nAst1: Progress and Concepts Thread

[Joseph Upson]

two things:

sorry for taking so long to answer Dave, but if you want RPM to be the compare done to 34A, then in the TP hex editor(or any other you might use) change DDFF(5DFF) to $62 for RPM X 25(if you shift under ~6400) or $76 for RPM X 50. you can safely draw up to ~1 amp with that circuit, wouldn't try drawing any more than that, i've tested 1.5 amps for short periods of time without issues, but i wouldn't push it.

also, bugfix version coming VERY soon, i actually didn't realize i had two serious issues, one with the idle VE table, another with the 5200-8400RPM spark table. i have quite a changelog for this release, unfortunately i don't have it 100% tested yet, so no release ATM.

I just discovered you sent me a nast1 version of the 8F code. I believe I was distracted by the problem I was having at the time that has since been corrected. Is there anything special I need to do to start working with it as I don't see any flags available to select the extended/ported tables and disable the original table to avoid any conflicts.

[robertisaar]

IIRC, i put an updated version of the enhanced 8F code on the TGPforums a while ago, it probably is a little more stabile than the one i sent you a while ago, you might want to look into it.

also: here is nAst1 1.05, check the release notes people, lots of hidden info. i don't have all of the mode 4 stuff tested yet either, but just be careful with the ones that sound like they could cause issues, like commanding 5* ATDC spark advance, stuff like that. i'll have to document that later.

[Joseph Upson]

IIRC, i put an updated version of the enhanced 8F code on the TGPforums a while ago, it probably is a little more stabile than the one i sent you a while ago, you might want to look into it.

also: here is nAst1 1.05, check the release notes people, lots of hidden info. i don't have all of the mode 4 stuff tested yet either, but just be careful with the ones that sound like they could cause issues, like commanding 5* ATDC spark advance, stuff like that. i'll have to document that later.

I did get 1.05 from you and am currently doing a desktop calibration of the VE tables to try out on the car soon. I'd probably be a little scared to run all out with it at this point.
I finally got the driveablility problem under boost figured out. I suspected spark blow out but didn't know why. Once I realized the very same motor had run 7 psi detonation free with 212 deg inlet temps on mid grade fuel and a poor tune I knew I shouldn't be having problems with 180 degree inlet temps, on 93 octane, a decent tune and alcohol injection.

When I compared an old tune in TP5 I realized the undiagnosed bad knock sensor had caused me to chip away at boosted timing to the tune of as much as 9 deg and it was when those spots were hit abruptly the sudden delay in ignition was probably allowing the spark to be drowned out as AFRs were dipping into the low 9s and has gone as low as 8.

Since that was cleared up the car is running like mad and although it's still in the car the alcohol has not been necessary for 10 psi but is set to come on at about that point where a slight drop in power is felt because I have a pre and post turbo nozzle. The post turbo nozzle is placed after the air temp sensor so the apparent ~30 deg drop in inlet air temps at 11 psi are the result of the pre turbo nozzle.

Timing has been brought to within about 3 deg of the original timing at 7 psi but after that is about 5 deg lower because initially the engine had twin turbos and wastegates with 7 psi springs and wasn't intended to see more than that at the time. Now a single 35 mm 7 psi wastegate can't handle the boost bypass so the engine has climbed accidently into the 12 psi range without a lick of detonation. Fortunately I anticipate the problem and setup the boost multiplyer and spark table to accomodate this which they have done.

The car has run very strong since the problem was resolved and what's left to do now is get on it with the cam retarded. I sent you a message a while back explaining that when I activate the VVT module the exhaust note mellows out a little more and the tone drops a little and the AFR drops about .4 while idling. Hopefully a positive difference can be felt under boost. I have to finish wiring in the Zeitronix WB unit so that I can use the logic circuit to activate the VVT at around 4000 rpm. I'll make a trip to radio shack for a 5 volt relay to set it up.

You've been more than a great help. Appreciate it.

[brian89gp]

Apologize if any of this was addressed. This is a long post to take in at once and I might have missed some stuff.

-use the 5V MAF's that everybody else except for GM uses. Many more options and a 5v signal is probably easier to scale then a frequency

http://www.sr20forum.com/technical-information-library/149465-maf-size-part-s-whp-output.html

-drop the boost multiplier idea for the VE/spark table. The tables themselves are more then capable of handling it and those values are redundant
-use a single MAP

I got the 2 BAR MAP code for you that I did a while back for $DF. Might be good for comparison purposes. Ran it for over a year with boost so I know it works. I got it work fine with a single 2 BAR MAP, thus saving the A/D channel for other purposes.

[brian89gp]

Found this nice tidbit:

http://www.diy-efi.org/pipermail/gmecm/2006-April/001689.html

And then this:

http://www.diy-efi.org/pipermail/gmecm/2006-April/001691.html

Too bad his website is no longer up. Thank god for the way-back machine though.

http://web.archive.org/web/20050321021500/http://home.cc.umanitoba.ca/~umfreedd/index_files2/ecm.html

[brian89gp]

This post and everything below are for BHAK/BCFA $DF


The dummy $4b value came from running the car and looking at the normal pressure value. Took that and hard coded it.


delete the AC pressure transducer
stock:
C720 @2 ldaA #$00
C722 call LF04E
C725 staA L014C
C728 ret

D067 ldaA L014C
D06A cmpA #$D0

E74B ldaB L014C
E74E cmpB 226, X

F0BB @2 ldaB L014C
FOBE cmpB 0, X


modified:
C720 01 @2 nop
C721 01 nop
C722 01 nop
C723 01 nop
C724 01 nop
C725 01 nop
C726 01 nop
C727 01 nop
C728 39 ret

D067 86 4B ldaA #$4b
D069 01 nop
D06A 81 D0 cmpA #$D0

E74B C6 4B ldaB #$4b
E74D 01 nop
E74E E1 E2 cmpB 226, X

F0BB C6 4B @2 ldaB #$4b
F0BD 01 nop
FOBE E1 00 cmpB 0, X

[brian89gp]

2 BAR MAP sensor (notice I reused L014C from the a/c sensor) The idea is you have your stock 1 BAR value, then you have your 2 BAR value. Everything uses the 1 BAR stock, you change the tables you want to use the 2 BAR (L014C)

stock:
9907 ldaA #$30 ;load a/d channel
9909 call LF04E ;a/d read routine
990C staA MAP ;store result
990E staA L0054 ;store result

the others are similar, just a different a/d routine is called


modified: (due to legacy, the 2 BAR MAP sensor is on the a/c sensor pin, #$00 instead of #$30
9907 call FE2E ;jump to routine
990A nop
990B nop
990C staA MAP ;store result
990E staA L0054 ;store result

9CC9 call FE2E
9CCC nop
9CCD nop
9CCE staA MAP

C51E call FE3C
C521 nop
C522 nop
C523 staA MAP



FE2E ldaA #$00 ;a/d channel
FE30 call LF04E
FE33 staA L014C
FE36 ldaA L014C ; load 2bar
FE39 cmpA #$7E ; see if it is 1 or 2 bar
FE3B bcs @+3 ; if <7A then jump (1 bar)
FE3D ldaA #$FD ; else load max 1 bar value
FE3F ret
FE40 ldaA L014C ; if 1 bar
FE43 lslA ; multiply by 2
FE44 ret

FE45 ldaA #$00
FE47 call LFD5D
FE4A staA L014C
FE4D ldaA L014C ; load 2bar
FE50 cmpA #$7E ; see if it is 1 or 2 bar
FE52 bcs @+3 ; if it is 2 bar
FE54 ldaA #$FD ; load max 1 bar value
FE56 ret
FE57 ldaA L014C ; if 1 bar
FE5A lslA ; multiply by 2
FE5B ret

[brian89gp]

2 BAR spark table
stock:
A0B1 ldaB MAP3
A0B3 ldX #$8129
A0B6 ldaA NL_RPM_INDEX


modified: (notice it is using L014C variable now) This is messy and doesn't need a call anymore, it was legacy from previous versions when I had 2 MAP sensors running at once.
A0B1 call fe27
A0B4 nop
A0B5 nop
A0B6 ldaA NL_RPM_INDEX

FE27 ldaB L014C ;2bar var
FE2A ldX #$8129 ;spark table
FE2D ret


The kpa scale is roughly this now: 10,21,33,45,57,69,81,93,105,116,128,140,152,164,17 6,188,200

[brian89gp]

The BPW boost multiplier. I would not suggest doing it this way. Extend the VE table to 2/3 BAR, perhaps 32 cells wide, and do all the boost fuel tuning in the VE table itself. Provided here only for reference.



stock:
B7D7 @1 stD L008F
B7D9 stD L0159
B7DC ldaA CTS2


modified:
B7D7 @1 call LFE4A ;F77 table addition
B7DA nop
B7DB nop
B7DC ldaA CTS2

FE4A stD L008F ;bpw
FE4C ldX #$9620 ;F77 table
FE4F ldaA L014C ;2 bar variable
FE52 ldab #$79 ;offset to start at, $80 = half
FE54 call LF6FC ;2d tbl lookup
FE57 ldX #$008F ;bpw
FE5A call LF7F8
FE5D lslD
FE5E stD L008F ;bpw
FE60 stD L0159
FE63 ret


The BPW table is located at #1620 in the binary file, it is a 1 column x 11 row table. The rows are 100kpa to 200kpa in 10kpa increments.

[brian89gp]

Independent baro sensor. Waste of an a/d channel but here for reference. (unless you have spare ones, then why not)

Take note that the BARO sensor is on the original MAP a/d of #$30 and the new MAP is on the original a/c transducer a/d of #$00


stock:
F1E1 ..............bunch of stuff

F27C


modified:
F1E1 ldaA #$30 ;a/d channel
F1E3 call LFD5D ;a/d read routine
F1E6 staA BARO
F1E8 clra
F1E9 staA L01F4
F1EC ret
F1ED-F27C nop

[SappySE107]

Good to see you around again Brian

[pocket-rocket]

Good to see you around again Brian

I'll second that

[brian89gp]

Howdy Ben.

Holy tables batman! Well, I guess you can't ever complain about not having fine enough RPM resolution anymore. How hard is that to tune?

I hate to suggest it (because I know what you would do with the tables....48x48...2304 cells/bytes...) but I would drop the boost BPW modifier and boost spark retard tables and convert the main VE and main spark table to 3 BAR. More accurate that way and easier for a less experienced person to just pick it up and start tuning. Probably would have to move down to 32x32 tables just to get them to fit.

Can the P4 PCM's read the extra space in 512 chips?

[SappySE107]

Its not too bad but its a lot more specific for RPM vs MAP with all those cells and it can become a game of wits if you chase every cell individually.

[brian89gp]

I see a large block of free space between 1FEB and 10E0 (3851 bytes) on the stock AXKK bin, enough to fit three 32x32 tables (VE, Spark, MAF AFR) with 800bytes left to spare. 32x32 should be enough for decent RPM resolution and 3 BAR.

Or if you so choose, enough to put your two 48x16 VE/Spark tables in sequential order. Would make it easier in the tuner to have one large table vs three of them.

Not nitpicking I swear, just brainstorming.


edit:

If you are concerned about loss of resolution, you could make the first 16x32 (or 48 if you prefer) the 1 BAR area, column 16 would be 100kpa. The second 16x32/48 would be 100-300kpa (column 17-32)

And thank you greatly for bringing IDA to my attention.