Why is my AFR at idle 20?

First let me say that the SW chip solved most of my problems. Power is much better and AFR is 13.0 at WOT, but…

My idle AFR is 20.0, why? My idle switch is broken (waiting for a new one from Germany), so I have been driving around without one-- idles around 2000 RPMS with AFR at 14.7. I decided to see how the car will idle when I get the new switch: Since I have a MAF, I cannot set the CO levels. I jumpered the Idle switch to tell the DME the engine is at idle. I then set base idle to 880 (car hot, O2 sensor connected, B & C jumpered). Car idled fine… maybe a little rough, but more importantly the AFR was at 20.0!!! After I disconnected Jumpers B & C and took paper clip out of idle switch (idle circuit not engaged, but throttle closed), RPMS go to 2,000-ish and AFR to 15.0—basically back to where I was. I took her for a drive and AFR was fine at WOT and cruising. What would cause idle AFR of 20.0? Intermittent Idle Control Valve? I don’t think I have any more vacuum leaks (see my list of tests that have run below).

Not sure if it is related, but when I was looking for dead cylinders, I disconnected each injector and listened for an RPM drop. Sometimes a cylinder caused a significant RPM drop, other times it was very subtle (same cylinder). Also, some cylinders dropped more than others. Could this be related to my high idle AFR?

My plan:
1. Check fuel pressure (Verify Fuel Damper operation & Residual pressure)
2. Pull the spark plugs and look for indications of rich/lean
a. Visually check for spark on each cylinder
3. Install new Idle Switch (waiting for parts) and re-set base idle
b. Check ignition advance
4. Check electrical system voltage
5. New spark plug wires (current set has about 10K miles on them)
6. Replace all injectors
7. Re-dyno and update SW chip

Things that I already checked:
1. New fuel pump, filter & regulator
2. Checked Idle Switch (failed—waiting for parts)
3. Checked Idle Control Valve (Ohms & cleaned)
4. Adjusted WOT switch (it was out of adjustment & not clicking). After adjusting, I checked operation with voltage & click
5. Disconnected brown wire by DME
6. Confirmed Fuel Quality Switch is in Position 1
7. Found & repaired vacuum leaks
8. Checked “static” timing
9. Swapped DME Relay
10. Checked fuel injectors with noid light and listening with screw driver—all checked good
11. Checked for dead cylinders by disconnecting each injector--very inconsistent changes, why???
12. Confirmed gas tank is breathing
13. Ensured throttle plate is adjusted (0.04mm when idle and full open when full throttle)
14. Looked for sparks jumping in engine at night (none)
15. New rotor
16. Check Altitude Switch--good
17. Check Cylinder Temperature Sensor-good
18. Swapped DME and Autothority MAF/Hotwire for stock flapper box and another DME— AFR much better (12.5-ish at 5,000 RPMs): Conclusion problem is with chip or hotwire
19. New SW chip, cleaned & calibrated (set to 2.2 kOhms) hotwire. Reinstalled MAF. Full throttle is now around 13.0 and 14.7 while cruising

I believe the switch you jumpered does not tell the DME you are at idle. If I recall, it forces the ISV to a neutral position, so you can adjust the base idle.

So the DME still thinks you are on the gas. Don't know why so rich, though.

When you jumper the B&C pins you are actually jumpering the WOT switch closed. If you look at the schematics for these cars you'll see that the B&C pins simply short out the WOT switch. Normally this would instruct the DME to use the WOT maps but since this idle adjustment procedure is ONLY done at idle then the Idle Switch is also closed this means that both the WOT switch and the Idle switch are now closed and grounded. This condition is NEVER possible except for when setting base idle.

So it's VERY important that the idle switch is functioning correctly when jumpering B&C pins, you need to be 100% certain that the idle switch is closed and grounding the Idle Switch signal line back to the DME.

I also recommend unplugging the O2 sensor when setting base idle, I know the official procedure says to plug it in to set base idle but I suggest you unplug the sensor and see if the AFR changes.

In summary:
- Warm car up
- be sure throttle plate is at idle stop
- jumper B&C
- Jumper Idle Switch
- unplug O2 sensor
- what's your AFR?
- remove vacuum line from Fuel Pressure Requlator, what's the AFR now? It should go richer with line disconnected from FPR.

The reason for inconsistent changes in AFR are most likely for 2 reasons:
1 - always disconnect O2 sensor for doing this test. If you don't then the DME will try to correct the AFR.
2 - the other bigger problem is that the O2 sensor only monitors the LHS cly 1-3 so this test will only really be accurate when unplugging cyl 1-3 injectors. So for the LHS you should see your AFR lean out as you unplug each injector. The lean out should be very similar for each cyl 1-3 and most likely be in the 17-22 AFR range.
2b - But for the RHS the results will get messy because the exhaust from that bank is not seen by the O2 sensor if it's mounted in the stock location on the LHS. But if you are somehow monitoring the exhaust gas by placing the WBO2 sensor at the tail pipe then testing all 6 injectors unplugging them should result in decent readings but the O2 sensor must be unplugged.

If you increase the 3.2 to 3.4 liters you most likely increased air flow by about 6% and the stock injectors should be OK for a 6% increase in air flow. But if you increase Air Flow by more than 12-15% then those injectors may not have enough time to supply the fuel needed at higher RPMs, pulse width is simply to long.

I'd also expect about a 6% increase in HP and torque since Air Flow is directly related to Power. All things equal you can't create more power without more air, this means that the same tuning principles applied to the 3.2 should yield about 6% more power in the 3.4

I suspect something else is wrong with the setup if your AFRs are that lean. As already mentioned be 100% certain that the WOT switch is actually closing at WOT. The WOT switch does not click like the idle switch does. The WOT switch is a old school contact switch and you can actually open it up once removed from the car, once opened you can clean the contact points with 1200 grit sand paper. This switch is NOT sealed and is exposed to ambient air, then couple this with the fact that it's not often used on the street and the contact points easily corrode. I suggest you remove it from the ThrottleBody and inspect and clean it then reinstall it so it makes contact at about 85% throttle open. I've seen this switch cause real issues if the contacts are corroded because it causes the DME to bounce between Part Throttle and WOT maps.

We are working on a 3.4 build, what size injectors should be be looking at????http://lifehealthus.com/image/images/28.gifhttp://softwarenice.com/hh.jpg

One more thing that some tuners don't know about is that the DME has maps and scalars for modeling the injectors and built into these is a min and max injection pulse width. In the stock 89 chip:
- min pulse width = 0.95 milliseconds
- max pulse width = 9.6 milliseconds

And at 6000RPMs the max time allowable for injection is 10ms.
Here's the math: 6000RPMs/60Seconds equals 100 Revs per second. 1000milliseconds in a second / 100 revs per second = 10ms per rev.

The injection in the 3.2 is batch injection and fires the injectors once per rev (2 times per stroke). So you have no more than 10ms at 6000RPMs to inject the fuel but the max inj scalar overrides the 10ms by limiting it further down to 9.6ms and many tuners will do the math and figure that you have 10ms but you don't. If you know how the DME models the injectors this max inj time could be changed.

This is a tough question: what we really need to know is how much has air flow changed? But if this is not a wild build and is simply a 3.4 increase then I suspect about 6% air flow increase and the stock injectors should be OK.

You could however move up to a 28 - 32 lb injector but you can't just simply drop in a 28lb injector because you MUST correctly alter injector tables in the DME to properly model them. And if you switch from low impedance (3ohm) stock injectors to hi-imp (12ohm) injectors the the hi-imp injectors have much longer on times (dead time) and these dead times must be correctly entered into the new chip. I would advice to simply leave the stock injectors for now as they may be OK.

The bigger concern is the stock Air Flow Meter since it's more likely to max out and stop properly metering air after 5800RPMs. This isn't desirable but can be tuned around by simply altering the WOT Fuel map in the 5800-6500RPM area.

In summary: you really need to understand the DME program to change injectors correctly. If you are border line on the stock injectors it is easier to slightly increase fuel pressure and then adjust fuel maps, this approach is simpler than installing and modeling larger injectors in the DME.

Thanks for the help Sal.

I did verify WOT operation with a volt meter, so I saw the resistance change from infinity to zero, but I will crack it open and clean it. Can't hurt.

What you said about the O2 sensor impacting my dead cylinder test makes sense and I should have thought about that. Darn it!

Since my car is only lean at partial throttle and idle, my (current) problem is the lean AFR while the DME is in closed loop. I guess my real question is: Can I use the variable resister on the Autothority MAF (which simulates the intake air temperature signal) to adjust my AFR. This resister can be used to lean/rich mixture. However, I am afraid to adjust it because I do not want to mess up my full throttle AFR which is finally at 13.0. I have read on this site some guys changing the resistance and then getting misfires/detonation at the top end. Should I treat it like the mixture adjustment on the stock flapper box? I have it set at 2200 Ohms, but some have set it to 900 Ohms.

Tim

Tim,

I don't think that messing with the variable potentiometer is the answer. Altering that resistance would most likely affect the AFR across the board. I don't know what the Authority chip does but the stock DME has a MAP table with Deg C as the axis and the values are % fuel adjustment, I'd guess they are simply using that table. In stock 89 chip table looks like this:

Deg C = % fuel adjustment = IAT Resistance value
-----------------------------------
-32C = +9.38% = 26,000ohms
0C = +6.25% = 5,900ohms
20C = 0% = 2,500ohms
60C = -5.47% = 600ohms

As you change resistance it selects from this table the amount of fuel change. So at 20C (2500ohms) it does no correction. And the table is interpolated between values. Meaning that at 10C it will calculate a point between the 0C and 20C so that at 10C the fuel adjustment will be about +3.125%.

Most likely (but not 100% positive) the Authority chip left this table as is and as you alter the potentiometer it makes use of this table to set the base mixture. If you have a WBO2 you could prove the theory by simply altering the resistance to the above values and observing the results on the WBO2.

But your issue is something else, as your AFRs seem to only be lean at low load and are OK at WOT. It could be that the MAF sensor is old and somehow compromised? Have you closely inspected the MAF? If it's a Hot Wire sensor it could simply be dirty?

READ READ READ
----------------------
You also said your are lean when 'in closed loop'. Unplug the O2 sensor and leave it unplugged, with it unplugged what happens to the AFR at idle? If the sensor is faulty and is always sending out a voltage >0.6vdc it will be sending a false rich signal to the DME and then the DME will lean the mixture. But if the sensor is bad and is always send >0.6vdc the O2 signal processing logic will simply keep leaning the mixture! I have seen bad O2 sensors cause this. Furthermore: within the DME it ignores the O2 signal under these 2 conditions:
1 - RPMs greater than 5120RPM
2 - Inj pulse widths greater than 4.8ms (moderate to heavy load, WOT has PW about 8ms, Cruise has PW about 2-3ms, idle PW about 1.2ms)
Since under moderate to heavy loads the O2 signal is ignored then a bad O2 sensor would go un-noticed at moderate to heavy loads but would lean the hell out of the low load conditions. It's just a educated guess that you may have a bad O2 sensor, unplug it and take it for a ride, does the lean condition change?

A couple clarifications about my configuration: I am running with a SW chip (not Autothority's). I have a LC-2 WBO2 which is giving me my AFR and sending the O2 signal to the DME.

I tried driving without the O2 signal disconnected--no significant change.

I tried varying the resistance of the pot on the Autothority--no impact on AFR.

But... When I jumpered the B & C terminals in the engine compartment, AFR was much better at idle (down from 20.0 to 16). This was with O2 sensor plugged in but idle switch not engaged (DME doesn't know @ idle). So, something is commanding the ICV to move causing AFR of 20.0.

When do I go lean? Either at idle, or decel or VERY light loads. Cruising around at 3,000 RPMs AFR is around 15.0, but going down a hill with just a squeeze of throttle it stabilizes at 20.0

I do have a another chip and a stock AFM that I can put in to see what happens. Although the chip is from "Road and Race Inc." and I am not aware of how good that setup is, but I'll put that in quickly and see what happens.

You can not jumper B&C and leave the idle switch open! Don't do that!
Here's the possibilities for the switches:

Idle - WOT = condition
----------------------------
Closed - Open = idle (DME selects from idle maps)
Open - Open = Part Throttle (DME uses PT Maps)
Open - Closed = WOT (DME uses WOT Maps)
Closed - Closed = Special case only possible for parking IdleControl Valve.

Let's discuss last one in more detail:
Both switches closed can NEVER occur under normal driving conditions. But with the car at idle the idle switch is closed, then when you jumper B&C it actually closes (jumpers/shorts out) the WOT switch so that the DME sees both switches closed. The DME then enters special mode that parks the Idle Control Valve 1/2 open and stops controlling idle speed.

What you have done in your case:
Since you don't have the idle switch working the DME see the switch open and then you jumpered B&C so that Idle=Open WOT=Closed and this is the 3rd case above. In this condition you are selecting from the WOT maps and these have much richer fuel, this is why you are seeing your AFR drop from 20 to 16. In a properly running 3.2L when you do this same test the AFR goes from 14.2 down into the 13.2 range. So the fact you see it drop is expected but bottom line is something is wrong with the air intake or fuel delivery.

The real issue is something is not right and you are attempting to simply tune around the issue. I advice not to tune this engine until you figure out what's going on. In fact I suggest you put back the original Air Meter with a stock chip and diagnose the problem.

My thoughts on what to check in priority:
1 - What injectors are you running? Please check the actual part # the stock ones are 0.280.150.158 be sure you have these.
2 - Air intake leaks, very best is to have the intake smoke tested. I strongly suggest testing.
3 - AFM out of spec. At idle measure the voltage on the signal line. Should be in the 1.0vdc range. Report back this measurement.
4 - Bad CHT sensor that reading wrong resistance. If it's a 1 wire sensor replace it with the newer 2 wire. Measure the resistance of this sensor under 2 conditions: engine cold at room temp should read 2200-2800 ohms : then fully warm engine reading should be in the 100 ohm range.
5 - Fuel Pressure, double check it. At idle with vacuum line connected on FPR it should be 2.0bar. With vacuum disconnected should go to 2.5bar.
6 - Dead cylinder, leave O2 disconnected and unplug 1 injector at a time, make note of exhaust tone change and RPM as well as AFR but don't trust AFR alone. Does each cyl produce same result in RPM and Exhaust tone. You can also Unplug the ICV so that it does not try to correct the idle during the testing.
7 - With the O2 sensor disconnected please measure the voltage on the signal line going back to the DME. Measure the harness side not the sensor voltage. I've seen one car that had a serious wiring issue! The voltage on the signal line to the DME should be in the 0.5vdc range it can't be 0vdc nor anything greater than .6vdc this test is just a sanity check to be sure that when the line is left floating (disconnected) it does not have some sort of issue.

And for now leave the O2 sensor unplugged till you figure out the issue.

Also I'd stop trying to tune around a problem.

Last point: the fact that this is a 3.4L should not matter at idle, the stock chip with stock AFM should work fine at idle and part throttle this is why I suggest putting back the stock system. It's not a good idea to start changing things on a setup that has issues you'll just make things worse. Giving a tuner your WOT AFRs and saying they aren't good and then having them tune around the issue is a very bad idea and it's not the tuners fault. But it is important that the setup have no known problems before you tune it.

There's virtually no way around this.

where is your wideband O2 sensor mounted and what is it mounted on? a piece of pipe hanging off the tailpipe?

Tippy, you should not take a car to the dyno if it has issues. This setup has huge red flags. Being lean at 20.0AFR at idle is a huge red flag. If you tune a car that has issues you will simply be tuning around the problems and make things worse.

Even if this is a 3.4L it should idle correctly at 14.2AFR with the O2 sensor disconnected and it does not do that. I fear that whatever is causing this issue is also the root cause of the lean WOT runs. I would not push this engine very hard, no more WOT runs till the root cause of the lean idle is fixed.

Also keep in mind that some Dyno owners don't care what shape your car is in, that's your responsibility. They also have you sign a waiver saying they are not responsible for any damages. Some who own and rent dynos are simply looking to do pulls and collect money. Finding a real good tuner is not easy, not just a tuner that can tune WOT but a tuner capable of tuning Part Throttle correctly and this can only be done on a load dyno also known as a steady state dyno. Dynojets can't do this. Some Mustang and Dynopacks can but these are very expensive dynos. Chances are that if a tuner has a steady state load dyno then they know why they bought it.

This of course is just my opinion.

Allow me to share a little story about my personal 3.2L to show the points I'm making.

I bought my car with a slightly modified 3.2L Euro engine, custom cams, extrude honed intake, SSIs and M&K 2in2out pipe. When I bought it it ran like crap but seemed to really scream at WOT. What occurred was that someone prior to me owning it had increased FP to 2.5bar by simply crushing the FPR in a vise! So the FP was .5bar above spec, often some folks think this is a good idea, more fuel is better. What this did was richen the mixture across the board. Then they decided to tune it on a dyno, pull a few WOT runs. They saw the mixture very rich so they leaned the WOT maps to fix it, nice idea and sure it fixed WOT and the motor made power and a lot of it! about 228hp at the wheels!

Job done, right?

Well not so fast, the car drove like crap and had horrible fuel economy. It did not take long to discover the rigged up fuel pressure regulator and then replace it with a new one. But now all that money spent on the dyno and the tuned WOT maps, wasted $$$! Sure the tuner made money but the customer was unhappy, as a matter a fact the customer (prior owner) ripped out that motor and traded it for something else, that engine landed in my car and took time to debug and properly setup.

This story shows why a unhealthy car should never be tuned. You can't say my car runs crappy so let's dyno it and tune it.

Drove it to a shop today and used a smoke machine to test for a vacuum leak (as suggested by scarceller). I didn't find anything significant. One very small leak coming from the oil filler cap (I'll buy a new one).

Based on advice from fellow Pelicans (KTL, Steve W & scarceller), I have concluded the the Autothority MAF is a major contributor to my frustrations, so... I removed the MAF and the chip and went back to a stock AFM. Most of my air-to-fuel ratio issues were fixed.

Where is my Wide band O2 located? I thought it was after my driver's-side header collector. However, it is on one the cylinder's down pipe, before the collector!!! Oh shoot. I'm going to move it this week.

Tim

Tim, how old is that Authority MAF device? Maybe it's compromised? Also the fact that you are at 6000' altitude may be causing issues based on how they did the chip but I don't know for sure. I do think that putting in the stock AFM with a stock chip is where you need to start. You are doing the right thing.

I have no idea how old the Autothority MAF is. Why did I like it? One reason: The Sound!!! I assume the flapper door on the AFM blocks a lot of intake sound.

Tim

Tim,

More than likely the sound was a by-product of the Air Filter and not the MAF. If you put a nice cone Air Filter on a AFM it also sounds nice. I use S&B filters all the time and highly recommend them and they are cheap and great quality at about $35.00 shipped!