Fuel/Spark tuning for LH 2.4/EZK with TunerPro!

[Hagar17]

cripes ditch win2k already!

[beepee]

Originally Posted by klr142

I'm not sure(I'm mostly a Mac person). It says Tunerpro.exe is not a valid Win32 application.

You've got something seriously wrong with your tunerpro install. Uninstall, re-download, and try again.

Bump: added Ostrich install.

[klr142]

Originally Posted by Hagar17

cripes ditch win2k already!

Did you miss the part where I said the laptop is from 1998? 233mhz processor and 4gb hard drive can only go so far, lol.

Originally Posted by beepee

You've got something seriously wrong with your tunerpro install. Uninstall, re-download, and try again.

Bump: added Ostrich install.

I'll try it again when I'm back home.

And man, you're all over this!!!

[2fast4u]

ok lemme start the scary questions:

When I double click on RPM limit it is set at 39. Is that just a value for 6500ish rpms, or is it saying my rev limit is 39 rpms?


nvm, holding the mouse over it explained it.

[klr142]

Open up the rpm table that is included in beepee's "kit". It tells you what that number corresponds to in rpm.

EDIT: I tried a fresh install of TunerPro v5, same message. :( I won't be tuning on this laptop anyway, but yeah.

[JW240]

Originally Posted by ipdown

JW, may be speed rating is wrong? Most memories have access time in ns (e.g. 27C256-70, that 70 is access time), if this chip has slower access time LH will have trouble reading from it.

The ones i have are 70 ns access time, i'll try your suggestions.

[beepee]

Originally Posted by 2fast4u

ok lemme start the scary questions:

When I double click on RPM limit it is set at 39. Is that just a value for 6500ish rpms, or is it saying my rev limit is 39 rpms?


nvm, holding the mouse over it explained it.

I've added explinations for most of the values in the comment field; the XDF for EZK is still pretty rough (though it is fairly complete as far as I can tell) so I will be working on that next.

Anyone with WinOLS, or who happens to have an alias of ipdown or blabla, who would like to add tables, please step up! :-)

Also my RPM equation is an estimate; the values are not exact but they _should_ be within 100 RPM of reality. I'm not even going to get into how I determined that equation but I am sure that there is a simpler/better one.

What sucks is that I can't figure out how to make TunerPro do conversions that aren't simple math. That's why the RPM table is included; every time I try and tell it that RPM = 60*e^(0.0215834714*(255-x)) it gets wonky and doesn't even come close to a correct value.... ugh.

[beepee]

Originally Posted by JW240

The ones i have are 70 ns access time, i'll try your suggestions.

I think the highest I have that works was less than 20, so...

*opens google calculator*

9xx LH2.4 runs at 12.5 Mhz

(1/12 500 000) second = 80 nanoseconds

I guess 70 should work, but it is cutting it close. I'd shoot for 40 or less.

I am told that the 5xx ECU only runs at 6Mhz so it might work just fine if you have one of those.

[linuxman51]

hm, what's the deal with the timing numbers? they don't seem to make sense...

[beepee]

Originally Posted by linuxman51

hm, what's the deal with the timing numbers? they don't seem to make sense...

Could you give me more detail? I haven't played with my EZK yet; I'm convinced I'll asplode my motor.

Which table are you looking at, and how is it 'off' ?

[linuxman51]

all of them. shows max advance anywhere of like 17 degrees. I'm willing to accept conservative timing, but even that is excessive ;)

[beepee]

Originally Posted by linuxman51

all of them. shows max advance anywhere of like 17 degrees. I'm willing to accept conservative timing, but even that is excessive ;)

I feel ya. If the EZK is anything like the LH box, the tables are based on clock speed and tooth count... so I'll see if I can find the clock speed.

I have no idea where the X/8.7 formula came from; maybe one of the SAAB boys put it in there. It was in the XDF when I downloaded it.

And I really need to step up and get some ASM skills.

[linuxman51]

If I can usurp my buddy's ostrich I might have a little more to add to this today or tomorrow.

[beepee]

Originally Posted by linuxman51

If I can usurp my buddy's ostrich I might have a little more to add to this today or tomorrow.

That would be excellent. If you get a chance before I do, crack open your EZK see if you can get the speed of the crystal on the board; it will probably help a lot with the maths... I'm at work right now and the only picture of it I can find has a test mark obscuring the numbers...

[linuxman51]

Originally Posted by beepee

That would be excellent. If you get a chance before I do, crack open your EZK see if you can get the speed of the crystal on the board; it will probably help a lot with the maths... I'm at work right now and the only picture of it I can find has a test mark obscuring the numbers...

I posted all that info years and years ago, I think I had the crystal speed for ezk in there as well. not all of the info is correct but the hardware stuff was pulled basically from datasheets off the part numbers from the car(s)

[beepee]

Found it!

http://forums.turbobricks.com/archiv...p/t-41499.html

[JW240]

Originally Posted by beepee

I think the highest I have that works was less than 20, so...

*opens google calculator*

9xx LH2.4 runs at 12.5 Mhz

(1/12 500 000) second = 80 nanoseconds

I guess 70 should work, but it is cutting it close. I'd shoot for 40 or less.

I am told that the 5xx ECU only runs at 6Mhz so it might work just fine if you have one of those.

iirc the stock chips are -250, so 250 ns. the 70 ns should be more than fast enough. I'll try the different way of programming this weekend, see if it works. should be cool for E85/gasoline. the 27SF512 is cheap on ebay, directly from china, ~32$ shipped for a 13 pack.

[ipdown]

All my 9xx ecus have 12 MHz cristals.
LH 2.4 timer is running at 500 KHz (in 950,962,967 ecus which I have disassembled),
rpm signal from EZK is two pulses per revolution.

The following formula is from motronic, but seems to work for the EZK ignition tables:
X * -0.375 + 60

Reg. rpm conversion approximation - 60*e^(0.0215834714*(255-x)) - would be great to
find the exact formula. This is what LH does when converting timer stamps to convert them to one byte values (this is 8051 assembly):
Note there is map lookup in assembly code, the map content can be seen below in C code

    LOOP:

    ;; R1 = 6

    ;; R2 = DELTA_CRC_HI - low byte of (last rpm pulse time - new rpm pulse time, in 500Khz)

    ;; A  = DELTA_CRC_LO - high byte



        DEC R1

        RLC A

        XCH A,R2



        RLC A

        XCH A,R2

        JNC LOOP



        MOV A,R2

        RRC A

        RRC A

        ANL A,#0x3F

        JNC DONE



        INC A

        JNB ACC.6,DONE



        INC R1

        CJNE R1,#0x08,DONE

        SJMP TOO_LOW_RPM



    DONE:

         MOV DPTR,#1x64_MAP

         MOVC A,@A+DPTR

         XCH A,R1

         SWAP A

         RL  A

         ORL A,R1 <-- done

Very roughly translated to C this looks like this:

    int translate_RPM(int real_rpm)

    {

       int map[64] =

       {

           0,  1,  1,  2,  3,  3,  4,  5, 

           5,  6,  7,  7,  8,  9,  9, 10,

          10, 11, 11, 12, 13, 13, 14, 14,   

          15, 15, 16, 16, 17, 17, 18, 18,

          19, 19, 20, 20, 21, 21, 22, 22,

          22, 23, 23, 24, 24, 25, 25, 25,

          26, 26, 27, 27, 27, 28, 28, 29,

          29, 29, 30, 30, 31, 31, 31, 31

       };



       // timer T1 is clocked with 500 KHz,

       // RPM signal from EZK is doubled

       int v = ((500000) / ((real_rpm) / 30));



       for (int i = 6; v < 65536; --i)

          v *= 2;



       v /= 1024;

       return map[ (v % 64) ] + (i * 32);

    }

This looked to me as some weird logarithmic function, and is very similar to one found here:
http://www.dattalo.com/technical/the...7s%20Algorithm

[linuxman51]

Originally Posted by beepee

Found it!

http://forums.turbobricks.com/archiv...p/t-41499.html

wooooow that was a long time ago. I'll re-verify EZK megahurts at lunch.

[linuxman51]

Originally Posted by ipdown

All my 9xx ecus have 12 MHz cristals.
LH 2.4 timer is running at 500 KHz (in 950,962,967 ecus which I have disassembled),
rpm signal from EZK is two pulses per revolution.

The following formula is from motronic, but seems to work for the EZK ignition tables:
X * -0.375 + 60

editing the data in the xdf, this looks about right

for those doing it through TP, the equation is ((X * .375) * -1) + 60. for whatever reason, multiplying by a -.375 made everything zero.

[beepee]

Originally Posted by linuxman51

editing the data in the xdf, this looks about right

for those doing it through TP, the equation is ((X * .375) * -1) + 60. for whatever reason, multiplying by a -.375 made everything zero.

Change it to -0.375 and it works. TP's equation parser sucks

Whoa ipdown, thanks for that! I am glad my approximation formula is useful. Looks like we will never have a straightforward formula for RPM though.

I have just re-polished the EZK XDF and posted it, check it out. If you all approve, i'll update the zip...

http://dl.dropbox.com/u/2074311/Volv...-beepee-v2.xdf

Let me know what you think and if I am moving in the right direction... Thanks.

[linuxman51]

Originally Posted by beepee

Change it to -0.375 and it works. TP's equation parser sucks

Whoa ipdown, thanks for that! I am glad my approximation formula is useful. Looks like we will never have a straightforward formula for RPM though.

I have just re-polished the EZK XDF and posted it, check it out. If you all approve, i'll update the zip...

http://dl.dropbox.com/u/2074311/Volv...-beepee-v2.xdf

Let me know what you think and if I am moving in the right direction... Thanks.

Man it's your zip and xdf, you don't need our approval to update your stuff ;)

Checked the crystal on a 153(i think) ezk (2.4 ezk...) and it's 8 mhz.

grabbed my moates burner at lunch, bouta get that software spun up on the new laptop

[beepee]

Originally Posted by linuxman51

Man it's your zip and xdf, you don't need our approval to update your stuff ;)

Checked the crystal on a 153(i think) ezk (2.4 ezk...) and it's 8 mhz.

grabbed my moates burner at lunch, bouta get that software spun up on the new laptop

Well I like feedback, if what I am doing is only useful to me, then what's the point??

I'll update stuff now.

[beepee]

Just added a huge step-by-step to the OP. Let me know if it helps anyone; I went from memory for lots of it so I am sure there will be corrections...

Time to finally install the crank sensor UPS says is sitting on my porch, yeehaw!

[2fast4u]

Amazing. Seriously man. Awesome stuff. Truly set up for idiots. Can't thank you enough.