Thanks for sharing all of this, your thread is full of very valuable information 🙏👍👍
maybe I missed the posts, but could you détail a little bit more about your PCV / Catch can setup ?
I plan running a 6AN line from the front of the block (from the balance shaft inspection hole), and of course two lines from the valve cover.

 

I start discussing the PCV system at the link below so as you work down the post I show a diagram for how it works.
https://www.evolutionm.net/forums/ev...l#post11908671

I doubt you need anything more than a single -10 fitting off the valve cover to keep block pressure in check, which can be had at the link below.
http://www.radiumauto.com/10AN-Male-...d-In-P940.aspx

You can put the intake manifold PCV to the front valve cover nipple and place a good catch can in between. You can then tie the -10an off the back of the valve cover to the turbo inlet with a good catch can in between, which essentially mimics the stock setup.

 

Thx for your reply !
i’m currently running the « stock » system but with two Saikou michi OCC.
mon my second engine I wanted to add a vent from the block and also try to figure out how to properly set the system without using the intake manifold vacuum which is only off boost vac source (and get rid of the pcv valve), maybe by using « pre turbo » vac source only (as the oem system on boost) or maybe even better using a pump like you.

 

You don't want to get rid of the intake manifold PCV because that is the best source for vacuum off boost and your car spends most of it's time under that condition. Therefore, I would keep running the intake manifold PCV with the Saikou can. If you go with a single -10an press fitting at the the back of the valve cover for the turbo inlet PCV, then for the intake manifold PCV you just need to add a one way check valve to the can and route that line to the front valve cover nipple.

 

Ok I get It, it’s « inverted » vs factory. I’ve already that 10AN fitting. So basically I’ll run a third can for the block fitting, which I will probably route to the turbo inlet.
I’m still on new block pre assembly, but it take that opportunity to make things clear about PCV system « routing ».
again, thx for your help.

 

yes, just switch ports on the valve cover and add in a one way PCV valve to the intake manifold PCV can. Again, not sure you will need to use the inspection hole as additional venting so perhaps just use that hole to take a pressure reading. If you want to tie in the inspection hole then no need for an additional can and just tee it into the turbo inlet can.

 

Thx a lot, I’m going through your threads, and learn a lot a very useful details ....

 

You can learn everything not to do.

 

So the EVO has been running well with the 4.08 final drive except that it has a slight whine on decel. I did a test to see how the final drive affected spool and to my surprise didn't seem to do anything. Below is the 4.53 (blue) and the new 4.08 (red) done in 3rd gear. I also went back to a fresh set of OEM side engine mounts and the car idles so damn smooth compared to the stiff side mounts I had, which rattled the car to death at idle.



I have everything sorted out for now and hope to get it to the dyno once I finish the turbo swap on the GTR. I have a pair is custom Gen II GTX3071r turbos made and my buddy CNC'd the turbine housings to open up the wastegate hole and unshroud the valve along with making larger custom inconel wastegate pucks. These turbos are known to creep 2-3psi with this setup so I hope the larger gates eliminate this issue.




Because I hate making changes to my setup I decided to order a long runner intake manifold from a G4JS, which arrived today. This should bolt up to the EVO head and I will spend some time cleaning it up and welding a flange to convert it to my 70mm TB. The runners are actually very large at the plenum and then taper down to the flange. I am curious how much low end torque I can gain with this intake vs the stock Curt Brown ported. Probably going to hurt some of the topend power, but for $50 it will be worth the exercise. The turbo swap will take most of my time because I have to do a few custom hoses and redo part of the turbo inlets, so I won't be able to focus on this until late next month. The plan is to do a before and after on the dyno and of course, lots of data.

 

What is the spool up on this turbo?

 

A little update on the EVO and GTR. Did the turbo swap on the GTR and the car put down 749hp@wheels on 93 pump gas tuned by me using AccessTuner.




I then decided to upgrade my GTR transmission because rebuilding the EVO transmission made me a tranny expert. It actually wasn't as bad as I thought although it was definitely way more involved vs the EVO trans, but the end result is that it is working flawlessly.








For my next GTR project I am building a 4.15L stroker motor in the hopes to make about 900hp@wheels in 93 pump with a set of GTX3576r turbos. At this time I am collecting all the parts and plan to build the entire engine myself, but paying a shop to line hone for the crank and balance the rotating assembly. I'm essentially going to build a complete motor on the side so I don't have the scavenge anything from my stock motor, but this will take some time because these parts aren't cheap.




Now on to the EVO...I removed the front O2 sensor signal off of my wideband controller and went back to using an OEM type front O2 sensor. The problem with the wideband controller is that it goes through a calibration cycle when you start the car so for about the first 30 seconds the ECU doesn't get an accurate reading and the car doesn't run properly. I was going to purchase a new OEM sensor, but at $400 I went with the NTK 23550 for $80. That change alone has resolved the issue and I can now start the car after a hot restart and it runs as it should. My original OEM sensor lasted 6 years before I had to replace it because the threads stripped trying to remove it from the O2 housing. The $40 Bosch sensors work well, but on my setup they last about 1 year before they die due to the heat so I am hopeful the NTK is more resilient.

I'm also going to take the car off of SD and go back to the factory MAF. The SD tune runs well, but there are quirks with SD that make it do random weird things and I just want to car to run more like stock. I know at 600hp@wheels the factory MAF screen can be a problem due to deforming the aluminum air straightenerr, but I have a few tricks I think will resolve that issue. To remove the MAF screen take a flat head screwdriver and run it along the edge of the plastic piece holding in the MAF screen. Do not pry with the screwdriver and simply run it along the edge and it will break the bond.





Once you break the bond you can then remove the plastic cover and pop out the air straightener.





The factory MAF straightener is 10mm thick and has a cell diameter of about 2.5mm. I found a company that sells aluminum honeycomb plate that is also 10mm thick, but with a cell diameter of 6.5mm. I will simply cut it to shape and swap it into the MAF housing, but I think this should be sufficient to survive the flow when making 600hp@wheels.
https://www.etsy.com/listing/9240150...e_recs_3&crt=1

If the above mod doesn't work then I'm going to have a buddy of mine CNC the part below, but integrate a 4.25" diameter air straightener and have it made out of stainless. It will be located further from the sensor, but other Karman vortex MAF housings have the air straighter a good ways from the sensor.

 

Everyone seems to say SD is the bee's knees, and I'm gathering parts to convert. What are the drivability issues you haven't been able to work around, or what is better about MAF based tuning at 600hp?

 

Well done on the GTR dyno results on pump 93.

It is not the 4.9 bosch sensor's fault, on the idle and start up issues, but the problem relies on the stock ecu hardware and specific .bin/rom patch you are using. It should not take 30 seconds to be calibrated by the ecu and get a proper reading out of it. Back in the day where open source ecuflash roms were still only on early stages and on MAF only I also replaced the stock 02 with a wideband bosch one and I had no issues, that was on 3plug JDM evo 8 ECU. I use the same sensor on the Power FC D-jetro and I have no issues whatsoever with it. The bosch sensor is of the latest tech and construction and it is as fast as the oem narrowband o2 sensor, as per my experimentation on both of them on the Power FC. There are enough car manufacturers out there that use a wideband sensor for closed loop instead of a narrowband one, and some of them use this specific bosch one. The same would be if you were using the bosch sensor for closed loop function on an aftermarket ecu such as Link, Motec, Haltech, HKS, Syvecs, AEM, etc.

Do you have the wideband connected directly to the oem ecu or do you have it connected after its controller?

The honeycomb mesh of the stock MAF is not a problem nor a restriction at 600whp on a Dynojet. You really did not need to remove it or worry about it. But if you do want to go through with the modification, know that you do not need the honeycomb in the first place , the MAF will do fine with it removed. If you wish to replace it, it will make no difference whether you use a 1cm thick or a bit higher or lower than that, nor does it have to be at 2,5mm , it can be 2 , 3, 4 i.e, Also the distance from the sensor does not make any difference either. Many hotwire i.e type MAF sensors do not use a honeycomb air diverter. The honeycomb is mostly there on Karman type sensors as a safety net.


All the best with your projects,




Marios

 

On the EVO, whether you have a MAF setup or SD setup, they are both running in SD mode when under boost. So we are really only talking about the tuning differences when the car is in closed loop operation, which is typically for all off boost conditions. The factory setup runs MAF in closed loop up to a predefined Load and TPS value where it then goes into open loop and you tune fuel using the High Octane Fuel map, which is essentially a SD VE table. In most instances a MAF setup will more accurately measure the actual charge density entering the motor vs a SD setup, which is why most people say MAF runs smoother for off boost driving. The factory MAF has baro compensation, which can be added to the SD setup, but most don't add that capability.

So why do most switch to SD, it is because of the MAF housing itself. At 500whp the MAF screen can get compromised so it is often removed. Once removed, some claim the MAF no longer works properly so might as well go SD. From data posted, at 600whp the MAF housing itself doesn't seem to be a significant restriction assuming you remove the aluminum MAF straightener. The most I've seen posted was 680whp through the factory MAF housing. Assuming the MAF housing is a restriction, I am also exploring the use of the 7MGTE Karman Vortex sensor in the Lexus SC400 MAF housing. It scales from 0-2100HZ vs 0-1600HZ for the EVO, but from what I can tell the housing itself is larger than the EVO. It also doesn't have built in baro so that is something that would have to be added. However, for now I am going to pursue making the factory MAF work at 600whp and with an air straightener.

Regarding my specific problems, the engine often stumbles at low load combined with low rpm, which I have tried to resolve within the VE table. The car will sometimes and randomly dip throttle when coming to a stop when pressing in the clutch. Again, this is random so I have not been able to resolve. Finally, I have never been able to resolve cold start driveability issues. The car will fire up just fine at any low temp, but the car stumbles for the first 30 seconds of driving unless I simply let the car warm up for about 1-2 minutes.

 

I was using the 0-1V analog output of the wideband controller into pin 71 of the ECU so it has nothing to do with the Bin file. The 0-1V analog output from the PLX wideband controller is junk until it goes through its calibration. The Innovate didn't have this problem because I could program a default voltage output (0.45v) until it completed the calibration phase. Anyway, what I am talking about has nothing to do with the factory ECU and is simply an issue associated with the wideband controller. I still use the PLX controller's 0-5V wideband output into Pin 73 of the ECU for data logging.

Regarding the MAF screen, because I have no personal experience using it at 600whp I can only go by what I read. I never viewed the screen as a restriction, but there are plenty of posts showing the screen collapsing for whatever reason at less than 600whp. However, I appreciate your providing your experiences. In the case of the Karman Vortex MAF, I prefer to run the straightener to get a more precise measurement due to how it works, but I understand it can be removed and you can compensate by remapping the MAF table. The potential downside of removing the screen is for a fixed amount of actual charge density entering the motor you will likely have a wider range of corresponding frequencies.