets filled and we'll see what comes of the mpg.

Thanks for the hole response Brian.



Sal, more thanks for your intel and recommends. I'll recheck the adjustments with the engine warm tomorrow. I've come off a long rebuild with the last bit of it being this fuel mileage puzzle. When this is sorted, am not enrolling in any other 911 project for a while. I'd like to enjoy the car for a while without a wrench in my hand. Wide band is on future list.


Thanks Dave.

If the ECU's still apart, take pics of the solder side as well, in particular where those power transistors with the big heatsinks connect to.

Might as well.

Looks like you're making good progress!

I did, post #39. ECU is back in car now. If weather looks good, will drive to Islamorada today to check mpg. If so, post results after doing so.


http://forums.pelicanparts.com/support/smileys/wat6.gif
.

Great thread and diagnostics going on here. What great Pelicans we have offering up great technical support. Love this place. :)

I bought the AEM setup, made it portable with clips for power. Used it on porsche, motronic VW vanagon, one day i will install it permanently.

Ah, sorry, haven't kept up with all the posts !

Any defects, cracks in the solder where circled, look closeup ?

http://forums.pelicanparts.com/uploa...1714848515.jpg

It was poorly soldered before. When doing the soldering, bend the tabs and increase the soldering area.

P... I looked but in hindsight, I could have looked more intensely now that you raise the point. With the ECU back in the car, goal now is to get a handle on where the MPG is at. Later, and if issues persist, I'll dive back in the ECU. Thank you for the push in this regard.


Good eye Dave. Bending for increased contact area makes perfect sense. Whomever did this board obviously didn't concern themselves with this. Intel appreciated.
.

MPG check today…

First crank start. Idle begins at 1200 rpm. Can’t call idle perfectly stable—hunted for 1 cycle down to 1000 and up to 1300 then stabilized at 1200. Runs at 1200 for some seconds, then slowly settles to 800. Filled tank at Shell till pump itself cuts off. 93 oct. Background notes: Tire pressures cold: 25 F / 29 R (Found tires wear better—more evenly across width—with lower than recommended psi. I realize there is the question of which is more economical, higher tire pressure for less fuel use, or lower pressure for longer lasting tires?) Oil now and after engine’s break in: Motul 20 - 50 synthetic. (Break in was with Motul mineral.) Gearbox: 5 spd. Engine mods just done: 964 cams set to 1.45, SSI exhaust, Dansk 2 in 1 out muffler. Sound is slightly more robust than stock but no drone and not loud.


http://forums.pelicanparts.com/uploa...1714863455.jpg

Went downtown Miami to Islamorada and back. Returned to exactly the same Shell fuel pump that I filled with. Did same refill allowing pump to cut itself off so fuel used should be spot on.


http://forums.pelicanparts.com/uploa...1714863612.JPG

Miles driven.


http://forums.pelicanparts.com/uploa...1714863640.JPG

Refill gallons.


Miles put on are highway for the most part with sections of stop & go in Miami and in the Keys. Only a few heavy footed accelerations were made during this test. Highway cruise = 60-80 mph. MPG = 27. Good news (so far.) No idea what the AFR is. Am inclined to think that without an Ox sensor, there’s nothing telling the ECU what the burn status is. Am guessing the ECU must default to some set measure of fuel delivery—total guess so do not quote me. Manual check of AFR obviously needs to be made. Thinking this through, it seems sensible even with an Ox sensor installed to check the AFR.

Mission today was to drive the fuel tank down immediately to see if there is or isn't an improvement from where this fix started, that being 11 mpg (city.) Clearly things are headed in the right direction. Now, and with last fill up just made, city mpg test starts. Will take several days before next refill. Update when city number is had.

Just received the new Ox sensor. Setting it aside for the time being---not installing it.
..

Looks about right. O2 sensor will bring milage up default map is rich from what i have read. I just did a 500 mile trip and got close to 30 mpg keeping revs at 3200 mostly freeway, some stop and go traffic.

Helpful graphs;

http://forums.pelicanparts.com/uploa...1714929680.jpg

O2 sensor connected - Lambda ~ 1.0

Helpful!?!?! I'll be candid, I see abstract art Dave. Is there any explanation to enlighten those who don't make head or tail of those images?

Warm idle adjustment. Jump BC. After a few increment adjustments & rpm checks, idles at 800 after a total of 1 1/2 rotations clockwise from starting point. Must say it's a good place to be making progress... thanks to you guys.

Sorry for not explaining what was posted. As can be seen from the diagrams, Lambda (1.0) is the focal point, and represents the optimal fuel mixture point for emissions and fuel economy.
The O2 sensor when connected attempts to achieve a Lambda of 1.0.

By using the graphs, one can easily determine what happens when Lambda is greater or less than 1.0 to all the key variables of an internal combustion engine (ICE).
These variables are fuel consumption, torque, carbon monoxide, and hydrocarbons. As one can see from the graphs, it's best to have a good working O2 sensor system.

Now it makes sense. Thanks Dave.

If an Ox sensor is serving to optimize the fuel mixture, why run an engine without it?
.

https://i.pinimg.com/originals/c1/a0...36aba5e792.gif

Because some people dont adequately utilize their brain...

So according to you all owners with 911 ,racing or not, without catalyst and oxy sensor have problem with their brains,dude kinda insulting don`t you think?;-)))

Ivan

No sir. I stated some.

You interpreted all.

Never mentioned a "problem with their brains."

I stated they don't adequately use them, like for critical analysis.

The context wasn't "without catalyst and oxy sensor..."

It was with cat, and running without an O2 sensor resulting in a computer trying to calculate optimum running conditions, but unable to do so.

Just some subtle, but significant differences 🙂

well,i suppose you answered to Karl question
"If an Ox sensor is serving to optimize the fuel mixture, why run an engine without it?


well..i `m guilty that i did not adequately use my brain past 40 years;-)))That is just fine with me......

Ivan

Question about running engine with or without Ox sensor is not to offend anyone or create conflict. Am simply attempting to makes sense of how best to skin the cat I have. Clearly my engine is running better without the Ox connected than with it---notably this "running better" being without the Ox sensor I supposedly screwed up by soldering in the splices.

There's now a new twist to the puzzle...

http://forums.pelicanparts.com/uploa...1715206980.JPG

http://forums.pelicanparts.com/uploa...1715206980.JPG

Looked at new Bosch Ox sensor I just received. This is same part number (13918) I’ve been running for years.


http://forums.pelicanparts.com/uploa...1715206980.JPG

Then looked at Ox in car for a comparison. Electrical prong that should be on car’s side… is inside Ox sensor connector—wire has snapped at prong. Exactly when this occurred is unknown. This adds a curious twist to the situation. Now have to consider if this has anything to do with the fuel issue or is this just a good find and otherwise inconsequential to what's really wrong?


http://forums.pelicanparts.com/uploa...1715206980.JPG

http://forums.pelicanparts.com/uploa...1715206980.JPG

To answer that, prong returns to where it belongs. Then reconnected this Ox sensor—this has the solder-spliced wires that Bosch says “don’t do that.” Started engine and after it lowered to 800 rpm… not a horrible idle but not a smooth one either. Gen light is ON. Engine's struggling. Rev’d throttle slightly and engine sputtered. Shut down and disconnected the Ox. Restart… drop to 800 and smooth as silk. Rev’d same as before and no issue. As pieces of this puzzle come together, the soldered splice seems increasingly to blame.

City mileage test without Ox continues.

Seems to me you already proved it runs better without an O2 sensor than with the soldered one connected.

It's not clear which connector got broken, the one on the car or the one on the new sensor?

Will these cars run without an 02 sensor? Yes, although probably a little rich.

Will they run better with a working sensor, also yes. When these cars were produced this was proven, state of the art technology.

In all the discussion on this thread I haven't heard a good reason not to use it.

It's impossible to claim that the car "runs better" without the O2 sensor when it's connection is so compromised that it is doubtful that a quality signal is actually reaching the dme. It is logical to claim that this engine does run better without a likely faulty O2 circuit, and a doubtfully correct emissions setting given the changes to components made without excluding other variables.

Both Kroon and Sierra Madre Collection offer the harness side connector that can replace the broken connector.

1. The 911 3.2 uses a heated O2 sensor, that's two additional connections (white wires - 12V engine running) at the sensor.
Without the heater, the sensor might not function properly at low RPM.
2. To determine whether the O2 will function properly;
a. Use a voltmeter to test the O2 wire from the DME ECM (pin 24). It should be about .50V NOT connected.
b. With the engine warm, measure the O2 sensor (not connected). It should be at a voltage between .20 & .80 at idle.
3. To check that the DME ECM's O2 circuitry is functioning properly;
a. ground pin 24 - should take with in a minute to start to make black smoke
b. use a AA battery and connect the plus pin to pin 24 and ground the negative pin of the battery,
the engine should start to idle very low (200 - 300 RPM).

Brian... engine does "run better" without the Ox sensor given the condition mine was and now is in---somewhere along the way the single wire broke at the prong. That is the male side of the connection that belongs on the car side of the connection. There's a bracket mounted to the car's frame that holds that prong along with the 2-wire connection positioned just above the prong.

Though the 1-wire (signal I assume) broke away from the prong, it's possible it still made a connection. Maybe sometimes, all the time, or not at all. And so my reconnecting that break and testing how the engine ran with that fix. Not well is the answer and so the finger now pointing more clearly to my soldering the splices being the problem here.


That's the way I see it 917. You are phrasing things with great clarity. (My wording was not as precise as it could have been.)

Thanks for the note on Kroon and SM. I'll look into that. For the time being, I've soldered the prong back onto the single wire and run that wire / prong through some tubing. Threads on the prong turned tightly into the tubing used so the prong is securely positioned (not to be pushed in by the female connector.)


Appreciate the diagnostic notes Dave. Impressive knowledge on your part but I don't know this stuff like you do so need a bit more info to make sense of it.
.

Related info:

"You can just use a digital voltmeter (dc setting) and read the voltage off the O2 sensor. 0.2 volts and below is too lean, and 0.8 and above is too rich. When you get it to fluctuate in between at idle, you will have it just right, for it is a fine line just to get it in there. Make sure you have the car at full running temp before finalizing the adjustment."

--- Steve Wong here on PPF




http://forums.pelicanparts.com/uploa...1715277518.jpg

http://forums.pelicanparts.com/uploa...1715277518.jpg

--- Harold here on PPF



Harold used some gadget from DataQ to get the data on his laptop. Have PM'd asking him what that gadget is.
.

It does appear that Sierra is selling Kroon's product, which is useful since you don't have to wait for importing it.

Here's how to splice the new to the existing - turn down the volume because the music sucks!

https://youtu.be/GL8cJ6CJY8o?si=zNwAxW2HuoKdAN7g

http://forums.pelicanparts.com/uploa...1716632221.jpg

My solder-extended version that has been installed, and new stock unit that is same Bosch part (13918) without my altering it.

Scope tested both of these. Correct AFR for 3.2 per Steve Wong is when signal ranges (fluctuates) between 0.2 and 0.8 v with engine at operating temp and at idle. (Less than 0.2 is lean. Greater than 0.8 is rich.)

Scope settings:

Volts - each vertical square = 0.1v.
Time - each horizontally square = 0.1 ms.

0 volt level is indicated by the small arrow at the far left on the data screen.

Scope is connected to sensor’s black wire. Ox sensor is NOT connected to the ECU/DME so the AFR here is without input from the Ox. A few days ago, the Air Flow Disk was turned back to its original position, and ECU/DME fuel adjustment switch turned back to 0.

Engine’s idling at 800 rpm and is at 210 dF.


<iframe width="560" height="315" src="https://www.youtube.com/embed/OJqAJXkr1ug?si=0EHXjcEBFhTkblgy" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

Ox with soldered wires spliced in.


<iframe width="560" height="315" src="https://www.youtube.com/embed/tTTFZNX3qrk?si=jn7u_TwXvjYx8RvO" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

Same Bosch sensor without being modified.


Have not driven with the new Ox sensor connected to the ECU/DME. City mileage test without Ox connected remains in progress and don’t want to “contaminate” results. When city mpg test is completed, will connect new Ox sensor (to ECU/DME) and repeat the Islamorada highway drive, then the city mpg tests. Will also scope test the newly installed Ox sensor at speed.

.

This was posted in post #70! The one referenced must have read post #70.

Dave... You did say the same as Steve. I honestly got lost with the connect to "pin 24" thing. Don't intend to be rude but I was. Pardon me.

See, he too reads and learns from the Forum.

For the record, please correct my quoted values, not .02 and .08 volts, the correct range is between 0.2 and 0.8.

Reference the original post from 2001 that I was quoted from:

http://forums.pelicanparts.com/porsche-911-technical-forum/54799-mixture-adjustment-84-3-2-a.html

Or you can just use a digital voltmeter (dc setting) and read the voltage off the O2 sensor. 0.2 volts and below is too lean, and 0.8 and above is too rich. When you get it to fluctuate in between at idle, you will have it just right, for it is a fine line just to get it in there. Make sure you have the car at full running temp before finalizing the adjustment.


And yes, I see that Loren, or ahem, I mean "Dave" has also been properly searching for, reading, and learning from my posts in this forum, as he quoted my values correctly!!!

And yes, I see that Loren, or ahem, I mean "Dave" has also been properly searching for, reading, and learning from my posts in this forum, as he quoted my values correctly!!![/QUOTE]

:D:D

http://forums.pelicanparts.com/support/smileys/wat6.gif

Thank you Steve. Corrected.
.