Adamw

This is how I normally look at it:
 
Logged values list.mp4

For everyones elses benifit, I have tried helping this user for several days last week via the tech support ticket system.  
If you overlay a "good run" with a "bad run" and line up a point where the RPM & MAP is the same, you can see the injector PW is also the same.  RPM & MAP is a good indicator of air flow.  So, we can say the airflow is the same, the ECU is commanding the same injector PW, if air and fuel flow is the same, the result should be the same AFR.  The fact it is different would suggest some external influence or injector behavior that is not repeatable. 
For this reason I have suggested the first thing he needs to do is fit a fuel pressure senor and verify the IAT calibration is correct (since this seems very low).  
Overlayed logs below.  Coloured traces are a good run, pink traces are the lean run.

If you dont want to pull the cluster out, then you will either need to use the relay coil or buy a commercially available tacho booster.  I have always used these with good sucess:  https://www.ebay.com/itm/142171581596  Or since you are in Aussie, this is the same device with a different sticker: https://www.justraceparts.com.au/jrp-tacho-booster-signal-adapter
This booster has to be connected between the ecu and the tacho.  The ECU outputs a 12V signal, but the tacho is originally designed for a ~50V signal.

Hi All,
The developers of RealDash have been working with us to make their app compatible with Link ECU's.  It is now compatible.
For those that havent seen it you can learn more here: http://www.realdash.net/
In my short play it seems to work well and has lots of cool features.  Lap timing, navigation, configurable alarms, live graphs, live streaming of data, music player, etc.  The app and a selection of basic "skins" are totally free, then there are some more fuller featured paid for "premium skins" that do more stuff such as data logging and may look cooler these are still relatively cheap ranging about $3-$10.  One example below.
Right now it wont work on the Atom, Monsoon, Kurofune or Black Storm as the serial stream is not enabled in those ECU's but we hope to enable that setting in the next couple of months.  Right now it will work in the Blue Storm, Xtreme, Fury, Thunder, all Plug-ins & Force.  It will also work on old G4/Vipec V series ECU's.
Connection to your tablet/smartphone is direct using your normal Link USB tuning cable, for phones/tablets that only have a micro USB port you will need a "Micro USB OTG adapter".   Instructions for setup and connection attached to bottom of this post.
 
 
Edit 18 Jan 2020;
Have added a zip folder with an example configuration for CAN connection using the generic USB or Bluetooth CAN Adapters (see realdash website for supported adapters).  This config also includes an example of how to send an onscreen button press to the ECU.
 
RealDash.pdf
 
Realdash CAN example.zip

Correct, multitooth position = cam.  trig 2 sync mode = cam pulse 1X.  Offset can be anything since the CAS can be stabbed in in any position.

It wont hurt to do a scope capture when holding at say 5000RPM.
 
No settings would need to change, but you would need to fit coils with built-in ignitors.  

I was replying at the same time as Vaughan, but I will leave my thoughts here too...
Most likely a lack of ignition system energy.  There are no cuts or limiters in the log.  Dwell is quite stable and no recorded trig 1 error so my initial feeling is unlikely a trigger issue.  
But please do a PC log so all data is recorded.  You dont even have MAP, lambda, or anything fuel related recorded in that log so very few clues to work with.
 

Go into the engine fan function and try swapping the active state on each fan one at a time.  Its possibly one of these is wrong.
Also, go to >analog inputs>MAF and set the input to off.  You will then be able to clear fault codes to get rid of the CE lamp.  

Its bouncing off overrun fuel cut because TP is zero but RPM is above the overrun deactivation RPM.
So, its getting too much air from somewhere.  I see the idle valve is sitting at 100% so possibly that is some of it.  Looking at the idle base position table in our Evo9 base map it is a bit messed up with large numbers at the hot end.  Try setting the whole idle base position to something like 50% then do a power cycle so idle valve homes again and then restart and see if it is happier.  

Also, remember for injector timing to have any noticeable effect your injector duty cycle needs to be very low - <35% is typically the point were I start to see little benefit (even at only 35% the injector is open for 250deg crank rotation).  For this reason in typical performance staged injection scenarios the timing of secondary injectors has not been a priority.

Im pretty sure the Nissan CAS shorts the ground pin to the body anyhow, so probably chassis is ok.    

I would say the main problem is the ISC speed lockout is set to "Driven wheel speed", then under speed sources you have the driven wheel speed source set to "Average rear wheel speed", but there are no rear wheel speed sensors assigned, so the speed lockout will never work.  This means closed loop will start winding the throttle closed anytime you are in overrun, then when you push the clutch in the throttle is too closed so the RPM drops and stalls. 
I suggest changing both driven and non-driven wheel speed source to "LF wheel".  Set ISC speed lockout to 10kmh.  A couple other changes I would suggest:
Change idle ign control AP lock out to 0.8%.   I would typically have idle target error on the axis of the idle ign table - not engine speed.  It should work ok as is provided you never want to change idle speed. E-throttle target table will probably work better without the big change in target below 1500RPM.  I would generally have the same value right across the whole top row - unless you have an engine which is slow to return to idle then I sometimes have a smaller target above say 1500rpm.  

Can you explain better what you are wanting to do?  It would be unusual to switch fuel tables for staged injection.  Typically you should think of the fuel table as a representation of the airflow through the engine.  Does airflow change when you switch injectors?

But why do you need a different fuel map when the staged injectors are working?  Is there something else that causes the VE to change?  Or are you using two different fuels or something?

Oh, so you are asking for a timing map for the secondary injectors?  Sorry, that is not possible with G4+.
 

It shouldnt do, but one possibilty is you may hit some cell that wasnt originally tuned so just keep an eye on that.  

You can use the Haltech device.  Note it is just a rebadged PCS unit so it may be cheaper to buy the original:  https://www.powertraincontrolsolutions.com/Performance_Aftermarket/Products/CAN_Modules/EGT_Module/A-EGT2000/  
You can connect it to the existing bus (along with the dash) or to the 6 pin port if you'd rather.  The dash has configurable bit rate so you can set it to match whatever device you choose to go with.
You dont need Haltech probes, any "K Type" thermocouple will work with any of these.  Suggest you get quality ones though, not generics from ebay/amazon/aliexpress.  The Sensor Connection make good ones for reasonably cost. 
Link dont currently offer a TC amp.  Probably the most common one I see Link users go for is the ECUmaster one due to its low cost, the PCS one is a bit more professional with billet case and proper thermocouple connectors.  AEM have a decent one too.  There are quite a few others but they either get a bit costly while not offering any more than those already mentioned or are only 4 channels.  

Fixed up a few things in your map.  Turned on idle ign, fixed the base position table and the E-throttle target. reduced RPM lockoutr and increased deadband, It should hopefully work better. 
The other thing you need to fix to prevent the stall is the wheel speed sensor.  It is not working in your log.  Without this the ecu will start closing the throttle trying to control idle during overrun and this may cause it to stall.   
FEWFEWELTWEEEEKZ idle changes.pclr

So the rear speeds never spike?  I would say most likely the GT101 is not happy with its target, they can be pretty fussy with target shape, size, squareness etc. 
You could try a ZF GS101202 or GS101201 or a Littelfuse 55505 in its place, these are direct replacements for the GT101 and are generally faster and less fussy.  

I just had some spare time to look at this request a bit closer today.  It appears basic Subaru DCCD functionality like some other ECU's offer should be mostly doable with the functions we already have available.  I dont know how well it compares to the OEM control strategy however.
Diff control is simple PWM. 
From what I can gather from a couple of "competitor brand" maps, the vast majority of the duty cycle control comes from a basic 3D table which is Speed Vs TPS.  On top of this main 3D table there are a couple of other conditions (probably only applicable to rally cars) to unlock the diff during handbrake slides and during left foot braking (we already have conditions available for GP PWM outputs but there are other thresholds needed for these rally conditions so we would need to involve a couple of virtual aux's to duplicate these well).  
The only bit that I dont have enough information about (or possibly I just dont understand the logic well) is it appears during braking an extra 2D map comes in to play.  Again i think we could duplicate this reasonably using virtual auxs to switch to a different table.  Neither of the maps I have appear to use any G or yaw sensor, steering or other driver inputs.
The DCCD-Pro seems to do things a little different to above and information on it is sparse.  I think when/if we add 4D Aux tables that will allow us to do it a bit nicer.  
The main hurdle I see is you are going to need about 5 years of track tuning to get all these tables dialed in...  
      

No, neither are really needed for GDI and both are difficult to determine.  Deadtime is less critical because the injectors arent exposed to variable battery voltage like they are in a port injected engine, they are powered at constant voltage from the high voltage transformers.  So although there is a deadtime, it is relatively constant (and small) so can just get "baked in" to the VE table.  All GDI injectors I have seen are still operating with quite linear response at idle PW. 

The 3 main problems are the integral on the E-throttle set to 0, this means any small long term position errors would not be acted on, the non zeros in the top row of your E-throttle target (this means base position is wrong), and thirdly in the log the idle status shows "MAP lockout" all the time so it was not trying to control idle at all.  It looks like you have since changed that map lockout value after the log was done however so that has already been fixed.  
I have made lots of changes in your map attached below, so it still may not work well yet, but it should be close enough that it will at least attempt control.  Load this map in and do another log from cold start - assuming the idle does control around target then leave it idling (and logging) until it is warmed up.  This will give us some data to fill out the base position table with some realistic numbers.
You can use the file compare function if you want to see what I changed.
 
rodrigo-datsun-Idle test 1.pclx

Many assume because there is a spark every crank rev that the otto cycle of the wankel engine is 360 deg - often it is thought of like a two stroke engine.  However many forget the Wankel has 3 combustion chambers and the rotor revolves at 1/3 of crank speed so it actually 1080 deg for a full otto cycle.
When you inject fuel into a fresh chamber on the intake stroke, it takes nearly a further 2 crank revs before that charge reaches the other side of the engine compressed at TDC ready for spark. 
So you cant just cut fuel and spark at the same time - There is fuel and air already sitting in the chamber from 2 revolutions ago...
The simplified sequence:  The rotary limit mode starts by cutting fuel first, but continues to spark for the next two events since there is already fuel in those chambers, then it cuts spark.  When RPM has dropped below the limit and we want to restore power, then it enables fuel first, then waits 2 spark events before enabling spark again.  
This strategy is the same as the Mazda factory ecu's use, it is the safest for stock engines as very low risk of backfires or lean combustion events, but it is also relatively "slow" and granular so not commonly used for motorsport type applications where the seals have been replaced with something more robust and can handle a more conventional limiter strategy.  

Connect the AEM white wire to any spare AN Volt.  Connect the AEM black wire to the engine block.  

>Analog Inputs>GP Inputs.  You can use GP Temp or GP analog.  Set up the calibration in linear Cal 1/2/3.