[Schlok]

We need an AI ML interface to ingest all this knowledge and have it be the sole 'search point' for Camaro tuning/knowledge. Wow, this is some serious stuff going on.

[radz28]

06/08/23 Update (ramblings):

I've accumulated some more miles WITHOUT using the REFLEX's CLOSE LOOP (CL) feedback. I felt like I was starting to spin my tires a bit, in terms of progressing, and felt like I was confusing myself more than I was making consistent improvements. This was by no fault of the REFLEX, but because I am already at the threshold of my understanding (as far as I can tell - I KNOW I am very much, still, a newby) of HPTuners and the E92. Having never delved into what I have lately has pushed my understanding to limits I haven't yet experienced, but enjoy the learning, and the excuse to drive FURY more. I thought if I polished everything better, and gave myself some more time to think about the REFLEX, and correlations/balancing with the E92, I could turn it back on down the road and be closer to a better tune (even though it's still pretty safe as it stands). What's tripping me up is how my injector characterization (E92) effects fueling and how to best utilize the REFLEX, when I need it to come in. Watching Mike's videos/posts and Chris's progress (and re-watching and re-reading repeatedly) has helped to a point, but I'm kinda' off on my own after a certain point. I feel like I'm making a new discovery or realization (even corrections to what I THOUGHT I understood) every few days, which is fun.

A few more thoughts and musings and what I think are going to be improvements:

• What I've been noticing:
  It's been difficult to, consistently, see repeatable positive changes. I know part of it is because I'm always trying to improve my base tune file, and that includes anything from the TORQUE model to AIRFLOW model - the latter of which impacts my REFLEX tuning. Anyone might ask why in the Heck wasn't it already dialed in? Well - because I've changed the FUELING model by changing the INJECTOR characterization [33355] where I am. The WOT-part is still OK, and safe. I'm still farting around in the transitions from lower-to-mid-THROTTLE, but I continue to feel better about progress. Because the transitions happen around where I've made [33355] changes, it makes blending fueling between the E92 and REFLEX a challenge, because depending on PEDAL input, I can get different fueling at the same point in the MAF curve among the E92 and REFLEX. Then - I could have differences in POWER ENRICHMENT (PE), too. What to do?...
  
• LTFTs were moving around and I didn't notice:
  One of the by-products of changing the INJECTOR characterization was I noticed my LTFTs, naturally, moved around. Because I'm only altering part of injector characterization [33355] that controls higher-to-WOT conditions, that leaves two other profiles to blend into in a way. Because of this, I can see different FUELING for the same MAF frequency at different times. So - I've been focusing on PART-THROTTLE parts of the MAF, and keeping the REFLEX OFF for as long as I don't need the additional FUELING. I'm seeing that even light-THROTTLE, DESIRE FUEL PRESSURE (HIGH SIDE pressure, that is) can move into some cells of [33355], which is where the DI INJECTORS are starting to pull fuel, and I need to bring the REFLEX in. I've been slowly trying to wrangle LTFTs closer to a few, negative, percent from ZERO, up to almost 6000Hz, just because that's kinda' the cut-off I've been trying to reach. From my experience with DI-FUELING falling away around there under extreme circumstances, this is where I wanted to be sure the REFLEX could come in strong, if necessary (even though I've been told it's probably not necessary - I'm just stubborn, because I don't have a dyno' to play around on and don't want to take any unnecessary chances). SO - as long as I keep DESIRED PRESSURE to less than about 15MPa in the lower-THROTTLE areas, I'm trying to rely on the DI for FUELING, and trying to keep LTFTs as close as possible so they have as much flexibility to compensate as possible. That's probably all OCD, but why not try? lol...
  
• PE/THROTTLE PEDAL/AIRMASS relationships:
  For a long time, I've understood I needed to try to see if I could make reasonable relationships among these control-characteristics. Being newer to this level of tuning, it's been difficult to determine HOW to do this. I use GRAPHS and tools in the SCANNER, and have used tools and training I've taken, but haven't noticed, through any of these, that what I needed was already presented to me.
  
  I figured I needed to try to visualize and understand what the controlling relationships were that were shared between the E92 and REFLEX. It seemed one of the significant ones is THROTTLE PEDAL. Because I can change FUELING in the MAF curves, in the same cells, with the PEDAL, I distilled my focus to this, for now believing this was the keystone (at this point) in my tuning. The trouble was, I didn't readily understand how I could visualize that data in a way I could understand it, and thought it would take more time to try to find it than it would take to just drive, re-tune, and drive, and re-tune, etc... Then - I had an epiphany: graph PEDAL % over the DESIRED PRESSURE table. I could also compare, separately, PRESSURE, at the same points, to the PEDAL/PRESSURE, and see if I could make consistent/predictable comparisons to CYLINDER AIRMASS.
  
• Finally figuring out some tools to help visualize and compile data to move forward:
  I use GRAPHS all the time, but haven't created too many, because I haven't needed to (for the most part), until now. I tried to come up with a custom GRAPH that I could use to overlay THROTTLE PEDAL to DESIRED PRESSURE. After I did that, it started to become clearer where my PEDAL input would request what PRESSURE from the HPFP. After understanding that, it was easier to understand where in the PRESSURE table, and what PRESSURE, I was commanding. THEN - I could better understand how that could/would affect the MAF frequency in the curve easier. I DO have to use averages, though, so it's not a perfect blend, but I think it will do well enough for what I need to do. Only driving and logging will validate that, though. I think I SHOULD be able to get it a LOT closer, now, though. I was kinda' shooting in the dark before, even though I knew what was going in in the PRESSURE table. Not SEEING consistent THROTTLE relationships to that table was an issue.
  
  I also added some more CHARTS to give me an alternative visualization of all of that, so I could see what all of them was doing at the same time. I hadn't thought to use this tool for this job until now, but DUH. I use the CHARTS, too, all the time, but I couldn't wrap my head around how to try to show all that stuff until now. DUH... It's starting to show more consistency among my LOGS, so that's a good step in the right direction.

Moving forward:
From what I can tell now, a lot of the missing information I think I needed to put these relationships together are continuing to come together even better now. My tune, currently, is relatively smooth, and such that I don't feel the REFLEX-transitions come in an out much at all. WOT's still fine, but that was the "easy" part. I'm still finding near-WOT, from a dig, showed that my extrapolated VIRTUAL TORQUE TABLES (VTT) for E' were making the THROTTLE a little crazy. I could see that it wasn't opening near as much as I was requesting from the PEDAL, and it wouldn't stay near my request until near 40-MPH. Bringing-up the DRIVER DEMAND (DD) TABLES is fixing that, and I'm mostly within 5-10% of PEDAL commanded. I knew it would take some work, not having anything but a stock VTT table for E', so this wasn't a surprise. The DD changes are pretty predictable, so that's part of the reason why I'm farting around with that stuff, too.

This is still fun. At least I can get E' for about $2.99 around here, even though I'm only getting about 100-tank, LMFAO. Take that hippies!!! LOL.

OH - Also: ToohighPSI's billet lid fits under the stock ZL1 hood (obviously - with the shortest PI-plates Mike offers) just fine. I never followed up, and there was never a reason to doubt Mike, but investing what I have into that was a little nerve-racking, so I wanted to confirm fitment, to leave no doubt for anyone else.

[radz28]

07/10/23 Update:

A short and sweet one: I am pretty sure I've narrowed down where I have been stumbling.

I've reported that my switching point/transition from part-THROTTLE-to-high-THROTTLE has been difficult to smooth. It's been inconsistent and hard to really get the fueling to come-in with smoothness, and it was difficult to get any positive changes. It is looking like it's due to the table(s) this method directly changes, and it's looking like massaging the DESIRED FUEL PRESSURE [17071] table needs to be brought in, too.

FUEL PRESSURE MULTIPLIER - PROFILE 2/3
To reiterate - this method pivots around the FUEL PRESSURE MULTIPLER 3 [33355] table. This table is referenced for higher-THROTTLE/WOT FUEL PRESSURE relationships. The premise is that this table is what the ECM references for RAIL PRESSURE from about 16MPa and higher (this is refenced in the INJECTOR PROFILE tables [2348]). In order to reduce DI fueling, we would adjust the values in this table, to pull fuel out, so we can start to bring the PI in. I had from probably about 50% PEDAL/THROTTLE to WOT pretty well sorted from manipulating [33355], but my mid-THROTTLE transitions were still inconsistent, and I would see STFT/LTFTs correct either LEAN or RICH, for the same MAF frequency cells. If they weren't getting to almost +/-30% corrections at times, it wouldn't be so bad, but I could make changes to the E92 MAF, OR the REFLEX MAF, and couldn't get consistency. Then it occurred to me to scrutinize BOTH the FUEL PRESSURE MULTIPLIER 2 [33354]AND 3 [33355]...

I noticed that there was a trend where these inconsistencies took place, and that was where [17071] was commanding around 14-16MPa of FUEL PRESSURE. If I force the car to stay in this window of FUEL PRESSURE, the pressure would diddle in there, and I could get LEAN and RICH, in the same MAF frequencies, back-to-back. I could literally force it to be inconsistent, if I force the engine the stay within that range of FUEL PRESSURE. So - I started really looking into the MULTIPLIER tables, and saw that I created HUGE gaps in PRESSURE, right in this area, right where both of these tables hand-off control to the other, and back-and-forth. This gap was as much as 30%. The stock values, were, essentially, the same, for both tables at the same PRESSURES, so, like in a lot of tuning, smooth transitions should try to be maintained.

With this in mind, for my last drive, I smoothed this area A LOT, and it seems like the ECM is responding with predictability now. This last drive scans showed that I was almost ALWAYS LEAN, in these areas now. I didn't touch the E92 MAF, and only pulled a little fueling from the REFLEX, because I wanted to see what the STFT/LTFT trims would do, if anything. BUT - FINALLY - I was seeing consistency, and that was something I'm sure I could work with, which leads me to DESIRED FUEL PRESSURE [17071].

I could be wrong, but it seems like this table correlates to MAF fueling calculations. It seems like it enables the MAF to fuel by FREQUENCY AND AIRMASS/RPM. So - because you can force the engine to run at 5500Hz at low-THROTTLE in one scenario, and at high-THROTTLE (and RPM) in another, but get different fueling for each, AND maintain your desired EQ, this table commands different RAIL PRESSURE so you reach about the same MAF fueling for the airflow you are commanding. That was a run-on I just can't polish right now, but will later... Basically - it allows for consistent MAF fueling for different loads at the same MAF FREQUENCY. SO - because I know my E92 MAF curve to be accurate, and have smoothed my FUEL PRESSURE MULTIPLIER tables such that I can get consistency, I am moving on to modifying this table. I am confident in the need to do this because I have changed what the MULTIPLIER tables are reporting DI flow is to the ECM, and I have to compensate for this change. Because I have reduced DI flow for the MULTIPLIER tables, and I am seeing the ECM go LEAN at these FUEL PRESSURES consistently, I'm going to attempt to lower the DESIRED FUEL PRESSURE cells in these areas, so the ECM adds fuel and rid of these lean spots some. I'm not sure getting it as tight as the stock calibration is necessary (or possible with my experience), but if I can bring the STFT/LTFTs down to less than 10% in CLOSED LOOP, I'll call that a success. Seeing that I'm only getting large LEAN SPIKES in specific areas, I think lowering the values in this table, to increase the fueling, is going to take care of much of the lean spots. I've made those changes in my current tune file, and will soon see if I'm right. I'll post results of that after I can get some time on the road and validate, either way.

Inching closer and closer!

[radz28]

I can confirm, after the drive, with the DESIRED FUEL PRESSURE table changes, that my STFT/LTFT corrections are improving, and everything is still consistent. I'm leaving the MULTIPLIER tables alone, now, and focusing on polishing the PRESSURE table.

I'm more confident than ever that this is the turning point to successfully figuring out this method. A few more drives will flesh this out.

[laynlo15]

Originally Posted by radz28

I can confirm, after the drive, with the DESIRED FUEL PRESSURE table changes, that my STFT/LTFT corrections are improving, and everything is still consistent. I'm leaving the MULTIPLIER tables alone, now, and focusing on polishing the PRESSURE table.

I'm more confident than ever that this is the turning point to successfully figuring out this method. A few more drives will flesh this out.

Real progress and your almost there.

[radz28]

Originally Posted by laynlo15

Real progress and your almost there.

YUP!!! I'm pretty much done.

I updated my 2nd post with as concise a summary as possible. I still want to post up some screenshots, but that's going to be a little bit. I have a little mountain road driving I want to do, for final validation of those transitional THROTTLE inputs, but he's behaving and seems pretty happy as things go right now. I'm still waiting on Magnuson's darned certifications, which is why I just don't out too much yet, but I'll feel less anxious when that happens (hopefully soon...).

But - I think I delivered as best a description of what I could in that summary, and some more details, if anyone's interested in the logic, at the bottom of the 2nd post. Hopefully - it's helpful to someone. I have yet to find anyone else talk much, AT ALL, about this method... I've only seen it in a training video I paid for, and it didn't go into this much detail. Otherwise - this wouldn't have taken months, lol.

[JSH]

My tuner has spent two months trying to tune a PI LT4, and I'm not sure he's done yet.

[radz28]

Originally Posted by JSH

My tuner has spent two months trying to tune a PI LT4, and I'm not sure he's done yet.

I would NEVER try to give anyone ANY impression that I'm anything remotely close to a tuner (I'm sure you already know that from prior discussions and posts) but that is almost not surprising to me. Especially if it's remote. I'm sure my OCD got me carried away in doing stuff I might not have had to mess with, but I could not stand it not driving as close to stock as possible. It's was not difficult to put the car into a regime where it will buck/surge where I was having the most trouble. Because I live a sea level, but have close access to mountain roads (as high as 7000-feet), I didn't want any trouble with any of that, nor problems with my TORQUE MODEL causing a slipping tranny/converter or something like that. So far, trying to simulate those regimes by dragging the brake to enter some of those higher-load-lower-RPM-areas have yielded smooth transitions and my airflow model CL corrections are under control and relatively minimal. I don't think I'll get them perfect, partially because of the compensations for the PI, but I'm going to try. Come Winter, I'll be double-checking, and that will be the final validation of all this, IMO. So far - so good though.

If it's who I think it is, they'll get it nailed down

[laynlo15]

Originally Posted by JSH

My tuner has spent two months trying to tune a PI LT4, and I'm not sure he's done yet.

Is it a TooHighPsi PI? Or another brand?

[JSH]

Originally Posted by laynlo15

Is it a TooHighPsi PI? Or another brand?

TooHigh. In summary, he said the Reflex system is complicated and expensive, but I assume others have differing opinions on this system. Some might not see it this way. I am not referring to KingLT1 who also is one of my tuners.

[laynlo15]

I think the Crawford and Holley exceed the cost of the Reflex. Mike has an open line at THPSI and the tuner should have called for some help. There are bunch of really good tuners that are using the Reflex without issue and for sure not taking 2 months to tune it. If he is still having problems please have him call Mike.

[NicKey]

Originally Posted by JSH

TooHigh. In summary, he said the Reflex system is complicated and expensive, but I assume others have differing opinions on this system. Some might not see it this way. I am not referring to KingLT1 who also is one of my tuners.

I believe that you may be getting some miss information. First; This is the least expensive system we have yet to see on the market and next it is the least complicated management system that we have seen. Have your "Tunner" get a hold a Mike @ THPSI as he is a wealth of Knowledge regarding these systems and tuning them.

Also while it is true that it takes some time to properly tune this Set up, 30 days is not in the realm of reality. We tune pump gas to full e85 with three pulley combos along with trans tuning and street driving which is completed in two days.

[radz28]

The REFLEX is NOT complicated... lol. Has the tuner watched Mike's videos? If he follows those videos, I CAN'T see how it would be too complicated.

If you use all of Mike's hardware/information, all you're doing (in Mike's video) is calculating a percentage of MAF curve that you want to pull from the E92 so you can input the remaining into the REFLEX MAF curve. If you use Mike's Excel spreadsheet, the directions literally give you ALL the numbers. You PASTE the new E92 curve into the E92, and PASTE the new REFLEX curve into the REFLEX MAF curve. Then, you make the initial TORQUE calculation adjustments in the E92, and pretty much start driving (with safe EQ and SPARK table corrections) and work your way to WOT and up to whatever E' content you can run. It's largely all in the video tutorials he's put on-line.

AND - with the CLOSED LOOP WIDEBAND feature update back in April (or March?...), the REFLEX will help with the fueling, and show what corrections it's making. With that information, you can start to polish the MAF curve to get it closer and closer to optimal.

Is there something your tuner has said, specifically, that is giving him trouble? I mean, if you want to run 60% fueling from the E92, and 40% from the REFLEX, those calculations can all be made in Mike's spreadsheet. I know your combination is a lot more demanding than mine for calibration, but there aren't many definitions to set-up, in the REFLEX, to get what it needs to add the fueling where it needs to be. There's a simple MAF curve, like OEM, you put your values in, and define what you want your PEDAL POWER ENRICHMENT to be (like OEM), and define what you want the EQ to be (like OEM), and it should be good to go. You can rescale pretty much whatever you need to better match (not completely, but close enough from what I've needed to) the E92, so I don't know how it could be that complicated.

As I already said - I'm no tuner, so it took me a long time to get up to speed (as far as I can understand, lol). Given I have to tune on the road, that doesn't help either. But - if I can ?figure it out?, I'm sure your guy can. If nothing else, maybe, I'd make sure he understands there are videos Sitar has put out there to get started.

[JSH]

Originally Posted by radz28

The REFLEX is NOT complicated... lol. Has the tuner watched Mike's videos? If he follows those videos, I CAN'T see how it would be too complicated.

I shouldn't have posted. I could post the details of what he told me but that wouldn't make sense on TooHigh's thread.

For me personally, I'm not doing PI because my OEM short block must already be nearing the failure point of various components.