ECU logging post TMIC/TD05 install

[BSOD2600]

Ever since I installed my 07 TMIC and TD05, I've wondered if the various stock components are able to handle modifications. I've kept the boost level below fuel cut. This was done so I thought I wouldn't exceed the other stock parts and wait for the Stage 2 Revtronix chip to raise the fuel cut level in software.

Last week, I got the EvoScan USB subaru cable and software. To my dismay, it doesn't have the correct parameters programmed in the configuration file. Another member pieced together a starter file on the evoscan thread. While it was a good start, the conversions were mostly wrong. I looked through vrg3's b10scan source code and extrapolated most of the conversions for the various parameters. I also was able to obtain the custom datalogging software Revtronix uses -- which also works with the EvoScan cable! I also have a LC-1 Wideband O2 sensor hooked up (revscan is unable to connect to it for now).

After procuring a laptop, I was able to do a few WOT runs using both software logging tools. Here are the results plotted.

Round 1
Stock ECU, injectors, MAF, MAP. MBC set to ~11.5 PSI.

EvoScan








RevScan









EvoScanLogs.xlsx
RevScanLogs.xlsx

Notes

• Speeds are via the Vehicle Speed Sensor (VSS). Not entirely sure they're 100% accurate anymore due to my 4.11 5MT and 16" wheels.
• I purposely cut off the top/bottom of the graphs to concentrate on the small area where the numbers really matter during WOT. Download the Excel files for the full.
• For the most part, the data logged between evo and revscan seem to match up, so I'm fairly sure the data conversions are correct.
• No idea how revscan is obtaining/calculating the g/s measurement, but Mike has stated that ~155 g/s is the max of the hitachi MAF -- I'm pretty close to that.
• I know the hitachi MAF maxes out at 5V -- mine on many occasions is reading 4.9V, so I think I'm basically there.
• It would appear that around 18-19ms injector duty cycle for stock injectors is 'static', since mine flat line.
• 4th and 5th during WOT near the end of my testing, I was feeling 'pulsing' in the cars power. Looking at the PSI line in the graph, I suppose that's the culprit due to wastegate fluttering?

Conclusion
The TMIC and TD05 at ~11.5 PSI max out the stock components.

[Fkyx]

By "stock components" are you referring to injectors or ECU?

[gijonas]

What he said.

[wtdash]

By "stock components" are you referring to injectors or ECU?

MAF and injectors @ least, eh? You might look @ your plug gap, too.

[Adam West]

What is IPW?

So get the Jecs MAF in and try again?

[BSOD2600]

Yes stock: MAF, injectors, and MAP.

IPW = Injector Pulse width. Will of course throw on the JECS MAF, 440cc injectors and Revtronix chip here in a few weeks . Then will redo the benchmarks again and see where I'm at.

[Fkyx]

If you could run it with the 440's first and log some stuff, I'd like to see that. You'll have matched my setup after installing the grey tops.

[BSOD2600]

Yea, will do more logging with the 440's before I install the revtronix chip. Although, if the MAF is maxed out and commanding the injectors static, then it won't really matter much right? Just the AFR will get richer.

[Legacy777]

I would suggest throwing TPS signal on there. It's a good indication of what's actually going on.

Also, I'd suggest using different y-scales for various signals. Too much stuff on one graph really doesn't tell you much, so you may want to break them apart. Just a thought.

[BSOD2600]

I would suggest throwing TPS signal on there. It's a good indication of what's actually going on.

Can do, but will have to remove some other parameter. Otherwise data doesn't get logged frequently enough. 6 parameters is 2-3x a second. 19 parameters is once every 2 seconds -- far too little granularity to see what is going on.

Also, I'd suggest using different y-scales for various signals. Too much stuff on one graph really doesn't tell you much, so you may want to break them apart. Just a thought.

Can do -- also I did attach my Excel data files For the most part, all of the parameters are in the 0-50 range, so it works, IMO.

[Legacy777]

I tried d/l the files and I don't know if I was doing something wrong, or that they're saved in excel 2008. I get a zip file. Try saving them in excel ver. 2003.

[BSOD2600]

Yes, they are saved with Excel 2007. Users of older office versions need to download the Microsoft Office Compatibility Pack for Word, Excel, and PowerPoint 2007 file formats

[AWD_addict]

They open with Excel 2002.

Thanks for sharing the info.

[Legacy777]

I'll try the compatibility pack. IT blocked the d/l. I d/l at home, and am transferring to work. We'll see if that works.

[Legacy777]

Ok.....that's just weird. I had the compatibility pack installed. For whatever reason when I click to d/l those files or right click and save as, the files save with a .zip extension, not the .xlsx extension.

If I change the extension, it works fine, excel opens them and does the conversion.

Wonder if it's an issue with Comcast's webserver...

[BSOD2600]

Odd, I just tested downloading those files with Firefox 3 and no problems. Saved as an .xlsx and/or opened automatically with Excel. meh.

Anyways, I found out some of the parameters that RevScan logs, which b10scan/EvoScan does not have, are easily accessible . Some things of possible interest: Different coolant temp source, Spark Learn factors, Idle target speed, Fuel trim (long/short term), MAF Airflow!, and knock retard (short/long term). Hopefully do more logging next week once I get the additional o-rings for the 440's and installed. Few more days after that, install the JECS and Revtronix chip .

[Legacy777]

Hmm....wonder if those are actually available on the first gen ECU's.

It was my understanding that their a little primitive and didn't have that info available.

[Adam West]

Speaking of O rings Dr. Injector is a good place to get injectors serviced here in Seattle. They flow test and replace rings et al. very quickly while you wait during the week. Forget the cost but worth it for the peace of mind. The Renton store is the nearest to me...

http://docinjector.com/location1.htm

PM me if you'd need a hand in tuning. I'm just learning but would love to see what you are doing real time...

I live on North Beacon Hill, just South of the ID in Seattle.

Cheers,
AW

[BSOD2600]

Round 2
Stock ECU, 440cc injectors, MAF, MAP. MBC set to ~11.5 PSI. The ECU was NOT reset after the 440cc install. Approximately 200mi / 1 week has elapsed since their install. Watching the AFR's, the computer has adapted over the week since the initial install.

I was able to 'find' the ECU location which has the MAF airflow to use with EvoScan, BUT the scaling factor doesn't match RevScan. Get the idea though on the ramp up and what the MAF is seeing. I also switched from Vehicle speed to RPM + IPW (injector pulse width), so I could calculate the IDC (injector duty cycle). I am using the same hills/on-ramps as the previous data logging session (maybe not the same order though).

EvoScan









RevScan





EvoScanLogs.xlsx
RevScanLogs.xlsx


Notes

• It's very apparent on many of the runs, when the MAF gets close to ~4.8V, the injectors go static. The AFR flatline matching the MAF flatline shows this.
• Interestingly, in 5th WOT even with a ~70% IDC, the MAF is maxed out and injectors are going static.
• Appears the injectors (or ECU?) are not capable of going over ~80% IDC. Once the IPW reaches 18ms, the you're just dumping fuel.

Conclusion
The 440cc's help with some increased performance, but ultimately the ECU cannot properly manage them during boost conditions. Once again, the TMIC & TD05 are too much for the stock ECU/injectors/MAF to properly handle.


Up next for round 3: revtronix stage 2, autec maf, 440cc, 11.5 psi boost...

[Legacy777]

Good conclusions. I was pretty much maxing out the MAF with my TD05 running stock boost. That is really your hinderance, not necessarily the injectors.

As you saw, once the injectors go static, you lose all fine control of fuel, and you kind of just dump fuel in, and can adjust the amount via boost pressure, which is not really ideal.

All of this is pretty much what led me to go stand alone. The stock ECU is not capable of accurately controlling the engine with the proper range of sensors (i.e. primarily larger MAF)

[BSOD2600]

Round 3
Revtronix Stage 2 chip and Autec MAF installed. Once my injector air gap was fixed and the ECU reset, things work well.

Did a few passes along the highway onramps as previously. For some reason, this time EvoScan had issues connecting to the ECU and/or loosing connection mid-pull. After a few passes, I started to crank up the MBC. Eventually on one pass, I had a scary reminder that seriously need to upgrade my suspension and then called it quits for the day. Did a few more spirited drives on the city street in 3rd (or was it 4th?).

EvoScan


Lost connection with the ECU near the start. Grrr


Turned the boost up a little bit.


More boost increase and set to ~15psi where I want it.




EvoScanLogs.xlsx

Notes

• The ECU learning process after the chip is installed is a little rough. Initially, crappy idle, high EGTs and lack luster acceleration. Once the ECU does learn though, a very noticeable mid-rpm range acceleration and responsiveness.
• Make sure your O2 sensor is functioning properly and up to spec. Currently tracking down an issue when the engine is cold, which causes a really lean AFR with Mike -- appears the O2 sensor is weak/going bad -- it doesn't heat up fast enough before the ECU uses it.
• The chip does a decent job handling idle full throttle conditions with the new injectors and MAF.
• The claimed increase of MPG might be legit after all. I still need to run a few more tanks, but my last highway tank got ~25 mpg (which is up from my usual ~22 mpg).

Conclusion
After the initial learning hurtle, the chip is a very nice improvement. The 440cc injectors and Autoecs MAF are well within their tolerances for the increased boost levels. I'm going to keep it at ~15 PSI for now.

[asc_up]

So basically what you've found is that it's safe to run a TD05 16G at 15 PSI with the Revtronix Stage 2 chip (the one meant for a TD04), 440cc injectors and the JECS MAF?

[BSOD2600]

So basically what you've found is that it's safe to run a TD05 16G at 15 PSI with the Revtronix Stage 2 chip (the one meant for a TD04), 440cc injectors and the JECS MAF?

Yup.

Looks to be plenty of more room too, if one is brave enough . Highest MAF voltage I recorded was ~4.6. 440cc injector duty cycle is around 65-70% at 15 PSI. AFR's are reasonable around 11.5-12.5.

[Fkyx]

That's REALLY good news. Because I'll probably end up going this route soon after my EJ20G swap is finished. Thanks for this!

I imagine results won't be too terribly different running this on an EJ20G with all of the same modifications?

[asc_up]

Ok so now I'm confused. How much extra room did Revtronix leave during their tuning? The amount of air that a TD04 pushes at 15 PSI is significantly less than the amount of air that a TD05 pushes at 15 PSI. So is their tune ridiculously rich or something?

Just to make it clear, I am in no way questioning your findings. I just would have thought that a TD05 on the TD04 chip running the target boost for the TD04 would make the car run lean.