Jump to content

cj

Members
  • Content Count

    192
  • Joined

  • Last visited

  • Days Won

    17

Everything posted by cj

  1. cj

    G4+ Fury first tune 2gr-fe Turbo 92 MR2

    under ignotion main, change spark edge from falling to rising. Typically only honda coils and a couple other weird ones run falling edge coil trigger. The majority of coils work with rising edge. This should fix the coil temp and the failing to fire.
  2. cj

    Fury - What happens during idle?

    Oh one more thing, are you running E85 in this engine? you've defined the stoich ratio for your fuel as 9.8 but it should be 14.7 for petrol. This wont be hurting anything yet, but might once you start using closed loop systems.
  3. cj

    Fury - What happens during idle?

    Your ignition table is in MGP, and doesnt go into negative. This means your ignition timing at anything non-boosted is effective a single row lookup. I notice it runs pretty minimal vacuum anyway, but you should still have this drop down to at least -30kpa as it seems your engine would use this at idle. Your fuel table numbers are very small, and your master fuel number is quite large - this doesnt "break" anything as such but it makes it very hard to get any resolution in your fuel table. eg you can only move from 12.0 to 12.1 and this might be quite a jump in injected fuel quantity. Try changing master fuel trim to 0, master fuel to 5, and then select the entire fuel table and type *4 <enter>. this drop your master fuel value to ~1/4 of what it is now, and multiples the table values by 4. The injected quantity will be the same but its easier to work with the fuel table now. It looks like the reason it stalls is because it runs too lean. healthy idle is probably 0.9 lambda and 1.1 is probably stall. Post start enrichment is keeping you going to start with, and when it runs out, your engine goes lean and stalls. Try bumping up your master fuel table a little bit (maybe from 5 to 6 once you do the step above) until it can keep running once post start enrich goes to 0. Then once you know what "stable" idle looks like, you can reverse engineer the post start enrichment numbers to stop the idle creeping up at the beginning but this temporary high idle is not your biggest problem right now.
  4. cj

    MR2LINK G4 V2 V3 pinout

    The intercooler fan is controlled by its own controller - the other smaller box next to the ECU. You'll see there is a temp sensor on the underside of the rear engine cover that also wires to this cooling ecu. The taillight relay and stop light switch are just for primitive electrical load detection. Properly tuned closed loop idle or ignition idle control means they arent needed. There are a couple posts in this thread around the EHPS control on SW20 pumps. short version is it will work pretty well with no ECU inputs at all, but you can add them for additional control if you want. the vss signal that adds/removes assistance is comes from the dash (as long as you still have a factory dash)
  5. cj

    EZ30D JDM swap into Subaru 2.5rs

    Ah ok, the last one I had to fight with the fuel system wiring on was a legacy but they are usually pretty similar. Here are the relevant pages from the service manual relating to fuel gauge First thing i'd point out is that with you only finding 2 wires on the module you're looking at the sub module. so working backwards through that second diagram, BlackYellow is ground. As long as you can confirm ground on this wire where it meets the primary level sender this is not complicated. RedYellow is the low fuel level light, I believe this goes to ground when the level is low, but it doesnt participate in the level gauge anyway so not relevant to this issue. This leaves BrownWhite as the only wire leaving the sender pair. Blue is simply a join between the 2 and simply needs to be checked for continuity. Thinking about this, the way to verify whether the 5v pullup is coming from the ECU or from that "custom CPU" in the gauge cluster, is to: 1) verify the wiring to ground and between the 2 fuel senders. with the ECU and gauge disconnected you should be able to measure a resistance between the BrW wire where it would plug into and ground, and this resistance should change if you push the level sensors up and down. This confirms your wiring through the level sensors. 2) Disconnect all the plugs from the gauge cluster Verify that the relevant pins on the plugs show 12v and ground to confirm that the dash is "powered up" 3) disconnect the ECU from that circuit but connect the dash. Now with the key off you should see exaclty the same behaviour as above. With the key on, if you see the same 0V and varying resistance to Ground, then the ECU needs to provide the 5v pullup. If you see 0-5V ish as per the ECU function for pin B136-27 above, then the gauge is providing 5v pullup internally and your problem is either in the dash or on the main power/ground feeds into the dash. I also noted while checking these over that the fuel level senders and the speed sensor share some of the same wiring on the power/ground feeds, so maybe your problem is as simple as a fuse or ground wire not connected? quick check with a multimeter with key on will tell you if you see 12v bewteen the 2 non-signal pins of the speed sensor.
  6. cj

    EZ30D JDM swap into Subaru 2.5rs

    What model chassis are you using here? Some of the newer ones had a fuel pump controller and/or had the pump and the primary sender as one unit.
  7. cj

    EZ30D JDM swap into Subaru 2.5rs

    The fuel senders on most subaru's take 5v from the ECU, but the output from the sender to the gauge. Think of the senders like a pull-down to ground based on variable resistance. There are also 2 fuel sender modules connected in series and if the second one is disconnected, you will have no ground connection. join your 5v regulated output onto the BrB (BrW in your case) wire between the gauge and the sender and it should work. The speed sensor probably does the same thing and expects a 5v pullup somewhere in the circuit. Try wiring this into a DI on the ECU and setting pullup to ON. will fix your speedo + give you vehicle speed into the ECU. In the below diagram, C goes to ground.
  8. cj

    Temp signal issue

    Youve got some interesting electrical issues going on there. Way more than just the temp sensors. You battery voltage comes and goes. between 1:10 and 1:14 it drops to 0 half a dozen times, usually right when your temp sensors go into their minimum value of -40C. there is also a quite worrying spike up to 40.9V battery voltage when it comes back, and when this happens your sensor all add about 50degC to their readings so that voltage spike is clearly reaching the sensors. I'm guessing it arced a bit when reconnecting to voltage. I suspect your temp sensor variances are directly related to loose wiring, probably near something major given how many of the systems it hits an once - eg the ground on the engine block or the battery, or somewhere between your battery and the main fuse box. Its also possible you have some exposed wiring somewhere un-fused (like the wiring to the starter motor) and its contacting something and causing an intermittent short. Your lambda sensor drops in an out, and sometimes when its working its giving you insane values eg at 1:16 you apparently were at full throttle when your lambda went from 0.8's to a value of 8.something! (not afr, lambda...). On second look some of the dropouts are caused by voltage drops and it going through the startup process again, but the spikes up to very lean numbers while at full throttle dont make sense. At 1:07 it looks like your injectors cut out or something. You're still at full throttle but it goes crazy lean and your MAP + revs start to drop. are while not a wiring issue, you fuel table values are maxing out at 150 for large parts of the full throttle run. If this is on modelled, you need to check your injector flow rate config, that the configured fuel pressure matches the pressure set on the regulator, and that the correct engine capacity is configured. If its traditional, you probably need to double your master fuel number, and halve all the values in the fuel table (select the whole table and type " / 2 <enter>" Also, you very lumpy idle is probably related to your fuel table numbers near idle being so different. it moves from ~50 to ~90 between 1000 and 1500rpm and only moving about 20kpa.
  9. cj

    Nissan z32/rb25 ECT calibration?

    if you leave it overnight, does it show a value that seems within a couple deg of ambient? 70C isnt completely out of the ballpark if you have lots of coolant to heat up or a cold day or your fans are on etc, or if you've got something funny going on with your thermostat or the placement of the sensor.
  10. cj

    Lean warm/hot restart

    its off because "sensible" values for IAT fuel enrichment are already used as part of the modelled calculation. You *can* turn it on if you have a reason to do something "non-sensible", such as completely opposite of normal like add fuel at high IAT value at specific throttle opening, and just set the majority of the table to 0 except the points you want to change. The modelled fuel behaviour is still going on in the background. Think of it like an additional overlay table which you dont normally need.
  11. cj

    Fuel Table Cell Numbers

    Its just so you have some resolution to play with. There is no "magic" to having 50's in the middle. In traditional mode the master fuel value is how long the injectors open for at 100kpa if the fuel value is 100. Any value in the 100kpa line is basically what percentage of the master value is going to be used. master fuel trim is bascially just there to let to put decimal points into the master fuel number. ie you cant enter 10.5ms, but you can enter 10ms, and then 5% fuel trim. So 10ms and 10% fuel trim is the same as just entering 11ms and 0% fuel trim. If you're tuning from scratch I dont see a reason you'd use it. More likely it would be if you put in new injectors that were a few % off your old ones, or new engine parts that changed your VE just a fraction but equally across the whole fuel table
  12. cj

    Blown engine!

    Have you got some pics of the side of the pistons as well? It doesnt look like the sides are scratched up which is where you'd normally expect to see damage, just the front/back which is weird. Is the damage to them all on the fronts, or is it equal front and back. Is it all the pistons or just one? Have you got the crank out too? How do the thrust bearings look - aka does it look like the whole crank has walked forward/back?
  13. cj

    2JZ Electric Power Steering Pump

    The pumps come with a controller. I cant remember if its inbuilt or in a separate module right next to the pump but there is certainly a controller, it takes in voltage, ground, vss, and a signal the toyota manuals call PSCT which as far as I can figure stands for Power Steer Cutout. I've got an SW20 mr2 wired up to a fury and have it so the main power feed to the steering is via the factory relay (which may count as a "controller" anyway in that its not direct battery fed/not fed when key is off), plus a factory VSS input from the dash, plus I have the PSCT wire hooked up to the ECU and configured as below This means it doesnt provide assist when the engine is off, or above 20kph (the idea is to have it behave as a manual rack at anything above carpark speeds). Works a treat, and you can demo to the certifier that it must be "controlled" by switching that aux on/off in front of him and having the assist cut out. Only minor gremlin is that because of the way link ECU's seem to bring the outputs online, it very briefly ground's that output and applies assist at key on (for a fraction of a second) and you can hear the pump power up then down again. If this was deal breaker it would be simple enough to wire the aux out to a relay instead of straight to the controller
  14. cj

    KA24DE G4+ Atom ignition concern/question

    Can you post a pclr file as well? Are the injectors and fuel system stock? First suggestion would be have you tried giving it some throttle pedal while starting? Your MAP value drops by 5 or 6 kpa during cranking which is normal, but when you stop cranking, it's only very slowly recovering. Holding vacuum like this suggests that maybe your idle valve is blocked off or not working, and you're getting no air in at all. If these are stock injectors, you're running pretty low pulse width for cold start. Do you have any kind of crank enrichment turned on? your battery is a bit low but given it can read steady RPM its more a "good idea" to replace it than a certain fix. That dwell thing is a bit weird. without a pclr file I cant really guess at why its doing that and neither 0.2ms or 40ms seem like numbers I would expect to see in most cases.
  15. cj

    Fuel pressure sensor

    If you watch that frame by frame the value is jumping between about 48 and mid 50's. These seem like reasonable numbers but depending on your fuel system design and regulator you may disagree. You're looking at absolute pressure, so if you have the vacuum line hooked up to the regulator you should except to see a few psi fluctuation due to pressure variances through the intake. If you look at the logging screen with the graphs instead of the instant values it will be a lot easier to see trends, history and general "how much it moves about"
  16. cj

    0.18 hertz lambda oscillations

    That grounding isnt good, you've effectively joined the sensor ground to the battery negative, which will make ground offset problems like you quite possibly have worse. Try grounding it directly to the engine block at the same point as your main ECU grounds, and disconnect it from the sensor ground. The LC2 manual says it can draw up to 3A, which I'm sure is more than you can safely put through the sensor ground wiring so dont even try (cant remember the actual safe number right now). if you want to try something much easier but less likely to actually fix the problem, some engines create an EGR effect using VVT advance at cruise-ish MAP values. any EGR effect will cause lambda changes even with constant air+fuel input. You are seeing 2-3 degrees of VVT advance that has a very loose correlation with the LH VVT solenoid position Try setting this to 0 in the 10% TPS row at 1000-2000rpm just to rule it out as its much easier to test that re-wiring your lambda sensor, but you'll probably have to do that anyway.
  17. cj

    WRXLink V3-4 Cut Issues

    your pclr file shows 260 MAP limit, but your log file shows that it was using a 245 limit when you captured that log. Did you remember to store the config to the ECU after you made the changes (press F4 or ECU controls>store to ECU)? If you dont do this, and just save the map to your laptop, the ECU itself will go back to the old config on poweroff/on.
  18. cj

    Blown engine!

    Can you post that log here so we can take a look? The injector wouldnt cut out from a voltage drop. it *might* run marginally leaner while the voltage was lower. If the dead times were set even close to correct you wouldnt see much change at 6000rpm. eg if your injector open time ms is at 20ms, and you have 1ms dead time @ 14v, and you drop to 12v where you have should have 2ms dead time , but you still only have 1ms because of incorrect settings, your fuelling would drop by around 5%. This should be noticeable in the logs.
  19. cj

    Blown engine!

    it would make sense for it to run lean if it dropped to ~12v and your injector dead times were badly off for that voltage, and that might cause knock, but you would see the whole log go lean, and at high RPM dead times dont make that much difference. Knock control probably would have seen it too. If the log shows good AFR, and it also shows the timing to be sensible values, then i'd be looking at a mechanical failure in the engine. Oil and coolant level ok? head gasket maybe?
  20. cj

    WRXLink V3-4 Cut Issues

    Can you post your pclr file too? At 13:04 is the only point you hit MAP limit, and yes its still showing as 245 so i'd guess you've only adjusted it for certain ECT's. At a few other points (eg 12:36) your MAP appears to flatten out a bit at 230-ish but you arent hitting the limit - even if the limit event isnt in the log it would still have the -5 deg ingition trim active for a few seconds and we only see this at 13:04. At both these points you're at 100% TPS and the ECU is requesting 70.5% DC from the wastegate. This lines up with what you said earlier that some days its ok, and other days you get MAP limit behaviour. I'd say your tune (specifically the boost control settings) puts you very close to your MAP limit all the time (230kpa ish), and when something is a little bit different (air temp/atmospheric pressure, oil temp, heat in the metal of the turbo, maybe the gear you are in or the amount of weight in the car), it causes you to get 5 extra kpa and hit the limit. At 3:26, and everywhere else you are at light throttle, you are running _very_ lean. Anywhere in the log where you are over ~50% throttle the lambda value is good, and at idle its ok as well (a bit rich but a reasonable value, and the richness might be needed depending on engine mods). It looks like the fuel table at light load hasnt been well tuned. In fact if you swap your originally posted map to graphical view you can see the big dip in fuel values between idle and about 110kpa. In the pic below you can see the valley right along the -40kpa load cells, and pretty much everywhere in the red square looks like it needs to get bumped up by 3 or 4 points.
  21. cj

    Blown engine!

    sounds VERY unlikely. Are you sure you have cause and effect the right way around? several things failing in an engine cause cause the alternator to stop, or an electrical fault may have caused a voltage drop and something that led to engine failure. Do you know why you had a voltage drop in the first place? What type of engine failure, what type of engine, and what RPM were you at when it failed?
  22. cj

    RX8 DBW throttle pinout

    Foot position (accel pedal sensor) in each diagram, the top wire is 5v, middle is signal, bottom is sensor ground. TPS wiring. A&B are not listed as having a function here, and are the ethrottle +/- drives. +5v and sensor ground wires are shared between both sensors on the TPS.
  23. cj

    G4+ Fury first tune 2gr-fe Turbo 92 MR2

    Heres one ive made a few change to. Changes are: changed fuelling from multi-point group to sequential change engine cc rating from 3500 to 3456 (as per 2grfe spec) changed fuel pressure calculation to use actual FP sensor instead of MAP reference as it looks like you will be wiring up an FP sensor disable IAT correction - modelled fuel mode handles this anyway. WIth the turbo setup you may end up turning this back on later once you have boost + high IAT's to deal with. set your fuel table to the one from the monsoon base MAP as a good starting point for modelled mode, previously you had what looked like a traditional fuel table that wouldnt have worked well. changed trigger2 edge to falling as per link 2grfe help guidelines. There is nothing in the help files about trigger2 pullup on/off so if you have starting issues, try changing this and get a trigger scope to confirm what the voltage is doing. changed base timing trigger offset to 0 as per link guide - you should run the base timing procedure with a timing light & fuel+ign disabled to confirm this before starting however. Changed your tacho multiplier to 0.75 - this is the correction for if you still have a factory 4 cylinder tacho but a 6 cylinder engine, and assumes that if its a pre-beams dash you've done the resistor swap on the tacho itself so it will work with an ECU based tacho signal otherwise your tacho wont work at all. removed the +4% park/neutral throttle trim. Note that in ethrottle configs all percentage trims are added to the requested throttle opening percentage - ie 2.5% from target table + 4% trim = 6.5%, NOT 2.5% + 4% of 2.5 = 2.6%. Changed you idle base % table to sensible numbers for the above reason - 30% that was in there would have given you 5000rpm idle or something. This table is also where you do the majority of your tweaking of idle being a few hundred too high or low at certain temps. I havent touched the fuel injector settings as it sounds like you have the info for the already. Check if the manufacturer has a short pulse width adder table as well which can really help with idle/low load fuelling in modelled mode. Also note that your VVT target tables are both blank so VVT will appear to not work. I dont have example numbers for a 2GR but typically you'd start with what's in the OEM map if you have nothing else to use. your 2 exhaust cams (digital input 1&3) have copied the 170 VVT offset from the intake cam. This is likely wrong but wont stop it from running. once you have it running, run the CAM angle test to find correct values for these. Not a "start" problem, but is DI6 a speedo input? you need to set this to LR or RR wheel speed rather than GP speed. If you leave it on GP you can see speed in the logs if you look at this input, but it wont pass through to the other functions in the ECU (speed lockouts on idle etc) As previously mentioned, your CAL1 table for your oil and fuel sensor is incorrect and these 2 sensors wont read correctly. Correct numbers should be roughly 0.5 to 4.5 volts input, and the KPA output going from probably 0kpa up to the max limit of the sensor. You need to set driven wheel speed source to match the wheel you set DI6 to be, then set acceleration source to be driven wheel speed. (under chassis > speed sources) 1544485703_2GR-FETest9-28-18-modified.pclr
  24. cj

    Temp sensor inputs

    Yes you can, you just need to wire in a pullup resistor to 5v so that the AnVolt input sees between 0-5V depending on the resistance of the sensor.
  25. Have you tried turning on pullup resistor? Normally there is a 5v pullup signal coming from the dash, but if you have disconnected this you will need to supply a pullup signal from the ECU. Here is a speed sensor diagram for a late 90's legacy. The colours may be wrong but the pinouts are likely the same on most subaru's until they went to ABS driven speedo's in the mid 2000's. Is this how you have yours wired up?
×