Jump to content

Rozsko

Members
  • Posts

    232
  • Joined

  • Last visited

Posts posted by Rozsko

  1. 10 hours ago, Adamw said:

    I think you will be in bootloader mode if you dont have pin 4 grounded.  From the manual:

    uvyto6i.png

    Thanks Adam. As per the manual if it is in bootloader mode, then there should be a solid green LED, which is not the case for me.

    But that is definitely a good point. I 'll give it a try.

  2. Hey guys,

    Today I was experimenting a bit with my STi and ran into something weird that I can't get my head around.

    My intention was to do an indirect individual lambda measurement on all four cylinders, by disconnecting the injectors one after the other and see how lean the lambda becomes each time. I assume if all four cylinders work with the same lambda.

    While doing this test, all four times when an injector was disconnected, the lambda leaned out roughly to the same value, however the EGTs changed as well.

    For example when I disconnected inj#4, EGT#2 increased. See attached picture, log and map for more details.

    Do you have any idea what is causing this?

    Thanks,

    Béla

    screenshot:https://1drv.ms/u/s!Ale4oyMCOgLThYou9l_dlHBVUSR70Q?e=kJcjH8

    log: https://1drv.ms/u/s!Ale4oyMCOgLThYovPZ-3id31fZXaPg?e=1NYmcq

    map: https://1drv.ms/u/s!Ale4oyMCOgLThYowND06D62f_6at1A?e=SvW6AL

  3. Bringing back an old topic in the hope that someone can help with the serial configuration of Spartan 3 ADV (I think these new versions do not have Bluetooth any more).

    I just received my device and I am trying to connect to it through USB. Installed Termite and connected to the device, but whatever command I issue, there is no reply on the serial terminal. When I hit enter in Termite, there is a red led that lights up for a moment, so I assume the command is sent to Spartan, but nothing coming back. I tried it by only plugging in the USB cable and also powering it up via pin 1 and 2. No sensor is connected at this time, since I am just trying to verify (and potentially set) the settings of the device on my bench.

    From the past I know FTDI chips might be problematic to work with as a user, so I updated the driver as well to the latest and tried this with two different laptops with no success.

    Any tips are appreciated.

    Thanks,

    Béla

  4. Hi Guys,

    Quick question on signal voltage of cam sensors.

    Recently I did some crank-cam scoping and realized that the right hand cam signal is about 10V base instead of 5V that is on the left hand sensor.

    Initially I thought I had wrong probe settings, ie attenuator, but finally I did a quick check at the ECU, A11 pin (pin removed from the connector) and the ECU pin had 10.64V KOEO.

    Should I worry about that? In the past when I was doing the same cam-crank scoping, it was certainly not the case, and since even with removed pin it is that high voltage, it must not be a wiring issue.

    Thanks,

    Béla

  5. 6 hours ago, kaptainballistik said:

    I'm actually struggling with something similar ATM.. Comp on 2-4 is  the same, but 2 tends to foul. The STD excuse always seems to be the Subaru coils are Bleh. Which I don't agree with!

    In your case, its known for Subaru's to run a bit hotter ( The coolant is hotter there) on 4, so I would first look at adjusting the correction for Cyl 4. 

    When you say adjusting the correction, what do you mean? AFR? Because I already tried that, but the EGT decreased even further, no matter if I enriched or leaned out the mixture.

    The other thing I wanted to ask is about the normal spark plugs. Do you have any particular partnumber to try? I looked up NGK, Denso and Champion as well, but the first two have only standard heat range plugs, the latter one have colder but only with three ground electrodes.

  6. On 3/2/2022 at 10:19 PM, kaptainballistik said:

    Yeah, plug 4 looks a bit Shady!

    Running slightly tighter gaps?

     

    At idle, tighter gaps should not be required I assume.

    What is interesting though is that EGT is about 80-100C less in#4 then in the rest of the cylinders, which leads me to either severe misfire or some sealing issues with the cylinder. (Individual fuel correction in #4 did not have any good result. No matter if I added or removed fuel, EGT decreased even further, though I was expecting it to rise when leaning it out.)

    So I did some in cylinder pressure testing and the cranking compression is exactly 10 bars both in #2 and #4. The running idle pressure waveforms again look exactly the same in #2 and #4 and there is only a little difference between the two in peak pressure.

    The ignition waveforms look horrible though and it is worse in #4, but I don't think it is electrical, since I swapped the cables between #2 and #4 (and changed the firing order accordingly) and nothing changed.

    Also swapped injectors between #2 and #4, again no difference.

    So, every test I do is not unveiling any problem, yet there seems to be something wrong with #4.

    I recently replaced both cam sensor and the crank sensor and again no change. Crank sensor signal is nice as it can be (all peaks are above 5V or under -5V), and cam signals are as weird as they can be with Subaru.

  7. 34 minutes ago, kaptainballistik said:

    The key is the colour of the ceramic, and if there is a ring of carbon around the electrode.. and how much carbon is down the plug grove.the "non-copper plugs are hopeless unless you have a really good tune. In my STI's original spec it would always start on 3 cylinders with  the :"non copper" plugs. 

    On the other hand, Some people have Zero issues! (My Road WRX probably "needed" plugs 20K ago and its still ok.  

    I think its the combustion chamber design that the issue......

    #2 plug: https://1drv.ms/u/s!Ale4oyMCOgLThYUWUDxroZFmBDD8Vw https://1drv.ms/u/s!Ale4oyMCOgLThYUXY_WWmTxAJ698UQ

    #4 plug: https://1drv.ms/u/s!Ale4oyMCOgLThYUQpI7fegORN_sdKA https://1drv.ms/u/s!Ale4oyMCOgLThYURIlcVEgwUjvk2iA

    Now that looking back the pics, #4 is worse then #2.

  8. On 3/1/2022 at 12:34 AM, kaptainballistik said:

    Ahh, Switch to the Copper core Plugs until you get it tuned, Even the short copper plugs meant for the early heads work ok.

    How much Carbon fouling do you have?

    Well, I tried quite a few different plugs by now, so trying one more won't hurt. :)

    I have some carbon build up on the plugs, but I would not consider it excessive.

  9. 56 minutes ago, Alexduong88 said:

    ! so with the strain gauge with a amplifier that outputs 0-5V could you use the AN Volt 11 or AN Volt 12 on the expansion port? Or would I have to find a AN with a 5v? 
     

    Thank for the help

    The expansion port has +5V and sensor ground as well if needed.

  10. 11 hours ago, Adamw said:

    Although this car was all badged as "STI" on the exterior and interior and official registration, it had small brakes, no widebody, only 5spd, no DCCD, etc,

    What the hack?!

    Unfortunately our project car is behind schedule as well, so i wasn't able to test it myself, and I have no idea when we get there.

    On the other hand it is good to know the design is ready and is only waiting for production. Hopefully you can sort that out shortly.

  11. On 5/10/2021 at 12:32 PM, Adamw said:

    Still not much closer at this stage.  I dont want to make a guess, but based on the time frame of the last couple of plug-ins we have released I think it is still at least 6 months away.

    I know this was a guess only, but was wondering if we could know anything more on the G4X plugin for the 2008+ STi.

    Thanks

  12. 16 minutes ago, essb00 said:

    Only one CAN-lambda per CAN ID (950, 951, 952, 953 on your case)... Having different ID per device is the only way for the ECU to know from which device it is receiving data from.

    I was going to paste the link of a previous topic that I opened (actually here it comes: https://forums.linkecu.com/topic/13222-link-can-lambda-stream-setup/) where Adam explained how it works to prove that only 1 ID will be used by the Link modules, but I scrolled through it and realized he talks about one CHANNEL and not one id, as the ECU expects Lambda2 to be 951, Lambda3 to be 953 and so on.

    So, along with your thoughts, actually that answers my question. If the Spartan3 devices are assigned id 950, 951, 952,.. and all are set as Link CAN, then I can use multiple devices on the same CHANNEL that is set to Link lambda in the ECU.

    2 minutes ago, Adamw said:

    essb00 is correct.  You need to program the spartans with a unique ID each between 950-958, you do this either with USB or bluetooth depending on how old it is.  In the ECU you only set one channel to "Link CAN Lambda", it captures all lambda ID's.

    Just two seconds before my reply realizing how dumb I am. :lol:

  13. 1 hour ago, essb00 said:

    Spartan 3 has serial to USB connection, where you can send commands from computer (using terminal) to change parameters (CAN ID, etc.). You can see overview how to connect on Spartan 3 user manual. This though is quite complicated --- so it might be the best idea to request to have different CAN IDs pre-configured as you order 'em from 14point7.

    Thanks Essb00, but that is exactly the thing I want to avoid, since I am running out of available custom CAN frames.

    So, let me rephrase. Can I use 4 "Spartan 3" on the same ID (950) and assign Lambda1,2,3,4 the same way as you can do with "Link CAN Lambda"s?

  14. Bringing back this topic for a quick question after spending some time searching and reading.

    If the Spartan3 is set to Link mode on Id#950, then can I connect multiple units and set them up separately for Lambda1,2,3,4 the same way as the Link CAN Lambda? If the CAN Devices tab will only sow real Link CAN Lambdas (as per earlier comment), then I assume not, but wanted to check.

    If this is not possible, is there any plan to refresh the Link CAN Lambda to support LSU ADV sensors?

  15. On 11/12/2021 at 9:32 PM, Adamw said:

    I believe CAN bus changed after 2011

    As far as I understand from some Subaru documentation, the CAN only changed for the WRX, not the STi. We have a project in progress with a 2013 sedan STi where I was planning to hook up my WRXLink107 and re-pinning some of the connector pins, but unfortunately the project is on hold for some time now.

  16. 11 hours ago, Adamw said:

    Your misfire is probably unrelated to ignition energy if it is at idle.  Cylinder pressure is very low at idle so doesnt take much spark energy to start combustion.  You would typically see ignition energy related issues at higher loads.  

    1. Yes.  
    2. The suppressor provides a short ground loop path for the spark to return to the coil, this reduces emi.  This Adaptronic video explains the concept quite well, watch from about the 8:00 mark:  https://youtu.be/ZTDsm6b69Lk
    3. Again, unlikely to be sensitive to dwell at idle - you can give it more as a test.   
    4. VE and hence cylinder pressure drops away after peak torque, so since you need less ignition energy you can drop dwell back at higher RPM to give the coil more cooling time.  Electronics lifetime is greatly influenced by their average and peak temperature.
    5. Not in your case.  Probably overkill, but they are good coils, originally designed for a mercury two stroke as a CDI replacement so provide good energy at even short dwell times.  They do have 3 isolated grounds, one is meant to go to the head, 1 to sensor ground and one to battery ground so that can make wiring a little more involved to do it right.  Having said that I have seen many people just ground them all to the same lug and get away with it too.  
    6. OEM style carbon core leads are still the best by far for emi supression.  Spiral wound often have quite low resistance and generally arent as good.  You would typically see trigger related issues if there was an emi issue.  

    Thanks Adam.

    On the suppressor capacitor, as per Andy it is meant to be used for the old two pin coils, where the +12V is the common connection for the primary and secondary winding. So does this mean the Subaru COP coils are the same internally even thought they are 3 pin coils?

    And one more thing on the IGN1A wiring. So on the 107X G4X, the ground pin from the coils go do D26 on the ECU which is a normal ground. So the sensor ground pin from the IGN1A then should really be hooked up to the sensor ground pin on the ECU (if I want to go by the book), right?

×
×
  • Create New...