nAst1: Progress and Concepts Thread

[robertisaar]

and closer still.... now the code is ~75% functional.

with spark cut enabled, functions as a spark cut for ~2 seconds, then it flips back and forth between cut and no cut for about a second, then goes back to full cut again...

quite confusing, but there's only so many places in the code that can modify dwell, i'll just have to breathe on more of them?

[robertisaar]

p..... progress?

tested up to 12,000RPM, was still getting 50RPM increments... so i'm not sure what was going on there. perhaps just logging only on the 100RPM bytes?

anyways, tested 4 BINs this time, 1 step fuel vs 1 step spark vs 3 step fuel vs 3 step spark.

the fuel BINs tested as i expected them to(hard to remember switch combinations and P/N vs D when using 4 circuit decoder), so those were fine, now i'm stuck with the dwell being cut correctly roughly half of the time... fuel is delivered at all times, so if you manage to accelerate past your step 3 limit with spark cut, it will continue to deliver fuel. this shouldn't really be possible unless you're experiencing some incredible pre-ignition.

so now i just have to figure out why dwell is alternating between normal and cut on roughly every other engine cycle and it will be ready for testing.

[robertisaar]

bacon is pretty nice. i could always draw some ASCII bacon in the calibration and should any of it be altered, have the program refuse to run the engine.

[pocket-rocket]

Bacon.

That is all.

Sent from my HTC One X using Tapatalk 2

[robertisaar]

pulled the testbench out of hiding(elizabeth hates it when i essentially take over the living room floor) to test a few things...

RPM increments of 100 above 6400 RPM? i didn't experience this. i ran up to 6900 and was still seeing 50RPM increments. above that speed, my pulse generator is way too touchy, i need to build another one that is more linear.

new spark cut code is promising, but after it still failing this time, i stared at the code for even longer and figured out why dwell was still being commanded. so, slight tweaking of my current revision and assuming it goes well i'll patch your BIN if you want to test it. assuming it works for you, i'll add it into the next release. spark cut(if selected) SHOULD work with either the normal 1 step or 3 step rev limiter, i'll have to check out both during the same session to be certain.

i get the feeling that there was something else i tested, but i've forgotten what. trying to work on only one thing at a time to streamline updates is easier said than done.

i do have a new accumulating trip economy calc in progress, it actually uses the PCM to do the majority of the work and then spits gallons consumed and miles traveled over the ALDL to have tunerpro do the final calculation. while i could have the PCM do all of the math, i'd like to see seperate usage and distance numbers. REQUIRES the extra RAM chip.



assuming this works out, then i'll add in the "MAP above this, always PE" option/scalar, test until functional, add in the T56 reverse solenoid, test until functional, probably release that as a new version. afterwards, i know the "AE while in PE" issue still exists and can probably tackle that, possibly in the same release?

am i missing anything else pressing?

[robertisaar]

well, we can calculate it manually to verify... find a high-load sample, post the BPW and RPM and if in double or single fire.

working backwards, 80% at 6800 should be around 7mS if double-fire, 14mS if single-fire.

[caffeine]

I've been meaning to ask, is the duty cycle value in the datalogs accurate? On a WOT pull to 6800 the log said my injectors were at 80% duty cycle, and I was only at 9 psi. I'm running the 52lb (@44.5 psi) injectors at 50 psi right now. AFRs were steady between 11.5 and 11.7.

Just put a new rear end in my car so I'll be back to tuning very soon. Still waiting on my spark rev limiter .

[ericjon262]

WANT, but at the same time.... might as well go with the 3900 and start with 240/240 instead of 210/215. either way, they both flow so much damn air the way they'll repond to positive manifold pressure is going to be exciting, to say the least.

go with the TDC! you know you want to!

2 swaps I like/would like to see, LT5, and LQ1

[robertisaar]

WANT, but at the same time.... might as well go with the 3900 and start with 240/240 instead of 210/215. either way, they both flow so much damn air the way they'll repond to positive manifold pressure is going to be exciting, to say the least.

[ericjon262]

Rwd lq1...

DO IT!!!!!

[robertisaar]

if i were to charge for nAst1.... somehow..... nobody would want to pay what i would be asking. the amount of hours i have into correctly documenting A1 is kind of staggering... then adding in the stuff i have so far to make nAst1..... then the hours i've put into converting the DF BINs into nAst1 compatible.... and eventually i want to get the 2.8 and 3.1 5 speed W-body BINs from 6D translated.... then i might do the TGP 8F stuff.... i don't want to attempt to estimate that time, i'll just end up depressed.

i like seeing these old, "outdated" PCMs be the driving force behind some rather badass builds. i mean, 13.xx range N/A V6 isn't by any means slow, i can't imagine how fast some of the boosted builds might go.... i'm still wanting to make a turbo LQ1 build, but i'm afraid it will become an endless money-pit of assploded transmissions and always having to compromise aspects due to being FWD.... if i had a 3rd/maybe 4th gen F-body or a 1st gen S/T truck, different story.

[ericjon262]

I'm def going to keep an eye on this so that if I get fed up trying to make OBD2 work, I've got another option. it's either this or an EBL Flash P4.

Thanks for doing all of this and keeping at it, I can only imagine what you have tied up in it...

also, FWIW, bl85c is on here too under a different username.

[robertisaar]

i think i gave out the info i used to make my initial version and that was used by bl85c(???) to patch into 6E or code59, i forget.... anyways, i see how he did it, not very different than what i did, in fact, he may have solved my inconsistency issue.

here is the code i saw:

      ;
        ; Disable EST
        ;
D722:            CMPB  LC500   ; Compare to RPM for Disable EST
D725:            BLO  LD74A   ; Branch if Lower
D727:            CLRA
D728:            CLRB
D729:            STD     L3FDC              ; SPK Dwell period cnt'r
D72C:            JSR     LF4EE              ; SHORT DELAY
                                                             
D72F:            STD     L3FE6   ; SPK Dwell Cnt'r 
D732:            JSR     LF4EE             ; SHORT DELAY
                                                             
D735:            STD     L3FE8               ; CURRENT EST FALL-LAST EST FALL
D738:            JSR     LF4EE             ; SHORT DELAY
                                                             
D73B:            STD     L3FF6               ; EST FALL CNT'R
D73E:            JSR     LF4EE             ; SHORT DELAY
                                                             
D741:            LDX     L3FEC              ; B Cnt'r last ref
D744:            JSR     LF4EE             ; SHORT DELAY
                                                             
D747:            STX     L3FE4              ; B Cnt'r Bgn nxt Dwell
D74A:            RTS

since i already have the "framework" portion done, it won't take me much time at all to add in a few more commands... maybe something i'll be able to do tomorrow while the wife is photographing a wedding.

[Mars]

So It seems that someone figured this spark cut thing out ? posted on thirdgen.org...

Is that patch available to everyone... ?

[robertisaar]

so does a 1930s vintage carbeurator..... doesn't mean i'll ever use one by choice.