[toohighpsi]

Originally Posted by cjperformance

Not sure what is going on here. Only thing I changed was the Lambda-KP from a value of 2.2 on my first run where I had no log, down to 1.5.

Well looks like everything is working correctly, injectors are firing at a dismal 2.0ms and lambda is doing all it can to correct - adding 5% fuel - which I'm guessing is what you've limited it to.

Quite simply you're 60% short of where you need to be, so start adding fuel. Not sure where you're at in the E92, but I'd expect those port injectors to be up around 6-8 ms to be giving you enough fuel.

Not sure why you're so far off - you didn't try to use the calculator more than once on already modified data did you? Just so you know, once the data from the E92 has been pasted from the calculator to the E92, you can't copy it from the E92 again and use it in the calculator - it will be double reduced.

If you want to post your curves we can look at them and get you figured out.

[KingLT1]

It doesn't look to me like whatever you pulled from the maf was correctly added to the Reflex. I believe The reflex curve needs to be 180 opposite of the E92. If you pull 30% from maf then you need to ramp in 30% from the Reflex. Being your PI injectors are only 2% tells me the curve wasn't scaled properly. Start adding fuel! Maybe Greg can send you his Reflex file to compare?

[cjperformance]

I feel like the MAF curve is fine, because on the first run, reading my AEM gauge, I am absolutely certain I was getting .80 lambda.

Keep in mind too, I'm getting out of it as quick as possible so these "pulls" are sub 6k short pulls until I make certain everything is working.

I will work on double checking everything.

I think the fact I was having trouble connecting to the reflex after the 2nd update had something to do with it. I feel like the software got uploaded, but it wasn't triggering the injectors.

Going to double check the values, reupload the reflex tune, and retest things.

[cjperformance]

Here's another section from that same earlier run.

[cjperformance]

Here's the bin file and spreadsheet. Everything looks right to me based on what I know...

Which may not be as much as I think!

I just keep coming back to.... the first run I was getting .80 on the AEM gauge.

Made a change to the Lambda-KP to change from 2.2 on first run to 1.5 on second run. I don't think that value changed anything, but something must have made the Reflex not trigger the injectors the second time out.

I reupped the first reflex tune this AM, but I changed the Lambda control min limit from -5 to -10. After I uploaded to the reflex, I pulled it back down again and included it in the zip. Then after I shut the car off, pulled the battery terminal for 30 seconds to ensure a fresh start the next time out. Will try to get a new short run today.

[cjperformance]

I'm willing to concede that I'm an idiot and what I swear I saw on the AEM the first run was a fluke or I hallucinated. ????

I did another run and saw about the same thing as the prior run, but this time I had allowed more correction (10%) of with the Lambda and there was an improvement, but still way off.

So let's just throw away what I THINK I saw in the very first run, and just go with what we've got here.

Next time out I'm going to allow a lot more auto correction in the Lambda, and also add about 20% to the reflex MAF table.

[cjperformance]

Looks like when I took 10% out of PE in the E92.... I may not have correctly added it back to the MAF.

In any case, unless someone thinks otherwise, I'll continue to try to correct the imbalance on the port side.

[KingLT1]

Now we are getting on the same page because honestly I was scratching my head when you said you saw .80. lol

[cjperformance]

Originally Posted by KingLT1

Now we are getting on the same page because honestly I was scratching my head when you said you saw .80. lol

I'm seriously questioning reality (or my sanity) because I was watching it like a hawk! And I also recall noting on the log (that I didn't save of course), that actual lambda was lower than the targeted!!!

So with nothing to prove or disprove my version of reality, just going to say .... it never happened, lol...

I'm really giving those piston rings gaps the ol' shake out though... sheesh.... prob safe to say not an issue on my engine.

But I have been doing very short pulls. I'm sure a long pull would toast those pistons real nice.

[radz28]

This is the same kind of stuff I'm tripping over. It's mostly my fault. LOL...

[cjperformance]

Is adding 20% to the reflex enough? If the reflex only supplies (or supposed to supply) 30% of the fuel, and I'm 30% off target then should i add 60% or more to the reflex for the next test?

My plan was to multiply the existing (copied from spreadsheet) maf values in the reflex by 1.2. Then i was going to allow 40 or 50% lambda correction factor, but there should be a more scientific approach.

Going back once again, i believe i removed fuel from PE and added to the maf correctly, at least according to my understanding of the process.

I thought maybe if i had the injector size wrong, that could be the issue too.... but looks like they are 78 lb/hr injectors, and configured as such in the relflex.

In any case this should be correctable, but would be helpful to be able to figure out how i messed up the original calculations.

[radz28]

Originally Posted by cjperformance

Is adding 20% to the reflex enough? If the reflex only supplies (or supposed to supply) 30% of the fuel, and I'm 30% off target then should i add 60% or more to the reflex for the next test?

My plan was to multiply the existing (copied from spreadsheet) maf values in the reflex by 1.2. Then i was going to allow 40 or 50% lambda correction factor, but there should be a more scientific approach.

Going back once again, i believe i removed fuel from PE and added to the maf correctly, at least according to my understanding of the process.

I thought maybe if i had the injector size wrong, that could be the issue too.... but looks like they are 78 lb/hr injectors, and configured as such in the relflex.

In any case this should be correctable, but would be helpful to be able to figure out how i messed up the original calculations.

I wouldn't purport to know what the problem is, but only offer background and what I've experienced over about 6-adjustments so far. AND - I'm not yet running the CL on the REFLEX. I've only been datalogging on HPTuners, too. And - am still trying the alternate method I've mentioned. Up to now, and with what seems like some success - I've set-up my REFLEX exactly like Mike's videos, from beginning to end, except for the REFLEX MAF curve (nothing for fueling in the E92, except for the definition I'm testing, has been compensated for otherwise, including E92 MAF and PE). I've matched, as closely as possible, the THROTTLE and PE setting, too, to my E92 toon. I'm pretty much seeing what I'm commanding off the WB.

Maybe it's worth starting fresh with a non-CL software file (or disable the CL function, which I wouldn't know how to do and haven't yet looked at), using Mike's V3 spreadsheet, and see how close you are. Then - maybe hone a little more, and finally polish it off with the CL. Unless I'm doing something wrong (which I wouldn't be surprised), the WB readings seem to pretty much reflect any adjustments I make, to either the E92 or REFLEX MAF curves or PE multipliers. It seems like the REFLEX is following my commands, so I'm just chasing my E92 MAF fowl-ups right now, and fooling with the DSX/OEM pressure spikes/drops. Anyways - maybe it might be easier to start over, because I feel like Mike's spreadsheet was a pretty darn good starting point. At least when in OL, the REFLEX seems like it follows commands predictably. That being said, I'd try to get E92 IPW where you want it, and then add enough in the REFLEX to add the fuel to reach your EQ. As already pointed out - it seems like the data is showing the totality of the fuel you're missing, so I think adding/splitting that fuel to the E92/REFLEX MAF curves that amount is the answer. It'll be "better" to be too rich, than lean, so were it me, I'd add that 60% to that area of the REFLEX and maybe smooth it into the rest of the curve, and take small steps as you continue (as you already are), until you can get to a full WOT hit.

JMVHO. I'm still learning, too

[cjperformance]

I think I'll try again from scratch w/o doing the PE->MAF move and see what happens. Feel like that has to be the source of issues.

[cjperformance]

Not for sure, but I'm betting this is the issue. Trying to get ahold of Mike at THPSI, but they are having phone issues.

Even allowing 40% correction via lambda didn't do anything. Betting the injectors aren't firing.

While waiting for a callback, I re-applied the reflex software via the bootloader, and reuploaded the bin, but injector state still shows ENGINE_OFF. In one of the THPSI vids, I got a screencap where that should say ENGINE_RUN.