Hi everyone, I ran into an interesting issue while data logging today. In third gear holding at 7500 rpm via cruise control after about 10 or 11 seconds the injector duty cycle goes from the 40's to the 20's, AFR follows suit, MAF stays consistent etc etc.Tthe only real change is the Inj duty and Pulse. I'm not quite sure what the issue is, I was leaning towards a vacuum leak until I gave it another go to see if the issue repeats itself.

In the first 7500 run I stopped at 25 seconds due to misfire codes and high AFR. On the next log I decided to stick with it to 33 seconds, much to my surprise the injector duty cycle kicked back in at around 25 seconds and AFR's followed suit. Basically My duty cycle is going from an average of 40 down to 23 back up to 41.

If anyone has any guess as to whats going on I'd love to hear it!

Cheers.

 

Afrs ?

 

and Calc Load as well as other values. In otherwords, you need to post your log and also shat you are logging with etc. You might just be between an injector staging point or not even seeing the true injectjor duty cycle data. Cruising at 7500 in 3rd gear us not really a normal/anticipated/planned driving cycle for the Mazda engineers

 

Here you go!

Let me know if the zip is busted. Thanks guys.

 

Worked for me. I'm a noob to this, but >

I've done the same experiment at 5000 rpm. I didn't run into the AFR issue, but noticed that the oil temp absolutely soars over the coolant water temp when doing this (AFR ~ 12.9). Since you're already starting at 200 F CT, and at a higher rpm, the oil temps could be 250 or more. If you measure oil temp at the oil filter, it's at the *lowest* value = mesured at the front cover before going to the coolers, it's quite a bit higher. (With increasing rpm, the pressure of the apex seals against the wall contribute to a rapid rise in frictional heat that eventually dominates thermal losses.)

If I had to venture a guess, it would be that your engine is trying to seize. When that happens, the throttle is opened to maintain rpm. Is inj duty cycle referenced only to P1? If so, one may be cutting P1 and opening P2/S1. When the rubbing bit of the bearing gets knocked off, more oil flows through, releasing the motor into normal operation. Why the AFR changes, I don't know, but I'm sure the tuning charts are not done taking into account the extra pre-seizure friction.

Increased oil pressure via the Mazmart or RB mod may help, but IMO, you're on dangerous ground here.

 

woe--thats a lot to take in from that chart?

Maybe fuel pump probs?

 

Possibly... I'm not 100% on what calculations/sensors the computer references when it decides to cut the duty cycle.

 

Just took a quick look. From line 59-119 where it appears to jump up (lean) your average AFR is 14.56 and your average injector duty cycle is 24.44. From line 119 on where it appears to drop back down (rich) your average AFR is 11.9 and you injector duty cycle is 41.9.

The only other thing I see is that the Injector pulse width from line 59-119 is an average of 1.9 and from line 119-on it is 3.3. Still thinking................need food to calculate.

 

I don't know what went out first though.

Lean AFR reading causing the injectors to cut back or the injectors cutting back causing lean AFR.


I'm trying to see if I have an air leak or a tune problem.

 

It appears the issues pops up when the calculated load hits 60 and above. Once it drops below 60 the injectors pop back on.

You can see it more clearly with this edited copy my friend made me.

 

The injector duty cycle reported to the OBD port is basically a calculated made up kinda number once the secondary injectors come online. If you are just on the Primaries it sorta makes sense...but after that it isn't a number that is useful for anything.....

 

You're in open loop. AFR reading does not affect the injectors. Are your P2s scaled correctly? Your calc load and airflow does increase slightly in that range where it is leaner. Timing changes slightly as well if you look closely. Could be transition phase for the injectors. The injector duty cycle in the log is only for P1. The P2 injectors could be kicking in as you cross 60% calc load.

Never knew anyone to run cruise at that high an rpm for that long.

 

So should I try scaling down Bank 3 and seeing what happens?

Or just adjust the VE for that area?

 

How does the AFR reading not effect the injectors?

Sounds to me like you are saying latency?

7500 rpm cruise ... lol I know where that method comes from .. well at least for me it came from Kane.

I've occasionally gotten a flashing check engine light or a misfire cel while doing cruise logs 7500 and up ... I've never had enough brass to do a 9k cruise.

I've also never looked at the duty cycle ...

Oh and sub'd for interest.

 

You should have a procedure for checking your injector scaling /latency . IS 7500 cruise it ?

 

I believe once it goes open loop the MAF *AFR is irrelevant, the PCM reads directly off the Fuel VE table at that point.

I'm not 100% sure about that though.

 

I dialed in my 6k, 6.5k, and 7k. The injectors seem to be scaled ok due to inconsistent variations found in 6.5k (10%) and 7k (4%).

I adjusted the VE table and nailed it, I then went to run the 7500 cruise and it started this bullcrap.

 

I'm 99.9% true that is the case. Why else would it be called open loop? It's all based on the table at that point. There may be a handful of modifiers.... air temp and barometer maybe...

 

Uh, don't think so. Closed loop is where the output of the system in question "feeds back" to adjust the parameters controlling that output. In this case, the measured output is primarily the front O2 sensor with some influence from the rear. The fact that open loop has no such feedback doesn't mean that all kinds of other sensors and tables can't affect the system. (Just not the ones directly measuring output).

Open loop does mean that the measured AFR does not affect engine parameters.

I don't see though how the MAF reading can be "switched off" at any time other than starting.

 

WOW...

In open loop the PCM uses the MAF mostly..and modifies the values based on temp/pressure etc to set a calculated load value...that determines the amount of fuel to inject based on the values in the various fuel tables and modifiers

In closed loop..the PCM looks at what it thinks the lamba value SHOULD be based on the above...and will correct it to actual using the WBO2 sensor

 

um.. i was describing open loop, as was shady in relation to Volumetric Efficiency. I'm so used to using a MAP based system now I lose track of the crazy MAF details.

And just as I thought.... the VE table in ATR is calculated load and RPM. It's a shame Jeff isn't around... he'd likely yell at us then point out something like the VE table isn't used...

 

Where does calcLoad come from? It's a calculation. I don't see any obvious way to get that without MAF as an input parameter.

 

LOL Sorry I meant to say AFR not MAF, I was really burnt out yesterday. What you said above is what I meant.

I was basically trying to say my A/F ratio went up but it shouldn't matter because I was in open loop.

 

MAF is used in open loop, just not the Lambda feedback.

The Fuel VE Table, MAF, Inj 1-2-3, IAT, MaxCalc Load, and ECT are all in play at this stage.

If Inj 3 is good, the next culprit would be the Fuel VE, I usually don't log my 7500 cruise quite that long, just long enough to get a stable AFR. I would postulate that maybe the turbine may be impeding flow at that high rpm to the point the VE on your engine is lowered signifigantly due to exhaust pressure, or the BOV is lifting causing air to recirc and not make it into the motor, dropping load.

The fact the MAF voltage isn't changing makes me very suspect of the Fuel VE table. Like I said this weekend, it is weird, and I've never seen it do that before without any explanation in the log (like MAF volts).

 

So I guess at this point I should continue tuning and see if the issue presents itself later on down the line.