Fuel Throttle Enrich (VE) Type 4

[UNISA JECS]

Hi Matt I was wondering if you could have a look at the "Fuel Throttle Enrich (VE)" table for the B14 SR20DE, I posted a thread about this some years ago and I also came across a thread where you had fixed the tracing on the ER34 (sorry dont have the link) but then I compared the B14 and ER34 a noticed a difference with the "VE TPS Scale" and was thinking maybe you didn't apply a fix or finding that you had found with the ER34, just a guess cause the courser doesn't seem correct on the B14 SR20DE...

I suspect if this "Fuel Throttle Enrich (VE)" table worked correctly (trace) it could help me tune out a "transitional" light throttle boost leaning out instead of having to heavily lean heavily on the "AlphaN Increase Fuel" to get me the fuel I need in those spots, thanks in advance Matt...

[UNISA JECS]

Took a video recording of a playback (B14) along side the (ER34) Fuel Throttle Enrich (VE) map for comparison, also in the "Consult Log View Playback" you can see the "Adjusted TPS VE Index" doesn't follow the B14 VE TPS Scale but does follow the larger scale of the ER34

Even the ER34 RPM Scale seems better matched to my log then the B14 RPM Scale



On the Play back ER34 left side and B14 right side

https://youtu.be/WfWXu3ZweMc

[UNISA JECS]

Here's a copy of the log that in the video above

[Matt]

In the ECU itself the raw scales have no adjustment done to them:

R34: 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240 255

B14: 12 14 24 32 40 48 56 64 72 80 88 96 104 112 120 128

NEO ECUs (like R34) use a Boost sensor and TPS sensor combined to come up with the ATPS value. In the maptrace I actually have to double the ATPS value to get correct indexing (so from 128x2 = 256) since the actual ATPS readings only reach 128 (from where I pull the calculated value read in ECU code)

With the B14 ECU I found the opposite, where scales were half as you know, but the ATPS readings are double (reaching 255). So the adjusted TPS (ATPS) in your log file shows it going upto 208 is then halved internally to 104 to make it match the scales (and verify changes in the map match my trace, which is the main thing you are concerned with)

In both cases, I do not mess with changing the VE TPS scaling at all, but instead modify the maptrace to suit. This will mean however that B14 ATPS can never go higher than 255 (ATPS) value, even if setting its VE TPS table values above 128.

So question is, should the scaling for the B14 table doubled to display 255, when actual numbers are upto 128? I'm not sure on that one, but at this stage I would prefer to not adjust table scaling, but instead modify the maptracing which appears to be correct

[UNISA JECS]

The thing is the map tracing isn't correct on the "Fuel Throttle Enrich (VE)" at idle on the far left side its off by 1 vertical column and it will never push to the last far right columns and who knows where its truely at in between those points.

But I do understand the not wanting to change the scales as we learned in the past that wasn't a great idea for other tables and addition of TIM was a great addition, no doubt.

Now this was a video capture I did a long time ago back in 2014 when I was comparing the the standard unaltered B14 128 scale Fuel VE Table vs an even interpolated scale that went to 255 (Back then I never seen the ER34 standard 255 Fuel VE Table) so what I want to try naturally is try using the ER34 RPM and VE scales since those actually match up to my logs accurately so it appears at least more so then the standard B14 128 fuel ve table does and see if that cursor becomes accurate and usable.

This is old school back when I would change scales now I don't do that stuff but this VE tables go me thinking there might be an acception or something going on I dunno, maybe in this case the VE scales needs to be doubled I dunno but it seems to make more sense but this stuff is way over my pay grade lol
https://www.youtube.com/watch?v=1lb6kaK0STs

[UNISA JECS]

Took another video screen shot *read youtube discription*

Basically my VE cursor is in column "104" WOT but I have to make the changes in column "120" WOT to get any effective change to increase fueling

https://youtu.be/IHTi8jTaIyU

[UNISA JECS]

ve fuel table cheat sheet.JPG

Trying to make a little cheat sheet so im not blindy guessing where the cursor is, the RPM points are good but the ATPS will take longer to populate over much more tuning and logging

(209.23 KiB) Downloaded 2044 times

Also should point out I am not rescaling the VE RPM or (A)TPS scale because cause its useless, it will just move the cursor around but will still not match actual RPM points same holds true for the ATPS scale

[Matt]

I've had a look at the video. Does the change appear to be a 'static' offset from what you can tell? The difference between 104 and 120 on the indexing (16+offset) at the top, but is this consistent for down the lower end

0xB6DA: 9C 64 30 00 EB ldm #FUEL_VE,0x3064
0xB6DF: 9C 66 30 F0 EA ldm #TP_SCALE_VE,0x3066
0xB6E4: 9C 68 30 E0 EA ldm #RPM_SCALE_VE,0x3068

Indexing can be dependent on a flag 30AB#10

If the flag is set, then the ECU does some different calculations using addresses E098 x E087 and multiplies against TP (address 31AF) and then uses this value instead. Flag value can be seen in the memory monitor. From memory, the flag is set if there is a TPS fault (so load is used to index the table)

The tracing assumes the flag is clear (off). If the flag is clear, then the address comes from ATPS at memory address 314A/314B (where the LSB is discarded, so only 314B used for the index)

0xB724: AD 4A 31 lda ax,0x314A (ATPS MSB)
0xB727:
0xB727: 29 00 FF and ax, #0xFF00
0xB72A: 8D 6A 30 sta ax, 0x306A
0xB72D: 8D 5C 30 sta ax, 0x305C
0xB730: F8 sem
0xB731: AD 5D 30 lda al,0x305D
0xB734: 8D 72 39 sta al, 0x3972
0xB737: 20 BE BD jsr 0xBDBE (interpolation routine)

Apart from the above analysis, I'm unsure why the tracing appears to be 'off'

[UNISA JECS]

Well today I spent a whole day messing and tuning my P11 specifically on the VE fuel maps and I made some really good progress and got the tracers pretty much spot on, i'll need to take a clean log pull for you to look at though.

ve fuel table cheat sheet2.JPG

(234.83 KiB) Downloaded 2034 times

[UNISA JECS]

I've had a look at the video. Does the change appear to be a 'static' offset from what you can tell? The difference between 104 and 120 on the indexing (16+offset) at the top, but is this consistent for down the lower end

0xB6DA: 9C 64 30 00 EB ldm #FUEL_VE,0x3064
0xB6DF: 9C 66 30 F0 EA ldm #TP_SCALE_VE,0x3066
0xB6E4: 9C 68 30 E0 EA ldm #RPM_SCALE_VE,0x3068

It appears to be a static/linear offset, 7 through 108 got the cursor pretty much dead spot on accurate all over the map

I did notice that no matter what below 3,000RPM "AlphaN Fuel Increase" is needed to get low RPM fuel increases, this is allowing me to make my main fuel map match my actual wideband the more and more I fine tune it and makes changes to the "Fuel Throttle Enrich (VE)" table finally, its pretty awesome.

[Matt]

So RPM tracing is spot on? (had a thought that the scales are inverted)

RPM raw
0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225
RPM x 0.5
0 7.5 15 22.5 30 37.5 45 52.5 60 67.5 75 82.5 90 97.5 105 112.5

Your ATPS trace
- 7 14 20 27 34 41 47 54 61 68 74 81 88 95 101

Its got an offset of 9 at the bottom end and 20 at the top end, so isn't purely linear. I'm not sure how to yet adjust the calculations to 'fix' this

[UNISA JECS]

So RPM tracing is spot on? (had a thought that the scales are inverted)

RPM raw
0 15 30 45 60 75 90 105 120 135 150 165 180 195 210 225
RPM x 0.5
0 7.5 15 22.5 30 37.5 45 52.5 60 67.5 75 82.5 90 97.5 105 112.5

Your ATPS trace
- 7 14 20 27 34 41 47 54 61 68 74 81 88 95 101

Its got an offset of 9 at the bottom end and 20 at the top end, so isn't purely linear. I'm not sure how to yet adjust the calculations to 'fix' this

Ok let me see how I can best describe the RPM Scale and Trace

The RPM Scale values are actually wrong, the trace seems to be pretty much spot on with the "actual" values I listed below, but if you try and adjust the RAW values to fix this everything goes to shit basically

Something is not correct about the RAW values and the RPM Values, something is amiss, if you increase the RAW values the cursor gets way off track big time.

Nistune>>>>>>Actual Real RPM
Value>>>>>>>Values
5625>>>>>>>>~6300
5250>>>>>>>>~5870
4875>>>>>>>>~5350
4500>>>>>>>>~5000
4125>>>>>>>>~4600
3750>>>>>>>>~4200
3375>>>>>>>>~3850
3000>>>>>>>>~3500
2625>>>>>>>>~3200
2250>>>>>>>>~2700
1875>>>>>>>>~2300
1500>>>>>>>>~1950
1125>>>>>>>>~1500
750>>>>>>>>>~1000
375>>>>>>>>>~700
0>>>>>>>>>>>~0-350??? (impossible for me to measure)



https://www.youtube.com/watch?v=dszAAm3pej8

[UNISA JECS]

I swear that R34 RAW scale and RPM values match my cursor most closely then anything else i've seen with respect to other cars using the VE Fuel Table

sdfsfsdfsdfsdf.JPG

(289.81 KiB) Downloaded 2016 times

[Matt]

I'll have a play with this ECU on the bench next week. The code and routines between B14 and R34 appeared the same (including when it uses the load when the TPS/boost signal isn't available on R34)

[UNISA JECS]

I'll have a play with this ECU on the bench next week. The code and routines between B14 and R34 appeared the same (including when it uses the load when the TPS/boost signal isn't available on R34)

Hey Matt have you had a chance to mess with this yet?