Ignition Retard Limit
 

Hi Guys,

While looking for this knock stuff I came across the Ignition Retard limit in the ROM.

88590015 and 88580013 (9's???): 0x133e
96530006 and 94170015 (8's???): 0x173e

2 bytes, so uint16 scaling.

Now it defaults to 203 (or 0xCB), I found that by increasing it to 220 I was able to run -15* as opposed to the previous limit of -10*.

Now whoever disasm'd this ROM (Bez???) already found this, so I don't want to take any credit for it :)

Also I don't know the proper scaling yet, but if anyone figures it out let me know :)

Usual deal of test everything at idle - altmaps makes testing this pretty easy, just leave your standard map as-is, and then flick to altmaps with lots of timing retard at idle..

Cheers
D.

That retard limit was referred in this thread as well:

https://www.evolutionm.net/forums/sh...9&postcount=13

90550001 should have the same: 0x173e (uint16)

Have it in my 96530006 def - but not played with it yet (actually do not need such a big retard :)

Unless you're a ricer and like to shoot flames ... like me. :lol:

or want to make a custom ALS using altmap ;)
great find :)

Tephra, can you find the address for rom 88840016 or 17?

mine is set to 255 on rom 90550001 but when logged it still only goes to -10 . regardless of what you put in the map.

In 88840016 and 17 address is 0x133e, but changing the value gives not linear results ad tephra wrote. I never logger less than -15 changing only 0x133e...

try setting 0x2038 the same as 0x133e

0x2038 should default to 0xCB as well..

edit - 2038 not 2308 :)

Howdy dave
according to bez to get the timing to retard to anything below -10 you had to patch a few different places in the rom as there was a lot of code to stop the timing from dropping,
in the rom bez sent me he has full antilag, realtime tuning and speed density setup.
it would be excellent if you could incorporate these mods with your own to make a super patch,
if the above could be patched along with your own mods it would mean the stock ecu would be as powerfull as any aftermarket ecu,
and as bez has stated he is fully willing to share his findings for the benefit of the community ;)

I agree it would be closer but I want more load and rpm columns ;)

Change it to 128 and watch the 36 counts of knock that appear from nowhere ;)

yeah because 128 would actually be limitting the retard so much that you are going into advance there... ie not allowing advance to drop below 10* or whatever...

90550001 uses 0x173e (or 0x173f unit 8) (by the way those address's work for all E7, E8 and E9 roms)

So would 0x2038 workout to be 0x2438 for 90550001 and all other e7 and 8 roms ?

:D er, ive enough knock thanks,,think il sharpen up that chainsaw :lol:

Amen to that!{thumbup}

173e doesn't work for all rom's - my first post has what 173e is in evo9 land :)

I looked at the code for 90550001 and its totally different... doh

Interestingly in 90550011 the 0x173e is 0xBC not 0xCB

Sorry didn't make my post clear.

0x173e unit 16 or 0x173f unit 8 works for all evo 7 and 8 roms.

0x133e unit 16 works for all evo 9 roms.

Has anyone checked to see if changing these numbers actually moves the timing? It may be something to do with crank position referencing? Maybe you are allowed -15 but it may have moved the base timing +5. I know for certain the trigger system itself can support ALOT more retard than -10. I have used ~30ATDC on other ecu's for certain ALS'.

BTW 173E is correct on the 7GTA too.

my car definatley run worse on -15 over -10... I would say its working...

Ok I had a play around with 88570008-v5.10fix-mods on my mates Evo 9 Fq320 today.

Setting 0x133e to 220 did allow -15° to be run, setting to 237 allowed - 18° and 255 allowed it to clip -22° and we had some nice flames :D

Changing 0x2038 seemed to stop things getting more violent so I left it at it's default of 195.

you mean 0x133e for 88570008 yeah?

Yeh thats the 1 (sorry was late at night here when I posted :o

Finally I tried it: it seems that adding to 0x133e and 0x2038 the same value I can get 1/3 of ignition retard for every unit added. For example I added 21 to both (so they are set to 0xE0) and I logged -17 degrees (21/3 = 7). If it is proportional we should get at max (0xFF - 0xCB) / 3 = 17 more retard, so -27 degrees should be possible. Not tested yet, but I don't think we will get more than -20...

Anyone tested something more?

EDIT: wrong thread

I had reason to look at this for a chap today and spotted this thread from the dimm past - thought I would do an update here.

The addressing is valid for:
88570008
88580013
88580014

<scaling name="RetardHex8" units="MAX Retard Degrees - hex" toexpr="x" frexpr="x" format="%02X" min="0" max="260" inc="1" storagetype="uint8" endian="big"/>

<scaling name="RetardLimitMitsu" units="Spark Retard, Degrees ATDC" toexpr="-5+(x-160)*256/728" frexpr="160+(5+x)*728/256" format="%.1f" min="-200" max="255" inc="0.3" storagetype="uint8" endian="big"/>

<scaling name="RetardLimitMrFred" units="Spark Retard, Degrees" toexpr="61-x*89.66/256" frexpr="(61-x)*256/89.66" format="%.1f" min="-28" max="61" inc="0.3" storagetype="uint8" endian="big"/>

<table name="Ignition Spark Retard Limit #1 Degrees - MrFred" category="Timing" address="133F" type="1D" scaling="RetardLimitMrFred"/>
<table name="Ignition Spark Retard Limit #1 Degrees - hex" category="Timing" address="133F" type="1D" scaling="RetardHex8"/>

<table name="Ignition Spark Retard Limit #2 Degrees - MrFred" category="Timing" address="2039" type="1D" scaling="RetardLimitMrFred"/>
<table name="Ignition Spark Retard Limit #2 Degrees - hex" category="Timing" address="2039" type="1D" scaling="RetardHex8"/>


For 88590015 Evo9
<table name="Ignition Spark Retard Limit #1 Degrees - MrFred" category="Timing" address="133F" type="1D" scaling="RetardLimitMrFred"/>
<table name="Ignition Spark Retard Limit #1 Degrees - hex" category="Timing" address="133F" type="1D" scaling="RetardHex8"/>

<table name="Ignition Spark Retard Limit #2 Degrees - MrFred" category="Timing" address="205E" type="1D" scaling="RetardLimitMrFred"/>
<table name="Ignition Spark Retard Limit #2 Degrees - hex" category="Timing" address="205E" type="1D" scaling="RetardHex8"/>



RetardLimitMrFred is an approximation but gives a close result with the degrees showing as negative (-) meaning ATDC.

RetardLimitMitsu is the true scaling with the result in degrees ATDC.

level=3 just limits when ecuflash will display it, ie developer/user/etc...

good work guys....

so what is the lowest anyone has had on the 7 rom (90550001)

.....me want fire:D

-10

so they don't work on 90550001 ?

man so many things to look at.. sigh!!!

More tests done with 88840017. At the end, modifying both 0x133e and 0x2038 with same value with rules as above, I can get no more than -20 degrees of retard. Any try to go below -20 in timing map will become a +50 in log.

I haven't checked the new setting on 90550001, as my car only lived for 18 miles and died again :(

Once it is up and running propperly I will try the new setting, and see if it will retard a little more {thumbup}

wtf - again? what happened?

The oil line to the oil cooler (push fit) came off, and it dumped enough oil and pressure to bring the oil pressure light on :(

oh... so no damage then (hopefully)

not having good luck are u? :)

Won't know until I replace the oil line, and put some oil in it, but Hopefully it will be ok.

Nah think my luck ran out at the end of last year.

Maybe next yr will be better for me :lol:

hopefully its fine :)

Cheers mate, I have my fingers and toes crossed.

Any news on this??
Tried it on the 88840017, as al lupo said, -20 seems to be the max. If you try to go lower, the logs show 50*.

There must be another place in the ecu code where it prevents you to go lower than -20 :(

bump

I suggested a user try 0xffff in the word value to see if it lets him retard more.

Not sure how you'd log it though or if there are other limits.

Didn't work out. In any case, even before reaching 0xff, the ecu allows you to run -20 (at 0xd4 if i remember correctly). Going from 0xd4 to 0xff, the timing limit is still -20.

Bringing this thread back up....

I have the 96530706 rom, and i changed the retard limit upto 255, and it is only retarding -10 degreees.
has anyone else had success getting it lower than -10?

with the launch maps, i was able to build 15psi at 5500 rpm. that compares to 6-8 psi before.

Same here. I'd like to hear an answer to that question.

BEZ had also commented on this on the ative matrix forums

http://www.aktivematrix.com/forum/viewtopic.php?p=4366

i dont see how anyone else got lower than -10..

Tephra,
maybe the rom that you got from BEZ had a few other things modified in it to allow you to run -15 degrees?

With my ecu 96530006, i change adress 0x173e to 230 and no gain in log to -15 or -17 degress, not work this patch for evo 8, other solution please!! Only -10 :crap: i need -17 minimun.

I change also this offset, 0x173f to 255 but... only -10º, not work 0x173f and 0x173e to 255, tephra any solution?? :helpme:

what do you need -17 degrees of timing for?

True Antilag. It will be generally in the range of 15-25* retard to ignite the fuel in the exhaust manifold. On a Euro car doesnt it already have the SAS factory system?

Hello Mellon! more time without speak with you jeje all good friend?? Best Regards this spanish boy jeje

Yes is for real ALS, minimun is -17º, but ecu is only to -10º :crap:
In europe, SAS only un RS version, GSR version without SAS system. :mad: With throttle kicker is the solution and retarding to -17º o r -20º, this ecu is very bad, this limitation is ugly.

ok, I figured that's what you guys were getting at but didn't see it mentioned or just missed the point.

Any solution for delimiting this restriction-ignition-retarding this ecu??

bump

Tephra any info for this problem with retarding in ecu more -10º? Thanks! {thumbup}

The scaling in earlier ecus is: x * 90/256 degrees, referenced to the CAS pulse rising edge (75.77 degrees BTDC)

0xCB(#203) * 90/256 = 71.37 - 75.77 = -4.4 degrees

That gives a limit of -5 degrees not -10, I'm not sure if you have to factor in the 5 degrees of base timing, that could give the expected -10 degrees.


Edit:
This is incorrect scaling, see post 60.

:eek::eek::eek: Translate for humans please no for programers jejeje

Based on that math isnt less more? Or am I looking at it upside down?

See post 60.

I get it now, thanks Ceddy.

Setting the limit to ~180 actually limits timing retard to -5 degrees. Setting it to 203 or more allow -10 degrees. Evo8 EDM 96530706.

So this logic is incorrect.

Good looking out Biggy.

Thats an error on my part. There are like 10 different scalings for Ign Timing the ecu uses. I was looking at the wrong one.

People called this Tim61 in previous ecus.

x = 256 * (61 - degAdv) / 90

degAdv = 61 - (x * 90 / 256)


This would make:
220 = -16.34 degrees
203 = -10.37 degrees
188 = -5.09 degrees
173 = 0.18 degrees


If someone with an Evo ecu could confirm this scaling correct, it would be helpful.

Thats nice Ceddy, looks the goods to me, well done.

My EvoIX Edm to -20 degrees

http://www.youtube.com/watch?v=wuy6c...embedded#at=29

Pug did you use the Scaling I posted in message #60?

And did you change only one address?

88590015 is 133E, stock value 00CB
Not sure if address is universal or different in other ROMs.

Have you tried lower then -20?
The maximum possible retard should be -27, but not sure if there are additional checks that would limit it before that.


Sorry for so many questions, I don't have an Evo to test on.

Thanks.

Using 235 as the retard limit on an EDM IX i've logged -22° before, however it was only for a second or so, before it changed and held -18°

I have not tried to lose timming -20 in the map, I'll try another day ...

My ecu took 88840617 ALS (88840013 stock) max retard 1 and 2 = 233


The explosions are very strong!

did anybody try to change 0x2438 (uint16) and get any success?

96530706

hmmm interesting

any luck with this on 88570008?

if someone guides me in the correct direction i can figure it out .... hopefully!

Bump! {thumbup}

good info

stock ecu has came ALONG way

90550001 timing retard limit to 255 or -29 gives -10 deg logged no lower :(

So are we going to get a table to be able to make this adjustment? If we get a table, I am willing to test the heck out of it. I am not keen on going into the MUT to make change.

hey guys. is there a table for setting that limit? I'm on 94170015 on tephra 5.10 and I can't find it. Do I neet to update to v6 or v7??

I've been sniffing around the subroutine that sets the ignition retard limit, and it appears that in the newer ROMs, there are two ignition retard limit tables that are used for two different operating conditions. Not sure yet on what conditions these are, but here are the two tables for the 8859XX15 and the 9653XX06. They may both need to be set more neg for ignition to retard further under all conditions. Note that the max possible neg value is about -28 deg. Should be easy for people running newer JDM, EDM, and ADM ROMs to find the additional table by comparing the offset between the two addresses.

8859XX15:

<scaling name="IgnTimingECUa" units="deg" toexpr="61-x*90/256" frexpr="(61-x)*256/90" format="%.1f" min="-30" max="50" inc="1" storagetype="uint16" endian="big"/>
<table name="Ignition Retard Limit #1" category="Timing" address="205e" type="1D" level="1" scaling="IgnTimingECUa"/>
<table name="Ignition Retard Limit #2" category="Timing" address="133e" type="1D" level="1" scaling="IgnTimingECUa"/>

9653XX06:

<scaling name="IgnTimingECUa" units="deg" toexpr="61-x*90/256" frexpr="(61-x)*256/90" format="%.1f" min="-30" max="50" inc="1" storagetype="uint16" endian="big"/>
<table name="Ignition Retard Limit #1" category="Timing" address="2438" type="1D" level="1" scaling="IgnTimingECUa"/>
<table name="Ignition Retard Limit #2" category="Timing" address="173e" type="1D" level="1" scaling="IgnTimingECUa"/>

88570008-88570014, 88580013-88580015, 88840016, 88900006

<scaling name="IgnTimingECUa" units="deg" toexpr="61-x*90/256" frexpr="(61-x)*256/90" format="%.1f" min="-30" max="50" inc="1" storagetype="uint16" endian="big"/>
<table name="Ignition Retard Limit #1" category="Timing" address="2038" type="1D" level="1" scaling="IgnTimingECUa"/>
<table name="Ignition Retard Limit #2" category="Timing" address="133e" type="1D" level="1" scaling="IgnTimingECUa"/>


9417XXXX has only one ignition retard limit table:

<scaling name="IgnTimingECUa" units="deg" toexpr="61-x*90/256" frexpr="(61-x)*256/90" format="%.1f" min="-30" max="50" inc="1" storagetype="uint16" endian="big"/>
<table name="Ignition Retard Limit #2" category="Timing" address="173e" type="1D" level="1" scaling="IgnTimingECUa"/>

great work mrfred....

now anyone can help me find it for 8857?

What address do you have now?

Very nice mrfred,

Seems that there is no difference between JDM, EDM or USDM for the tables.

I just found only one RA map with -28.6 for the second table (where the others have -10.4).

mrfred any chance you could look at 90550001 for a second table have the table at 173e , :helpme:

Awesome! I'll try this out on 9653xxx6. Thanks mrfred, you are ON FIRE lately! :D

found this earlier in the thread

That's an older ROM. I suspect it has only one table.

can we have -28 on the second table?

-28.6 is the limit I think for the second table

That's the limit for both tables, although there may be other factors down the line that prevent timing from reaching -28.6 degrees.

the first table says 25.8, is that correct? And the second says -15 now?

Both are -10.4 deg on stock 8859 and 9653 ROMs. Don't know what ROM you have.

address? i dont think anyone has even started to look at the limit for this rom...

where do i start?

sorry for being a noob :)

What's your exact ROM ID?

With the second table at -28.6, my car wanted to rev uncontrollably upon cold startup. I turned it down to -20.6 and its more normal. First is at -28.x.

Did you by chance log timing?:D

What would be the table adress for 88580714 V7, Original ROM 88840016 on EDM Evo, thanks...

I have rom 88570008.

Mine says 28 for the 1st, -10 for the 2nd.

Should I change the 28 to -15?

Updated my earlier post to cover most all newer non-USDM ROMs.

thanks MrFred...

88570008 - jdm ix

Added to my earlier post.

9417XXXX has only one ignition retard limit table:

<scaling name="IgnTimingECUa" units="deg" toexpr="61-x*90/256" frexpr="(61-x)*256/90" format="%.1f" min="-30" max="50" inc="1" storagetype="uint16" endian="big"/>
<table name="Ignition Retard Limit #2" category="Timing" address="173e" type="1D" level="1" scaling="IgnTimingECUa"/>[/quote]

I tried it out today. Generaly no retard below -10. In the last log I got -12.7 in only one cell. Attaching my maps and two logs. ROM is 94170015.

Do we need to richen the fuel cells or can we let 14.7 AFR ?

This will never work because we will need a special separate map for TPS when the ALS will be activated (jacked open throttle).

The TPS Retard value from this so called TPS retard map must be a retard value from the base ignition map.

With the tables that we have at this moment we can`t run ALS properly.

Only tephra maybe he can give birth to this options using a tephra mod patch for the cars with ALS system available....

The TPS Retard parameter must be taken from the TPS Retard map.

Probably other criterion must be met in order to make the ALS working.

When the ALS will be active a different closed throttle condition is allowed for.

With the ALS program enabled the ignition timing must be modified by the TPS Retard map.

Some other maps must be available in order to run ALS (Spark Map Limit and Fuel Limit).

A proportion of injection events and of sparks will be disabled by the limiter
functions (limits will be taken from the FUEL and Spark limits maps).

The Spark Limit map dictates the level of ignition events removed vs. engine speed and throttle position. The Fuel limit map dictates the level of fuel injection events removed vs. engine speed and throttle position.

The TPS Retard map will dictate the level of ignition retard to be applied vs. engine speed and throttle position.

Probably the OEM ECU can be programmed to run a proper Anti Lag setup but this must be made by adding a special patch to the rom.