I keep uncovering things from DA3 that improve the way the car drives with this Miniram. It's astounding how so much of the DA3 (a completely different ECM) ports over to 8D and runs so much better. The base code must be very very similar.

In this instance, the tables Slow O2 Filter Coefficient vs Airflow and the O2 Integrator Delay vs Airflow.

Here's 8D (top) vs DA3 (bottom). Granted the DA3 is using two O2 sensors, so i just grabbed one side. Also the number divisions in the table is different, so I just interpolated between values to fully populate into 8D.

Not exactly sure why it's running better... I don't know for sure what these tables do. Also not even sure how to describe in words how it actually is better... but the seat of the pants feel vs my foot on the throttle... it just feels "crisper" and more responsive. Like somehow the transition out of AE fueling into normal fueling feels more crisp... if that makes sense.

Everytime I think I've optimized it, it gets a little better when I find something new... lol

 

Are you running the DA3 timing curve, your own, or did you interpolate this as well...?

- Rob

 

Actually a hybrid between the DA2 BAFL Corvette and the 8D AXCN Corvette. I found that the L98 was too conservative at high load and the LT1 was too conservative in the low-mid load. So I essentially merged the two into a hybrid table and it works very well.

 

Absolutely agree, and great work by the way. I remember when I siamesed the Tuned Port Injection system years back, the known bins were useless to work with. Reviewed $EE, started changing things to compensate and it made a huge difference in how the car felt overall.

Good job.

- Rob

 

I travel for work on occasion and have to rent cars... so when I drive the late model cars, there's a part throttle feel to it that's just super crisp and nimble (not sure how else to describe it).

And I had always thought that there has to be a way to get my SBC to feel like that... I figured that an engine is an engine for the most part (notwithstanding the overhead cams, and all the other modern tech)... but that it mostly had to do with the tuning.

Turns out I was right, as I've finally gotten to the point where my throttle feels just like those late models.

 

It is always intriguing to me seeing such high SA values as found in the LT1s and L98s and similar. If I didn't have so many false knock issues I would be a lot more inclined to up the ante on my SA. I've taken a conservative approach in that regard solely due to the false knock.

 

Have you read through some of the recent threads on false knock? There are ways you can determine one way or the other.

Believe me, the added timing is well worth the effort.

As a reference, i run 9.8:1 compression with a cam at 215@.050 with a 114 LSA, on 91 octane gas. I'll get about a 1 to 1.5 retard kicking in every now and then, but I'm ok with it since I know the ECM is taking care of it and I never hear any audible pinging.

 

I sure have. Me and you discussed it some in one of your threads. I went with the same delco sensor you put in as well. Torqued to spec. I am fairly confident a majority of mine that is being picked up is driveline noise. I still have a pretty mean clunk in my umi torque. And fairly confident I have some input shaft bushing/throw out bearing noise starting.

I see a lot of counts pop up all at once in areas I wouldn't expect to get counts or on shifts. For example I doubt it is true knock at 1600rpm 50 map with 27* spark...

I wish I had just gone straight to running a 90* fitting with some thread tape to quiet it down some.

 

My main point of reference on my more recent "false knock" effort was a previous situation years ago where I was attempting to run ~10.1:1 compression... such that the knocking became very audible when the ECM ran out of knock retard allocation in the tune.

During my diagnostics, (since I knew what to listen for) I had turned off the knock retard and was gently getting into the throttle to see if I could hear anything. When I didn't, I would get into a little bit harder, and then harder, etc... And when nothing manifested, that was a big data point that it was false knock. And, like you, I was getting knock retard in areas that made zero sense.

I just think these modified engines simply generate a lot of "noise" with roller rockers and cams with more aggressive ramp rates, headers, etc, etc... it ends up playing games with the knock sensor.

 

Yeah, I have been scared of knock for years now too after I had detonation ruin some pistons when I was 19 with this car. I blame it fully on lack of experience and knowledge of what I was doing...but still makes me fear going too far with SA. I think I would struggle to hear the pinging too between loud exhaust system and having pretty bad tinnitus in my ears.

 

Spark plugs can show indications of detonation. That's also something I looked at. If you google the subject, you can find what I'm referring to (unless you've done that already).

 

Yep, I pulled plugs mid summer to inspect. I didn't see any deposits or signs of detonation.

 

they have these things called mechanic ears, or chassis ears. guys on the suv boards use them to find bad drive line bearings and random noises. I don't have any experience with them, but maybe could stick a mic next to the knock sensor and see what it's hearings?

 

We were having discussions about knock awhile back elsewhere, and the answer was always found in the usage of gasoline and why it is the culprit, and why everyone kept dancing around varying concepts of having the best way to detect and stop this byproduct of gasoline from occurring caused by its' hydrogen radical. Looking at different concepts, or revisiting some such as Ion Sense Tuning, and Crank Angle Misfire detection, and of course a simple relocation of the existing knock sensor up and away from suspension and exhaust components. I can tell you for certain (in my case) that e85 has always been the best possible answer to keep knock, real knock, completely out of the equation. Don't get me wrong, I switch back and forth with the fuels on the fly, but in my case... you can only imagine how the datalogs look running pump gas with solid lifters. I'll keep spark retard limited to 2* because I know it's not real, and 2* pull isn't that much to make a huge difference when I'm on it, but the timing pull will at least give the user a chance at confirming if it is in fact real or not either way when reviewing the datalogs. I also relocated my knock sensor at the top of the block, just in front of the bellhousing flange, similar to the Buick guys...

- Rob

 

I've got my allowable knock retard set at 2 as well, but only between 0 and 50kpa. After 50, it's 5. But looking at my datalog, I only ever get about 1-1.5 deg at 70 and above on occasion. So I'm pretty sure I have my situation well under control

 

top of the block huh?

I'm guessing already a threaded hole there? passengers side?

the wonder for me always is, 406 never came with a knock sensor to swap over. still running a 305 sensor.

​​​​​

 

No I had to drill and tap, I put it right behind the oil fitting that the earlier models used for the oil pressure sensor, as that is my turbo oil feed, right behind the feed and by the bellhousing itself where the ID numbers are. I essentially redid the wiring for the EBL system and now have my knock sensor wire and O2 sensor wires leading back there. Was in the works of working with RBob on creating a race oriented XDF (to kinda cut to the chase for EBL owners, race only, no emissions etc, to simplify it for them), but that of course will never come to fruition now.

- Rob

 

ahh if I ever have the engine out and apart I might think about relocationing it then. doesn't sound like an in car mod.


I find myself wondering if Rbob was getting up there in age? for someone like me that just drifts in and out as my projects are always all over the map and I get side tracked. he was just always here. only time I really talked to him was a few pm's about the old north star coil pack mods. he was interested in offering parts to make it easy, but in the end didn't see the point as it didn't really solve any inherent problems and guys that wanted coil on plug were going ls ecm's anyways.

 

He kinda went above and beyond the scope. I told him years back he shoulda just created one ECM for all, one XDF for all, and make harness adapters depending on the wiring that the system was going into, with the XDF turning on and turning off whatever needed to be for the users application. This way people wouldn't have to re-pin, they could just buy a pigtail adapter pre-pinned, install, and be on their merry way. He was so damn precise though, wanted it to be exactly as GM intended, albeit Flash oriented. Meh but I digress. Sorry for the hijack ULTM8Z.

- Rob

 

No problem! Actually if it wasnt for RBob theres no way i would have gotten my Miniram dialed in.

He definitely did go above and beyond in support of this forum.

 

More!

This one wasn't so obvious in terms of finding it... Tunerpro doesn't have this table in the DA2 and Tunercat doesn't have it in 8D.

Loop Closed Param, INT Delay Factor .vs. % Adjusted SLOW o2 Voltage Error

Again, I'm not sure exactly what it does... It seems like how long the ECM waits to update the integrator vs how far off the O2 reading is from stoich. But the engine does seem to like the DA2 BDZL version more than the 8D AUJP.

8D above, DA2/3 below... the error units are also different... TP gives it in terms of mV, whereas TC gives it in counts (there's probably a conversion formula).

The improvement from this wasn't that easy to discern.. with the Ostrich, I was able to switch back and forth between 8D and DA2/3, but it wasn't obvious. I'm going to stick with it though since the general trend seems to be that DA2/3 seems to run better with the Miniram.

 

Keep in mind that these tables were originally configured by GM with vehicles employing catalytic converters and slower responding O2 sensors with less resolution overall, which was anything but. I rarely touch the INT and Proportional Gains for members because its pretty easy to tune around it, but it my case, I have the INT and O2 swing points so tight the system could literally squeeze a diamond from a lump of coal. Makes for a more enjoyable tune in my opinion, and cuts to the chase.

- Rob

 

I think I finally captured them all. Because these are mostly not available in the DA2/DA3 xdf in TP, I'm working exclusively in TC.

In fact the one in my previous post above actually is in TC, except it's name is a little different.

I ended up emailing TC to get a correlation between the DA2 and 8D tables....




So with this, I went back and converted them all to DA2/DA3

And now that I have it fully converted over, just to convince myself I'm not imagining things, I made an equivalent bin with the 8D values. Nice thing about the Ostrich... you can swap back and forth with the push of a button.

Yeah, the DA3 values runs way better.... not so much generating more power necessarily, but mainly with the "feel" of the throttle and the driveability. I'm not sure how to describe it... the 8D feels a tad more sluggish than the DA3. The DA3 feels much sharper in terms of the seat of the pants feel vs how my foot moves the throttle. I was monitoring the WB and it's hard to detect a significant change in the AFR's.

 

I'm actually curious if this is a Miniram thing or would it be applicable to other systems, like even the factory TPI.

If anyone wants to try it... I ended up cataloging all the values into an Excel table. It actually correlates to the TP tables as well as TC (in case the user doesn't have TC). Simply grab the DA3 values and copy/paste into the corresponding 8D table.

 

As I am studying up on my closed loop knowledge I thought I would look into these to see how EBL Flash port mod compares. See below. Once I get my closed loop tune dialed in a little bit more I might start playing with these to see how it likes them. These are currently all set to values found in the base bin I started with (3006 - 5.7l TPI, 6-spd, alum heads (Port Mod)) except for the PRP - Gain vs o2 Error which I reduced based off my increased injector sizing.

 

Took another look at this and created a test bin with what I believe to be the correct values to implement DA3 based values into my EBL tune. Do these look like I converted them correctly based off of your spreadsheet?

 

Yeah, those look right to me. The question though is how these behave in the EBL. Curious to know...

 

Found another parameter...

Loop Closed Param, CL Gain Factor Multiplier (as titled in SAUJP v7 xdf). Hex location 0x49B. I'm hunting down as many of these "gain factor" type parameters and tables as I can...

Again, not sure what it does (and it doesn't have a description in the XDF)...

But it's .625 in the factory 8D. I set it to 1.0 (the maximum value) to see what would happen

Definitely another significant improvement in throttle response (it just keeps getting better and better). Again, I would NEVER suspected the 7730 ECM could be made to run like this.

And I didn't want to say this when the EBL was still an active product (all due respect to RBob... God rest his soul)... But I wonder how much of the improvement people noticed with the EBL was due to the 7730 being held back by some of these more obscure parameters that GM stuck in there.

Almost like GM was trying to make the 7730 more emissions oriented rather than performance oriented.

Prior to all of this, It always felt like the ECM was fighting me on some level trying to get everything dialed in (it always felt like there was a lot being left on the table). It ran good but nowhere near on the level its running now.

 

Out of curiosity. What do your o2 cross counts look like if you were to graph them out?

For example mine have gone from this:


To this:



This is with running the settings I changed in the post above. Here are more of my PRP settings listed out as well as my o2 targets.




My next drive I am reducing my PRP - Gain vs o2 Error a little bit and increasing my Prop Gain Multiplier for Idle to 2%. I've noted my idle cross counts aren't very defined, with a slightly hunting INT. Overall, I think I am making good progress on running closed loop. Have even started diving into AE tuning.

Edit - Worth noting my wideband is showing this at cruise:



To me this is indicative of too much prop gain at non-idle? This is what pushed me to reduce PRP - Gain vs o2 Error to try out next drive. I think the swing is a little too much up down on AFR.

 

I actually idle in open loop.

But overall, I haven't gotten into this stuff anywhere near what you're doing. All I know is by seat of the pants feel the car is running way better than before.