Jump to content

Adamw

Moderators
  • Content Count

    4,795
  • Joined

  • Last visited

  • Days Won

    258

Everything posted by Adamw

  1. Ok, sorry for the delay, I just had time to look at your file closer and give it some thought. Basically I think you have it set up pretty right as is. It should do what you have described above. I dont think you will need the 4D ignition that I mentioned as I was mistakenly thinking you also wanted a variable amount to torque reduction between 40 & 85%TP. I suspect the 90% cut you have set in the driven upshift cut will be way too much and will feel like you have hit a wall. Start at about 40% and see how it feels. You will also need to experiment with the torque reduction time and introduction time - my feeling is it will probably want a shorter reduction time and longer introduction. Also, if you update to the latest firmware you wont have to use DI1, you can trigger the gear shift event directly from Aux 1 or a virtual aux.
  2. The next test I would do is this: First make sure error low is set to 0.0V and error high is set to 5.0V on the AN Volt that is set to MAP. Next, unplug the MAP sensor, use a paperclip or similar to bridge the 5V wire to the signal wire at the MAP sensor plug, if ECU and wiring is ok the ECU should show 5V. Try the same again but bridge the signal pin to the gnd pin, ecu should show 0V.
  3. Try this, I got all of your list in there except battery volts. 190803Tune with Haltech CAN.pclr
  4. There is no setting that can disable Inj1 in isolation. If the injector clicks with the injector test then that proves wiring and ECU is ok. Assuming there are no trigger errors then Inj 1 should be commanded PW when running. Have you checked compression?
  5. Adamw

    Mazda MX5 VVT

    If you open the set base timing screen again, is the new value still there or has it returned to zero?
  6. Adamw

    Mazda MX5 VVT

    Are you sure you hit enter when changing the offset? The field should turn blue if you have done it correctly.
  7. Yep, the branded gaugeart devices are locked to that particular brand ECU in the firmware. The Link CAN is flexible enough to be able to duplicate the Haltech stream however there is a limitation of six ID's (haltech uses many more in their full stream). If you attach a copy of your tune and a quick list of the channels you would like to be able to display I should be able to set it up (may not be able to give you the full channel list that is avail in the gauge).
  8. Yeah, that will probably work although it will need a bit of trial and error on the aux output table to get the voltage output right.
  9. I have seen this before on a 2JZ, as per CJ's advice, change to custom PID, usually just halving the P is all it needs. I suspect it may be caused by oil viscosity as I have also seen many control well with the default settings. The original factory recommended viscosity in these engines is quite thin, 0W/20 or 5W/30 from memory.
  10. I never got around to doing that video sorry. There are a couple of good Motec webinars that will be worth watching just to learn the basic fundamentals of CAN. I will give a bit of info here to try to help make sense of it. Post any questions you have after having a go. First, the old Link "generic dash stream" and the new Link AIM MXS strada dash stream are basically identical, the only change made was the scaling of the Lambda channel - this was changed to allow the dash to work out of the box with both G4+ and older G4/G3 ecus without a config change. So if you want to see how the message is formatted, look up the Generic dash stream in the help file. Most commonly in OEM applications CAN messages are simple single frame messages, what that means is every frame is sent with a unique ID. A frame is usually 8 bytes of data. In motorsports it is more common however to use what is called sequential or compound messages so you can send multiple frames with just 1 ID. The Generic dash/MXS stream uses the compound message technique. In this stream, there are 13 frames of data sent using just 1 ID. To allow the receiving device to know which frame is which once it gets to the other end, you use one of the data bytes in each frame as an index or identifier. To confuse matters many manufacturers also use different terminolgy to name all these parts - the formats, ID's and even how bits and bytes are numbed, etc so it often can be a bit of a challenge to understand for a noob at first glance but it is actually quite an easy communication technology once you know the basics. Most good software nowadays has some kind of visual representation of the message which makes things easier. Some examples of the differences I will show below. AIM call a compound message "enable row counter", some manufacturers call it a multiplexed message. AIM call the identifier/index byte a row counter. You will also see in the pic below their bit numbering is done differently to Link's too. I have attached a configurable version of the AIM stream for the ECU and the dash. To see this in the dash you import it into the "CAN Builder" which is the seventh button from the top left in RS3. Edit, added later: I forgot to mention byte ordering. A byte can represent a decimal number from 0-255. Stuff like ECT for instance that only goes from 0-120C can use just 1 byte. If you need to transmit a number bigger than that then you need to spread it over more than one byte. 2 Bytes allows 0-65535 so is enough for most automotive stuff. When you have more than 1 byte then there are two ways you can place those bytes into the frame, you can have the big byte first (called Most Significant first, Big endian, or Motorola format) or you can have it last, which is refered to as Least Significant first, Little endian or Intel format. The Link MXS stream uses LSB, you can see on the left side of my screenshot above, AIM call this "Low to High". LINK_Generic Custom_@20170524_021339_007952.xc1 Configurable Generic Stream MXS.lcs
  11. Adamw

    Mazda MX5 VVT

    Actually looking at it again I was party wrong in my last post, the NC uses the Mazda 3 or suzuki modes. Your engine should use the MX5 VVTI trigger mode and the offset should be around zero. 4 teeth on the crank and 3 teeth on the cam.
  12. Adamw

    Knock Audio

    If you open up windows sounds is your tuner nerd showing up with a green check mark and as a default device like below?
  13. Adamw

    Mazda MX5 VVT

    The trigger offset is not known! Hope that helps. There are potentially 2 variations of the NC trigger pattern. One uses the "MX5 VVTI" pattern as expected, offset is normally around 0 (but you should always confirm with a timing light). The other has the cam signal inverted and uses the Suzuki swift V2 triggermode, the offset is unknown for that engine.
  14. For the EGT's you will need a thermocouple amplifier, I would suggest a CAN bus based one. The ECU master EGT to CAN is relatively cheap and seems to work well. The EGT amp and the dash can both connect to CAN 2 in the main B plug so you will just need a couple of superseal 1.0 sockets for that. I also suggest you query Boomslang about how there loom takes care of the high current TPS signal for the GTR 4wd computer, our GTR plug-in has an extra amplifier circuit added to the adapter board to handle that, a wire-in ECU obviously wont have that. I know haltech sell a seperate black box to do this task but only for the R34, the R32/33 uses a different voltage range so that wont work. I can give you a circuit to make something, but it should really be their job if they are selling a plug & play harness.
  15. Adamw

    Coil pack dwell settings

    Ok, sounds like you have it under control then. If it fouls a plug again after you replace it then you could try a little more dwell, it might need a little more to "burn through" the contamination.
  16. Adamw

    Coil pack dwell settings

    Is it running sequential or wasted spark? I have heard that particular coil doesnt like being run wasted spark and will overheat quickly if run that way. I have used one with good success but it was on a relatively low powered engine so it wasnt working very hard. I dont remember what dwell I used, but I would start around 2ms and only give it more if it misfires.
  17. It sounds like you are on the right track, gear shift control is probably the best way to handle this. Although gear shift control will always give a fixed amount of torque reduction. A possible improvement may be to add a 4D ign table also to add some variable retard based on TP or boost or similar. We need a little more detail to help though. You mention no torque reduction is needed below 40%TP, and you need torque reduction above 85%TP, but what do you want to happen between 40& 85%? Also, you you update to the latest firmware, there have been a number of significant improvements since 5.6.4 that you are running. Knock control processing speed was improved significantly, CLL now has 3D tables for the trims, more virtual auxes etc.
  18. Note there was a mistake in the pinout in the older versions of PC Link. It is correct in the lastest version. Ign 1 = A21&22, Ign 2 = A18, Ign3 = A14, Ign4 = A10. For Ign1 you can use either of the pins. Correct. Correct. If the new sensors are hall effect then you will also need to change that setting and probably turn the pull-ups on.
  19. Yes. Yes, splice the sensor ground out to as many sensors as you like.
  20. Correct, pull-up voltage is not a problem, as long as it is more than 1.5V the ECU will see the signal. The sensor doesnt care so much about voltage on the signal pin, just current.
  21. Adamw

    Which ECU?

    Yes, any of the ECU's will be able to handle that input and it sounds like you could probably also use the ECU to do the job of the SGI5E. Note the Xtreme will do a V8 in Sequential and Direct spark also. Just the speed inputs will be your consideration with that. The Fury with 6 ign drives could still do COP too, they would just be fired in pairs. Yes you can load any G4+ map into any model G4+ ECU. However, it is a fairly basic engine, no VVT, no E-throttle or anything else too extraordinary, it wont be too strenuous to start from scratch if you cant get his file.
  22. Adamw

    Looking into Link Fury

    I agree with Rosso here, The G4+ ECU's have pretty flexible CAN but unless you have a very good understanding of CAN and are capable of sniffing and decoding the original system you are going to have a hard time duplicating it with the Link.
  23. Adamw

    G4 Fury on Rx7 FD3S

    Correct Wiring diagram below. The ECU should control relay #2 & 4. #1 is controlled by the AC unit and #3 is controlled directly by a thermoswitch. I assume G/B means it is either a green and black or gray and black wire that you need to connect to Inj7.
  24. Try these: r31 New CAN.rcg R31 New CAN.pclr
  25. Your throttle control doesnt look bad to me. You might find a small improvement but I dont see much to worry about in your log. A little less P & D and a little more I would probably work better, try p=7, I=0.1 & D=45.
×
×
  • Create New...