2.3 vs 3.5 software - engine running rough on 3.5 good on2.3

[driftshop]

e16 with 800cc map.e23

e16 converted to e23 running well

(24.54 KiB) Downloaded 1040 times

Hey Guys,

I have been using the 3.5 software for my 1JZ project and I have been chasing my tail trying to get it to run properly and idle well without fouling plugs and causing issues.

Scenario - cold or hot start - if i rev it to 2500rpm it will run very rich (in the 10 AFR) and start misfiring. It clears up after 5 seconds of flaring the throttle only to run rough at idle and stall.

Temporary solution - I've downgraded from 3.5 to 2.3 and re-mapped the fuel map and some parts of the ignition map and it runs as if it was an OEM ecu without any issues. I used the 1.6 (DOS) ECU file to upgrade to 2.3 and it runs excellent. Once i upgrade the 1.6 -> 2.3 -> 3.5 I get rough idle and overfueling issues as described.

I would like to have the best of both worlds - latest 3.5 software with the reliability of the 2.3 map. Can anyone help me with what I can do to make it run properly in 3.5 software?

Please see the files attached. The fuel maps may not be the same, however I think the secret to the problem is there is missing data from the 2.3 map that is not transposed to the 3.5 map due to their different values and setup.

[driftshop]

"Your message contains too few characters."

[xpro]

mine is the same, runs sweet on 2.3 and when i switch over to 3.5 my car runs like a pig. Any ideas please?
i tought when upgrading the software all the settings are automaticaly upgraded but stay the same??
thanks

[Scott@FP]

Set up your firing order and inj sequencing and try again.

Oh, and the Ign AT and ET comps are very strange.

[J&R]

Driftshop,
When running 2.3 or lower it askes you to put in a value of of 0.0 when running MSD Or Simimlar types of ignition systems. When doing this the ecu is put into a default digital square wave pulse at 10ms in duration when seen on an oscilloscope.

When upgrading to 3.3 or higer you Have to review this and it sets the Dwell to 1ms and your engine runs like a Dog.

Set your Dwell to 10ms or if your not sure of 10ms, 5ms does suffice, but to be the same as it was previously was in the square wave of 10ms and see how it goes.

As it says in the read me files Please review every setting as going from v2.3 to v3.3 is a huge change in software and the two were never ment to be compatible. But were made somewhat compatible for simplicity sake.

See how you go.

x pro read above and change your settings accordingly

[driftshop]

Thanks for the feedback guys

Scott - thanks for picking that up. What should they look like can you please provide an example? They were on 1.6 version ECU software file that was upgraded to 2.3 and the values are in % and not degrees. So at 50 AT the timing will be retarded by 10% for that value (around 1.8 deg for at 250kpa). Does this sound ok?

J&R, thanks for the tip. Wayne at Meridian motorsport did recommend to write down all the values and change the parameters. I will spend some time on it and see how it goes. My ignition is a Motec CDI running in wasted spark mode. It made 382rwkw with the 3.5 software installed, but it fouled plugs on hot and cold start and idle revving until the idea of running the 2.3 software came to me before I went to sleep one night. I have not fouled any plugs and it runs very well, but I'm lacking the functions of the 3.5 setup that I need to run my non-motec analogue sensors. I've got a feeling that you are right on the money with your suggestion

I will RTFM

--
----
Costa

[Scott@FP]

Thanks for the feedback guys

Scott - thanks for picking that up. What should they look like can you please provide an example? They were on 1.6 version ECU software file that was upgraded to 2.3 and the values are in % and not degrees. So at 50 AT the timing will be retarded by 10% for that value (around 1.8 deg for at 250kpa). Does this sound ok?

Costa

Good point, I didn't notice the ign trims = % not degrees. The -30 jumped out at me and I thought "-30 degrees?". -30% shouldn't make it run that bad, -30 degrees would.

Other items that seemed odd were the IJPU value of 10 ,seems high for a MAP comp'ed turbo car with a single set of injectors. v 3.x allows values of up to 400 in the fuel table so IJPU can be reduced further than previous versions, but can't remember which version had the increase to 400, if it was v2 to v3 or prior. Regardless if your table values and IJPU were the same it should run fine.

Typical IJPU setup for a 500ish cc/min injector would be 3 or 4, 1000-1200 2 or 3 on a 1:1 fuel MAP Comp and stay below 400 max in the table. No reason you can't use IJPU =10 but you gain main fuel table resolution with the smaller IJPU.

Fuel Decel may be messing with you, log fuel decel, if its active while running normally set all fuel decel parameters to zero. Accel/decel were one of the things changed from v2 to v3.

Firing order. Usually 1-5-3-6-2-4 (15 too young, 36 too old, 24 juuuust right ) for inline 6's, but don't know for sure on your engine.


There's something somewhere messing with ya, look deep enough you'll find it.

[Dragon]

Also, you could have a look at the Y axis on Cranking and Post Start Comp tables.
Using "F<ET>" or just ET as Y axis may be easier and more conventional.

Are you sure that Injector Current is set correctly? It may make a difference.

[driftshop]

Ok I changed the dwell to 10 and it is working well - not as good as 2.3 but certainly not fouling the plugs as much. I'm getting richer mixtures across the board with the current fuel map which is odd.

Other items that seemed odd were the IJPU value of 10 ,seems high for a MAP comp'ed turbo car with a single set of injectors. v 3.x allows values of up to 400 in the fuel table so IJPU can be reduced further than previous versions, but can't remember which version had the increase to 400, if it was v2 to v3 or prior. Regardless if your table values and IJPU were the same it should run fine.

Typical IJPU setup for a 500ish cc/min injector would be 3 or 4, 1000-1200 2 or 3 on a 1:1 fuel MAP Comp and stay below 400 max in the table. No reason you can't use IJPU =10 but you gain main fuel table resolution with the smaller IJPU.

IJPU was 10 from the previous map that ran with 550cc injectors on the 1.6 DOS software (lol), however your suggestion will help with running 1600cc+ injectors when I choose to run E85 after I make a custom top feed rail and plenum setup 4x the resolution would be great for the larger injectors!

Fuel Decel may be messing with you, log fuel decel, if its active while running normally set all fuel decel parameters to zero. Accel/decel were one of the things changed from v2 to v3.

I'll check this out over the weekend. I've currently upgraded to 3.5 and ran your 10ms dwell suggestion. It is running better, but the map is about .5-.75 AFR richer on the exact same fuel map as on the 2.3. I have not checked other parameters for their consistancy. I do think there is more fuel being added on accell/decell in 3.5.

Firing order. Usually 1-5-3-6-2-4 (15 too young, 36 too old, 24 juuuust right ) for inline 6's, but don't know for sure on your engine.

There's something somewhere messing with ya, look deep enough you'll find it.

Good way to remember an inline 6 firing order 1JZ./2JZ is 15 36 24. Its kinda fun chasing this hiccup, especially when progress is being made and I have a 2.3 map that works flawlessly as a backup.

Dragon: The injector current is the same for each revision, but maybe the 3.5 software is more sensitive? I'll play around with the cranking and post start comp tables. Think that in the translation the values were very high causing a 10+ AFR mixture that would not burn properly with the previous 1ms dwell.


I appreciate the feedback, bit by bit the upgrade will be sorted out. Every problem I fix today is one less on race day!

[RossB]

There is an odd difference between your V2.3 and >3.5 configs. Did you change the axis setup for the Fuel individual culinder table? It is not the same in the 2 configs. It could be worth converting the file again and putting your changes in to a newly converted file. Also make sure that you are using the latest version of V3.5 ECU Manager. The only thing that shouldn't convert up properly are the PID parameters for idle control.