KenyonKanye

Hey everyone, so I've been working on getting my car to idle with my setup and have run into some issues. I don't know if it matters but I'll start off with a basic list of mods:
Evo 9 MIVEC head
4G64 block (2.4L)
10.0:1 compression
Speed density (no MAF, GM IAT & Omni 4 bar MAP)
AEM X-series wideband
Stock intake manifold
... I think that's all that really matters for this? Let me know if you need more info.

My overall issue is that the car won't idle/stay running. It's a brand new motor & setup, that is completely different from anything that's ever been in the car so I don't have any previous/base map to go off of. I'm just working on getting it idling, not trying to drive it or anything until idle is dialed in. And once idle is dialed in I'm going to progress extremely slowly in tuning the rest of the map. Not that any of that particularly matters for my current problem.


I'm using ECUFlash Tephramod v7 with the ROM ID 88591715 and EvoScan v2.9

NOW! Let me list out my questions and then I'll explain my thought process.

• Being a 2.4L block and considering Volume is an important part of the speed density equation, is there something besides the MAF Scaling, SD MAP Sensor VE and SD RPM VE tables I should adjust to compensate for that?
• When tuning for idle and using the LiveMap, can I use Load 1Byte at MUT 41? That appears to have mostly sane values, where-as the LoadMUT2Byte at MUT 00 does not have sane values at all..
• How do I log the Manifold Air Temp used by SD? Like the one from the SD MAT Scaling for IAT Scaling Table? It seems like it would be MAT Scaled at MUT 44 but that shows a constant of -10 (the formula is x-40). I am logging IAT at MUT 11 which shows sane values (formula is x-40 again), and it is showing 27/28 deg C which is about right for ambient temp.
• Is MAP Scaled at MUT 45 the kPa that would correspond to the SD Map Sensor VE table?
• What other things should I be looking at in order to get the car to idle?

So, my thoughts currently are assuming I can use Load 1Byte as the Load value, my AFR based on Load over RPM looks like this (this is with me using the throttle to try and keep the car around 1k RPM):


So, the AFRs aren't great with the dramatic fluctuation to 17.8/16.9 but, generally speaking I would think they're good enough that the car at least has enough fuel to idle.

It does seem weird to me that the idle load goes so high, I would think it would be around 40-60 LOAD %, but I also suspect that the higher load value is due to it being 2.4L and the MAF not being scaled for that? I don't know if that is true, and I was messing with the MAF scaling trying to get the load range to move, but that was when I was looking at this based on LoadMUT2Byte, which has values that make no sense. You can see the values in the log file I attached (which is the same log file the screenshot came from). At one point it showed a load of 260 %... Which also leads me to believe that I have that stuff configured wrong in ECUflash, but as far as I've read the only reason to use 2byte load for logging is when doing high boost stuff and your 1byte load goes above 255 %.

You can also see the timing advance in the log file, but it's mostly at 18 degrees with some brief fluctuation as low as 5 degrees and as high as 20 degrees. Also the car was only running for 18 seconds in total, so all of this was very brief.

I was logging knock and it never went above 0. I also did verify that the ISC Steps were changing, they mostly hang out around 75 (I don't think I was logging them in this particular go). The wideband gauge shows within .1 of the value logged from what I've seen while looking at both. Also my TPS reads 17-18% when fully closed and 100% when WOT, but as far as I've seen on here, that is relatively normal?

What else should I be looking at to get my car to idle? My thoughts are that it has to do with the high load values, but I haven't been able to get them to go down, so I don't know what other tables to look at.

Sorry this is such a long post, but I was trying to give as much information as possible about where I'm at and what I've been trying so far. Happy to provide more info if needed though! I really appreciate you taking the time to read this

P.S. While I keep the car running by using the gas pedal to stay around 1k RPM, it sounds pretty good/happy. A little lopey but it has GSC S2 274 cams in it, so that doesn't seem too abnormal.

240Z TwinTurbo

iTrader: (25)

I didn't read everything you wrote because it was a novel, but please read this thread I created.
https://www.evolutionm.net/forums/ec...onversion.html

The VE table does not reference load (1byte or 2byte), but references MAP KPA. This is only true for the VE table as the ignition table and MIVEC do reference load. The correlation between the MAP (KPA) and Load is defined using the KPA:LOAD table. When you make a change to the VE table based on MAP (KPA) it modifies your High Octane Fuel map with the coorelation between MAP(KPA) and Load being defined by our KPA:LOAD table. Use the values I have for your High Octane Fuel map.

Regarding your fueling, if your original tune was based on a 2.0L and now your are 2.4L you need to account for the increased displacement and its affects on fueling by lowering your Injector Scalar by 20%.
2.4/2.0 = 1.20

So take your current injector scaler and multiply it by 0.8 to get the new injector scalar. If that is too rich then raise the injector scalar until you get the idle AFR in a good range. If starting from scratch set your VE table to 100% everywhere.

KenyonKanye

So, I get that VE doesn't reference load, but rather helps determine it. Mostly my question is, at idle a load value of 120% seems like a lot, which in my mind means I should lower the value in the first 2-3 boxes of the SD MAP Sensor VE table, since that is the one that has MAP kPa:LOAD. But, I am also not sure that 120% is necessarily wrong or bad on SD for this, it just seems too high.

Is there a different kPa:LOAD table besides the SD MAP Sensor VE one?

Is MAP Scaled at MUT 45 the kPa that would correspond to the SD Map Sensor VE table? For some reason I initially thought that was logging Hz, but I just looked again and it's kPa and it's logging as 12 the entire time the engine is running. If this isn't the one I should be logging, which one should I log?



Nothing is the same. Motor size, cams, injectors, turbo, pre-turbo intake, intercooler, SD, fuel pump, etc are all different from before. Not to mention the tune I do have from the car previously was on an older version of tephramod that used the smaller resolution tables.

But also, my AFRs for the most part at 1k RPM are 14.7, they're just not very smooth yet, I just don't quite know why. At least to my understanding the ones in my screenshot are relatively close to what they should be for idle, if they're not what should they be? Or just smoother?

240Z TwinTurbo

iTrader: (25)

You DO NOT look at Load when tuning the VE table. You need to make sure your formulas are accurate in EVO SCAN and ECU Flash depending on what MAP sensor you are using and you MUST data log MAP in KPA so you know where to tune the VE Table. The KPA:LOAD table provides the correlation between MAP (KPA) and LOAD. So when you see a MAP (KPA) of 330kpa this corresponds to a given load that you defined in your KPA:LOAD table. This KPA:LOAD table tells the ECU what LOAD column to adjust in your High Octane Fuel MAP to alter fueling based on the MAP(KPA) you are reading. Make sense? If your idle VE is 120% then your values will only be higher under boost. I would adjust your injector scalar until your idle VE numbers are in the 80% to 90% range.

No, why do you ask?

You need to search setting up the MAP sensor based on whatever MAP sensor you are using. I have the OMNI 4bar so it has a specific formula it requires vs a different sensor.

14.7 at idle is what you will get if O2 feedback is active. You need to look at the STFT to see how much is being added to make the motor sit at 14.7. MIVEC will also play a role so you need to isolate your MIVEC map around idle so it isn't alternating between no advance and advance. This goes the same for timing as well. Getting the tables setup correctly with the proper formulas is a PIA and took me 10's of hours reading and help from forum GURU's. If you don't have the sensor setup correctly along with your LOAD formulas you are going to be chasing your tail for a long time.

KenyonKanye

Okay, also just a note I am using the 2D maps, not the 3D ones. Doesn't matter for this but just throwing it out there. So, I need to log MAP kPa, but is there a way to log the LOAD that is calculated based on the VE table? Like, once this is setup correctly, how do I know what LOAD range I am in in-order to tune fuel, ignition and mivec? Isn't that the 1byte load value I was referencing earlier? Or is it something different? Or can you not log this calculated LOAD value?

Quote:

If your idle VE is 120% then your values will only be higher under boost. I would adjust your injector scalar until your idle VE numbers are in the 80% to 90% range.

By injector scalar do you mean the injector scaling table that is cc/min?

Also, how does the injector scaling affect LOAD? My understanding of LOAD is that it is the amount of air *going into* the cylinder, which would be calculated prior to calculating how much fuel to add. Is that incorrect?


Just making sure we're talking about the same things.


I also have an Omni 4bar. If you look at the log I shared in the first post, the 2nd to last column is the Omni 4bar reading in psig, which makes sense if you're thinking about it as vacuum/boost but it doesn't make any sense when I convert the psig value into kPa. Could you tell me what your EvoScan setting/item is for logging the MAP in kPa, like the MUT address and the formula?


In the log I posted, I am also logging the O2 feedback STFT (and the other trims) as well as O2 Sensor voltage. So far all of the trims haven't read anything besides 0. The O2 Sensor voltage is weird though and goes 0-5v instead of 0-1v like a narrowband should. Could you tell me which EvoScan settings/items you're logging for O2 feedback STFT, as well as the O2 Sensor voltage if you happen to be logging that? The values that I'm logging don't really make sense, but there shouldn't be any settings in ECUFlash that affect the MUT locations or formulas for these particular things, as far as I'm aware.

I am also not sure if the engine has run long enough for it to calculate a STFT, in total between me trying to start it and get it running, it's probably only been running for 120 seconds at most.

As for the LOAD formula, which formula specifically are you talking about? The MAP kPa one? Or is there another LOAD formula I need to mess with in EvoScan?

Thanks for your help! I am mostly trying to make sure I am logging everything correctly right now and also that I understand how the tables relate to everything else. I have read all of the sticky'd posts about SD as well as setting up the MAP sensors and things like that. Most of it seems like it's logging right except for the couple of things above.

240Z TwinTurbo

iTrader: (25)

Do you run a GM IAT sensor for the SD? Just noticed you are using 2D vs 3D. I am inclined just to send you my 2D and 3D ROM's so you can use them as a starting point since I already set them up for the Omni 4Bar, GMIAT and Wideband O2 sensor. Send me a PM with your email. You can reference my files to see what addresses and formulas I use or you can simply use my map as a starting point and make changes based on your MIVEC and timing curves.

KenyonKanye

Yea, I am using a GM IAT. Could you also send me your EvoScan config? I'll send you a PM, thanks!

KenyonKanye

So, after messing with it some more, I've made some pretty decent progress. The load values are sane now around 37-45%. The car will now start and "idle" for about 10-15 seconds on it's own consistently.

However, I am still having some trouble with my logging. Oxygen Feedback Trim (STFT) still has never read anything other than 0. And I'm not sure what's going on with the Oxygen Sensor voltage. The original formula was
but that was giving me weird voltages between 0-5+v. I re-did the formula based on 0-255 ADC, and came up with
which actually gives me reasonable voltage values, but I don't know what the ECU thinks it is... I checked the voltage at the ECU pin with a multimeter (I don't have anything to log voltage values) and the voltage appears correct there.

The car is still rich, I've scaled my injectors back ~20% (they're PTE 1000s). I scaled it from 731 cc/min to 597 cc/min but it didn't make any noticeable difference in AFR.. The AFRs are mostly around 12.0 +/- .5 and that was true both at 731 cc/min and 597 cc/min.

The car starts up sounds pretty good and then just decides to drop RPM and shut off, it drops RPM quickly & smoothly and never sounds like it's falling on it's face... It just slowly shuts off (like half a second maybe between idling decent and then off).

So, not sure what to look at for my STFT logging or the car not staying idling.

jozy

Nice. I would have just gone standalone

240Z TwinTurbo

iTrader: (25)

I sent you my 2D SD ROMs along with my tune when I was running PTE 1000cc injectors and I provided more detail in the email. I also sent the EVOSCAN file so don't mess with it and just pop it in the correct folder. You will need to adjust the dwell tables since I run the Bosch P100-T coils. My car is 2.15L vs your 2.40L so you will need to lower the injector scalar so multiply my number by 0.9. My timing is more aggressive so use your timing values. Also, my knock table filters are setup specific to my motor so use your values. MIVEC should be good so use my table. You should not have to do much to get it going since I did most of the leg work. If you blow up your motor because you used my file that is your fault so I take no responsibility for your results so use at your own risk.

XxOLIVERxX

wow! 240z you are on fire!

so what would one have to do with injector scaling when going smaller? say de-stroked to 1.8L? im running sd on a 4g93 na engine

i notice you did not mention the "MAF Size" setting. is there a reason for this? iv been using a definition that suggests setting maf size from 356g/s to 309 for a 2.3L engine?

your point about the KPA:LOAD table seems logical and im wondering if you could explain that some more..
so on a speed density rom the load and ve/kpa are separate but combined somehow?

if we set our kpa:load to 1:1 for every increment in theory the car should act pretty normal right?
by changing the right axis to match the settings defined in the XML when selcting the map sensor to use? i have seen others talking about changing the column on the left. tried this but confused myself and it didnt seem to help at all.. does this table represent a voltage that is tied to the maf or something?

i am trying to understand exactly how the fueling is calculated to work out what im doing wrong. i have tried adding more load and less load and i seem to never get good results

load /2 byte appears normal and stable, but i seem to have an area at :
-3000rpm 13afr no matter what im doing with throttle, regardless of load, kpa etc, always get 13. will be 14.7 bang on just before and just after with a random dive to 13 just at 3000 on the dot.
-4500-4700RPM 13afr that seems cannot be tuned out. fuel map says 12.3, VE table a little high on purpose to try hit 12.0 .
as it passes 4600 AFR map drops to 10.6, but no where in the fuel tables do i have a 10.6 value, when this happens it spikes lean into the 13s then aggressively drops.

i have tried changing maf scaling and smoothing, no dice unless i put stupid figures which will cause spikes when crossing those cells. every other area is looking pretty normal, but those particular spots are messed up.

hope thats not too much of a pain to read, and any help much appreciated

i feel there is definitely some areas of speed density that need further discussion!

KenyonKanye

You would scale the injectors larger. A good starting point would try scaling up by 10%, since your motor has 10% less air-ish.

So, the way the two tables work is that you scale the MAP to correspond with a kPa. After you have that scaled, then you're scaling kPa:Load, this is the Load value the car will try and use for fueling & timing and anything else load-based. It isn't necessarily a 1:1 relationship, since there is still a lot of calculations going on to figure out the true load, but that is how you move it around.

I had weird & inconsistent problems when using a 2D SD ROM, Once I switched to a ROM that was 3D SD, which means there is 1 table for setting your LOAD, and it looks similar to your fueling/timing maps things started working a lot better.

240Z TwinTurbo

iTrader: (25)

Assuming your original scaling based on a given displacement is accurate, then you modify your scaling in proportion to the change in displacement. This will get you roughly where you need to be and you must fine tune from there. So if you have the scaling setup for a 2.4L and then go to a 1.8L your scaling would increase by (1.8/2.4) 25%. This will only get you in the range of where you need to be so you may have to tweak from there.

I explain everything here...
https://www.evolutionm.net/forums/ec...onversion.html

You are scaling the boost sensor MAP (in KPA) to correspond to the Load calculated by the ECU. This relationship is not 1:1 typically so you have to log MAP (KPA) vs LOAD to determine this relationship. I explain this in the link above. On my setup from 0 to 100KPA the relationship is linear, but from 100KPA and up the relationship diverges. So I am 1:1 at 100KPA but the relationship at the top is 320KPA:400LOAD.





Not sure I understand if this is a question. Again, the link I provided above explains it all. The 3D VE will allow you to maintain fuel trims much closer to 0 as opposed to the 2D tune, but obviously takes longer to tune out the VE table. For 3D you still have to define the KPA:LOAD so that does not change. I recommend you set the High Octane Fuel Map to my values and then tune the VE table. You only need to log MAP (KPA), RPM, AFR (WIDEBAND), and STFT, which is enough to tune out the VE. ONLY for the VE Table can you use the MAP (KPA) so for the MIVEC, Ignition Timing, etc you need to log LOAD to determine where you are on the map. Again, this is explained in the link provided.