$8D new bit status definition
07-17-2006, 11:57 AM
#101
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
I know this won't help much, but I like to see the S_AUJP cal data and compare it to what I've got on occasion as I strive to find what seems to work the best.....
07-17-2006, 07:39 PM
#102
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
I think the way you are setting it up will work fine.
All of the S_AUJP items listed and then comment out the ones you DON'T want used.
The Value from your existing calibration would then be used.
If I got that right.
Biggest thing there is to ensure someone doesn't disable the EGR or CCP if they don't want to.
Seems very user friendly
UPDATE:
I fixed the bit display routine.
I numbered the bits 0-7 instead of 7-0 Duh!
Also included the knock light staus bits to aid with troubleshooting.
You guys have mail to try.
All of the S_AUJP items listed and then comment out the ones you DON'T want used.
The Value from your existing calibration would then be used.
If I got that right.
Biggest thing there is to ensure someone doesn't disable the EGR or CCP if they don't want to.
Seems very user friendly
UPDATE:
I fixed the bit display routine.
I numbered the bits 0-7 instead of 7-0 Duh!
Also included the knock light staus bits to aid with troubleshooting.
You guys have mail to try.
07-17-2006, 10:24 PM
#103
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Originally Posted by JP86SS
I think the way you are setting it up will work fine.
All of the S_AUJP items listed and then comment out the ones you DON'T want used.
The Value from your existing calibration would then be used.
If I got that right.
Biggest thing there is to ensure someone doesn't disable the EGR or CCP if they don't want to.
Seems very user friendly
All of the S_AUJP items listed and then comment out the ones you DON'T want used.
The Value from your existing calibration would then be used.
If I got that right.
Biggest thing there is to ensure someone doesn't disable the EGR or CCP if they don't want to.
Seems very user friendly
Once I get done testing I (or someone else)need to make a complete control file for the current version of the S_AUJP_Vx bin to be released. I would be nice to have a few people look it over after it is created.
BTW, the merge_bin tool was written to work with any size bin files up to 64K bytes. That is, it will work for 4K bins, 16K bins, 32K bins, 64K bins.....or anything in between. So the other guys doing source for other bins & ECMs can use it also. I did this mainly because I am still using the $58 BBZB and slowly working my way back to the $8D AUJP. I wanted one universal tool.
Originally Posted by JP86SS
UPDATE:
I fixed the bit display routine.
I numbered the bits 0-7 instead of 7-0 Duh!
Also included the knock light staus bits to aid with troubleshooting.
You guys have mail to try.
I fixed the bit display routine.
I numbered the bits 0-7 instead of 7-0 Duh!
Also included the knock light staus bits to aid with troubleshooting.
You guys have mail to try.
07-17-2006, 10:31 PM
#104
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
My vote is to have them all active so you get a "base" S_AUJP file.
Then as you progress with tuning you could comment them out and reload.(as you tweak existing items to get where you want them).
That way you can re-start over and transfer the things you like using the same control file. Nice progressive format that way.
Its been a good starting point to have by leaving everthing as the base file was, the SA has been a bit much for some combos but its easy enough to run the routine again with the line commented and use your existing table.
Then as you progress with tuning you could comment them out and reload.(as you tweak existing items to get where you want them).
That way you can re-start over and transfer the things you like using the same control file. Nice progressive format that way.
Its been a good starting point to have by leaving everthing as the base file was, the SA has been a bit much for some combos but its easy enough to run the routine again with the line commented and use your existing table.
07-18-2006, 09:32 AM
#105
Member
Join Date: Jun 2001
Location: Eh?
Posts: 391
Likes: 0
Received 0 Likes
on
0 Posts
Car: 1988 Monte Carlo SS
Engine: 5.7L TPI
Transmission: T5
Axle/Gears: 3.73
Originally Posted by junkcltr
EDIT: I like the MonteCarSlow NVRAM module. You will never have to worry about using memory locations again. It is easily the best hardware I bought for messing with ECMs. You might want to check if he has made more of them.
There will be more made soon... 
07-18-2006, 09:32 AM
#106
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
That sounds great guys! That merge tool is going to be very very useful from the sound of it.
Now back to your regularly scheduled channel....
Now back to your regularly scheduled channel....
07-18-2006, 09:57 AM
#107
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Originally Posted by MonteCarSlow
Neato to see someone using it for different stuff. There will be more made soon...
There will be more made soon...
07-18-2006, 11:31 PM
#108
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
I made progress on the merge tool. The code for it is complete. I made a S_AUJP_V3_control.txt file for it. I need to test and verify that the address locations are correct for the control file and then I will send it out for others to try/test. It will probably take a few nights to verify the control file with all the file diff'ing that is involved to test the address list.
Junk
Junk
07-19-2006, 12:06 AM
#109
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
Sounds like a great effort and huge job - we all thank you!
07-20-2006, 12:40 AM
#110
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Current Knock LED algorithm:
The knock light is lit when the ECM detects enough engine knock that it starts retarding engine timing. The stock AUJP is 1.9* or 3.8* before it starts to retard timing (and light the knock light).
Proposed Knock LED algorithm:
The knock light is lit when ANY engine knock is present.
Or would it be better to have another user selectable bit so that the end user can select which method they like better?
The knock light is lit when the ECM detects enough engine knock that it starts retarding engine timing. The stock AUJP is 1.9* or 3.8* before it starts to retard timing (and light the knock light).
Proposed Knock LED algorithm:
The knock light is lit when ANY engine knock is present.
Or would it be better to have another user selectable bit so that the end user can select which method they like better?
07-20-2006, 04:18 AM
#111
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
If not too much trouble, user selectable would be best. If only one choice is possible, I'd suggest lighting the KL when ANY knock is present or detected.
07-21-2006, 04:29 AM
#113
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
OK, point well taken. You're right, if it's not enough to cause any spark retard, there's probably no point in worrying about it.
H3LL, I get spark retard right now when I'm not even getting knock..... LOL
H3LL, I get spark retard right now when I'm not even getting knock..... LOL
08-15-2006, 11:39 PM
#114
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
Any updates or code to test on this?
Since we found out there is a residual voltage for the KL, would putting a resister in-line with the KL indicator light make any difference? If so, how big?
Since we found out there is a residual voltage for the KL, would putting a resister in-line with the KL indicator light make any difference? If so, how big?
08-16-2006, 11:26 AM
#115
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
AFAIK that should do it but am unsure of the value to use.
It must be due to the leakage current of two outputs combined causing this.
Might just do a measurement across that pin to ground and use a resistor that matches that level (or slightly higher)
I am still seeing the bits flashing rather than staying constant when in PE.
I am going to do another test this weekend if weather and time permits.
Going to put the entire set/clear routine in and see if the bit display is more consistent. I think even with the interrupts being disabled, the routine is clearing the word and is not able to reset them during the current scan.
The ALDL is grabbing samples with mixed results?? Shouldn't be but I'm skeptical. If I leave the bits set or clear until they are checked again it may indicate better but also leave them on/off an extra scan. Not sure, thats why I want to confirm it with the full sequence.
The other two testers I transferred cal to have not responded.
Z69' and I are both looking to complete this phase (V4) and move on to the next set revisions. Feature creep (ebb and flo has been holding up the release) Still might be an issue with the 8K. Need to get with him on it.
It must be due to the leakage current of two outputs combined causing this.
Might just do a measurement across that pin to ground and use a resistor that matches that level (or slightly higher)
I am still seeing the bits flashing rather than staying constant when in PE.
I am going to do another test this weekend if weather and time permits.
Going to put the entire set/clear routine in and see if the bit display is more consistent. I think even with the interrupts being disabled, the routine is clearing the word and is not able to reset them during the current scan.
The ALDL is grabbing samples with mixed results?? Shouldn't be but I'm skeptical. If I leave the bits set or clear until they are checked again it may indicate better but also leave them on/off an extra scan. Not sure, thats why I want to confirm it with the full sequence.
The other two testers I transferred cal to have not responded.
Z69' and I are both looking to complete this phase (V4) and move on to the next set revisions. Feature creep (ebb and flo has been holding up the release) Still might be an issue with the 8K. Need to get with him on it.
08-16-2006, 11:49 AM
#116
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
Thanks for the reply! I'll try to take a resistance measurement on that tonight and them go to RS tomorrow and grap one to test this out. As long as you guys don't think there's any way I can burn something out doing this. Would hate to do that by some ignorant move on my part! Haven't heard from Z, sent him email on another topic (not V4 related) a day or two ago, so he must be pretty busy. Haven't heard from J either.
Hope all is well on your end...
Hope all is well on your end...
08-16-2006, 05:21 PM
#117
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
Shouldn't make any more current by having the resistor. The side effect may be a dim light if its too much resistance. Just have to experiment with it.
Just don't want to go lower resistance and exceed the current limit.
Going ok on my end just been swamped at work (biggest year ever in sales, now I've gotto get the stuff built )
Of course the hobby is more important
Just don't want to go lower resistance and exceed the current limit.
Going ok on my end just been swamped at work (biggest year ever in sales, now I've gotto get the stuff built
)Of course the hobby is more important
10-09-2006, 10:26 PM
#118
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Sorry this took so long.....been busy testing the B_AUJP (boost AUJP) in the car.
The problem with the stock ECM driver (switch) is that the resistance when it is OFF is about 20K ohms. A "good" driver/switch would be in the 1000K ohm range. Since the 20K ohms is not high, then the LED is partially lit when the ECM driver/switch is OFF. Increasing the LED resistor will make the LED more dim when in the OFF state and also more dim when in the ON state.
The only solution is a generic light bulb or a new switch/ckt that is controlled by the ECM driver. The new ckt would have a higher OFF resistance for turning the LED OFF and ON.
Wow, it is great to have the AUJP back in the car. The idle and part throttle are night and day better than the BBZB code. I have the car running off of the DIY-WBO2 sensor right now using a 2D lookup table and it is working great. If the WBO2 fails then it switches back to the NBO2 (still installed).
I have a "learn disable" RPM for closed loop at a min. RPM right now because of cam overlap causing rich idle. Always had this problem. That is, I run open loop below 1000 RPM (CAL value). I want to run closed loop for all RPMs (like stock AUJP). I was thinking of adding a AFR vs. RPM vs. MAP table for closed loop. It would replace the stock O2 mv threshold values (median threshold, min. O2, max. O2, etc). It would allow different AFRs for RPM and MAP.
I was thinking of using RAM location 0x00F6 for the 3D lookup table result It is used by the AUJP for the O2 cool offset. The stock AUJP always has this set to zero after the stock AUJP 2D lookup. What do you guys think? Something that would be usuable in the S_AUJP4?
The problem with the stock ECM driver (switch) is that the resistance when it is OFF is about 20K ohms. A "good" driver/switch would be in the 1000K ohm range. Since the 20K ohms is not high, then the LED is partially lit when the ECM driver/switch is OFF. Increasing the LED resistor will make the LED more dim when in the OFF state and also more dim when in the ON state.
The only solution is a generic light bulb or a new switch/ckt that is controlled by the ECM driver. The new ckt would have a higher OFF resistance for turning the LED OFF and ON.
Wow, it is great to have the AUJP back in the car. The idle and part throttle are night and day better than the BBZB code. I have the car running off of the DIY-WBO2 sensor right now using a 2D lookup table and it is working great. If the WBO2 fails then it switches back to the NBO2 (still installed).
I have a "learn disable" RPM for closed loop at a min. RPM right now because of cam overlap causing rich idle. Always had this problem. That is, I run open loop below 1000 RPM (CAL value). I want to run closed loop for all RPMs (like stock AUJP). I was thinking of adding a AFR vs. RPM vs. MAP table for closed loop. It would replace the stock O2 mv threshold values (median threshold, min. O2, max. O2, etc). It would allow different AFRs for RPM and MAP.
I was thinking of using RAM location 0x00F6 for the 3D lookup table result It is used by the AUJP for the O2 cool offset. The stock AUJP always has this set to zero after the stock AUJP 2D lookup. What do you guys think? Something that would be usuable in the S_AUJP4?
10-09-2006, 10:49 PM
#119
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
The only solution is a generic light bulb or a new switch/ckt that is controlled by the ECM driver. The new ckt would have a higher OFF resistance for turning the LED OFF and ON.
I have the car running off of the DIY-WBO2 sensor right now using a 2D lookup table and it is working great. If the WBO2 fails then it switches back to the NBO2 (still installed).
I have a "learn disable" RPM for closed loop at a min. RPM right now because of cam overlap causing rich idle. Always had this problem. That is, I run open loop below 1000 RPM (CAL value). I want to run closed loop for all RPMs (like stock AUJP). I was thinking of adding a AFR vs. RPM vs. MAP table for closed loop. It would replace the stock O2 mv threshold values (median threshold, min. O2, max. O2, etc). It would allow different AFRs for RPM and MAP.
I was thinking of using RAM location 0x00F6 for the 3D lookup table result It is used by the AUJP for the O2 cool offset. The stock AUJP always has this set to zero after the stock AUJP 2D lookup. What do you guys think? Something that would be usuable in the S_AUJP4?
I have the car running off of the DIY-WBO2 sensor right now using a 2D lookup table and it is working great. If the WBO2 fails then it switches back to the NBO2 (still installed).
I have a "learn disable" RPM for closed loop at a min. RPM right now because of cam overlap causing rich idle. Always had this problem. That is, I run open loop below 1000 RPM (CAL value). I want to run closed loop for all RPMs (like stock AUJP). I was thinking of adding a AFR vs. RPM vs. MAP table for closed loop. It would replace the stock O2 mv threshold values (median threshold, min. O2, max. O2, etc). It would allow different AFRs for RPM and MAP.
I was thinking of using RAM location 0x00F6 for the 3D lookup table result It is used by the AUJP for the O2 cool offset. The stock AUJP always has this set to zero after the stock AUJP 2D lookup. What do you guys think? Something that would be usuable in the S_AUJP4?
We have OL idle control by RPM already in there thanks to Z'69. Works good when I get tired of trying to please the cam at idle. Mines not too bad but just rough around the edges and its begining to shake things apart. Will probably re-enable it this weekend.
I'm interested for sure. Might be a V5 issue just because we've delayed too much already on getting the V4 out there. Been toying with the sim NB on the LC-1 but havn't had time to really make progress. Finally got my AFR to remain mostly flat to 5500 RPM but now my VE is too close to 100 and will have to raise the FP and constant to give it another shot (Once I replace the water pump
)Having the table would probably make that easier.
Q'
Is that strictly a CL control you've devised or is it still active in PE?
Time to dive back in.
I need to get my mind off work anyway
Still have several months of BS to deal with and the good weather is fading fast.Back on the bit subject, I did replace the shortened code with the original full routine and "most" of the bit flashing went away. Seems like the display is stable now.
10-09-2006, 10:54 PM
#120
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
I'll convert over to a standard bulb, no point in complicating this just to get an LED to work. If a user wants a knock light, having to use a regular bulb instead of an LED is a pretty small inconvenience and can just be a condition of having a K light.
The capability to run open loop below a user settable RPM is something I've heard others ask about. If that can't be incorporated into V4, it would probably be a real good feature for the next release. Your suggestion on the WB with a failover ot the NB sounds pretty neat as well. You guys have an awful lot of stuff I've not seen or heard of LOL. Wish I could make the code changes like you guys. Guess it's time for this Old Phat Phart to maybe consider learning assembly. Gawd I hate the thought of that....
The capability to run open loop below a user settable RPM is something I've heard others ask about. If that can't be incorporated into V4, it would probably be a real good feature for the next release. Your suggestion on the WB with a failover ot the NB sounds pretty neat as well. You guys have an awful lot of stuff I've not seen or heard of LOL. Wish I could make the code changes like you guys. Guess it's time for this Old Phat Phart to maybe consider learning assembly. Gawd I hate the thought of that....
10-09-2006, 11:02 PM
#121
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
Come on, you have plenty of years left to learn new tricks.
10-09-2006, 11:11 PM
#122
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Good to hear you have the "bit" stuff working well. I guess the 'clear all' method doesn't work well with the ALDL timing loop. What pump shot? I just did some testing on one.....water injection. I added some code that does a PWM with a period of 200ms and 6.25ms resolution. I used the 730 ECM pin F3 for it. Spent days testing in the garage measuring PWM and water output. You can check out the results at the Power Adder board if you ever add water/alky injection. I hope to install the kit before the snow flies.
I don't have any code written for the AFR vs. RPM vs. MAP yet. I am in the research mode right now commenting the AFR and PID code. I noticed the AUJP you have a table mis-labeled. It is at location 0x84B4. It says that it is an integrator delay, but appear to be the non-idle O2 mV threshold vs. MAP table. Not too sure yet, I need to do more studying but it is worth taking a look at it.
I will probably just do the AFR vs. RPM vs. MAP without use in PE mode. I am still playing with the boost multiplier table so I don't want to introduce another variable yet. I enable PE Mode when KPA is over 100KPA regardless of TPS due to the turbos boosting at 34% throttle........stock AUJP would not normally be in PE mode there. I will probably start coding the AFR table stuff in a few weeks so it would definitely be a V5 thing. I want to try and get the car tuned better with the current code before winter.
The B_AUJP is pretty basic. It has a spark subtractor, a final BPW multiplier, boost MAP applied to the fuel eqn., PE mode when in boost, CCP and EGR disable code (both retain stock BLM operation), the knock light code, water injection BPW, DIY-WBO2 conversion to NBO2......basically all Spring and Summer time coding. All worth it though. It runs so much better than with the $58 code. The table and equation resolution is years ahead of the $58 stuff.
I don't have any code written for the AFR vs. RPM vs. MAP yet. I am in the research mode right now commenting the AFR and PID code. I noticed the AUJP you have a table mis-labeled. It is at location 0x84B4. It says that it is an integrator delay, but appear to be the non-idle O2 mV threshold vs. MAP table. Not too sure yet, I need to do more studying but it is worth taking a look at it.
I will probably just do the AFR vs. RPM vs. MAP without use in PE mode. I am still playing with the boost multiplier table so I don't want to introduce another variable yet. I enable PE Mode when KPA is over 100KPA regardless of TPS due to the turbos boosting at 34% throttle........stock AUJP would not normally be in PE mode there. I will probably start coding the AFR table stuff in a few weeks so it would definitely be a V5 thing. I want to try and get the car tuned better with the current code before winter.
The B_AUJP is pretty basic. It has a spark subtractor, a final BPW multiplier, boost MAP applied to the fuel eqn., PE mode when in boost, CCP and EGR disable code (both retain stock BLM operation), the knock light code, water injection BPW, DIY-WBO2 conversion to NBO2......basically all Spring and Summer time coding. All worth it though. It runs so much better than with the $58 code. The table and equation resolution is years ahead of the $58 stuff.
10-09-2006, 11:19 PM
#123
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Here is the simple disable CL at idle code:
Code:
CAL AREA
LIDLCLRPM: .byte 0x20 ; 0x20 => 800 RPM (RPM/25), if RPM < LIDLCLRPM then disable closed loop
CODE AREA
;--------------------------------------------
; SET CLOSED LOOP FLAG (Is LD2EF: in ANHT)
;--------------------------------------------
;; MODIFIED CODE
LD2F7: ldaa *L0058 ; get the RPM/25 variable
cmpa LIDLCLRPM ; RPM - 800 RPM, compare to 800 RPM (0x20 = 32d, 25*32 = 800RPM) RPM/25
bcs LD2FC ; branch if carry is set, br if RPM < 800 RPM
;; else, RPM > 800 RPM, allow closed loop bit to be set
bset *L0047,#0x80 ; SET CLOSED LOOP BIT b7, 1 = CLOSED LOOP ; CRef: 0xD2B1
bra LD2FF ; CLOSED LOOP LEARN QUALIFIERS (Is LD2F7 in ANHT)
LD2FC: bclr *L0047,#0x80 ; CLR b7, 1 = CLOSED LOOP ; CRef: 0xD295,0xD2A0,0xD2AD,0xD2CE,
; 0xD2D8,0xD2E2,0xD2F3
;-----------------------------------------------------
10-09-2006, 11:23 PM
#124
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
My water pump puked after about 6 overly non leagal tuning runs on the highway on saturday.
You must have an older version file.
I went through the O2 sections and fixed all that.
There's time to play over the snowy season so no big rush if you don't have anything yet.
I went through the O2 sections and fixed all that.
Code:
;--------------------------------------------------------
; FAST O2 R/L THRESHOLD vs MAP
; +/- L849C (0.053 Volts) to make X-Count window,
; Determines direction, resets Prop gains, Increment X-Counts
; TBL = O2 VOLTS * 266
;---------------------------------------------------------
; VAL O2 Volts MAP Kpa
;----------------------------
L84B4: .byte 0x82 ;=130, 0.489 20
.byte 0x8A ;=138, 0.519 30
10-09-2006, 11:24 PM
#125
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
<going back to check my latest XDF file>
Maybe, but it would probably take me several years to learn it, too....
10-09-2006, 11:27 PM
#126
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
Yep, looksto be the same spot Z'69 put it in. Is an easy way to do it.
This file has it using a variable in place of the hardcoded Idle RPM limiter so you're not stuck with the 800 RPM. XDF item can adjust it up where needed and still function. There was some talk long ago about just removing it but in cold climates the idle could roll up to bad levels so it was left in there but made adjustable in just one spot instead of two.
This file has it using a variable in place of the hardcoded Idle RPM limiter so you're not stuck with the 800 RPM. XDF item can adjust it up where needed and still function. There was some talk long ago about just removing it but in cold climates the idle could roll up to bad levels so it was left in there but made adjustable in just one spot instead of two.
10-09-2006, 11:28 PM
#127
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Assembly bites. I hate it. All my AVR projects are done with "C" using GCC. Today's compilers are more efficient than assembly programming. "C" is the best language for doing this kind of stuff once you know how it converts the statements to assembly. That takes some time to learn, but in the end it is better code. Nah, no need for learning new tricks. We all need to wait on that guy that said he was going to re-write the AUJP in "C"...forgot who said it. That will be a great day.
10-09-2006, 11:34 PM
#128
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
I thought for sure it was in there.
Shoot me over your current bin and XDF and I'll fix you up.
Need to get feedback on the final version anyhoo
(I thought you had the latest but we're slipping)
----------
I remember that...
Shoot me over your current bin and XDF and I'll fix you up.
Need to get feedback on the final version anyhoo
(I thought you had the latest but we're slipping)
----------
I remember that...
Last edited by JP86SS; 10-09-2006 at 11:35 PM. Reason: Automerged Doublepost
10-09-2006, 11:36 PM
#129
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Yep, looksto be the same spot Z'69 put it in. Is an easy way to do it.
This file has it using a variable in place of the hardcoded Idle RPM limiter so you're not stuck with the 800 RPM. XDF item can adjust it up where needed and still function. There was some talk long ago about just removing it but in cold climates the idle could roll up to bad levels so it was left in there but made adjustable in just one spot instead of two.
This file has it using a variable in place of the hardcoded Idle RPM limiter so you're not stuck with the 800 RPM. XDF item can adjust it up where needed and still function. There was some talk long ago about just removing it but in cold climates the idle could roll up to bad levels so it was left in there but made adjustable in just one spot instead of two.
Code:
;******************************** FINALLY ************************************ ; 4-16-05 found a typo in my file, CFF4 showed branch to LD004, should have been ; Branch to LD00C as in my original file. Must have been looking to much at ANHT! ; Reassembles and matches PERFECTLY !!!!!!!!!!!!! ; I still have to go back in there and add the corrections that Scott (Z69') ; has sent to me. ; Do cleanup work from when symbols were converted back to 0x and "*L" ; RENAMED JPAUJP.ASM ; 6 hours
I made the hardcoded RPM Limit a CAL variable. That way I can adjust it if I bump the idle up (during testing). I keep it about 200 RPM higher than my max. target idle.
10-10-2006, 12:16 AM
#130
Supreme Member
iTrader: (2)
Join Date: Jul 2003
Location: Dallas, TX area
Posts: 3,205
Likes: 0
Received 0 Likes
on
0 Posts
Car: 91 Formula WS6 (Black, T-Tops)
Engine: 383 MiniRam (529 HP, 519 TQ - DD2K)
Transmission: Built '97 T56, Pro 5.0, CF-DF
Axle/Gears: 4.11 posi Ford 9"
Any chance it is called "IAC Param, Command RPM Open Loop Offset" at 0x631?
I'll email you the XDF, etc. in the morning JP
I'll email you the XDF, etc. in the morning JP
10-28-2006, 11:33 AM
#131
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Just wanted to give a quick update about the stack shortening stuff.
I have been running the car for over a week with the stack shortened to fit the BAUJP in a stock ECM. It has been running great. Very happy to have it working without the NVSRAM module. I can use the NVSRAM module strictly for bench testing now.
What that means is that the SAUJP can have a bunch more stuff added that needs variables. About 0x01D8 - 0x01C2 = 16 bytes worth. No more worrying about clobbering the normally allocated stock AUJP RAM.
I have been running the car for over a week with the stack shortened to fit the BAUJP in a stock ECM. It has been running great. Very happy to have it working without the NVSRAM module. I can use the NVSRAM module strictly for bench testing now.
What that means is that the SAUJP can have a bunch more stuff added that needs variables. About 0x01D8 - 0x01C2 = 16 bytes worth. No more worrying about clobbering the normally allocated stock AUJP RAM.
10-28-2006, 02:35 PM
#132
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
Cool!
So to clarify, You replaced the stack address "test" location with $01D9 ?
That would leave 1D8 - 1C2 available.
Having 16 free would be 0x1C2 to 0x01D1. If you've been running at 0x1D8 then that is 22 free. (Even better)
So to clarify, You replaced the stack address "test" location with $01D9 ?
That would leave 1D8 - 1C2 available.
Having 16 free would be 0x1C2 to 0x01D1. If you've been running at 0x1D8 then that is 22 free. (Even better)
Last edited by JP86SS; 10-28-2006 at 02:39 PM.
10-29-2006, 12:29 PM
#133
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Yes, there should have been an 0x in front of the 16. 0x16 = 22d free RAM locations.
Search the asm file for 01C2 and replace with 01D8. There are three places where this needs to be done. I end up using almost all of it in the BAUJP with 2 bytes available. I use the two bytes for debug info like reporting async BPW, course engine airflow, etc. I change it depending what part of the code I am working on. I report the debug variable at the PROM ID location of the ALDL datastream.
The stack end address MUST be an even address. like 0x01D6, 0x01D8, 0x01DA, etc. The code will fail if it is not even.
Search the asm file for 01C2 and replace with 01D8. There are three places where this needs to be done. I end up using almost all of it in the BAUJP with 2 bytes available. I use the two bytes for debug info like reporting async BPW, course engine airflow, etc. I change it depending what part of the code I am working on. I report the debug variable at the PROM ID location of the ALDL datastream.
The stack end address MUST be an even address. like 0x01D6, 0x01D8, 0x01DA, etc. The code will fail if it is not even.
10-29-2006, 12:54 PM
#134
Supreme Member
Thread Starter
iTrader: (1)
Join Date: Apr 2004
Location: Browns Town
Posts: 3,178
Likes: 0
Received 3 Likes
on
3 Posts
Car: 86 Monte SS (730,$8D,G3,AP,4K,S_V4)
Engine: 406 Hyd Roller 236/242
Transmission: 700R4 HomeBrew, 2.4K stall
Axle/Gears: 3:73 Posi, 7.5 Soon to break
2 of those locations look to be RAM refresh and initializing to zero upon startup. I was thinking to leave them alone or the refresh could cause issues.
I just added L0067 (Old VE) as the PROM ID this morning
Remember to use the 3 or 4 to load the double value consistently.
I'm also looking at making the datastream longer to include some/all of these:
L006F Gram Air/Cyl
L006B Limited Air Flow
L01AF O/L AFR
Just need to find out how to work the checksum on the message.
Have seen that it shouldn't matter but just have the need to know.
Didn't know that, good to know though 
Remember to use the 3 or 4 to load the double value consistently.
I'm also looking at making the datastream longer to include some/all of these:
L006F Gram Air/Cyl
L006B Limited Air Flow
L01AF O/L AFR
Just need to find out how to work the checksum on the message.
Have seen that it shouldn't matter but just have the need to know.
10-29-2006, 08:35 PM
#135
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
I replaced all three stack_end locations. I don't see why you think the refresh would cause a problem. It reads a ram location and writes it. I don't know if the MCU has DRAM or SRAM so I do the refresh. I will have to read it a bit more to see what the refresh does in context to everything else. The stack never needs a refresh so it is probably SRAM.
For the airflow, the value you listed will max out at 255 gm/sec. I run up to 400 gm/sec so I dump out the course airflow which will read higher than 255 gm/sec.
For the async BPW
The need for an even stack address is because of the RAM refresh and clear. It was written to clear two bytes per read/write.
For the airflow, the value you listed will max out at 255 gm/sec. I run up to 400 gm/sec so I dump out the course airflow which will read higher than 255 gm/sec.
Code:
stx *L0064 ; INTEGER PART OF AIR FLOW ;; TEMP DEBUG stx DEBUG_RAM1 ; save to RAM, save the quotient (This is the MSBs of the AIRFLOW in GM/SEC LCC73: tab ; FRACT TO B Reg ; CRef: 0xCC6C
Code:
LCD74: brclr *L0037,#0x80,LCD7C ; BR IF NOT b7, 1 = ENGINE RUNNING ; CRef: 0xBE03,0xCCDD
; ... else
;; TEMP DEBUG, Save the Aysnc BPW to report to the ALDL datastream
ldx #0x0000 ; clear the Async BPW reporting variable
stx DEBUG_RAM1 ; save to RAM
brset *L0045,#0x80,LCD7F ; BR IF TPS ACCEL ENRICH ON BIT SET
; ... else
LCD7C: jmp LCE02 ; EXIT (Goes to LCDFA in ANHT) ; CRef: 0xCD74
This is the actual Aysnc BPW update
;; TEMP DEBUG, Save the Aysnc BPW to report to the ALDL datastream
stx DEBUG_RAM1 ; save to RAM
jsr LE4BF ; Short delay (Is LE4B7 in ANHT)
ldaa #0x04 ; bit 2
jsr LE5DF ; TRIGGER ASYNC FUEL OUT (Is LE5D7 in ANHT)
ldaa #0xFB ; not B2
jsr LE4BF ; Short delay (Is LE4B7 in ANHT)
jsr LE5EA ; RESET ASYNC TRIGGER (Is LE5E2 in ANHT)
;-----------------------------------------------------------------------
LCE56: ldd L012E ; get accum fuel (Noted as L012E+1 in ANHT) ; CRef: 0xCE3D 
Thread
Thread Starter
Forum
Replies
Last Post
Thirdgen89GTA
NW Indiana and South Chicago Suburb
4
11-26-2015 11:04 PM
