Jump to content

DerekAE86

Members
  • Posts

    423
  • Joined

  • Last visited

  • Days Won

    11

Everything posted by DerekAE86

  1. Yeah being able to select how the Disable Input functions would be great. I had also initially hoped the Disable Input would just disable the updating of the table until that condition is met, rather than disabling the table being applied to the fuel calcuation also.
  2. Like Kris has said use a analog input. With your pullup resistor going to the ECU's sensor 5v reference. This method worked pretty well. However just be aware that the first 200 seconds of the "unfilered" result is _very_ sensitive. Just sitting in the car, or wiggling around adjusting your position makes the fuel level jump around significantly. Even filling up with fuel the sloshing in the tank takes ages to settle down. Felt a bit self concious sitting at the petrol station putting 5l in at a time to build a good calibration curve and having to wait a min or so each time for it to stablize hahaha.
  3. That would be between the Link CAN Lambda module and the Sensor itself yeah? Because it if there was a wiring issue between the CAN Lambda module and the body it's unlikely the module would detect or report back errors at all. I still get Lambda Temp reported back and it still appears to be heating correctly (ramps up to 780ish quickly) so it doesn't look like a supply voltage/current issue.
  4. Ah ok. So this is just a internal failure of the sensor and nothing external in the wiring/power supply?
  5. The included sensor in the kit died today giving Error 54 / Excessive Pump Current. The reading would be around a Lambda of 10.000 if it read at all. But looking at the log it was also giving Error 24 sometimes. Was just curious what that error indicades? The engine has been consuming a bit of oil so I'm about to rebuild it with new rings/valve stem seals etc. So my guess is excess oil in the exhaust would have killed the sensor? Though some insight on Error 24 would be nice just to make sure I'm on the right track and I don't have some other issue lurking.
  6. Thanks Vaughan, that's exactly what I was after.
  7. I've wired in a fuel sender via a 100ohm pull up to 5v. I didn't put any capacitor in the circuit because I was planning on smoothing it via a mathblock. I used av(a,200) so it averages the value over a decently long time. However I noticed that when the car is first started the GP Output starts at 0 before jumping up to the actual value (assume theres a brief period before the 5v kicks in?) This means the fuel level starts at 0 and slowly ramps to the actual level. Is there a way to put a delay in the initial check so the value starts at what the actual level is?
  8. DerekAE86

    Cosmo 20b

    I'm not familiar with it at all. But cars of this generation are typically pretty basic so it's easy to completely replace the ECU. If there is any other module or feature that relies on an output from the factory ECU, then as long as that signal is known the Link should be able to simulate it.
  9. I wonder if engines with higher compression can overcome this a little more? Or perhaps adjusting the timing along with the lambda target could help?
  10. I've been toying with the idea to setting up a Lambda offset for cruise condition to lean out the engine. Something like if the RPM is within +/-200 rpm of the cruise zone with TPS delta is less than a certain amount for around 30s for example it turn on the Lambda offset table. I know there are some manufactures that do this already. Apparently Mazda's Skyactiv leans out to 65:1 which is like 4.400 Lambda - But I assume the engine is specifically designed with this in mind. Has anyone played with lean burn for low load conditions before? Is there any tips for determining what your engine will safely achieve? I assume it'll be influnced by many factors; na/turbo, compression ratio, port/direct injection, how efficient your cooling is?
  11. DerekAE86

    Cosmo 20b

    Why would you want to? Link ECUs are a full replacement to a factory ECU, not a "piggy back".
  12. https://www.blocklayer.com/rpm-gear You can fill out all your gear ratio/final drive/tyre size information here and then calculate the rpm/speed to ratio to use.
  13. Depends on how far you want to go. I'd go for the MonsoonX since it'll cover everything you need + a little extra for future expandability. But anything above that will just give you more options and freedom to really play and add more features, inputs, outputs and modern things like DBW, etc. https://linkecu.com/which-ecu/
  14. Attach a copy of your tune so we can see how you've configured it
  15. What display? And how are you planning on outputting an AN or DI to it?
  16. I assume your display is a CAN display? Why not just send the Lambda parameter directly to it?
  17. Your sequence of events is lacking. Try the following; Key on to IGN Plug USB in Open PCLink Connect to the ECU (F3) Confirm it is connected by pushing the accel pedal and seeing the TPS value in PCLink reflect this Make an obvious but inconsequential change to the tune file. (EG. change the Tach output max RPM to 1000rpm higher than it currently is) File -> Save to save the tune file to your Laptop (give it a unique name so you can identify this tune file later) Store the tune to the ECU (F4) Disconnect from the ECU (F3) Close PCLink & Unplug USB Key to off Now turn the Key back to IGN Plug USB in Open PCLink Connect to the ECU (F3) Confirm the change you made is still present File -> Compare to open the tune file you just saved and compare it to what's currently in the ECU This will confirm the file you saved is the same as what's in the ECU. Just to make sure you're aware; when you connect to the ECU PCLink will display the tune data that is currently loaded inside the ECU. There is no need to "Open" a tune that has previously be saved on the laptop unless you want to overwrite the tune in the ECU with the file you're opening.
  18. All that "Saving" does is save the tune file to your Laptop so you can view or edit it offline. To make sure the tune settings are written to the ECU you must Store to ECU while the ECU is powered and connected. It may not have successfully Stored if you accidently power cycled the ECU by turning the key to ACC or maybe the USB cable being used it too long/poor quality and the ECU is losing connection to the Laptop.
  19. You are cranking the engine over while doing this, yes?
  20. With the ITB/Quad Throttle setups without an Idle Valve you'll have to adjust the throttle stop to keep the butterflies slightly open so they can bypass air. It's a very fiddly process to really dial in because you'll have to balance this between the throttle stop adjustment and the air bypass screw adjustment (the screw that sits on the mating face for the trumpet). Honestly I would recommend just getting an Idle Valve if you have enough outputs to run one. It will make the entire process much nicer. Also if you haven't got the engine running yet I would suggest re-checking your timing offset etc once you get the car to a point it will idle.
  21. Remember to store to ECU after you've made changes to which parameters get logged.
  22. In theory this should work for sending the RPM signal when the engine is actually running, then nothing when the engine isn't running: Setup on a Transmit User Stream @ 500kbit/s 20hz rate on ID 708. Also if the speed doesn't work, you can try sending speed via ID 176 and 178 (@ 500kbit/s 10hz) with the following for both: 176 has Front R and Front L, 178 has Rear R and Rear L. This is the ABS wheel speed data before it's processed by other modules.
  23. Would it be possible to get a copy of this spreadsheet? I'd love to fiddle to get a better understanding of how the VE fuel equation spits out the final ms value.
  24. I've been tuning mine with the source as BAP and tweaking here n there with a 4D table for Idle Valve% to help. I would like to try changing the source to MAP but I assume that would require quite the adjustment to the main fuel table. Do you have any tips that I could use to get it in the ballpark? I assume there wouldn't be any difference to the fuel table for high TPS loads since you'd be reading pretty much atmophere already. But the Idle/Low load ranges would change quite a bit. I think I read that having source=MAP the fuel doubles for a doubling of the manifold pressure? So if it's already correct for 100kPa full load, at 50kPa idle for example would you halve the current VE%?
  25. I think Adam's approach is certainly the most elegent here since you're utilizing your brake and clutch to do this. But if you do put in a dedicated momentary push button this is how I've done mine: To explain; I ran out of digital inputs so am using an analog input with a pullup resistor and a switch to ground. So 12v when switch is open, gnd/low when switch is pressed. I then output the switch to a GP PWM to allow it to be multipurpose. This lets me have different functions when the engine is in a kill state vs running and different TPS levels. This could be tweaked to whatever use case you want. I'm using engine kill because this forms part of my anti-theft. Then for the latching logic; if the the GP PWM is in the "100%" range for more than 2s the virtual aux goes active. Then if the switch is inactive AND the virtual aux is already active - meaning you can let the switch go and it'll stay active. This means as soon as you touch the switch again it breaks the AND logic instantly and unlatches it.
×
×
  • Create New...