e40 experts wanted, VE issue

[hymey]

I tuned a VZ Maloo today LS2 with e40 in SD. When I first went mafless in this car it ran 25% lean. This was with stock Open loop tables. I then configured the open loop tables so the pcm would always command 14.7 regardless of IVT or ECT. I did the base settings and began to build a VE table. I found the VE to have an irregular shape. So I looked into the other airflow tables and modified them only to find the other VE modifiers had no affect at all on the VE including the barometric tables . The only table to have affect was charge temperature blending. However still, this did nothing to help.

The VE has values of around 100 at cruise 60-70 idle area at 2800-3200 140%! 3600 140% WAY to rich it blends back to 80% then stays at 80-90% so the whole topend VE is less then low end!!!. I figured it was charge temp blending working off dyn.cylair so I altered it somewhat and then made it a flat 0.100 something I commonly do in the e38. It maxes at 150g/s and reads 1.00 there after meaning it works off ECT for blending but at lower rpm it reads less. What I have found in e38 is this changes the mixture by a factor and does so in the e40 but with the e38 works off SAE.MAF only so it basically doesn't work in SD and goes from the zero reading. Weather or not my flatlining of it has affected the outcome I am unsure but doubt it very much as it reads the same low and high so should change little.

The huge step in the VE has created a misfire at 3200 99kpa I have smoothed the ve off to help but still there slightly. The huge step looks very abnormal and shows something wrong in VE calculations. If the dyn cyl.air is working in place of SAE MAF with charge temp blending then the sudden increase to 1.00 factor may explain something but the jump in VE is over 60%! these errors cannot be fudged with other tables also the need to reduce ve down low and have it highest at peak torque where it should be.

Is there a need to go to a 2 bar COS to fix this bug? as I have looked at all other options to "fudge" the huge error in the VE with no outcome other then charge temp blending which is just for minor IAT ECT blending not for correcting major errors.


cheers

[redhardsupra]

you sure your IFR table isnt completely insane? that would do it easily. Also, I dont know about austrialian models, but all the E40s i've seen have their VE tables in GMVE, not the plain old VE percentages.

[hymey]

good point could easily be an ifr issue i will work on those , but i will also post the tune so ppl can see the crazy ve.

thanks for your advice

[JezzaB]

Post the tune up Joel. Would love to take a look. I havent personally had this problem

Jez

[hymey]

OK I think the issue is to do with the B2001 or B2002 tables Dynamic Air mode max and Dynamic air mode min. Which turns SD on or off basically. When mafless in e38 it doesn't matter what these are set to but with the e40 if I set them to 100 the car will not start. Standard they are around 3200 right where the VE drops off! If I set them to 8000 it doesn't make a difference although I know they are having effect as setting them to 100 doesn't allow the engine t run. As you can see from the VE there is a crazy drop off at around 3000 at this time the pcm is trying to obviously changing some kind of mode casuing a hiccup then drastically altering VE at these values. 100% VE at cruise and then 80 up top, what is going on!

I spoke to swingtan and he agreed there may have been a lack of "fine tuning' with the e40. So far other then the one above I have found two issues at least with this OS. The IVT is all screwed up it actually reads negatives when the engine is warmed up. and positives when cold so this needs work. Also in the injector IAT table B1213 it is meant to use IAT and is labled so but it's link is SAE.ECT?

The above issue with Dynamic cyl air is still present and the pcm is looking elsewhere above 3200 rpm as I have dropped the VE by over 50% so the car commands the correct pulse width above that rpm. I don't know where its looking but I have gone around in circles looking at all the VE modifiers(another thing I should mention is none of these tables under airflow actually work apart from charge temp! not even baro tables). So basically I have gutted the tune inside out and aslo applied the cos. But the Main VE error is an obvious one as this table works just it looks elsewhere for somekind of correction, and before I do anything else with it anyone feel free for suggestions. As I have tried the obviously and simple ones.

[ringram]

That table looks nice.

Have you checked fuel pressure to make sure that isnt messing you around?

[redhardsupra]

That table looks nice.

Have you checked fuel pressure to make sure that isnt messing you around?
yup, that's what i was just thinking. also everything that affects the fuel pressure, like voltage at the pump.

[hymey]

hey guys, thankyou very much for your replies I am taking all this info on board for further testing. fuel pressure is OK. I would like to mention that with the MAF plugged in the issue is non existant. I will go through the logs(will post one) the error is not fuel pump, but the injector pulse width, it goes from 12 back to 8 then back to 12 in a split second. Which causes the hiccup. I have a HPT cable aswell so I will try that to reduce the variables. I don't think the VE looks right personally. It wont look like a LS1 table as the map and rpm columns are ass about but to have peak VE at 3000 then max torque at 4500 where VE is lowest doesn't seam right. If the fuel pressure is an issue it could cause abnormalities in VE and I will double check this as your helpful advice suggests. Basically I am trying to gather as much info and suggestions as I can before starting again with the car and I am hoping to fix it with efilive before going to my HPT cable, In which that scenario I will full flash the factory tune in with efi then pull it out and start again with HPT. Purely only for diagnosis and a suggestion. I hope it is a silly error on my behalf and not a major issue. Will post a log tomorrow, so you can see what is happening.

Cheers

Joel

[hymey]

sorry to add this efi fans but just some hindsight on this matter. This is a friends SD tune in a GTO. and it shows zone limits. These are adjustable in e38 I am unsure if its the cause but changing this changes the shape of the ve drastically. It is limited to 3000rpm by this table in HPT, this isnt available to change with efilive so I will use HPT to test, the tune works great in this car(the GTO) and its untouched but still could be a problem with the car I am working on. You can see the normal looking VE and the table next to it which I would like to alter. Hopefully by using both programs I can find the cause which will inevitably help others if they get stuck like me.

[Blacky]

You can adjust the RPM and MAP zones in EFILive.
B8020 thru B8023
Regards
Paul

[hymey]

Hi blacky the b8020 and b8023 are not in this calibration. Did you check the tune? I can't access these.

cheers

[Blacky]

My bad, I looked up an E38 calibration instead of an E40.
I've sent the thread to Ross to take a look. If they are in the cal he'll know where they are.
Paul

[hymey]

jezzab thinks that I have changed something that shouldn't have been touched. So taking a clean slate forcing SD and fixing the OL tables to 1 lambda from 0.1 other then that leaving all the airflow tables stock and just tuning the VE. Keep it simple will keep you posted.

[ssvolvo]

Check your wideband. Have seen plenty of Botch LSU4 sensors go "narrow".

Specifically if it was left in the exhaust system not plugged in.

Most stock E40 GTO VE tables {B0200} are close right from the get go.

[GMPX]

sorry to add this efi fans but just some hindsight on this matter. This is a friends SD tune in a GTO. and it shows zone limits. These are adjustable in e38 I am unsure if its the cause but changing this changes the shape of the ve drastically. It is limited to 3000rpm by this table in HPT, this isnt available to change with efilive so I will use HPT to test, the tune works great in this car(the GTO) and its untouched but still could be a problem with the car I am working on. You can see the normal looking VE and the table next to it which I would like to alter. Hopefully by using both programs I can find the cause which will inevitably help others if they get stuck like me.Because the E40 has a real VE table this should not be a problem, the E38's zone adjusts do of course have a major impact on the calculated VE.

Cheers,
Ross

[hymey]

OK I think I have changed something I shouldn't have when the car comes back in the morning will try another tune I made up.

SS Volvo. Wideband is working fine. I was plugging the maf back in to see if it would run ok and the afrs read fine. Also this gets used generally on a daily basis and calibrated and I have found it to be fairly accurate. It never gets left in a car unplugged.

I will try the stock table with some of the other initially fuel trims went to +25%!! So it is a bit lean off the bat. I will see what a typical tune does ie just unplugging the maf first then go from there. Keep it simple.

thanks for your reply guys

Joel

[hymey]

Hey guys, this shows its neither spark or afr but simply a glitch somewhere. At 3170 pulse width is 9.4ms exactly 100 rpm later at 3270 pulse width is 19.5ms. At the time MAP is 101kpa so it is looking basically at the same cell in the VE. SAE.TP is also holding 88% which is 100% ETC.TP. Leading up to 3170 the afr gradually goes leaner as pulse width goes from 12 to 9, it starts leaning off from around 2800, at this stage g/cyl goes haywire. I can look at afr all day it didnt matter what timing or fuel I put in the pcm still commanded less pulsewidth at 3000rpm regardless of values in the VE, there after they required reducing by a big margin to meet the required pulsewidth of around 12 ms at WOT. I always look at pulsewidth when tuning it is basically the "carburettor jet" of efi so it doesn't lie.

Hopefully I can sort it tomorrow.

during misfire
http://i139.photobucket.com/albums/q295/princy3010/part1.jpg

directly after (ie only 100 rpm)
http://i139.photobucket.com/albums/q295/princy3010/part2.jpg

The values are increased by a long shot as you can see by the tune in this area. High VE numbers are required at 2500rpm and low by 3500rpm. I tried using the same VE values from bottom to stop @100kpa but it ran very lean at 2000rpm and still had the hiccup at 3170rpm.

Time to start again and hopefully end up with a near factory VVE as SS volvo pointed out and some others. A lot of stuff appears not to work in the e40 but some tables must have a dramatic effect. fingers crossed anyway.

[hymey]

Sorry for late reply guys I fixed this as soon as I got the car back. Basically the open loop tables are populated with 0.10s as you can see from the tune I changed these to 1.00 (ie 14.68) throughout the open loop tables. Once learning the good use of correct OL config in e38 and getting excellent success with those tables like cold and hot start with IVT tables to prevent lean run on hot start mafless(works a treat). I figured I would try this with the e40. And well, it didnt like it setting to 1.00s upset the VVE calcs. I am unsure why but I believe the math formula in OL is taking more effect then it did with the e38. Most people leve these untouched which is basically all i did all back to stock OL including the 0.1s which I thought would not help the SD situation. So very lucky the e40 does not suffer the "lean hot start mafless" like its big brother which then requires comprehensive restructure of the OL tables more so IVT which will richen the mixtures on start up until for IVT temps are reached in a minute or two. Anyway lucky that we dont have to play with IVT in the e40 as the intake valve temp figures are all screwed up they rotate go into the negatives by a long shot reach zero after about -250 deg C! then climb back up again. Weather or not anyone has ventured into the e40 like we have done with the e38 I am unsure. Sometimes best to leave things stock lesson learned but also remember that going deep into the pcm especially e38. going out on a limb has fixed a LOT of major issues many infact some people are unaware.

cheers

Joel

[JezzaB]

Glad you got it sorted. Like any ECM, all little tricks are found from making small little changes from stock and seeing cause and effect. Bet the e38s you didn't jump in and make 200 changes to values. They are a strange little ECM. But I like em!

Good stuff Joel

Jez