Volcom86

This is pretty common due to more load creating more EGT.  Subaru's are especially bad as the long primaries exaggerate the affect.  There are gear trim tables to compensate for the difference between gears.    I also noticed a related effect on our Subaru test car - if you had done a little gentle driving around a built up area for example and the manifold cooled down a little, the first "WOT pull" you did after that would never reach boost target. 

from my experience, shortening the WG arm/ more preload is detrimental to what you are trying to solve. boost control will be even harder down low but little improvement up top
more spring would help (if duty doesn't) provided the turbo isn't maxed out like Vaughan said. 

I would start by increasing the wastegate duty in the table as adjusting the mechanical system will affect all of the values in both of your boost tables whereas adjusting the setting in the ECU can be easily reversed and is much quicker to do.

the actuator preload will affect your minimum pressure (like the wastegate spring rate will) where a higher preload or stiffer spring will hold the gate shut for longer.
In terms of the spike and reduce the high point is because of the increase in wgate duty and the drop of will probably be the gate being pushed open more by the higher exhaust pressure by the wastegate

Try feeding it slightly more wastegate duty in the areas it drops below what you want, wouldn't be surprising if it is trying to push the gate open a bit more up top. If adding more wastegate duty doesn't provide more boost then either the turbo is out of flow or the wastegate needs to be held shut better.

I see few more things.. 
Your cam control is set to EJ20.
At 9:46 .. you have a steady throttle, Accel Enrich is done (9:44) and yet you are running rich... 
 

lower your Accel Enrichment clamp it reaches about 36% thats why u got rich spots when hitting the throttle, I've seen couple of voltage drops I would check them out as well
I suggest u populate a 3D table for the injectors deadtimes since you got a FP sensor this will help to keep u on point as well.

Yes that is pretty normal, it is due to the the fuel film evaporating from the port walls.  Basically the opposite effect causes the lean patch when you open the throttle (if no accel enrichment was implemented).   Some ECU's have a wall wetting model or a "decel enleanment" feature to improve it but we dont at this time.    
I like Shane T's demonstation:  https://www.youtube.com/watch?v=LTXPF-6M1D0
 

There are a combination of factors.  It looks like Post start enrichment is quite a bit of it, when that disables around the 50sec mark your lambda is back to about 0.85.  So you can reduce that and its hold decay time quite a bit by the looks which will help.  IAT fuel trim should normally be off for modelled mode, although it is not having a big effect in this case.
Otherwise I would guess the charge temp approximation is a bit too heavily biased to ECT in the idle area.  I would not recomend adjusting that at this stage as it will effect the whole tune.  If you want to pull a bit more fuel out for cold start I would turn on a 4D fuel table, put ECT on the axis and remove a little that way.  

Im having trouble spotting the cause of the different AFR.  For similar conditions the commanded injector PW's tally up to to what my rough theoretical calculation would suggest is needed to achieve the same AFR for the difference in fuel stoich.  There's a few other differences that throw a spaanner in the works however so dont make the comparison easy. ethanol temp, charge temp, etc.  
Does the problem get worse at say 0% eth?  
My only other thought is the ZT3 wideband doesnt have an analog ground, that can often cause the AFR to read different when for instance a fan or similar turns on and causes a small ground offset between ECU and chassis.   Im not confident that is the problem in this case - but something to consider since two fans are on in your fan is on in the 11% point of interest shown above vs no fans on in the 15% point of interest.    

Yes, if the fuel table has been tuned using a lambda with a significantly wrong calibration, the the fuel table will need to be re-tuned.

I think that has just masked the issue a bit.  Since your base fuel pressure setting was 300Kpa, but you actually have more like 345kpa, when you change to "MAP referenced mode" the ecu thinks your fuel pressure is lower and just increases pulse width all over.  
I dont really see any obvious signs that your fuel pressure regulator or fuel pressure sensor isnt working correctly.  Differential pressure appears to be well controlled during transients.  
In the pic below the original log is the coloured lines, and we are looking at the same lean spot area as you pointed out originally above.  The overlayed pink lines are from the new log, the closest operating conditions I can find (MAP is just a tiny bit higher).
You will notice in the pink traces the ecu thinks the differential pressure has dropped to 300kpa, so it has increased the injector PW from the original 2.89ms up to 3.62ms, but if we look at the pink lambda trace there is still a lean spike in exactly the same area - just not as bad since you have like an extra 30% PW going in.  You will notice the pink lambda trace is richer pretty much everywhere.  

I would say some of it is due to accel enrichment, you can see below you only get about 10% enrichment and allowing for the slight delay in lambda response your leanest bump happens right around where the throttle is still moving.  However, even after that lean bump the AFR still stays quite a bit leaner than target for the whole WOT bit so there is something else wrong in that area too.  
I would add about another 10% to the 2500 & 3000rpm cells of the accel clamp table and see if that helps.

Ok, it looks like your dash set up is wrong, you had no receive set up on CAN 2 and you had the Link RS232 receive turned on.  So try the file below, I had disabled the RS232 and loaded a CAN receive template.
Volcom86.pdc

Can you attach your plex config.

I dont see much wrong in your ECU set-up.  Can you give us some pics of the CAN wiring just in case there is something obvious there.

Yes they will work out of the box using the generic dash stream - or the Plex CAN is fully customisable if you want to send something that isnt in the standard stream.  
To connect to the plug-in CAN port you will want a CANPCB cable and a CANF plug.

Ah yes, now I look closer some of it is caused by closed loop lambda.  Its pulling lots of fuel out in cruise conditions then when you come to idle it cant remove that trim quick enough.
I would change the CLL RPM low lockout to about 1600RPM so it is disabled at idle and when coming into idle.  The min Eff PW that I mentioned above will probably help a lot too.  

In general I think it probably wants a bit more accel enrichment all over but if if drives ok without any flatspots or hesitation then i wouldnt worry about the general low load cruising stuff that your first pic shows..
The lean spike in your second pic at the 9:23 mark I would be more concerned about since it is at WOT and peak torque RPM so it is going to be pretty prone to knock around that area.  Its hard to tell if it really is lean there or if it is something like a misfire that makes the probe read lean?
On the short burst you gave it just prior to this one (~9:21) it didnt go anywhere near as lean.  The main difference I can see from a fueling perspective was due to it being from lower RPM so got more accel enrichment.  So you could try adding a bit more accel enrichment above 4000 to see if it has any influence. Generally you dont need much above 4000, but it depends on injectors etc.

 

The response from your lambda sensor is very slow, it takes over 5 seconds for the lambda to reach minimum after the minimum fuel trim is applied.  So this oscilation is caused by having the control rate much faster than the system response time.  5 seconds is pretty extreme, usually it is more like 0.5-1.0 second at idle so it may indicate there is a physical set up issue such as the sensor too far down the system or it may be a controller or sensor issue.
So firstly drop the CLL rate down to its minimum (1Hz), if there is still oscillation after that then you will have to reduce the gains too.  

Simon, seriously mate, take the Horsepower Acadamy courses. You already seem to have a fair idea what you are looking at. HPA will simplify the whole process for you, break it down into a couple of main areas  and give you the skills and confidence to slowly sort out this nightmare properly yourself. Not only will you gain a fuller understanding of fuel and ignition requirements, all your trims and enrichments, boost set up etc,  they also do full video demonstrations on all aspects of tuning on many different platforms and those videos are available to watch over and over.

Hi Simon,
I didnt see your pclr file when I posted that info. After having a second look with the pclr loaded as well I can see what you mean about the fuel pressure. The pressure itself isnt technically a problem as 3.5bar is a pretty normal pressure level, however the flow rate, deadtimes and short pulse width tables are taken from the 3bar table in the ID1700 data instead of 3.5bar. The fuel flow number is probably the only one having a real impact here. Unfortunately you cant really change either these parameters or the fuel pressure without re-checking the entire fuel tune. Changing the fuel pressure down at the regulator is more likely to be acceptable given you have everything FP sensor controlled, but its still something you want to be careful with as you could cause it to go leaner under load.
I can also now see what you mean about the -12% ish CL correction to your idle fuelling. you could try dropping the fuel table values for both table 1 &  table 2 at these cells & smoothing it into the cells near it. Better yet would be to disable CLL for a while. Go drive around smoothly for a bit (so there is constant load in each cell for a few seconds) then at the end of the run use the mixture map feature to adjust mostly table 2 (current E85 mixture means you are interpolating 90 odd % out of fuel table 2 so its really the only 1 that matters. Turn CLL back on after you've done this.
CLL isnt active at cruise because the RPM high limit is set to 1500rpm. That being said, it still looks to me like at cruise RPM & TPS your lambda values are pretty close to target - between 0.94 & 1.0 which seems ok. If the CLL high rpm threshold was bumped up to around 3k or 3.5k this would probably correct itself. You would probably want to drop the TPS high threshold down to 50 or 70% so you dont run CLL under full throttle. It seems your fuelling is reasonable at full throttle, although there isnt much time spent at WOT in the log to be sure.
For the cold start issues, you can try just upping the max idle valve % up to say 70 and see if this improves things. Your idle is a little below target at this point too so this might fix that.
regarding the feeling of it still having power when off throttle, this wont be lambda related. Only things that cause this (apart from air leaks) are faulty/too small recirc valves, idle screw set too far open, idle valve numbers too high, or ignition advance too high. Easiest one to change is dropping the values in ign1 & 2 table for the row that represents your off-throttle MAP value. The lower you drop this the less power will be generated by the engine to "keep pushing the car". How long does this effect last for? your fuel cut is set to kick in after 0.2seconds which should cut power a lot.
Your cruise log @ 26:54 shows fuel pressure dropping off a fair bit. You'd need to hold it at that level for a few seconds to see if it stablised or kept dropping off. Maybe the lines are too small? maybe low amperage at the pumps if its factory wiring?

VVT mode is set to Evo 9.  I should be set to "Subaru AVCS EJ20".

I just had a quick look.  There is some pretty odd stuff in there. 
It looks like it has been tuned for flex fuel but there is only one ignition map and no other corrections for ethanol content.  Usually E85 can take a heap more advance than petrol so what fuel was that ignition table tuned for?  VVT looks like the offsets are set incorrectly.  The RH inlet cam is showing 10deg advance when it is in the fully home position.  That means it will be getting 10Deg less than what is commanded. The VVT and ignition tables both have big "holes" in them around the cruise area that is going to make it drive pretty horrible. Boost control not working great at all. Fuel control is not great either.  I would say this is due to there only being 1 VE table set up, which doesnt usually cover flex fuel well enough especially when injectors havent been flow tested on both fuels. I think the saving grace is you are running it on relatively high eth %, which is relatively forgiving.  I wouldnt give it a hard time especially on petrol until it is looked at by someone more competent or interested.  If you want to PM me the tuners name I will pass it on to the right people to consider.
 
 

I think you need to consider using a different tuner. 
There should be plenty around in Victoria?