Which way to turn 3.2 air flow disk?

[brianlay]

The replacement sensor comes with a grommet
https://www.pelicanparts.com/search/?q=Oxygen+sensor

[spuggy]

Originally Posted by Discseven

Bosch language about drawing air "through the wires" suggests the wires being air conductive. Or is this marketing?

It's not special wire, it's just stranded and to a spec that'll support the required air flow.

The specs for the air requirement are detailed in the tech spec documents that Bosch provide. It's tiny, because the sensors don't require very much reference air.

But they do require some - or they don't work...

Automotive Applications

It was developed by Robert Bosch GmbH during the late 1960s under the supervision of Günter Bauman. The original sensing element is made with a thimble-shaped zirconia ceramic coated on both the exhaust and reference sides with a thin layer of platinum and comes in both heated and unheated forms. The planar-style sensor entered the market in 1990 and significantly reduced the mass of the ceramic sensing element, as well as incorporating the heater within the ceramic structure.[1] This resulted in a sensor that started sooner and responded faster.

The most common application is to measure the exhaust-gas concentration of oxygen for internal combustion engines in automobiles and other vehicles in order to calculate and, if required, dynamically adjust the air-fuel ratio so that catalytic converters can work optimally, and also determine whether the converter is performing properly or not. An oxygen sensor will typically generate up to about 0.9 volts when the fuel mixture is rich and there is little unburned oxygen in the exhaust.

The sensor does not actually measure oxygen concentration, but rather the difference between the amount of oxygen in the exhaust gas and the amount of oxygen in the air.

[mysocal911]

Here's a timing chart indicating what timing advance the "performance chips" do;



As can be seen from the 20% engine load ignition advance over the stock value, all performance chips advance the timing.
One tuner advances as much as 25+ degrees over stock at some RPM. This is very problematic potentially causing engine detonation, damaging the engine.
The 911 3.2 lacks knock control which was added on the 964/993 engines plus twin plugs. Without knock control these excessive timing
conditions can lead to detonation when;
1. using low octane fuel (who knows what one really gets at the pump)
2. climbing a high mountain grade on a hot day

It's significant that many lack any real understanding of the above facts!

[proporsche]

all this makes me wonder why do you even use oxy sensor??My 911 had it ,took it off the 1st week when i got it back in 1996..car runs a like champ...plus my gas consumption is excellent..you said your friend has the Co test machine, what is the co mix? When was the last time you adjust it?

Ivan

[spuggy]

Originally Posted by Discseven

I have opened the ECU and to my eyeballs, it's clean and everything looks in order. Am not an ECU expert so my view is entirely superficial. I did not look at any chip in particular. Nor did I see any notes or stickers on anything. I'll open it up again for this. Could you post a pic of or direct me to what / where I should be looking?

Look at the big chip on the board. It'll be either a 24 pin (early) or 28 pin (late) DIP. If there's no stickers on this chip and it doesn't have a little quartz window exposing the silicon die inside, then it's very, very, likely to be the factory chip.

Most aftermarket chips are EEPROMs (erased by applying UV light through the quartz window so they can be re-programmed - which is why you put the sticker over the window, even though it'd take a long time to erase them in sunlight). It's not impossible to get a PROM blown with a custom program - however, this is more expensive in small quantities (read: "you're not the factory") than just using an EEPROM, which can be erased/re-used over and over.

I'm not saying aftermarket PROMs for Motronics ECUs can't exist, just that I would bet money that there are very, very few - if any.

Originally Posted by Discseven

ECU mapping would be next on the menu.

Address the O2 sensor first.

If it were me, and there was something wrong with the ECU, I'd buy a plug-in Motronic replacement that was programmable. IIRC, Rasant used to provide a kit that used MoTeC, and RaceBoy on this forum can supply VEMS stand-alone ECUs for 911s & 944s that plug into the factory Motronic harness.

Stand alone ECU options for a 3.2
VEMS Plug-n-Play - quick review


Either approach gives you an ECU you can tune in real time, with a laptop, and other features you'd expect from technology built this century. Like support for wideband O2 sensors, modern injectors etc etc.

[mysocal911]

Originally Posted by spuggy

Look at the big chip on the board. It'll be either a 24 pin (early) or 28 pin (late) DIP. If there's no stickers on this chip and it doesn't have a little quartz window exposing the silicon die inside, then it's very, very, likely to be the factory chip.

Actually, that's not correct. Few if any 911 3.2 DME ECM memory chips were mask programmed, i.e. programmed when the memory chip was ordered by Bosch.
Most all the memory chips used by Bosch were EPROM, allowing Bosch the capability to update engine tweaks to the ECM requested by Porsche.

Originally Posted by spuggy

Most aftermarket chips are EEPROMs (erased by applying UV light through the quartz window so they can be re-programmed - which is why you put the sticker over the window, even though it'd take a long time to erase them in sunlight). It's not impossible to get a PROM blown with a custom program - however, this is more expensive in small quantities (read: "you're not the factory") than just using an EEPROM, which can be erased/re-used over and over.

That's not correct, EPROMs are erased by UV and EEPROMs are a flash type memory.

Originally Posted by spuggy

I'm not saying aftermarket PROMs for Motronics ECUs can't exist, just that I would bet money that there are very, very few - if any.



Address the O2 sensor first.

If it were me, and there was something wrong with the ECU, I'd buy a plug-in Motronic replacement that was programmable. IIRC, Rasant used to provide a kit that used MoTeC, and RaceBoy on this forum can supply VEMS stand-alone ECUs for 911s & 944s that plug into the factory Motronic harness.

Stand alone ECU options for a 3.2
VEMS Plug-n-Play - quick review


Either approach gives you an ECU you can tune in real time, with a laptop, and other features you'd expect from technology built this century. Like support for wideband O2 sensors, modern injectors etc etc.

Waste of time and money, and not good for a future 911 3.2 resale. Besides, most all 911 3.2 DME ECMs typically take less than 30 minutes to repair, if a problem occurs.

[pmax]

Originally Posted by Discseven

Good point P. I have opened the ECU and to my eyeballs, it's clean and everything looks in order. Am not an ECU expert so my view is entirely superficial. I did not look at any chip in particular. Nor did I see any notes or stickers on anything. I'll open it up again for this. Could you post a pic of or direct me to what / where I should be looking?

As already mentioned, look for these 24/28 pin flash memory chips ...

if there's a label on it and the manufacturer's still in business, they can be contacted for further details.


https://forums.pelicanparts.com/porsche-911-used-parts-sale-wanted/1121894-carrera-3-2-aftermarket-dme-chips-28-24-pin.html#post11730846

Good luck !

[mysocal911]

Originally Posted by pmax

As already mentioned, look for these 24/28 pin flash memory chips ...

if there's a label on it and the manufacturer's still in business, they can be contacted for further details.


https://forums.pelicanparts.com/porsche-911-used-parts-sale-wanted/1121894-carrera-3-2-aftermarket-dme-chips-28-24-pin.html#post11730846

Good luck !

They are EPROMs, i.e. one of these, either 16MB/32MB/64MB. The original Porsche ones, not Mickey Mouse "performance " chips, have just a printed number.

[proporsche]

if you have this chip ..it is still factory.if you have an early Carrera it will be the short one..this one is for 87-89

Ivan

[911obgyn]

You can check O2 sensor output while running, should be .3-.8V moving between those. Can give you a ballpark CO reading.Between meter ground to sensor wire.

[Discseven]

Yesterday: Adjusted idle setting with warm engine. Jumped BC and turned idle screw 1 ¾ turns counter clock wise. Idle now at 1,000 rpm. (Shut down and left test drive for Wednesday.)

Today / Wednesday: Cold starts on first crank. Terrible idle right away. Bouncing. Then sometimes holding at 800… then 1,000… then 1,200. Almost stalls and catches itself a couple times… finally stalls. Restarts on first crank. Test drive… warmed engine to 210 dF. Pulls clean in first and second gear. Held steady throttle in 3rd gear heading to highway and after a second… bucking. Canceled highway. Heading home. Warmer engine gets, worse it runs.

Thought about pulling Ox sensor and having a look at the wires to see if they are stranded or solid. Doesn’t make sense to fiddle with this anymore. New heated sensor ordered.

Will return idle screw and air flow meter’s disk-spring to their original positions. Leaving ECU fuel adjustment on #3. Will open ECU to check chip status. After new Ox is in… test drive.

Will report on run of engine. Then check mpg.

Update - From the grapevine word is this oxygen sensor inclusion is needless! Easy enough to disconnect it and see what happens.

Appreciate the input. Is now too much to respond to as I like to do. Know that I read it all.
.

[scarceller]

Be aware that a cyl that is not firing will cause MPG issues if you have the stock O2 sensor connected. Here's why:
A cyl not firing is a O2 pump, the O2 in the air fuel charge goes for the ride straight out the exhaust.
This extra O2 is then mixed with the exhaust from other 6 cyls.
Then the exhaust runs past the O2 sensor and the sensor sees all this extra O2 and thinks lean condition.
But in fact the other 5 cyls are not lean they are burning just fine.
What happens next is the O2 sensor sends this false lean condition to the DME.
Now the DME tries to fix the lean condition by riching the mixture to try and correct the false lean condition.
It's very common to see MPG drop by 20% when this occurs!

Are you certain all 6 cyls are correctly firing?
And do you have the stock O2 sensor plugged in?

You really need to get a WBO2 gauge to properly diagnose what the AFR is at idle, part throttle driving and WOT.

[scarceller]

If cold start is hunting for idle up and down it means mixture is to rich. The reason it hunts is because rich mixture causes engine to want to rev up and as it revs up it then hits the fuel decel fuel cut limit that's set at around 1300RPM. When the idle hits 1300RPMs or so the DME instantly cuts fuel and now the idle speed drops like a rock. And the cycle repeats.

So why are you running rich?

To diagnose these issues you need to diagnose fully warm and you need a decent WBO2 gauge so we know the exact AFR.

Stop messing with the AFM spring tension, you are likely trying to tune the AFR around another critical issue.

Changing the spring tension is a very bad idea because it effects AFR a lot more at high air flow rates than it does at the low rates.

If anything, move the other adjustment so you move the carbon contact point further down, this will lean the entire air flow rates evenly.

I can help further if you PM me, it's hard to diagnose here on forum. I can help and if we fix the issue come back and report the fix.

Yes, please unplug the O2 sensor and leave it unplugged during all diagnoses, you only plug it back in once you fixed the core issue. And if you don't have a Cat converter you don't need the O2 sensor at all.

Originally Posted by Discseven

Yesterday: Adjusted idle setting with warm engine. Jumped BC and turned idle screw 1 ¾ turns counter clock wise. Idle now at 1,000 rpm. (Shut down and left test drive for Wednesday.)

Today / Wednesday: Cold starts on first crank. Terrible idle right away. Bouncing. Then sometimes holding at 800… then 1,000… then 1,200. Almost stalls and catches itself a couple times… finally stalls. Restarts on first crank. Test drive… warmed engine to 210 dF. Pulls clean in first and second gear. Held steady throttle in 3rd gear heading to highway and after a second… bucking. Canceled highway. Heading home. Warmer engine gets, worse it runs.

Thought about pulling Ox sensor and having a look at the wires to see if they are stranded or solid. Doesn’t make sense to fiddle with this anymore. New heated sensor ordered.

Will return idle screw and air flow meter’s disk-spring to their original positions. Leaving ECU fuel adjustment on #3. Will open ECU to check chip status. After new Ox is in… test drive.

Will report on run of engine. Then check mpg.

Update - From the grapevine word is this oxygen sensor inclusion is needless! Easy enough to disconnect it and see what happens.

Appreciate the input. Is now too much to respond to as I like to do. Know that I read it all.
.

[Discseven]

Sal... Appreciate all the info.

Bottom line here: all things point to the soldered extra wire I spliced into the Bosch Ox length of wires in order to extend them to reach the SSI bung. Discovery made (in thread) is those wires should NOT be soldered according to Bosch. Adjustments I made will be returned to the original settings. I'll disconnect the Ox and test drive the engine. If good performance, will check the milage. Ultimately I'll use a friends tail pipe meter to check the air fuel ratio.

You asked about the engine running on 6 or 5 cylinders. Rebuild was recently completed and aside from this milage issue and now the engine running poorly given my adjustments, there's no doubt all 6 are firing. That said, thanks for explanation in this regard.

Your offer to PM assist. Much appreciated / Thank you. I believe the situation's under "control" for the time being. We'll see if that bears out in a few days

[Discseven]

Update

Today: Started engine with existing adjustments that proved faulty yesterday. Start on first crank. Erratic idle... Shut it down.

Returned Air flow disk to original position. Disconnected Ox sensor. Start on first crank. Even idle. Jumped BC and adjusted idle screw clockwise 1 3/4 turns (back to original position.) Remove jumper. Shut down. Start up on first crank (slightly warm engine.) Even idle at 800. Test drive... warm engine to 210 dF. Test pulls... to redline without hesitation every time. ...Making progress.

With faulty settings (from and tested yesterday,) engine would not tolerate being driven in high gears around 2,500 rpm without bucking. No bucking in high gears at that rpm level now after returning original settings and no Ox sensor.

Checking fuel milage next.

Input from you guys... priceless.
.

[scarceller]

Observing cold start will tell you a lot about the overall mixture as well as idle switch and ICV operation. At next cold start do as follows and observe behavior, a properly running 3.2 should behave as follows during cold start:

First, 84-85 cars idle at 800RPM while 86-89 at 880RPM
I assume if it's a 85 car it has the 85 stock chip.

- Engine dead cold
- Do not touch gas pedal during all testing
- Start engine
- Observe, does it hunt for idle at initial startup? If so it's likely running rich
- Does it sputter and cough? Lean condition
- At initial startup it should idle between 1100-1200RPMs
- Let it warm over next 3 minutes, do not touch gas pedal
- As it warms the idle speed must drop on its own to 800RPM assuming 85 DME
- After 3-5 min it should be idling at 800

If the cold start behaves as above your mixture is in the ballpark and we know your idle switch and ICV are controlling idle properly.

Report back on results for that testing.

[scarceller]

Adjusting base idle speed with BC jumper and setting base mixture with 3mm allen key on AFM should only be done with fully warm engine.

You need a decent WBO2 gauge to set mixture and the WBO2 sensor must be installed in the exhaust Bung not the tailpipe. Sensors in tailpipe will read lean.

Fully warm engine at idle your target AFR is 13.8 to 14.4 AFR at idle. If you can get idle set correctly usually all other engine conditions fall into place.

A properly running 3.2L should hit these AFR targets, all testing to be done in open-loop mode with no stock O2 connected
idle - 13.8 to 14.4
PT 3000RPM steady state flat road in 4th gear - 14.4 to 14.8AFR
WOT - all WOT conditions MUST be below 13.0AFR the stock chip is tuned rich so likely in the 11.5 to 12.5 range. But you want to be in safe zone below 13.0

Enjoy

[Discseven]

^^^ Thank you Sal. I did not set idle with a warm engine but idle did adjust to 800. Though not stone cold on the restart after adjustments, idle and performance was on target---stable idle and with the engine warmed up, a consistently clean pull through rpm/gears up to red line for multiple goes at it. This is good news but fuel consumption is where this journey started---mpg has yet to be checked.
.

[Discseven]

Opened ECU/DME and I didn't get to far. Appears to me there are soldered connections that don't enable the 2 boards to be easily separated to look "inside." Rather than screw something up messing with it, and given the engine running well, am leaving the chip question unanswered for the time being. If someone can tell me how to separate the boards, I'll go back in.

Pics of ECU/DME:















Rotary fuel switch that was in position 0... Is now in position #3. This was the first adjustment done a number of days ago. No change was noticed. Supposedly this position results in 3.9% reduced fuel flow.

[scarceller]

Moving the FQS from 0 to 3 does pull fuel out, about 3% leaner. But doing this is often a red flag that something else is off.

I did not ask, what altitude are you at? If Miami then sea level so no need to discuss.

DME boards can be separated, watch my video:
https://youtu.be/GAo01_454IY