944 turbo running rich under boost

[johnb]

I did run the stock map and saw nearly identical results. The WOT tables are nearly identical between the stock map and the Kolken map.

I have not pressure tested the intake, but with the new air filter, both the stock map and the kolken map show more of what I expected of a brief AFR of 10 when the WOT map kicks in then ramping up to AFRs in the 12s under full boost.

What did make an impact to my results was replacing the air cleaner. It was the one that came with the car and I replaced it with a fresh K&N. That allowed me to get better AFRs and a noticeable increase in power. The thesis is that the air cleaner was a restriction and at WOT wasn't flowing enough air to match the injected fuel.

This seems to be borne out by dropping the fuel pressure from 3bar to 2.5bar with no change in the WOT AFRs with either the performance or factory tune. After the new filter both maps show an AFR that drops when the WOT kicks in, then builds into the 12s as boost peaks. I assume that is more expected behavior.

So all good now? The AFM measures air intake after the filter, so in theory the injector pulse would still be matched properly to whatever air is getting through the filter.

Yes, I believe it is much better.

With regards to the question on the AFM, my understanding is that once you get on the WOT map at some point the AFM is no longer used and the ECU assumes some airflow rather than measuring it. In that case a restriction does matter. My thesis is that when running on the partial maps using the AFM the AFRs were reasonable as it was measuring the actual flow through the filter. That doesn't apply when you are at WOT at some point. The ECU assumes you're getting some amount of air beyond what it can measure or is measuring with the barn door.

My research on this WOT behavior is unclear from all the forums, but it is consistent with the data and some of the discussions earlier in this thread. Someone previously called at above 80% throttle position you enter WOT, but it's not clear to me if the AFM is also ignored.

But the bottom line is with the improved breathing through the filter I am no long pegged at an AFR of 10 for the entire boost cycle.

What I do see is that how I ran the boost pulls, going WOT at a relatively low RPM in 2nd or 3rd, causes the car to drop to an AFR of 10, before climbing back up as the boost builds to something more reasonable (in the 12s at peak boost). I suspect that under a more normal pull of rolling into the throttle versus WOT at low RPM that steep drop at WOT won't be as pronounced.

I haven't done any extended runs given this is a track car and I don't have access to a dyno, so I'm not sure if I am really dialed in, but I think I have a good enough starting point to run at 944fest and collect more data.

I haven't read the whole thread yet, so I hope I'm not being irrelevant here, but I felt I had to point this out: it's not true that the AFM is ignored at WOT.

The AFM signal is used at all times to figure out the base pulse width. That should get 14.7:1. Then the fuel maps are used to adjust the AFR away from 14.7:1, and for this step it's true that the WOT maps only use RPM and don't use the AFM.

Maybe that doesn't matter in your case but it's something that people say from time to time and I just wanted to set the record straight!

[superbad_austin]

So all good now? The AFM measures air intake after the filter, so in theory the injector pulse would still be matched properly to whatever air is getting through the filter.

Yes, I believe it is much better.

With regards to the question on the AFM, my understanding is that once you get on the WOT map at some point the AFM is no longer used and the ECU assumes some airflow rather than measuring it. In that case a restriction does matter. My thesis is that when running on the partial maps using the AFM the AFRs were reasonable as it was measuring the actual flow through the filter. That doesn't apply when you are at WOT at some point. The ECU assumes you're getting some amount of air beyond what it can measure or is measuring with the barn door.

My research on this WOT behavior is unclear from all the forums, but it is consistent with the data and some of the discussions earlier in this thread. Someone previously called at above 80% throttle position you enter WOT, but it's not clear to me if the AFM is also ignored.

But the bottom line is with the improved breathing through the filter I am no long pegged at an AFR of 10 for the entire boost cycle.

What I do see is that how I ran the boost pulls, going WOT at a relatively low RPM in 2nd or 3rd, causes the car to drop to an AFR of 10, before climbing back up as the boost builds to something more reasonable (in the 12s at peak boost). I suspect that under a more normal pull of rolling into the throttle versus WOT at low RPM that steep drop at WOT won't be as pronounced.

I haven't done any extended runs given this is a track car and I don't have access to a dyno, so I'm not sure if I am really dialed in, but I think I have a good enough starting point to run at 944fest and collect more data.

I haven't read the whole thread yet, so I hope I'm not being irrelevant here, but I felt I had to point this out: it's not true that the AFM is ignored at WOT.

The AFM signal is used at all times to figure out the base pulse width. That should get 14.7:1. Then the fuel maps are used to adjust the AFR away from 14.7:1, and for this step it's true that the WOT maps only use RPM and don't use the AFM.

Maybe that doesn't matter in your case but it's something that people say from time to time and I just wanted to set the record straight!

Do you know if there is a source that calls out how WOT actually works? This is what is called out in the 944 tuning guide here on carpokes:

"This is the map that attracts a lot of attention when ‘tuning.’ When running a 944 Turbo at 15 or 16psi, knocking (aka pinging, preignition, or detonation) is a critical concern. Gasoline consumes oxygen when it burns. At an AFR of 14.7:1, in theory, pump gas will have just the right number of oxygen molecules to burn all the gas, not too many, not too few. It’s also the sweet spot for minimizing the various emissions, so the car is designed to run at this AFR most of the time via its closed-loop O2 system. However, at full throttle the closed-loop system is turned off and tuners end up adding more fuel to the mix. The extra fuel helps cool the intake charge and makes it less prone to knocking. Up to a point, extra fuel also helps make more power. A good starting point is to shoot for about 11.8:1 at wide open throttle from the onset of boost all the way to redline, with little to no knock. Note that the factory let's the AFR lean out a bit on the top end, most likely mirroring the torque curve produced by the relatively small factory turbos."

I know the O2 sensor is turned off, but as I understood it that happens well below full throttle. I assumed this meant the AFM is also turned off, but reading the above I assume that means the O2 loop.

[Tom]

The KLR monitors the position of the throttle position sensor. When the throttle is ~65% open, the KLR sends a 'full load' signal to the DME by grounding the wire going to pin 3 on the main DME connector. When that signal is grounded, the DME switches from the part throttle maps to the full throttle maps and at the same time disables the closed loop O2 adjustment. Does that help?

[superbad_austin]

The KLR monitors the position of the throttle position sensor. When the throttle is ~65% open, the KLR sends a 'full load' signal to the DME by grounding the wire going to pin 3 on the main DME connector. When that signal is grounded, the DME switches from the part throttle maps to the full throttle maps and at the same time disables the closed loop O2 adjustment. Does that help?

It helps, but to explain my behavior I'm trying to understand if the barndoor measurement is still used in the WOT map. If the barn door isn't used, my hypothesis of a restriction in the air filter makes sense, the barn door isn't measuring the restriction and feeding the air flow to the ECU - the ECU is assuming some rate of flow.

If the barn door is used at WOT, there could be an issue with the spring in the barn door and its calibration reporting that more air is flowing than I'm actually getting.

I believe the restriction in the air filter is real given my A/B testing, I'm not sure if I should also be chasing the barn door measurement.

[johnb]

The KLR monitors the position of the throttle position sensor. When the throttle is ~65% open, the KLR sends a 'full load' signal to the DME by grounding the wire going to pin 3 on the main DME connector. When that signal is grounded, the DME switches from the part throttle maps to the full throttle maps and at the same time disables the closed loop O2 adjustment. Does that help?

It helps, but to explain my behavior I'm trying to understand if the barndoor measurement is still used in the WOT map. If the barn door isn't used, my hypothesis of a restriction in the air filter makes sense, the barn door isn't measuring the restriction and feeding the air flow to the ECU - the ECU is assuming some rate of flow.

If the barn door is used at WOT, there could be an issue with the spring in the barn door and its calibration reporting that more air is flowing than I'm actually getting.

I believe the restriction in the air filter is real given my A/B testing, I'm not sure if I should also be chasing the barn door measurement.

The airflow meter is not used for the WOT fuel map. I think your conclusion about the restriction makes sense. I don't know about the brief dip into the 10s you mentioned when transitioning to WOT though. You might still have some troubleshooting to do.

I'm not a tuner but for what it's worth I would not spend much time worrying about AFM spring tension unless the car can be seen to run incorrectly in stock configuration and absolutely everything else has been ruled out. Springs don't wear out nearly as much as the internet seems to think.

Intake leaks, bad wiring (especially grounds) and mismatched FPR/chips are all really common issues and can be hard to pin down.

[Tom]

The KLR monitors the position of the throttle position sensor. When the throttle is ~65% open, the KLR sends a 'full load' signal to the DME by grounding the wire going to pin 3 on the main DME connector. When that signal is grounded, the DME switches from the part throttle maps to the full throttle maps and at the same time disables the closed loop O2 adjustment. Does that help?

It helps, but to explain my behavior I'm trying to understand if the barndoor measurement is still used in the WOT map. If the barn door isn't used, my hypothesis of a restriction in the air filter makes sense, the barn door isn't measuring the restriction and feeding the air flow to the ECU - the ECU is assuming some rate of flow.

If the barn door is used at WOT, there could be an issue with the spring in the barn door and its calibration reporting that more air is flowing than I'm actually getting.

I believe the restriction in the air filter is real given my A/B testing, I'm not sure if I should also be chasing the barn door measurement.

As I understand your hypothesis, it isn't consistent with how the system works. See my and John's posts above. The barn door (aka AFM voltage) is indeed used at WOT, and the system is not assuming a rate of flow independent of the AFM signal. The barn door is used to create a base injector pulse width. Once the base pulse is calculated, it then gets adjusted for things like the maps, FQS, o2 feedback, etc. The maps (idle, part throttle, and wot) just modify the base pulse, and the base pulse on a running motor is derived using the AFM voltage. When the KLR detects 65%+ throttle, it causes the DME to modify the base pulse using the WOT map (rather than part throttle or idle maps) and also turns off adjustments based on the O2 feedback.

[johnb]

The KLR monitors the position of the throttle position sensor. When the throttle is ~65% open, the KLR sends a 'full load' signal to the DME by grounding the wire going to pin 3 on the main DME connector. When that signal is grounded, the DME switches from the part throttle maps to the full throttle maps and at the same time disables the closed loop O2 adjustment. Does that help?

It helps, but to explain my behavior I'm trying to understand if the barndoor measurement is still used in the WOT map. If the barn door isn't used, my hypothesis of a restriction in the air filter makes sense, the barn door isn't measuring the restriction and feeding the air flow to the ECU - the ECU is assuming some rate of flow.

If the barn door is used at WOT, there could be an issue with the spring in the barn door and its calibration reporting that more air is flowing than I'm actually getting.

I believe the restriction in the air filter is real given my A/B testing, I'm not sure if I should also be chasing the barn door measurement.

As I understand your hypothesis, it isn't consistent with how the system works. See my and John's posts above. The barn door (aka AFM voltage) is indeed used at WOT, and the system is not assuming a rate of flow independent of the AFM signal. The barn door is used to create a base injector pulse width. Once the base pulse is calculated, it then gets adjusted for things like the maps, FQS, o2 feedback, etc. The maps (idle, part throttle, and wot) just modify the base pulse, and the base pulse on a running motor is derived using the AFM voltage. When the KLR detects 65%+ throttle, it causes the DME to modify the base pulse using the WOT map (rather than part throttle or idle maps) and also turns off adjustments based on the O2 feedback.

Yeah, now that I think about it I take back what I said in my last post, a restriction really shouldn't be capable of changing the AFR. It's true that the AFM is not used as an input to the WOT map, but since the map value just multiplies the base value, it should still get you around 11.8 even with a restriction.

[johnb]

It might be worth clarifying a few details here to help understand what a restriction would actually do - because it would cause things to be "wrong" in a certain way, just not in the way the OP is experiencing.

Obviously the fueling is controlled by the length of time the injector is turned on for (i.e. pulse width). So you might expect the map cells to contains something that translates to a pulse width value for the injectors. But that's not the how they work. Instead, the injector pulse width is calculated from a formula using rpm and airflow, for all driving modes (idle, part throttle, and WOT). This pulse width is calculated to achieve 14.7:1.

The values that the maps contain are multipliers that adjust the AFR away from 14.7 (generally richer). So map values are generally in the range of 1 to 1.24. Multiplying the base pulse width by 1 does nothing. Multiplying it by 1.24 pulls the AFR down to ~11.8.

The part throttle maps use rpm and load (derived from airflow). So the AFM signal ultimately gets used twice in determining the part throttle fueling - once to get the baseline fueling right, and once again to adjust the AFR. The idle and WOT maps only use rpm to adjust the AFR. But that doesn't mean "they only use rpm to determine the amount of fuel". Remember the amount of fuel required for 14.7 AFR is already determined from the actual airflow signal, without the use of maps. The maps are only used to decide by what percentage the AFR should be enriched.

So as @Tom correctly said, no assumption is being made about airflow under WOT. But you might say that an assumption is being made about boost under WOT. So what happens when that assumption is wrong, let's say because of a restriction?

Someone at the factory must have decided that 11.8 AFR is the appropriate AFR for max (~10.8 psi) boost. They expected the system to reach this level of boost somewhere between 3000 and 3500rpm, and there's no way for the DME to measure this. So the WOT map cells for that rpm range contain the value 1.24, which increases fuel to around 11.8 AFR. If you have a restriction, then you don't get 10.8 psi at that rpm, it'll be some lower pressure. Let's say you get only 8psi. But as long as you don't have a leak, your base fuel pulse will still be correct for the actual airflow you have, and it'll still get multiplied by 1.24, so you'll still get 11.8 AFR. But that might not be appropriate for 8psi, it might be too rich for that level of boost. So you will get "incorrect" results in this case - you just won't get a different AFR then the map calls for.