Page 2 of 5

Posted: Thu Aug 27, 2009 2:54 am
by RomChip200
For 192 values story, log file already provided above: NIStune_2009-07-10_2007_02.csv

Will test my daisy chain of LC1s.

Crashes were due to beta previous versions of nistune, 0.9.12b is fine.
Matt wrote:also not sure with the innovate. we used the LM-1 other night on Petes laptop whilst tuning the R34 with no issue and I'm tuning/monitoring my R31 with the LC-1 on a semi regular basis

can you provide a log file (File-Configuration-Debug log) so I can look at this?

the 0.9.13 has logworks 3 and should support 2 chained devices (still to grab the LM-1 and test simulatenously with my LC-1)

when are you getting crashes? running XP or Vista? I dont get them often here so its hard to determine whats causing it. I'm only aware of crashing associated with log player (raised last week) and freeze ups during auto-probe com ports (suspect its on laptops with devices like bluetooth etc where probing com port causes windows to hang and nistune suffers - so working on a work around for that)

Posted: Thu Aug 27, 2009 10:44 pm
by Matt
okay 0.9.12b fixed it ... good thats where i found the bug causing crashes between 0.9.8 - 0.9.11 beta versions

sorry i mean a debug log when you do File-Configuration-Debug Logging and it ends in .log

Posted: Fri Aug 28, 2009 1:27 am
by RomChip200
Ok, will do debug log.
Matt wrote:okay 0.9.12b fixed it ... good thats where i found the bug causing crashes between 0.9.8 - 0.9.11 beta versions

sorry i mean a debug log when you do File-Configuration-Debug Logging and it ends in .log

Posted: Tue Sep 08, 2009 2:02 pm
by Matt
Just to let you know I've just gone over your address.zip file

Address entries in the base are common to all ECUs. If there is something not in a paritcular ECU then we 'zero' out the addresses to remove it

Problems in tables with the Z32 have been fixed. Still looking at the CA18 ones with the mixup in table descrptions. I'm checking it against the code to make sure

Posted: Tue Sep 08, 2009 6:17 pm
by RomChip200
Check the code.
On my side, It's a cross-check b/w the code and the tests in real life.
At least it solved my cold/warm startup problems on E85 by enriching the right tables.
Whe indentifying first COLD_START_ENRICH, and WARM_START_ENRICH tables in the code, you deduce FT_INJECT and AS_ENRICH

AS_ENRICH,&H3EB0,16,1,16,1,After Start Enrich vs Temp
FT_INJECT,&H3E80,32,1,32,1,First time Inj vs Temp
COLD_START_ENRICH,&H3EC0,16,1,16,1,Cold start enrich vs Temp
WARM_START_ENRICH,&H3EA0,16,1,16,1,Warm start enrich vs Temp


Matt wrote:Just to let you know I've just gone over your address.zip file

Address entries in the base are common to all ECUs. If there is something not in a paritcular ECU then we 'zero' out the addresses to remove it

Problems in tables with the Z32 have been fixed. Still looking at the CA18 ones with the mixup in table descrptions. I'm checking it against the code to make sure

Posted: Sun Sep 27, 2009 6:57 pm
by eduardo
H3EC0? Isn't this wrong? Most maps on Z32 ecus are located at the end of the code in 6500+ hex addresses.

Any new found things? Like constant for EGR and Boost solenoids?

Posted: Sun Sep 27, 2009 7:30 pm
by Matt
that doesnt look right
C42B : 96 B6 " " ldaa X00B6 (TEMP_IDX)
C42D : CE FE B0 " " ldx #WARM_START_ENRICH
C430 : F6 14 5D " ]" ldab X145D
C433 : C1 41 " A" cmpb #$41
C435 : 24 03 "$ " bcc LC43A
C437 : CE FF 30 " 0" ldx #COLD_START_ENRICH
C43A LC43A:
C43A : BD 80 EF " " jsr L80EF

Posted: Sun Sep 27, 2009 8:18 pm
by RomChip200
CA18DET here as mentioned earlier in this post !
RomChip200 wrote:
AS_ENRICH,&H3EB0,16,1,16,1,After Start Enrich vs Temp
FT_INJECT,&H3E80,32,1,32,1,First time Inj vs Temp
COLD_START_ENRICH,&H3EC0,16,1,16,1,Cold start enrich vs Temp
WARM_START_ENRICH,&H3EA0,16,1,16,1,Warm start enrich vs Temp

Posted: Sun Sep 27, 2009 9:00 pm
by RomChip200
What doesn't look right !?

The temp threshold is 15°C.
145D is initial engine temp.

If initial ambient temp is above 15°C, ECU takes warm start table as input (assumes the engine is very closed to be warm), otherwise it takes cold start table.
I verified it in real life.
cold start table is richer than warm start table.
Matt wrote:that doesnt look right
C42B : 96 B6 " " ldaa X00B6 (TEMP_IDX)
C42D : CE FE B0 " " ldx #WARM_START_ENRICH
C430 : F6 14 5D " ]" ldab X145D
C433 : C1 41 " A" cmpb #$41
C435 : 24 03 "$ " bcc LC43A
C437 : CE FF 30 " 0" ldx #COLD_START_ENRICH
C43A LC43A:
C43A : BD 80 EF " " jsr L80EF

Posted: Mon Sep 28, 2009 1:46 am
by Matt
dont worry, mixing up CA18 and Z30 addresses

I checked the CA18 stuff and all looks good.

I have some more things to check here (this is not verified)
#Ignition neutral timing table (32 byte)
IDLE_TIMING,&H39B0,32,1,32,1,Idle Ignition Neutral Timing

### Idle timing (TPS=on) ###
IDLE_TIMING_GEAR,&H3B10,8,1,8,1

#Neutral timing
IDLE_TIMING_NEUTRAL,&H39B0,16,1,16,1,Neutral timing

### Non-Idle timing (TPS=off) ###
WARMUP_TIMING_TEMP_MIN,&H3FD4,1,1,1,1
WARMUP_TIMING_TEMP_MAX,&H3FD5,1,1,1,1
#WARMUP_TIMING_TEMP_OVER,&H394F,1,1,1,1
#WARMUP_TIMING_COLD_START_TEMP,&H3FD6,1,1,1,1
#WARMUP_TIMING_TEMP_DELTA,&H3FD1,1,1,1,1
#WARMUP_TIMING_RPM_DELTA,&H3FD2,1,1,1,1
WARMUP_TIMING,&H3B90,8,1,8,1

#Load/RPM indexed timing map references
#LOAD_GEAR_TIMING,&H38E0,8,1,8,1,Timing table vs RPM/Load
LOAD_GEAR_TIMING_TP_MAX1,&H3FDA,1,1,1,1
LOAD_GEAR_TIMING_TP_MAX2,&H3FDB,1,1,1,1
LOAD_GEAR_TIMING_RPM_MAX1,&H3FDC,1,1,1,1
LOAD_GEAR_TIMING_RPM_MAX2,&H3FDD,1,1,1,1


#BF4B/9B and BF4C/9C are the O2 lean/rich voltage test points
#B5D0 is 8 byte table indexed by TP stored in 117B
#BB56, BB58, BB59, BB5A, BB5B are constants used for enrichment determination
#B820 RPM indexed table
#B8B0 previous TP indexed table
#B8C0 temp indexed table
#B8D0 RPM indexed table

Re: Z32 256 adress file

Posted: Mon May 23, 2011 11:40 pm
by RomChip200
Hello Matt,

Makes a long time :-)

I need to add some variables in the ADR File to be able to change easily their values.
Typically:
DATA:F407 AFR_min_coef: fcb 90 ; ...
DATA:F407 ; Minimum AFR coef: 90%
DATA:F408 AFR_max_coef: fcb 110 ; ...
DATA:F408 ; Maximum AFR coef: 110%

Do you have a generic format available for additional custom variables or tables ?

Re: Z32 256 adress file

Posted: Tue May 24, 2011 12:00 am
by Matt
okay thanks. generic format.... Just need to know units, conversion and min/max values

Re: Z32 256 adress file

Posted: Tue May 24, 2011 12:08 am
by RomChip200
I don't care.
Just something like:
# my new variable
MY_NEW_COEF,&HXXXX,1,1,1,1,Self explanatory variable name

This is just to ease the direct manipulation of binary values .... through Nistune
By adding such variable, you are supposed to know what you're doing.
Matt wrote:okay thanks. generic format.... Just need to know units, conversion and min/max values

Re: Z32 256 adress file

Posted: Thu May 15, 2014 11:05 am
by Matt
FYI Adding to the next version. Just been going through old posts!

1.2.76 and knock

Posted: Wed Jun 07, 2017 8:47 pm
by RomChip200
Matt wrote:okay the addresses were updated in the last release

as for these

AFR_LEARN_TP_SCALE,&H7BA0,8,1,8,1,AFR learn TP scale
KNOCK_LEARN_TP,&H7F74,1,1,1,1,Knock learn TP threshold
KNOCK_THRESHOLD_MIN,&H7FDA,1,1,1,1,Knock min TP threshold
KNOCK_THRESHOLD_MAX,&H7FDB,1,1,1,1,Knock min TP threshold
PRVR_TP_LOW,&H7FC3,1,1,1,1,PRVR TP low threshold

Two of these KNOCK_THRESHOLD_MIN and KNOCK_THRESHOLD_MAX already exist but the others needed to be added. Those will be in the address files/software for the next release
Something wrong in v1.2.76 in BASE_RB20_VG30_256_E.ADR for VG30DETT:
KNOCK_THRESHOLD_MIN,&H7932,1,1,1,1
KNOCK_THRESHOLD_MAX,&H7931,1,1,1,1
==> this corresponds to nothing on VG30DETT ECU

Before:
KNOCK_THRESHOLD_MIN,&H7FDA,1,1,1,1,Knock min TP threshold
KNOCK_THRESHOLD_MAX,&H7FDB,1,1,1,1,Knock min TP threshold

Now:
#Knock retard
KNOCK_THRESHOLD_MIN,&H7932,1,1,1,1
KNOCK_THRESHOLD_MAX,&H7931,1,1,1,1
KNOCK_RETARD_LIMIT_TP_IDX1,&H7FDA,1,1,1,1
KNOCK_RETARD_LIMIT_TP_IDX2,&H7FDB,1,1,1,1
KNOCK_RETARD_LIMIT_RPM_IDX1,&H7FDC,1,1,1,1
KNOCK_RETARD_LIMIT_RPM_IDX2,&H7FDD,1,1,1,1

KNOCK_RETARD_LIMIT_TP_IDX1 is the former KNOCK_THRESHOLD_MIN
KNOCK_RETARD_LIMIT_TP_IDX2 is the former KNOCK_THRESHOLD_MAX

Seems KNOCK_THRESHOLD_MIN cannot cohabit with KNOCK_RETARD_LIMIT_TP_IDX1.
Introduction of KNOCK_RETARD_LIMIT_TP_IDX1 disrupts the use of KNOCK_THRESHOLD_MIN.

So, still using the legacy KNOCK_THRESHOLD_MIN/MAX makes the indexes wrong for the knock tables, refer to the pic.
Nistune_knock_indexes.jpg
(287.22 KiB) Downloaded 3747 times