Injection phase / end angle for 32-bit?

redwagon · **Posted:** Wed Dec 09, 2015 1:38 am

Please excuse me if this has come up before, I couldn't anything pertaining to the 32-bit ECU.

I see the 16-bit (Carberry) guys have this functionality defined but wonder if it can found in the 32-bit ECUs. It's very helpful when dialing in larger injectors, especially for tip-in, to be able to control the crank angle at which the injectors stop firing. Any sign of this logic in the code? Thanks.

jkopinga · **Posted:** Sun Dec 20, 2015 10:30 pm

Bump!!

As Redwagon pointed out it would be very helpful to be able to adjust injection end angle when using different than stock injectors and/or manifold combinations.
Different spray patterns and droplet size will make these droplets travel faster or slower depending on the size which you can cater for with a different injection end angle.

EG. On my standalone 330* works best at idle being gradually advanced to 150* at 6k or so. Since I map in injector milliseconds I can clearly see quite large AFR changes when changing end angle while spraying the same amount of fuel. I tune end angle for the richest afr at several rpm points and then remap fuel. This will make for a much more responsive and higher mileage setup.

Jasper
CS Racing
Taiwan

redwagon · **Posted:** Wed Dec 30, 2015 5:40 am

Really? No interest in this?

td-d · **Posted:** Wed Dec 30, 2015 3:09 pm

Not so much no interest, as no time - been moving house for the past 2 weeks, I'll try and have a look if work doesn't catch up with me too quickly from next week on.

redwagon · **Posted:** Thu Dec 31, 2015 1:20 am

td-d wrote:

Not so much no interest, as no time - been moving house for the past 2 weeks, I'll try and have a look if work doesn't catch up with me too quickly from next week on.

Thanks buddy. I think the big injector crowd will gain a lot from this.

td-d · **Posted:** Thu Jan 14, 2016 9:55 am

Been trawling through, I just don't have enough of that area of the rom well defined to really see what's going on. But...

This table from the 32bit group N looks suspiciously similar to the injector timing table in carberry (x axis is mmHg). In the MAF roms, it's a straight 0 across the full table (and another table with straight 180's in it), which also rings a bell.

If you want to try and play with the table, I can define it for you - but it's really a shot in the dark.

redwagon · **Posted:** Mon Jan 18, 2016 9:05 am

I would love to try it out mate. I'm still fighting with this problem. If I add enough fuel to avoid stumbles blipping the throttle in neutral, I get a rich stumble under load.

td-d · **Posted:** Tue Jan 19, 2016 6:02 am

Alight - these are the 3 tables that are related in the the 400U rom, I suspect it is table 1, I do NOT know if the scaling units are indeed correct, I have used the same notation as Carberry - but it seems congruent with what Jason discovered in the MAF roms. Table 2 and table 3 are blended using the map ratio.

Add this to your RR definitions:

Code:

<table type="3D" name="Injector Timing Related 1" category="Alpha Fueling - Injector Timing" storagetype="uint8" endian="big" sizex="16" sizey="2" userlevel="4" logparam="unkn" storageaddress="0xCA5F4">
   <scaling units="Unknown" expression="(x*1.40625)-1.8e2" to_byte="(x+1.8e2)/1.40625" format="0.000" fineincrement=".01" coarseincrement=".1" />
   <table type="X Axis" name="Manifold Pressure" storagetype="float" endian="big" logparam="unkn" storageaddress="0xCA5AC">
   <scaling units="psi absolute" expression="x*.01933677" to_byte="x/.01933677" format="0.000" fineincrement=".01" coarseincrement=".1" />
   </table>
   <table type="Y Axis" name="Engine Speed" storagetype="float" endian="big" logparam="unkn" storageaddress="0xCA5EC">
    <scaling units="RPM" expression="x" to_byte="x" format="#" fineincrement="50" coarseincrement="100" />
   </table>
   <description></description>
</table>
<table type="3D" name="Injector Timing Related 2" category="Alpha Fueling - Injector Timing" storagetype="uint8" endian="big" sizex="16" sizey="9" userlevel="4" logparam="unkn" storageaddress="0xCA678">
   <scaling units="Degrees BTDC of Intake Stroke to END Injection" expression="x*1.40625" to_byte="x/1.40625" format="0.000" fineincrement=".01" coarseincrement=".1" />
   <table type="X Axis" name="Engine Load" storagetype="float" endian="big" logparam="unkn" storageaddress="0xCA614">
    <scaling units="g/rev" expression="x" to_byte="x" format="0.00" fineincrement=".01" coarseincrement=".1" />
   </table>
   <table type="Y Axis" name="Y Axis 0xCA654" storagetype="float" endian="big" logparam="unkn" storageaddress="0xCA654">
    <scaling units="RPM" expression="x" to_byte="x" format="#" fineincrement="50" coarseincrement="100" />
   </table>
   <description></description>
</table>
<table type="3D" name="Injector Timing Related 3" category="Alpha Fueling - Injector Timing" storagetype="uint8" endian="big" sizex="16" sizey="9" userlevel="4" logparam="unkn" storageaddress="0xCA76C">
   <scaling units="Degrees BTDC of Intake Stroke to END Injection" expression="x*1.40625" to_byte="x/1.40625" format="0.000" fineincrement=".01" coarseincrement=".1" />
   <table type="X Axis" name="Engine Load" storagetype="float" endian="big" logparam="unkn" storageaddress="0xCA708">
    <scaling units="g/rev" expression="x" to_byte="x" format="0.00" fineincrement=".01" coarseincrement=".1" />
   </table>
   <table type="Y Axis" name="Y Axis 0xCA748" storagetype="float" endian="big" logparam="unkn" storageaddress="0xCA748">
    <scaling units="RPM" expression="x" to_byte="x" format="#" fineincrement="50" coarseincrement="100" />
   </table>
   <description></description>
</table>

td-d · **Posted:** Tue Jan 19, 2016 6:05 am

This is what they look like:

redwagon · **Posted:** Tue Jan 19, 2016 11:21 pm

Oh, I notice that Table 1, the one that's scaled to MAP, has only two rpm rows where the 16-bit one has 13. That seems odd, given how unlikely it is to hit 29psi MAP at 800 or 1,000rpm...

Also, would it be possible to port that to an ECUflash deffo?

td-d · **Posted:** Wed Jan 20, 2016 12:28 am

The group N 32 bit rom also has multiple lines, but it's different in the MAF Roms - just like the 16but MAF Roms (which only have 1 axis - i.e. 2d). Don't forget with interpolation, anything above 1800 hits that second line (similar to the 2d table in the 16 bit MAF roms). I'll put it into Ecuflash, no problem.

td-d · **Posted:** Wed Jan 20, 2016 3:14 am

Code:

   <scaling name="Injector_Timing_1" units="Degrees BTDC of Intake Stroke to END Injection" toexpr="(x*1.40625)-180" frexpr="(x+180)/1.40625" format="%.3f" min="0" max="255" inc="1" storagetype="uint8" endian="big"/>
   <scaling name="Injector_Timing_2" units="Degrees BTDC of Intake Stroke to END Injection" toexpr="x*1.40625" frexpr="x/1.40625" format="%.3f" min="0" max="255" inc="1" storagetype="uint8" endian="big"/>

   <table name="Injector Scaling Related 1" category="Alpha Fueling - Injector Timing" address="ca5f4" type="3D" level="1" scaling="Injector_Timing_1">
      <table name="Manifold Absolute Pressure" address="ca5ac" type="X Axis" elements="16" scaling="psi"/>
      <table name="Engine Speed" address="ca5ec" type="Y Axis" elements="2" scaling="RPM"/>
   </table>

   <table name="Injector Scaling Related 2" category="Alpha Fueling - Injector Timing" address="0xCA678" type="3D" level="1" scaling="Injector_Timing_2">
      <table name="Engine Load" address="0xCA614" type="X Axis" elements="16" scaling="g/rev"/>
      <table name="Engine Speed" address="0xCA654" type="Y Axis" elements="9" scaling="RPM"/>
   </table>

   <table name="Injector Scaling Related 3" category="Alpha Fueling - Injector Timing" address="0xCA76C" type="3D" level="1" scaling="Injector_Timing_2">
      <table name="Engine Load" address="0xCA708" type="X Axis" elements="16" scaling="g/rev"/>
      <table name="Engine Speed" address="0xCA748" type="Y Axis" elements="9" scaling="RPM"/>
   </table>

redwagon · **Posted:** Wed Jan 20, 2016 4:48 am

Ah, okay, thanks. Makes sense noone bothered posting the whole of the MAF ROM tables in the Carberry thread.

td-d · **Posted:** Wed Jan 20, 2016 6:49 am

Great - please post results - I think if these are the injector timing tables, it would a very useful discovery for those running larger injectors as you noted.

redwagon · **Posted:** Thu Jan 21, 2016 3:40 am

td-d wrote:

Great - please post results - I think if these are the injector timing tables, it would a very useful discovery for those running larger injectors as you noted.

Not real results to speak of, but working on that. I'm working on the assumption (yes) that decreasing the values in Table 1 will shift the injection end point earlier. I'm starting by rescaling the table to suit my loads and boost pressures and reducing by 10* (or as close as possible) and going back to zero (whatever that is) at 2k4rpm, where I'm not having problems. This seems to make things a little better. LTFTs so far seem unaffected. The test will come when I try again to reduce my tip-in enrichment and see if the bogging comes back.

RomRaider

Forum rules

Injection phase / end angle for 32-bit?

Who is online