Adamw

Do it in volts as I said. You can temporarily set the analog input function to "voltage 0-5V" and the ecu will tell you the voltage, you do not need to measure it with a standalone tool. Fill the tank from empty in some regular volume increment and note the voltage at each volume.

I agree and I have been meaning to arrange that, keep forgetting. I will ask the admin now.

Yes, if the ecu that we are doing the mod to is one of our older ones with few spare anvolt inputs they will also add some flying leads for extra analogs.

Provided the level sensors are 2 wire, then wire them in series, you will then get the average of both sensors. For the setup, set the analog input as a GP input, probably cal 7/8/9/10 is the most suitable, Volts vs %. Empty tank, check the voltage on ANVolt, this is your 0% voltage to input into the cal table axis (my example below is 0.6volt for 0% full). Repeat for other increments such as 25%, 50% etc. You could just do 2 points calibration (empty and full) but it will unlikely be accurate in between.

Can you confirm you have the disc on the right with a single gap. I see DIY autotune have one with 2 gaps for some other nissans and your trigger 2 is showing a pattern more like this wheel would make.

I think you must be doing the triggerscope wrong as none of them show any waveform. Are you clicking the capture button while the engine is cranking (not before)?

To tune the knock system you really need audio knock detection equipment and experience of what to listen for. Once you have the threshold somewhere close to following the engine noise profile, then you have to purposely make it knock to make sure it is detecting the knock and you also need to make sure it is not detecting knock when there is none. When you have audio knock gear then you would usually use that for knock detection when tuning the ignition table anyway - rather than just using it to set up and test the onboard knock control system then change to using the onboard knock for tuning the ign. The experienced human ear with decent audio equipment is typically pretty good at discriminating knock from regular background noise with no setup, where as an electronic system will need significant tuning and testing before you can rely on it. A bit more info in reference to your comment "Is the stock sensor reliable/sensitive enough to watch a time plot of the engine noise/knock and adjust timing based on that?" Unless you are using a fairly low octane fuel I wouldnt expect a NA gen 1 1UZ to be "knock limited" - in other words you will reach the point of best ignition timing before knock occurs. So for example lets say under cruise conditions it may make best torque with say 43deg advance, but if you had knock detection equipment you might find you need to advance as far as say 55-60deg before knock occurs - so if you used "knock" as the metric for tuning the ignition table then you are likely going to end up well beyond peak torque values and probably no better off than being 10deg too retarded but well away from the knock limit. Knock detection is definitely a nice safety tool to have while tuning, but on its own it wont give you any indication of where your ign timing needs to be.

News from software team: "I think he can resolve this by deleting the settings file, settings.json in the appdata folder. its just up one folder from where the logs are". Unfortunately I think that means you will lose any non-default settings and the layout if you havent exported it or saved it in a PDM config.

Ill just pass this on to one of the software guys for advice.

No you cant really use the knock sensor for tuning ignition, the knock control system needs to be set up and tuned before it will be capable of reliably detecting knock.

In the early days of G4X we did have a conversion tool, but there has been so many new features and strategy changes since it is no longer useful. It is best to open up both PC Links side by side and start at the top of the settings tree then work your way down a folder at a time. As KOracing suggests the import/export to clipboard function is useful. The only tables that dont transfer well is some of the custom cal tables, these need the axis set up to match before importing.

Are you sure? Our wiring diagram shows the injectors are wired to 6 individual ECU pins originally - even on the older distributor models. I dont know what thread you are talking about where I suggested mulipoint group - but that suggestion may have been due to something like an atom or old storm ecu being used that only had 4 injector drives so couldnt do sequential. But if you are sure yours are wired in 3's then yes use multipoint group mode rather than sequential. Did you actually check the timing with a timing light? Can you give us a log of it cranking.

Yes I understand how the ecu works. At the instant prior to you turning CLL off you have the CLL pegged at the max clamp adding 15% extra fuel than is commanded by the fuel table. At the instant you turn off CLL that 15% extra fuel is removed and you go back to the PW that is commanded by the fuel table. Hence my comment the "ECU pulls 15% out when you turn CLL off". Im not saying it is removing 15% compared to the fuel table - Im saying you are getting 15% less fuel after that time point you turn it off compared to before. The O2 sensor voltage starts to slowly creep back towards normal after that - it takes nearly a full minute to reach a stoich voltage whereas a narrowband response is usually quite fast to mixture changes if the sensor is working correctly. Maybe it is carbon fouled at that point and it takes a little time for the carbon to burn off after the excess fuel is turned off? I dont know what is causing the lean reading and slow sensor response, but the ECU is doing everything it is programmed to do from the data I can see.

SPWA can be positive or negative. What you are telling the ecu is how much longer you need to open the injector for to acheive the quoted volume/flow. Below is an example. I have ploted two lines on the graph of flow rate Vs effective PW. Blue "ideal flow" is what the injector should flow if it were perfectly linear (straight line). The orange line is what it actually flows in real life when tested on the bench. You can see for example at 1.00ms PW, the measured flow line is above the ideal flow line (injector is flowing more fuel than it should), so in this case the ECU would have to reduce the PW to get the ideal flow (negative SPWA). Then at 0.50ms the measured flow is below the ideal flow, so at this PW the ecu would have to increase the PW to achieve the ideal flow (positive SPWA).

The ECU pulls 15% fuel out when you turn off the CLL, so possibly the rich misfire disappears so the sensor starts reading correctly again. Try reducing the dither voltage or the trim limit so it cant go so rich.

You need to use 1FZ trigger mode, then check the base timing once changed. If still no go then add or subtract 360deg from the offset. I have never worked on one so I cant tell you what the offset is. In fuel settings, Zero out the short PW adder table and set injection mode to sequential. Change fuel system type to MAP referenced.

Im not sure what your question is? The reason the CLL is adding max fuel and reaching the max clamp is the oxy sensor is sitting at 0.1V all the time, the ecu has added 15% extra fuel but the oxy voltage output hasnt changed. You can see the effective PW has increased by 15% so the ecu appears to be doing everything its meant to be. So why the low voltage from the oxy sensor? - I dont have a lot of experience with narrowband sensors but I guess it could be something like an air leak or small misfire so there is excess oxygen present - or there is some problem with the sensor temperature/heater power supply or possibly the sensor has been poisoned with silicone or coolant.

There is no terminating resistor in this case (well only one inside the ecu). For a single CAN lambda this works as there is little traffic on the bus and the bus is generally short so there is little chance of reflections causing errors. With little traffic, even if there are errors there is instant opportunity to resend. Here is a note from the CAN manual:

This is not an engine Im intimately familiar with, but likely the "Monsoon Honda K20 Modelled Fuel CLL" map in the PC Link Basemaps folder will have a lot of the basics needed. Connect exhaust cam sensor to trig 2 and intake cam to a DI.

TPS is always best for accel enrichment. Firstly turn on asynchronous injection with a min PW of 1.0ms, this will reduce the amount of accel enrichment needed. Start with something like below. This is from an engine using modelled fuel equation so the cold correction is all 1. If you are using traditional fuel equation you will often need more cold correction for a cold engine. Tune it warm initially anyhow and add cold correction later only if needed (flat spots etc when cold). Check with the engine idling what your TPS delta is and set your deadband a little above this. You can add the parameter "Accel Fuel" to a time plot to see when and how much accel fuel is being added. If you see accel fuel being added at idle or other steady state conditions then increase the deadband.

Yep, this used to be done quite commonly. MSD white wire to Ign 1, Ign mode set to distributor, spark edge set to rising. Im pretty sure most of the current "dual sync" distributors are hall effect. They also make what they call "cam sync" distributors and most of these seem to be reluctor.

Yes those coils have built in ignitors so they can be controlled directly by the ecu.

Try these, done in a bit of a rush so may not have it all correct yet. Also, I found a bug in labelling the keypad buttons so button 10/11/12 inherit some of button 1 label for some reason and cant be changed. So the exhaust valve control function may look a bit confusing since they are using "starter button count" Sacha V2.pdmc Sacha V2.pclr

You can just unplug or turn off the cam sensor once running if you think that is the issue. Im pretty sure we already tried that earlier on but cant remember.

Can you give us some fresh logs of it cranking, a trigger scope and a capture from the standalone scope if you can. What symptoms do you have?