batch fire of injectors - VLT

[skylinegtrhr]

Hi is there any news regarding CA18DET?

We have some cars with big turbos and injectors and I'm not very fond of doing them with this potential risk.

[Matt]

Not yet ... one at a time needs to be done. I've spent over 5 hours just on this one tonight based on the location stu has provided

This is what I have worked out so far:

symbol 0xBFD4 BATCH_WATER_TEMP1 (40degC)
symbol 0xBFD5 BATCH_WATER_TEMP2 (40degC)
symbol 0xBFD6 BATCH_WATER_TEMP3 (30degC)
symbol 0xBFD7 BATCH_WATER_TEMP4 (14degC)

symbol 0xBF60 BATCH_RPM_INDEXED_TIMING_TABLE (30degBDTC upto 3200 then 25degBDTC)

* based from stu's information
symbol 0xBFDB BATCH_TP_CHANGEOVER (96)
symbol 0xBFDA BATCH_TIMING_TP_CHECK (56)

; Timing to use
symbol 0xBFE0 TP_SELECT_TIMING1_1 ; 0
symbol 0xBFE1 TP_SELECT_TIMING1_2 ; 7
symbol 0xBFE2 TP_SELECT_TIMING1_3 ; 7
symbol 0xBFE3 TP_SELECT_TIMING2_1 ; 0
symbol 0xBFE4 TP_SELECT_TIMING2_2 ; 7
symbol 0xBFE5 TP_SELECT_TIMING2_3 ; 7
symbol 0xBFE6 TP_SELECT_TIMING3_1 ; 0
symbol 0xBFE7 TP_SELECT_TIMING3_2 ; 0
symbol 0xBFE8 TP_SELECT_TIMING3_3 ; 0

These have been added the software. The timing set must be offset from the other table. I dont really understand how it works yet

; Batch operation
; RPM limit 1 (<1800rpm)
; TP check (96)
; Timing low rpm 1
; TP check (56)
; Timing low rpm 2
; else
; Timing low rpm 3

; RPM limit 2 (<3400rpm)
; TP check (96)
; Timing med rpm 1
; TP check (56)
; Timing med rpm 2
; else
; Timing med rpm 3
;
; RPM limit 3 (> 3400rpm)

; TP check (96)
; Timing med rpm 1
; TP check (56)
; Timing med rpm 2
; else
; Timing med rpm 3
;

[bachig24u]

it almost looks almost like a 3d table to make the decision of changeover.
rpm, water temp, tp ??

[Matt]

Thats the way it looks. I would like to put smaller 3D tables in the software (for knock count and for this) so it displays correctly

However changing the values did nothing to the switch points on the bench so far

[bachig24u]

maybe rename this thread for CA18 as the VLT can't do sequential as Brett originally was asking.

[skyline_stu]

The constant I mentioned earlier was relating to the VLT ecu. I made no mention of CA18 info.
The VLT ecu doesn't run sequential injection, as there is only 2 injector drivers.

This is how the VLT injection mode changes over. The values are for example only

Injection starts off as a batch mode. (typically LOW load/RPM)
The calculated injection time in this example is 6ms - NOT TO SCALE !!!

Bank 1 ______|------|______________ 1 injection, of 6ms,
Bank 2 ________________|------|____ per engine cycle
<-----------------------------> 720' rotn.

Then the TP, RPM reach the threshold change over point.
Water temp (the mentioned constant) is another
Injection then changes to simultaneous injection.
Bank 1 _______|---|______|---|______ 2 injection cycles, of 3ms,
Bank 2 _______|---|______|---|______ per engine cycle
<-------------------------------> 720' rotn.

Still 6ms of injection total time, it just happens twice as frequently for half as long.

I believe JECS did this to make the injectors appear 'smaller' at low RPM to give good tune-ability at
light loads and to also allow the correct end of injection point to reduce HC/CO outputs.

The constant that I mentioned earlier is the temperature needed to allow this 'change of injection mode'
to happen. You can lock the change over out point and run batch until high RPM
(But I do not know the limitiations of the maximum injection time in batch mode)

The reason for finding this out was that it really caused some major issued with a sequential vapour
gas injection vehicle, as the LPG ECU measures the injection time per injection pulse - yet it forgets
the 2 turns of the crank for a single engine cycle. The MotorGLP ECU (LPG controller) gets confused
as load is calculated from a single injection event, not a single engine CYCLE. This causes a severe
lean point in the gas injection when the injection mode changes from batch to simultaneous.

Stu

The attached picture is of the LPG 'fuel map' table layout.

[Matt]

I've added VLT/Z31 to the release for this 1.2.81
http://www.nistune.com/mantis/view.php?id=92

CA18DET still to be investigated

[Mattward]

This is just what I need. Thanks Matt