EVO SCAN Run Need Help

[deepsouth]

Hi guys, as the topic suggests I had a play with the evo scan cable after some niggling driver issues, was doing my goo until it sorted itself out.
So I logged a run and had a look and things are not adding up in my mind.
According to the tutorial videso I should do a flat out foot the flat 3rd gear run while logging then pick the 100% tps data and read that.
So I did that but I have a few things which make no sense to me.
Firstly the RPM was waaaaay off, my rpm guage was showing 6500 before I hit a boost cut or something, the data says max rpm was 5400
The second thing is neither my coolant temp or my o2 sensor changed, neither did my fuel correction.
This all leads me to believe the car is stuck in limp mode or something...
I'll upload the csv data and see if anyone can interpret it with a bit more knowledge than me.
What was disturbing was the knock sensor stuff, it was all over the place which makes me think I had a fair bit of knock going on

https://www.dropbox.com/s/hgbhxsv9ktghvoo/EvoScanDataLog%203rd%20gear%20full%20throttle%20run%20to%206500RPM.csv

https://www.dropbox.com/s/4f2hx13vm31ml5g/EvoScanDataLog_2014.05.26_20.50.32.csv

So if someone who knows whats going on could have a look a let me know that would be much appreciated.

Thanks

 

[MDK87]

Miodrag is the man you need.
His username is vr401.

 

[vr401]

Your temperature is in Fahrenheit. Converted to C is 85 which is good.
you did push it to 6000rpm but you backed off to 70% tps lol
oxygen sensor reading is funy. I have to check whick one you have to log for mixtures.
you don't need to logg everything. More you log slower it gets.
Do a run in third gear from 3000rpm to 7000rpm and log only timing, rpm, knock, injector duty, O2 sensor, airflow.
Keep in mind that some of parameters are not correctly logged in evoscan on rvr.
When I get home I'll get screen shot of what to log.

Ps. What model is your rvr and what cable are you using?

Cheers

 

[deepsouth]

Hahaha Thanks VR401 I feel stupid now lol! Didn't see that F, I'll change it.
It's apparently an HSGR N23W I'm using the 1.3D Cable.
I say apparently as the colours are all off but everything else fits.
Where did you see the 6000RPM? The car cut in and out around the 6500 RPM mark not sure if it's a fuel cut or something else but I cannot get to 7000 atm. Boost was not creeping and the factory boost sensor is not connected due it being melted into a tiny blob...
Okay I'll change the log parameters to those you suggest and do another run tonight when I get home.
The car is running crazy rich, the mechanic is suggesting coolant temp sensor but if those temps are spot on then it obviously can't be that. I replaced the O2 sensor about 6 months ago so it should be okay I assume.
I was hoping running the evo scan would at least me know what the heck is going on with the car, one thing I noticed with the plug in, my check engine light was flashing at me, it was something like 3 slow then 3 fast, last time I did this it was a throttle sensor or something apparently, but couldnt be a 100% sure as the only error codes I could get were from the US mitsubishi evo site.
Thanks heaps!

 

[MDK87]

Good old Imperial units - forever causing confusion. :lol:

 

[penngwyne]

It's apparently an HSGR N23W I'm using the 1.3D Cable.
I say apparently as the colours are all off but everything else fits.

On the manufacturers plate
Model should be N23WG
Classification SRHF for auto or SNHF for manual

 

[vr401]

Is it an auto or manual.
Quick looking at the loggs shows faulty O2 sensor reading and that's if you ticked right one.
what boost is it running because your injector duty cycle is around 85% wot which is a bit too high for stock setup?
Another thing that may happen is that it's over boosting and it hits boost cut causing injectors to shut and O2 sensor to read 0 voltage.

I'll send you that pic. tonight. Do a third gear pull from 3000-7000 rpm and post it here.

Cheers

 

[deepsouth]

Hi VR401 it is a manual.
The boost guages shows a constant 8-9psi with no visible fluctuation.
The factory boost sensor is not connected as the previous owner melted the plug into a nice little ball so have just t'd the boost guage off the lines going to it for now.
Yeah the no change in O2 sensor readings has me suspicious but having replaced it less than 6 months ago I'd assume it should be ok...
I have finally found the coolant temp sensor that my mechanic reckons is the major cause of the poor running so will pick that up and install tomorrow.

Once I get that pic I'll do the run, he said to log as well from a cold start as the coolant temp also did not move from the 180F. It was a cold wet night last night but sitting in the driveway it didn't move at all.

THANKS HEAPS VR401 it is greatly appreciated.

 

[vr401]

For some reason your injector duty cycle was cut to nil at one point just bellow. It should not happen unless ECU is getting faulty signal from sensor which in your case probable is oxygen sensor and temp sensor.

This is how it should look beside knock at the end that i took care of.

Oxygen sensor reading at wide open throttle should be around 0.9 to 0.95 volts which from my experience is around 10.5 to 11 AFR ratio.
Lower oxygen sensor reading is means leaner mixture.

Do a log with settings in evoscan same as on pic. bellow and post it.

and this

 

[deepsouth]

Hi VR401,

I changed the settings and have completed both runs.
I logged from dead cold (sitting for over 4hrs in rainy cold tassie weather till warm), then drove to my run area, gave it two back to back to runs 3rd gear 3000-7000rpm.
I pushed through the cuts there was a cut around 5,500, 6000, and 6500 then till 7000 no more on the first run, on the second run there was a cut at 5,500, 6000, 6500, and 6800 ish and 7000.
Not sure what it is that is causing the cuts hopefully the logs will help.
Also you'll notice I had to keep the engine rpms up from cold start or else it will die.

Logging fewer things certainly helped I had no disconnections this time, I also did the see O2 sensor and temp sensor change so maybe it's something else causing the terribly rich running and poor cold starts....... ????

The first run starts at row 4259 and the second run starts at row 4540
I can't thank you enough for your help man, it's greatly appreciated!!

https://www.dropbox.com/s/etxaj6ivg1mxg3x/EvoScanDataLog%20Log%20from%20cold%20to%20warm%2C%20two%20back%20to%20back%203000-7000rpm%20runs.csv

 

[deepsouth]

Screen shot of settings too
https://www.dropbox.com/s/hxdyt4wgtwmmyq3/evoscan%20settings%20screen%20shot.docx

 

[deepsouth]

Interesting going through the log it looks like the O2 sensor stopped sending signals or lost signal as there are patches there when the revs kick in that the o2 sensor is not logging, wonder it that is the contributing factor for the car cutting out....

 

[deepsouth]

SO further updates after going through the logs last night, it appears my injectors are getting shut down throughout the rev range, wonder if it's an electrical issue as it occurs around the same time the O2 sensor logs stop... it certainly is a fuel cut and can clearly be seen now, unless the car is overboosting and i'm not seeing it on the guage causing them to cut??
Logging is so much fun, just trying to determine what you are seeing is a whole other kettle of fish....
Might try swapping out ecu's as I have a spare ECU and see if anything else happens...

 

[vr401]

does your car blows a lot of black smoke and uses a lot fuel?
To me this is the problem.
your oxygen sensor is not working correctly. For some reason reads a lot less which means for ecu that the fuel mixtures are lean and then ecu pumps in more fuel. ecu does it like that more and more until it sets the fuel trims to max. Once you hit your car WOT at higher revs, as you can see in your log, injectors are running max duty of 104% which means that ecu is trying to get more fuel in to compensate for mixture.
On the other side your airflow Hz reading is very high and for stock boost there is no way it can reach 1600Hz so your car either overboost without your knowledge or you have very bad boost leak somewhere.
Is your car idling good?

 

[vr401]

BTW, yes, it looks like you're hitting boost cut which is actually airflow per revolution calculation where your ecu sees too much airflow and it shuts injectors so you feel like it cuts the boost.
Check your piping for cracks and your boost at manifold.

 

[deepsouth]

Hi VR401,
Yes to all your questions, its blows lots of black smoke on startup and on hard boost, it idles like crap and uses HEAPS more fuel than it should.
I had a look at the car today, couldn't see any loose piping or cracks but that doesn't mean they are not there...there are ball bearings and bits of cable tie blocking off holes so it could be leaking from those areas.
I spoke with a guy today to get a new exhaust, intake and intercooler and turbo piping made up for it so that will at least go a long way to ruling out boost leaks.
I replaced the O2 sensor not long ago, wonder if the wiring to the sensor is suspect?

 

[vr401]

Hi VR401,
Yes to all your questions, its blows lots of black smoke on startup and on hard boost, it idles like crap and uses HEAPS more fuel than it should.
I had a look at the car today, couldn't see any loose piping or cracks but that doesn't mean they are not there...there are ball bearings and bits of cable tie blocking off holes so it could be leaking from those areas.
I spoke with a guy today to get a new exhaust, intake and intercooler and turbo piping made up for it so that will at least go a long way to ruling out boost leaks.
I replaced the O2 sensor not long ago, wonder if the wiring to the sensor is suspect?

If you can make simple pressure tester and try to pressurise the piping from turbo to inlet with 20-30psi you should be able to hear it easy.
With O2 sensor, was it original replacement or you had to hack the wiring to connect. I know there is 2 or 3 wires O2 sensors or something so it may not be right.
Test the car like this.
Disconnect the batery for 30sec. After connecting go for third gear pull but it has to be in first 5 min after reconecting batery. Logg everything i said to you before and post it. This test will tell us how much ecu is compensating with fuel corections due to something being faulty. It allow us to see what's there before ecu readjust itself. If we see simular situation it leads to conclusion that there must be boost leak present somewhere.

Cheers,

PS.
Good to hear that you're enjoying your evoscan. After you sort out your mechanical issues i may start teaching you basic tuning

 

[4piknup]

↑ guru.


I have a thread here somewhere about my homebuilt leak tester. It cost about $4-5. You will find leaks everywhere.
Also available on ebay for around $30.

 

[blackvr]

Miodrag, the only issue I have with the oxygen sensor theory is that the factory narrow band sensor will only trim fuel in closed loop conditions. Generalizing, this is in light cruise and load conditions, and reverts to open loop " around " 0-1psi of load. This means very little air will be escaping from the air plumbing.

The ECU is not using the narrow band oxygen sensor for any calculations in open loop. The sensor is still reading a voltage, and if it is very rich, it should read around 0.90V or higher. The narrow band sensor is only accurate around Lambda ie about 14.7:1 AFR.

Once in open loop, the fuel calculations are taken from the MAF sensor , air temps, revs and any other "adders". My feeling is there is a boost leak... especially when , like you noted , the Hz reading is so high. From my tuning, this would equate to 20 plus psi boost, and a larger than stock turbo . A fuel cut would definitely occur with this load in a stock ecu !! Evoscan tends to give funny duty cycle readings, and is heavily influenced by the injector scaling set .. I've had the same as well, and when comparing it to a commercial scan tool I have, the readings can vary.

Pressure test the system .. Especially look at the end tanks on the intercooler, as these are common to split and leak. it will be a big leak to get those sort of high Hz numbers

 

[vr401]

Miodrag, the only issue I have with the oxygen sensor theory is that the factory narrow band sensor will only trim fuel in closed loop conditions. Generalizing, this is in light cruise and load conditions, and reverts to open loop " around " 0-1psi of load. This means very little air will be escaping from the air plumbing.

The ECU is not using the narrow band oxygen sensor for any calculations in open loop. The sensor is still reading a voltage, and if it is very rich, it should read around 0.90V or higher. The narrow band sensor is only accurate around Lambda ie about 14.7:1 AFR.

Once in open loop, the fuel calculations are taken from the MAF sensor , air temps, revs and any other "adders". My feeling is there is a boost leak... especially when , like you noted , the Hz reading is so high. From my tuning, this would equate to 20 plus psi boost, and a larger than stock turbo . A fuel cut would definitely occur with this load in a stock ecu !! Evoscan tends to give funny duty cycle readings, and is heavily influenced by the injector scaling set .. I've had the same as well, and when comparing it to a commercial scan tool I have, the readings can vary.

Pressure test the system .. Especially look at the end tanks on the intercooler, as these are common to split and leak. it will be a big leak to get those sort of high Hz numbers

It is correct that ecu doesn't use oxygen sensor for open loop but as i said, oxygen sensor still reads mixtures as voltage and as you said it should sit around 0.9v at WOT but in his case it shows very low voltage? To me it is impossible for the car to run such a lean mixture at WOT that his sensor shows.
The other thing i believe is right is the fact that once ecu sets fuel trims it will use offset compensation from high trim for open loop WOT when it's picking the raw value from table so if it has high trim set to maximum addition of fuel it will do that percentage to table value. For example if the trim is +5% then in my opinion when it reads for example 10.5 AFR from table cel it will add +5% to it making it 9.75 AFR.
tThat is why we sometimes tune tables to different number like 11 to actually get 11.5 for example.

Btw: yes, i was pushing 1800Hz with big 16g/20g at 20psi so his stock should never be even close to it.