PDA

View Full Version : E83 specific pids



Tre-Cool
April 30th, 2014, 12:45 PM
Guys, is it possible to get some e83 specific pids, just like the e38/e78 etc ones.

Ideally atleast the following pids:
Air Flow Grams/Cyl - using the sae.maf and calculated g/cyl is not accurate
Air/Fuel Ratio Commanded - currently using sae.lambda which seems to work, however im getting enrichment that is much higher than anywhere in the tune. Also possible there are enrichment tables that are missing from the calz.
Intake Valve Temp - doesn't exist for e83 controller yet


I'll might come back with more once I start playing around with the car, but these are my need to have's. :-)

GMPX
April 30th, 2014, 02:56 PM
Actually no I'm afraid to say. The E83 ECM has an encrypted or compressed OS (not sure which it is) so we can't decompile the code to locate the memory address's for such PID's.

Tre-Cool
April 30th, 2014, 03:13 PM
is there any other pid I can use get the equivalency ratio?

found it weird that they have set 14.3 as the lambda value too

I'm happy to help in any testing you might want to try

GMPX
April 30th, 2014, 04:25 PM
SAE.LAMBDA should work, actually that is probably your only option.

Tre-Cool
May 6th, 2014, 12:30 AM
Okay, been nearly a week with the new car and left it stock for the first 3 days to log it in factory form.

Been running it on united's P100 mix (98+10% ethanol = 100 octane they reakon)

Stock they supposably run 10psi, it kinda did at first then it slowly wound it's way up to around 12-13. I'm guessing because it had been ran on cheap 91-95 fuel. Car was 2nd hand with just under 4k km's on the clock.

Anyway, I had a small play with it on the road and then onto the dyno, where I gained a solid 10-12% increase in torque over the entire rpm range bar the top end (which I believe is being restricted by torque management)
http://www.torqueup.com.au/downloads/Cruze/cruze-torque.jpg

On the road with the wideband shoved up the bum of the exhaust, actual is pretty dam close to commanded enrichment, however stock they run them at fat as .6 lambda!!! (as seen on the dyno graph)

Unfortunately my dyno is setup for rwd so I didn't have the cooling fan blowing for any of the runs but I allowed a fair bit of time to cool down between tunes.
http://www.torqueup.com.au/downloads/Cruze/cruze-lambda.jpg

Now, compared to the 1.4l turbo, they don't seem to hamper the tune too much in regards to making power initially, but now I feel i'm hitting a missing torque limit or 2 I'm stuck at making a consistent 280~282nm. Which might explain why the stock tune matches the max hp figure albeit at a later rpm.

What's interesting to note is that I haven't changed the duty cycle of the wastegate, yet im making 120-123kpa (17psi) of boost. I've only made a slight adjustment to the commanded power tables and I've played with it in extremes and settled it back down to close to stock. I've found that if you put in a large number in say the 80-100 tps% ranges, it essentially puts a dead spot/time delay at low rpm.
http://www.torqueup.com.au/downloads/Cruze/cruze-throttle.jpg

Now I've taken a look at an auto tune for one of these cars albeit in hptuners and they seem to have a whole bunch more torque tables which co-incidently have max torque figures around the numbers im hitting.

I've tried using leaving B7015 Enabled and putting in say 400nm for B7016 & I've also set it disabled and having the torque limit normal set to 2000 and lower to 1000nm(default) and there is no difference. in fact because it's not over-boosting I don't think either is doing anything.

So any chance of getting someone to take a look at the available torque management for the 12654173.calz

I can email in logs/tunes if needed.

joecar
May 6th, 2014, 02:26 AM
So did SAE.LAMBDA show the correct lambda (matching the commanded fueling tables)...?

Tre-Cool
May 6th, 2014, 02:38 AM
So did SAE.LAMBDA show the correct lambda (matching the commanded fueling tables)...?



yeah works fine, has from the start.

Tre-Cool
May 23rd, 2014, 11:55 PM
For anyone else in the future, there doesn't appear to be a torque limit. My "limit" is more to do with how much boost and therefore power the motor is making with my timing map & boost at 223kpa (minus 100kpa for baro) 280-296nm is pretty much where I get and fade off in the top end.

Just trying to get it to ramp the boost up a little more now is the hard part.

Anyway i'd suggest adding the following into your calc_pids file.

*CLC-00-2045
gcyl 0 1 0.2 "({GM.AIRPERSEC}*30)/{SAE.RPM}"

*CLC-00-2046
gcyl 0 1 0.2 "({SAE.MAF}*30)/{SAE.RPM}"

*CLC-00-2047
factor 0 3 .3 "{SAE.MAP.kPa}/{SAE.BARO.kPa}"

The below goes into the PRN area at the bottom of the file.

CALC.CYL_AIR F6F7 CLC-00-2045 "gcyl" "E83 ECM" "Calculated 4 Cylinder Air"
CALC.CYL_MAFAIR F6F7 CLC-00-2046 "gcyl" "E83 ECM" "Calculated 4 Cylinder Air Maf"
CALC.MAP-BARO F6F8 CLC-00-2047 factor "E83 ECM" "Map-Baro"

redhardsupra
June 10th, 2014, 04:52 AM
Actually no I'm afraid to say. The E83 ECM has an encrypted or compressed OS (not sure which it is) so we can't decompile the code to locate the memory address's for such PID's.
Actually you should be able to tell the difference between the two by measuring entropy on a fixed chunk of bitstream. If you send me an image, I can probably figure it out.

litreddevil
July 28th, 2014, 12:22 PM
why cant find sae.lambda in my pid's

GMPX
July 28th, 2014, 01:10 PM
Actually you should be able to tell the difference between the two by measuring entropy on a fixed chunk of bitstream. If you send me an image, I can probably figure it out.
Actually turns out we just needed to update IDA, they are using VLE on the CPU :Throwup:, not going to go searching for DMA PID data in that code mess.

Tre-Cool
July 28th, 2014, 04:34 PM
What's the chances if any of being able to find out if there is any calls to a specific pcm pin for the ethanol sensor or add them in? (there's a ton of empty pins on the ecu, but I imagine they need to be connected to the specific inputs within the ecu board like the maf?)

Efilive is also missing the capability to turn on the ethanol sensor completely (not that it matters at the moment), using hpt I can turn it on but the ecu then defaults to 100% ethanol and 255hz on the fuel composition input. I've tried blindly using a bench harness setup on my spare ecu and flex sensor on all the unassigned pins and had no change. Mind you I thought it was odd that even when I put the pin to the maf sensor signal pin it still didn't recognise or pickup the frequency.

I thought Perhaps because the ethanol sensor hz is only between 50-150 vs 1800+ for the maf it can't see it. but surely it would still read the lower values? I did just realize that I may need to drop the enable/fail frequencies down though.

Tre-Cool
July 28th, 2014, 05:10 PM
Also, there is a 223kpa max boost limit from the factory, Efilive and HPT have both incorrectly labelled the X axis based on ECT when in fact it's IAT based. IAT2 to be exact.

I've confirmed this by putting say 200 in the 0-40 ect range and left 60c as 250kpa. with 17 degree intake temps and 104ect max boost was 200kpa. change it so it's 250kpa for the lower cells and that's exactly what I get. well a little more but the map sensor max's out over 255kpa. :-)

The other thing I have learnt because I have started using the hpt cable exclusivily on this ecu due to the options available is in regards to the Fan's.

I have worked out that the 2 rows are actually independent of each other based on whether or not the radiator sensor option is disabled/enabled. (HPT has this an switch, so I guess efi could remove the 90 portion)

For example, Stock E83 calibration has radiator sensor disabled and thus desired fan speeds are controlled by the 105 column. but if you enable the radiator sensor switch the 95 column is the desired fan speed and 105 is ignored.

What's weird with the configuration on these Cruze tunes is the motors are fitted with an electric thermostat and DCVCP, yet there appears to be a phaser only installed on the inlet camshaft and it's not turned on. I haven't played around with the inlet cam phasing yet, but it would definitely be nice to be able to control the thermostat opening/closing.

I also created a cax file to disable the torque model bullshit they use on the pedal too which makes the car respond a LOT better. I'm now making over 150whp on pump 98 fuel with the bigger injectors, i'll be going back to stock ones this weekend again to see how they will fair with the increased boost. Probably not well I have a feeling. haha

http://www.torqueup.com.au/wp-content/uploads/2014/07/cruze-filter-power-1024x505.jpg

GMPX
July 28th, 2014, 11:36 PM
What's the chances if any of being able to find out if there is any calls to a specific pcm pin for the ethanol sensor or add them in?
Next to impossible, you would need a PCB circuit diagram to know where it ends up inside the ECM (CPU or other chip), it cannot be worked out from the code. With all the other ECM's we had the luxury of having a car on each ECM type using the sensor.