S13 CA18Det Fuel cut at 15 psi?!?

[Wasson Will]

Hi guys, Running a stage 3 ready S13, been playing about with nistune for a few years now and feel like I've got my head around it however I cannot get the car to run over 15psi at any revs, have ramped up the max ttp for overall and revs and scoured the all the other tables that could be causing it. Had the problem with my original MAF and thought that was the issue. Now have a GTIR 35 Slot MAF in place, the map is still a work in progress but the car is going well, until you raise the boost to 15 or above. Smoothed out the fuel map as this way causing massive surging but even though I am now pretty happy with the map it still stalls at 15. Driving it is now barely noticeable but the logs seem to show the MAF starts dropping off at almost exactly 15psi at any set of revs. Though it was a wastegate issue so have replaced the actuator with a dual action variable one, have had the same issue with the .5, .8, 1, and 1.2 bar springs, it is boosting up much faster with the heavier springs but still cuts off at 15. Finally thought looking at the logs it must be the Recirc dump valve lifting causing the Maf reading to drop, tried the 2 other heavier springs in this but still exactly the same response. Cannot think of any other mechanical factors that could be causing this.

Car specs are:
NGK Iridium BKR7EIX Plugs, new.
Drift cone air filter up to 500bhp
GTIR R35 Slot MAF in 3” tube, max 400bhp( Currently scaled to 80% to increase definition in usable range so approx 320bhp airflow)
GT T28 journal bearing turbo
Large front mount intercooler
Recirculation dump valve
6Boost exhaust Manifold
Large Turbo down elbow
3” Exhaust
Uprated fuel pump
Nistune Feature pack ECU
MS-IBC Variable boost controller
PLX devices, boost sensor, exhaust gas temperature and Wideband AFR through single multigauge

Have attached a couple of log files showing the drop off, one with the wastegate actuator set up as a one way where it tends to get a smoother response but is slower to build boost.

The other two are with the actuator double acting, build boost much faster but tends to drop off by around 1 PSI for around 1-2 seconds, however set to lower boost levels it maintains excellent stable boost after initial response. Have attached 2 files as I have the PLX devices in use and am having the daisychain connectivity issues when using the afr with other modules so one shows AFR and the other boost and EGT. All three are showing a full throttle run in third gear visible 5-10 seconds into the log.

Any ideas?? Know this is probably gonna be something simple but am out of ideas now...

[Wasson Will]

Oh yeah current map from the car.

[unex]

i dont see any fuel cut in the log, just lifting throttle.
If you can not reach bigger boost, maybe you have big boost leak. Have you done boost leak test?
Also, maybe you just have spark blowout issue.

[Wasson Will]

Ha ha well delayed response as the car has got other issues. Yeah have gone one cooler again on the plugs, however got bigger issues with overall running that I am currently trying to work out.. More of a head scratcher as I cannot get any of the PLX gear to work with the software so cannot log boost afr or egt whilst playing with it. Having changed nothing that I am aware of the car has been running richer and richer to the point of drowning itself out like a rockstar... Took it off the road running well to do the gearbox and it has been getting worse and worse ever since. No fault codes, has killed the narrowband Lamda it has been running so rich. Ready to chuck a match at the exhaust whilst it is running at the moment...

Definetly not me lifting off the throttle when was happening would start building up pulling like a train then pretty much bang on 15psi it starts bouncing like a kangaroo power no power, savage on the neck. I tend to lift off when it happens just as a reaction.

[Matt]

What PLX device are you using?
Does it communicate with their software?
Is the iMFD compatible device?

If you can select the COM port in the wideband window, also enable debug logging (Config > Misc > Debug logging) and then take a log of you trying to connect and post it here, I can check it out

With the problem where it is getting richer, if you take a log now (assume the tune same from 2 years ago?!) can you post that here? I would compare the injection time in two runs through on the tracer

Also have you checked your fuel pressure? Is that around 45psi and 1psi for each pound of boost?

[Wasson Will]

Yeah will use the log to try and see what is killing the the wideband connection but yeah it is the PLX imfd one touch setup so daisy chained sensors to the single digital display. Previously with EGT, afr and boost chained it would always always crash the narrowband so I seperated the AFR from the other two and would do alternate run logging either afr or boost and egt and it worked. Yeah the one touch gauge is working fine though only have the AFR connected to it currently.

This has got mixed over two threads now as this was my original problem but the richness issue, had issues with it after doing the gearbox which turned out to be an intermittent knock sensor fault which in turn caused me to cook the head (my fault obviously for driving it when I knew it wasn't right.) fitted an open port head when I fixed it rather than split port head that was originally(eco/torque head with butterly valves on 1 inlet valve on each cylinder.) Realised that when I did this had to change the fuel rail as well to fit the different inlet manifolds, sure when I comissioned the car after the original engine rebuild the Fuel pressure was 38/40 psi, is now at 45psi which explains the overall richness now. The vacuum port on the fuel pressure regulator has been disconnected since before I had the car presumably to suit the stage 2 chip that was fitted. You recommend reconnecting this then? So it increases pressure under boost rather than reduces pressure under vacuum?

[Wasson Will]

Right, believe both those logs should have it, debugging on. It seemed to knock out the ecu connection trying to connect to the wideband hence the two files. Tried changing the ports and aux 4 with both the afr1 and 2 files imported as csv files. When I open the file it goes back to the menu config screen but does not seem to acknowledge that the file is loaded.

[Matt]

Those logs are from your Log Viewer and end with CSV. You should have some in the logging folder ending with LOG. Can you post those ones also, or email to info@nistune.com

[Matt]

Also you are running 1.3.25 version?

[Wasson Will]

Yeah running the latest version, didn't seem to log any faults first time round have 4 logs now, 2 attached the other 2 are too big so will email them through. Also anyone recommend an aftermarket Lamda sensor for this engine. Cannot source genuine and currently have three aftermarket ones that do not seem quite right, fitted a generic Lucas LEB123 and that seemed to stay at .3/.35v until about 14 to 1 where it would go straight to 2.5 v

Now have a fuel parts LB1075 fitted which seems to work well is the map is very close, but weirdly when mixture goes below about 13/12.5 to 1 voltage spikes up to 2.5v again instead of going lower beyond half a volt. Have a Lucas LEB 5004 which is also car specific,haven't tried it yet as I was going to return it but now think it may be worth a try, see if it perfoms better than the current one which throws the short term fuel trims way out.

[Matt]

The logs are big because you are also connected to the Type 1 board (which is filling the log with data). That makes it more difficult to just see the PLX communications.

So for the debug logs, dont connect to the ECU, just the PLX so I can see those

So far...

Screenshot shows you are trying to connect to COM3

nistune-0701-1825.log
17:27:48.846 CWidebandCommon::OnButtonWideband()
17:27:48.846 WidebandMain
17:27:48.847 statusbar: Wideband 1: Valid Communications port not selected for wideband (few of these. No valid port selected)

nistune-0709-1048.log
(nothing wideband related)

nistune-0701-1826.log
(empty file)

nistune-0630-1436.log

13:37:39.259 CWidebandCommon::OnButtonWideband()
13:37:39.259 WidebandMain
13:37:39.259 CSerialSupport:Initialise(): Connecting to COM3
13:37:40.118 statusbar: Wideband 1: Connected to COM3
13:37:40.282 ReadLoop(): Timeout: Pending wait on read
13:37:40.305 ReadLoop(): GetOverlappedResult: bytes=9
13:37:40.305 RX:88 B5 91 FF 90 AD B1 FF 50
13:37:40.305 Tail not found at index
13:37:40.305 CPLXLoggerSM(1): Frame tail found at index 3
13:37:40.306 [7C][98][E6][03]
13:37:40.306 ReadLoop(): Waiting for 4 remaining bytes, only 0 bytes available
13:37:40.306 ReadLoop(): Pending wait on read 4 bytes
13:37:40.412 RX:94 54 D0 6C
13:37:40.412 Tail not found at index

Okay basically I'm looking at the data coming into Nistune and it does not follow the PLX iMFD protocol

// Start Bit (0x80)
// Sensor Address MSB (Bits 6-11)
// Sensor Address LSB (Bits 0-5)
// Sensor Instance
// Sensor Data MSB (Bits 6-11)
// Sensor Data LSB (Bits 0-5)
// Stop Bit (0x40)

I should see 0x80 .... 0x40 but the logging is not showing that.

1. Are you using the PLX iMFD cable supplied with the unit?
2. Do you have anything else connected to the laptop (showing as COM3?). Does COM 3 disappear in device manager if you unplug the iMFD?
3. Can you connect using the PLX Logger software okay to your unit?

If taking any more debug logs, only connect the PLX (not the Type 1). It takes a long time to filter through all the Type 1 data

Below is a screenshot of me connected

[Wasson Will]

Just tried connecting without the ECU to run the wideband independently, yeah the normal log seems to work fine the number of available channel is less than it used to be but guessing that is software development? The lead I am using is homemade as per the link on here somewhere, have two of them and both were working so I assume that they are okay. Maybe these are not suitable.

Not the com port is dictated by the lead I am using and cannot change it as far as I can see, swapping leads over changes it from com 3 to com 5.

[Wasson Will]

More shots

[Wasson Will]

What com port should it be on? Yeah if I unplug the unit it disappears from the list, I can change the com port as it appears in the windows config.

[Matt]

The logs you emailed me all showed ALM wideband as what you were trying to connect to (first screenshot) on COM5

After that following screenshots show PLX iMFD on COM3

I assume you are using the PLX iMFD wideband and not the older PLX R-series?

When you plug in the wideband, the COM port will display in device manager. It should match the same COM port in the drop list. So which unit is it? COM3 or COM5?

COM3 on some laptops might be internal modem or something else

Can you resend logs with the correct PLX unit and COM port selected?

Second wideband should be set to 'none'