Jump to content

MGV101

Members
  • Posts

    79
  • Joined

  • Last visited

  • Days Won

    2

MGV101 last won the day on May 18 2017

MGV101 had the most liked content!

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

MGV101's Achievements

  1. Thank you. Should I assume the default calibration of the internal 7 bar sensor to be 0 at 0.5v and 7bar at 4.5V without any offsets?
  2. Yes it calibrated just fine. I guess the 5kpa is based on the "default" calibration of the sensor? For example, maybe previouly the sensor was 3kpa above the "default" calibration and now it is 4kpa below allowing a 7kpa swing. But the strange this is, why did it suddenly change?
  3. The vacuum hoses checked out fine and if it was partial blockage or leak I doubt calibrating the sensor again would have fixed the problem. Whoever sucking on the tube must have strong lungs. MAP was at the low 30s on idle (before I recalibrated the sensor) when it should be at around 40kpa. Air density was reading 20% lower than it actually is, so I was getting 20% less fuel.
  4. Had an issue with an EVO8 suddenly running very lean at idle/light loads. The car was tuned around a year ago without any issues. Turned out the issue had to do with the MAP sensor calibration. With the engine stopped, the MAP sensor was reading around 7kpa lower than the BAP sensor. An offset of 7kpa will definitely have a huge effect on fuel calculation on high vacuum and a quick recalibration seem to have fixed the fueling issue. If this was an external sensor, I would most likely be checking for grounding issues or suspect the sensor was replaced by the owner. However, this is happening with the onboard MAP which is literally on the same circuit with the ECU with the same reference ground so there should be no signal/grounding issues. Could this be an indication of the MAP sensor going bad? Is there anyway to read the MAP calibration value in the map so I can make sure the calibration had not been tempered with by the owner causing the issue? Thank you!
  5. Thank you for the replies Adam and Vaughan. Using individual cylinder correct to check is indeed a much better way.
  6. On an distributor based ignition system, cylinder fired is dictated by the distributor correct? So even when the ECU think it is firing cylinder number 1, if the trigger offset is wrong by multiples of 180 degrees (for a 4 cylinder engine) it could actually be firing cylinder 2, 3 or 4. Since the firing intervals are evenly spaced out, the ignition timing would still be correct and a timing light won't be able to tell you that the ECU got the cylinders wrong. Although ignition timing will not be an issue but wouldn't this affect sequential injection causing injection timing to be wrong by 180/360/540 degrees? I guess to confirm whether the cylinder assignment is correct we could change the trigger offset by increments of 180 degrees while looking at the AFR. The offset with the richest AFR should be the one with the correct offset? I don't see this documented anywhere in the ECU Online Manual and searching online also yielded nothing so I just wanted to confirm my logic is correct before changing the offset by 180 degrees while the engine is running. (This would definitely be dangerous on a direct spark system) Seems like this would also apply to a wasted spark system too?
  7. Further update... After gapping down the plugs (from ~0.75mm to 0.6mm), we did a short drive and was still getting trigger errors on boost. Just noted the dwell time on the coils were set to ~5-6ms on the base map which seemed a little high. Have lowered it down to 4.5ms and lets hope that would help.
  8. Thank you for the detailed reply Adam. The ignition system is still stock. Looks like the coils had been recently replaced as they looked almost new (trying to confirm whether these are genuine stock coils now). Plugs are your typical IK22 spark plugs used in most JDM engines so they are definitely resistive type spark plug. The plugs were new but we measured a few of them just in case anyway and they all measured find at 3.XK ohm (don't remember the exact resistance). Ignition leads were new and came from here: https://mrretroleads.co.uk/product/lime-green-8mm-performance-ignition-lead-kit-zetec-black-top-kit-car-part-built/ There was no marking/branding on these "custom" leads but we measured the resistance and after factoring in the length, it does come close to the website's spec of 5.6Kohm per meter . Lastly, yes there was a capacitor near the coil just like the early Evos. The condition looked ok with minor weathering. We tested the capacitance and measured 460nF which was close enough to the 0.5uF marked on the capacitor casing itself. Without any other obvious faults/source of EMI, we are now gapping down the spark plugs from 0.8mm to 0.6mm and hope that would help. But yes I still agree the noise is most likely from the ignition system. Maybe there are damage to the stock signal wiring/shielding/grounding that is amplifying the noise. The car was in a very sorry/neglected state when it first came in. It's no longer on the dyno so it is difficult to consistently test for the trigger error again to confirm whether gapping down the spark plug have solved it. Only time will tell.
  9. I suddenly started getting trigger errors while upping the boost on a Mitsubishi GTO. It would happen at any rpm once boost rises above ~0.9 bar and would happen in around 1/3 of my ramp runs. It seems to be ok after I dialled down the boost again. I discarded most but two of my logs that had the trigger errors as there were way too many logs to save while at the dyno. As I returned home and started reviewing them, in one of log I noted something strange. Right before the trigger error, Trigger 1 Status showed "Test Gap" where normally Trigger 1 Status would only show "Count Large" and "Count Small". The crank trigger on a GTO is just has three large teeth and gaps with no missing teeth. Count Small is just referring to the higher teeth on the wheel after the falling edge is detected while Count Large is referring to the lower gaps in between as the rising edge is detected. I don't get such strange findings in my other logged run with trigger error but my logging rate was definitely not high enough to catch all trigger state so it might have still been present but the log just missed it. Why would the ECU suddenly be looking for a gap all of a sudden?? (The terms may be confusing here I assume the gap here is referring to the missing teeth and not the regular gaps between the teeth) Did that unusual behaviour of the ECU cause the trigger or did the trigger error cause this behaviour? Also another strange thing was, after the trigger error, the engine started to run extremely rough. This did not occur during other trigger errors (and this is probably why I had this log saved and not the others). However, after stopping the engine twice and restarting the ECU again everything was back to normal. Attached is a trigger scope take at ~3000 rpm on boost at steady state and also the aforementioned log. Screen shots also attached. There is also some noise in the trigger scope but is it sufficient to case a trigger error? trigg 3000.llgx trig stut.llgx
  10. What is the maximum current capability on the 5V rail of the AtomX? I might want to power a microcontroller directly off the AtomX's 5V rail which should draw no more than 100mA.
  11. Hello Essb, This is happening in open loop so integral gain should not matter at all (it's not even available in open loop mode). But before I switched to open loop, I did take a look at integral gain and it was set at 0.3 from the base map provided and I did try dropping it all the way to 0.05 to no avail. This was the point where I switched to open loop to eliminate any effect from closed loop control. Integral gain is effectively zero here. Anyway, here is the map as requested. start.pclx Anyway, after getting home and taking another look at the logs, I think I have found my issue. Seems like there is something wrong with the AC input. It is pulsing on and off every second causing the idle base duty to change.
  12. Just finished installing up a EVOIII G4X modified to work on a Mitsubishi GTO and started the engine for the first time. Everything seems to be fine other than the idle. Even after switching to open loop the stepper seem to be jumping around wildly from the base duty to 90% back and forth. I have confirmed it is not jumping in and out of idle mode by looking at idle status (it shows it is in open loop mode the whole time). Overrun fuel cut is also off, I can't find anti-stall but I think that is only for G4+ and G4X no longer have anti-stall setting? Strangest thing is even when I step on the throttle lightly to increase the rpm and have exited idle mode, the stepper is still jumping up and down causing the RPM to fluctuate. Attached is a log of the issue while I was playing around with different settings (switching from unipolar to bipolar, increasing/decreasing idle steps etc) but nothing seems to work. The only time idle is stable is during startup hold where the stepper is held at the maximum of 90%. EDIT: Attachments deleted as I ran out of space for attachments and have to delete some from my old posts.
  13. Thank you Adam! Yes looks like I have done the math wrong. Calibration Number = (Number of Sensor Teeth / Wheel Diameter) * 31.83. Note that wheel diameter is in meters (m) I was using cm instead of mm in my calculations.......
  14. Without an easy way to install a trigger wheel/disc on my front wheels, I am thinking of using the lug nuts at the rear of my hub to read my wheels speed. However, with only 5 lugs per revolution, the resolution will be extremely low at around 3Hz at 10km/h. I guess this should be "OK" for data analysis and traction control at high speed but is that enough resolution for launch control to work properly? or should I think of a better solution? Thank you! Edit: Actually, after calculating the calibration value, seems like there isn't enough sufficient resolution in the calibration to give an accurate enough speed. With a calibration value of ~3, just an increase/decrease in the calibration by 1 would result in a 25%+ change in speed......... The "Wheel Speed Correction" setting which should in theory allow me to find tune the speed is only availabe for Driving Wheel.
  15. Hello Adam, everything came in and I am impressed with the 2204A. After more digging in on how to use the oscilloscope, I've realized it has MUCH MORE use than I initially expect. Anyway, here are the wave forms I have recorded on the V-CAM sensor at idle and free revving at around 4000-5000 rpm. The signal looks perfect to me though and I really could not see any fault. I probably should have recorded the trigger 2 sync at the same time but I was too excited with the new oscilloscope and was too rushed to think of that. Do note that I have used the 20x attenuator for the 4000-5000 rpm waveform but have not set the probe to 20x in the software so the actual voltage should be 20x higher. Actually, the 20+V @4500 rpm looks pretty high to me. Could too higher of a voltage cause problems with the reading? I imagine it would be even higher at 6-7K rpm where I am having my issues. Idle 880.psdata 4500 20-1.psdata
×
×
  • Create New...