michael7810

Today I verified the fuel pressure is correct (37 static, 31 idle, 20 an hour after shutdown). While running the engine today I noticed the NBO2 was dithering between rich and lean like it is on Joe's YouTube video. Yesterday it was lean and pretty much a flat line on the OBD graph. It seems the rough running condition exists when the NBO2 is "stuck" on lean. Anyone have any thoughts as to what would cause that? Maybe the O2 sensor is bad??

Ftech9

Could you post a picture while graphing the NBO2 Volts and Injection Pulse Width sensors? The engine should be warmed up and idling on its own, also disable the unused graphs for clarity.

Regards,

Joe

michael7810

OK. I'll post it tomorrow

michael7810

Here's the graph just after cold start. Engine idling rough and popping thru the intake when revving up.

michael7810

I'll try again

michael7810

Here's the cold engine start graph

michael7810

And here's the warm engine after it started running correctly

Ftech9

Hi Michael,

I can see from the graph the NBO2 sensor is working reasonably well. The waveform you are seeing from the NBO2 is caused by the DME slightly increasing then decreasing the amount of fuel being injected. It is hard to see in your photo, but if you drive consistently on the freeway you will clearly see the Injection Pulse Width to have a similar waveform riding on it and in phase with NBO2 waveform. If the NBO2 sensor is stuck reading lean, then the question is why the DME is not increasing the injection pulse width to drive it back to the rich side. I suspect this is because the closed-loop fuel adjustments can only "trim" the fuel mixture so far and the DME has hit its limit for adding more fuel. Very lean conditions are generally issues like vacuumed leaks, low fuel pressure and/or restricted injectors.

Regards,

Joe

michael7810

What I don't get is what causes the problem to be intermittent. I checked the fuel pressure and it's good per Clark's garage. The injectors were all tested by Witchhunter and all the vacuum, fuel and water hoses replaced 8K miles ago. It's ran great up until June of this year. I suspect the DME temp sensor may not be 100% because it was reading 197 when the engine was warm but the dash gauge was only at the first white line (80C or 176F). Would it be a bad idea to set the FQS to position 1 and see how it runs? I know that would add 3.5% more fuel but I don't know what other issues that may cause.

Ftech9

The coolant temp sensor for the gauge in the dash is in a much different location that the one for the fuel Injection. Because of this, this they can read significantly different if the coolant entering the engine from the radiator is at a different temperature than the coolant leaving the engine to the radiator. If after sitting overnight the two sensors read similar temperatures, then you can be reasonably sure both sensors are working correctly. You could bump up the fuel with the FQS switch and see if that gets you into range. Depending on where you live, there may be higher ethanol content in the gasoline and therefore need the extra fuel volume to compensate. In any case, it is at minimum another valuable data point that could prove useful in planning the next diagnostic step.

- Joe

michael7810

I'll give it a shot. I live in Phoenix

michael7810

I set the FQS to 1 and it still has the "stuck on LEAN" condition when first started but the throttle response is much much better when cold than it has been in the 7 years I've owned the car. I has always been barely drivable until it warmed up. I was tempted to set the FQS at 3 but I want to drive the car first. I'll drive it this week and post an update. I also checked for vacuum leaks with carb cleaner and could not find any. What else could cause it to get stuck on LEAN when first started? Or is this condition really why the FQS was incorporated?? Finally, for future changes it would be nice to have the FQS accessible without removing the DME lid, it would be extra nice to make it adjustable via OBD software Thanks

T&T Racing

Hi Joe,
I have a Roque Tune NA MAF, supplied by Josh. I do not know MAF sensor p/n he used although I might be able to read the p/n. Let you know.
I support your effort in making the F9Tech compatible with Roque Tuning MAF NA.
Is your design going to allow for a Rogue,Tuning Chip, either 2.5L NA or 2.8L Stroker NA?
Is your chip going to be programmable for engine tuners to move the hp and torque curve and Air/Fuel ratio for various compression ratios and fuel octane to optimize engine performance for racing application?
I believe Rogue Tuning shifted the torque curve to a lower rpm than the stock OEM Porsche chip.
FRWilk had provided chips that moved the torque curve to a lower rpm and had the rev limiter set for 7000rpm.
For racing application, the rev limiter must be set for 7000rpm, as there are valve springs that the best cylinder head guru's use that are designed for 7000rpm, but not Porsche OEM..
Thank you,
Tom

michael7810

To recap: 87 Turbo with 132K miles - at almost every start, hot or cold, the NBO2 reading in OBD is LEAN for a few minutes. The engine appears to be running lean (i.e. I think the NBO2 reading is correct) because revving it is very sluggish and sometimes backfires thru the intake. After a few minutes the NBO2 starts to dither between LEAN and RICH and the engine revs OK. Fuel pressure is 37 static and 32 at idle, leakdown is good. I increased the fuel pressure at idle to 37 by disconnecting the vacuum hose and that made no difference. I've turned it off after the issue clears itself and restarted immediately and it goes back into the stuck in lean condition for a few minutes. I set the FQS to position 1 and that made no difference.
The injectors have 8K miles since testing good by WitchHunter. Plugs, wires, cap, rotor, fuel filter, air filter all have 8K miles since new. There are no vacuum leaks. On the road the engine runs fine, good throttle response, boosts to 0.7Bar, and pulls good to redline.
Can someone with a turbo and Sport DME w/OBD verify the condition I have is not normal?? Any ideas on what/how I can check other components?? Thanks

Ftech9

I just pushed FocusOBD v0.19.208 to the website. The big feature update for this release is the ability to monitor and adjust the ICV (Idel Control valve). This feature is still in Beta, so I will be making some final updates. The following two new sensors were added to the Data Grid tab:

Idle Control Valve Percentage (%)
This sensor displays how open the ICV is with 0% being fully closed and 100% for full open.

Idle Control Valve Milliseconds (ms)
This sensor displays the digital pulse duration the DME program is currently driving the ICV at. The specific pulse duration causes the ICV to open to a corresponding mechanical position. The mechanical range is about 2.5ms for full open to 9.7ms for fully closed.


New ICV Adjustment Procedure /w OBD+
This feature can be used to calibrate (i.e. “adjust”) the idle without the need to bridge Test Socket pins B&C to disable idle regulation. This is possible by doing the reverse of the traditional adjustment procedure where engine RPM is adjusted. Instead, we will allow the ICV system to hold the engine RPM constant as we adjust the valve to the correct set point. This is the same set point that is held constant when using the traditional procedure.

To check or adjust the idle/ICV calibration the following procedure is reasonable. I say reasonable because the factory recommended procedure involves special tools to measure the oil temperature and engine RPM. The OBD+ measurement for engine RPM is much better than the tachometer in your dash. However, like the tach in your dash it is derived from the ignition signal which is measured by an 8bit microcontroller (Intel 8051). Because of this, the RPM measurement may not be extremely accurate. I will soon implement a new RPM measurement that is based on the Flywheel "Mark" sensor and readout by the OBD+ 32bit microcontroller. As for the Oil temp special tool, I would think the DME coolant temp sensor would suffice for this. However, if I recall correctly, oil temp lags coolant temperature to some degree.

Do NOT bridge Test Socket terminals B&C for this procedure!

1. Disconnect the NBO2 sensor if equipped.
2. Turn off all unnecessary electrics in your 944/911 such as AC, headlights and stereo.
3. Warm-up the engine until the coolant temperature is above 190F (87.8C).
4. Adjust the 7mm idle screw until the “Idel Control Valve” sensors more-or-less stay bright green. This happens around 3.55ms for 944 NA and 3.57ms for 944 Turbo running stock programs.

Note: Adjustment cannot be accurately made with the cooing fans running, this puts load on the engine and effects both idle and ICV regulation. If your cooling fans turn on, you will need to wait for them to turn off before proceeding.


Adjustment Notes
It was a little tricky to get the ICV adjusted dead-on with my 944, the ICV control system is defenelty not perfect. I recommend initially getting is as close as possible. Then blip the throttle a bit and allow the ICV system to stabilize for a few seconds. Make another small adjustment, this may or may not immediately change the values, then repeat blipping the throttle (continue repeating as necessary). The ICV should stabilize with the OBD+ ICV measurement values bright green. I don’t think this needs to be exact, close should be good enough.


ICV calibrated at correct setting as shown in bright green.