Understanding the MK7 PCV function, retrofitting the MK8 PCV, and why catch cans are bullshit(IMO)

[DerHase]

Plus some additional info, clarification on the website:
https://www.datadrivenmqb.com/drivetrain/venturicrankcasetesting

Charts from the video

Initial baseline testing using a gauge...

And what I'm going to call "instrumented testing" using the MK8 PCV sensor and GM MAP sensor for loggable data:

 

[DerHase]

So some more stuff has been ordered to do further testing...

- A cheap APR-knockoff turbo inlet pipe via Amazon (if it's awful or doesn't fit, I can just return it)
- Another MK8 PCV valve
- A 1/4 turn ball valve to experiment with restricting the Venturi boost line (and also go from basic to full retrofit quickly for testing purposes)
- And an IS38

Not sure how that last one happened... though it probably won't be getting installed until later in the year. But bottom line I'll eventually be able to have IS20 vs IS38 data as well.

 

[DerHase]

For @yakev724 and @ITGUY who want to log in Mode 22 (not flashed via Simos Tools and Switchleg's high speed logging patch):

To log the SAI raw voltage in mode 22:

Name

Unit

Equation

Format

Address

Length

Signed

ProgMin

ProgMax

WarnMin

WarnMax

Smoothing

Enabled

Tabs

Assign To

SAI raw voltage

V

x/1000

%01.2f

0x40e0

2​

FALSE​

0​

5​

-1000​

1000​

0.7​

TRUE​

PCV

Corrected formula if used to read MK8 PCV sensor in PSI**:

Name

Unit

Equation

Format

Address

Length

Signed

ProgMin

ProgMax

WarnMin

WarnMax

Smoothing

Enabled

Tabs

Assign To

PCV Sensor Pressure

V

x * 0.0047717 - 14.795

%01.2f

0x40e0

2​

FALSE​

-6​

6​

-1000​

1000​

0.7​

TRUE​

PCV

For those on 3E HSL I log both just for the hell of it, but I have the pressure directly reference the mean voltage PID (which is factory smoothed already)... so adding both of these is necessary the way I have it done.

Name

Unit

Equation

Format

Address

Length

Signed

ProgMin

ProgMax

WarnMin

WarnMax

Smoothing

Enabled

Tabs

Assign To

PCV mean voltage

V

x / 6553.6

%01.2f

0xd0009cc6

2​

FALSE​

0​

5​

-100​

100​

0​

TRUE​

PCV

pcvvolts

PCV Sensor Pressure

PSI

pcvvolts * 4.7717 - 14.795

%01.2f

0xffffffff

2​

TRUE​

-6​

6​

-100​

100​

0​

TRUE​

PCV|0

**Regardless of which way you're logging, step 1 should be to "zero" the sensor with key on, engine off. When you start up Simos Tools, it might read +/-0.02 or 0.03psi, etc. IF it does, then look at the equation for the PCV sensor, and adjust the last value (14.795) up or down a little bit to adjust. Since that number is SUBTRACTED at the end, changing it to 14.775 will add 0.02psi to the gauge reading, and 14.815 will subtract 0.02psi from the gauge reading.


Anyway hopefully in the next day or two I'll have everything set up - I added a pin to my ECU harness so that I'll hopefully be able to log both crankcase vacuum AND PCV vacuum concurrently for testing purposes. Waiting on Switchleg to finish testing the latest patch to enable it.



Successfully logged the SAI voltage in mode 22:

Didn't get any pics of driving again, but verified the voltage to psi conversion above worked fine.

The reason for only keeping one PID in mode 22 is to keep the amount of stuff being calculated down as much as possible. 3E can handle tons of stuff, I'm up to like 180 things on that list that get logged all the time. Plus DSG things.

 

[yakev724]

Legend - my harness is on its way and just finished fabbing up the venturi to my charge pipe hose. Curious to measure the impact.

In case anyone needs this setup (though keep in mind not many aftermarket charge pipes have a larger silicone hose like EQT): bung fitting 1 fitting 2 hose

 

[DerHase]

Wanted to give a quick shout out to @chuncomunco and @tigeo for donating PCVs for testing (AP and BL respectively).

Things have officially gotten out of hand. I've now got:

(1) AH - cut open for science
(1) BL - cut open for science
(1) BK - cut open for science - my OEM valve

(1) AP - not cut open (yet). MAY try to run on my car if it physically fits and log crankcase pressures - though unsure if it will be directly comparable.
(1) BL - for on-car testing only, to be returned when done. Came off of tigeo's wagon, with zero problems so is "known good" to our knowledge.


(2) unmodded MK8 PCVs (one NIB)
(1) modded MK8 PCV (on the car, not in this pic, to be disassembled similar to the others with the cap modified for science (to be useable on the other 2 MK8 valves too)


Also got a $40 knock off TIP that looks like an APR reject. To be tested with at least the MK7 BL valve and MK8 unmodded valve to see if one is "better" than the other. I doubt there will be any loggable differences but who knows?


Also did some testing between cars with tigeo where we swapped Unitronic TIP and intakes back and forth logging a bunch of stuff. Going to test some more things this weekend and see what else I can come up with to compile it all in one central spot.

 

[DerHase]

So I threw @tigeo 's BL on last night to get some comparative logging vs the MK8 PCV in basic retrofit form.

As had been suspected: There's basically zero difference crankcase pressure-wise at least as noticeable in a 3rd gear pull. Same general curve and everything. The stock TIP is on for both of these pulls. The one with the BL (lighter colors) had a bit of wheelspin because it's cold and wet, so SP TC was working it's ass off... but since it cuts timing moreso than boost it doesn't really affect anything crankcase pressure wise.

The only real difference between the two valves is what happens at low rpms and light loads/idle. The BL idles at around -1.25 to -1.30psi and the MK8 is around -1.1psi in basic retrofit form.

Some driving around with the different stuff has made it apparent that what is a LOT harder to quantify is all the stuff that happens in transitions.

Here's a quick look at some of the testing we did... the Unitronic TIP did have a very noticeable impact on available vacuum, which showed up in the crankcase as well.

The Venturi as you can see just keeps everything dead ass steady. These are NOT actual "logs" being plotted, these are points pulled from logs at 3k, 4k, 5k and 6k rpms, after starting a 3rd gear pull at 2k rpms. Doing this for initial data points because I can structure it all on a single spreadsheet rather than building a giant model and everything like with the cooling logs.

Eventually will lay complete logs over and look more in depth, but this is relatively quick and easy to spot general differences.

 

[Torrid]

Is there benefit to going to the MK8 setup on a generally street driven setup? I ask only because if it needs replaced at some point other than the one time cost of the new pipe and plug, do you see it as a worthwhile upgrade. I have a stage 1 car occasionally driven as a hooligan, but I wonder if my now fixed oil cap issue is just a starting sign of the PCV going.

 

[DerHase]

Is there benefit to going to the MK8 setup on a generally street driven setup? I ask only because if it needs replaced at some point other than the one time cost of the new pipe and plug, do you see it as a worthwhile upgrade. I have a stage 1 car occasionally driven as a hooligan, but I wonder if my now fixed oil cap issue is just a starting sign of the PCV going.

I think it's a lateral move, but gives room to upgrade to the Venturi in the future.

I swapped a non-ventilated MK8 valve back in today, and doing more testing. Give me a week or two and I'll be better able to answer.


The MK8 valve seems to target a lower crankcase pressure in general, so when vacuum IS depleted from on/off transient throttle changes... It won't be sucking as hard to restore vacuum and pull oil upwards with it.

I think my ventilated valve was definitely screwing with accurate crankcase readings (as far as being able to compare with a BL PCV).

The BL targets about -1.3psi at idle or off throttle, and -1.1psi at speed.

The unmodded MK8 is around -0.8psi.

I DO think my ventilated valve might be applicable to the AH/AP PCV since it draws on the cam cover.


Also...

https://youtube.com/shorts/x8wvZBkviLQ?si=460Le_Ns61GNO8hf


I don't think that crankcases are going positive on track (unless really really bad TIP and turbo combo maybe? And blowby). I think the diaphragm opening up every time crankcase pressure spikes is 100% what is sucking up oil. It's a design flaw in the way the PCV passage is designed... But the Venturi fixes that.

 

[Torrid]

I'll definitely keep an eye on this thread to as I continue into mod on the car. I'm not going too crazy. I can't see a reason so far to go beyond stage 1, but I've thought of going with an upgraded intercooler if I hit any heat soak, but it hasn't happened yet. I'm beyond what the tires can handle until i do the wheel/tire/suspension stuff. That may change how far I go.

 

[DerHase]

Since I had the BL valve on, and needed to reinstall the MK8 valve I figured I’d do a “how to” for the main part of the retrofit.

Also a bit of a review on a $46 turbo inlet pipe I picked up for cheap testing.

Small look at the very end of what’s planned also.

Really just waiting on the new Switchpatch to be QA’d so I can flash the file and log both sensors together at the same time as that’ll make explanations a lot easier.

 

[krs]

Probably of no real value...

https://datazap.me/u/krsbr79/lvrscc...2?log=0&data=6-8-14-19-25-26-27-28&hg=8-19-43

I logged a few autox runs, just wanted to see what things were doing. Looks like an engine, running... I did have to idle well beforehand to get oil temps up, think that's the contributor to the higher than normal IAT out of the gate.

I don't have any issues with smoke anymore.

 

[DerHase]

Probably of no real value...

https://datazap.me/u/krsbr79/lvrscc...2?log=0&data=6-8-14-19-25-26-27-28&hg=8-19-43

I logged a few autox runs, just wanted to see what things were doing. Looks like an engine, running... I did have to idle well beforehand to get oil temps up, think that's the contributor to the higher than normal IAT out of the gate.

I don't have any issues with smoke anymore.

This is with the BL PCV correct?

Stock turbo inlet pipe?

 

[krs]

This is with the BL PCV correct?

Stock turbo inlet pipe?

BL, racingline TIP.

 

[DerHase]

So the PCV stuff kinda took a short back seat while the car was down for suspension work. And now VIR is < 2 weeks away so I'll be getting back to it soon, though doing a bit of testing and just setting the info/data aside for the time being.

In a nut shell @Switchleg is awesome and hooked me up with the new Switchpatch so I was able to add a pin to the ECU and data log both the crankcase AND PCV pressure sensors at the same time. I really wanted to wait for this because it makes doing pulls/comparisons SOOO much easier. Plus I'll be able to repurpose the ECU pin for other uses in the future too.

But for a little teaser/idea of what's to come:

Here is the Epman (bobo ass APR knockoff) TIP in Full (darker lines) vs Basic (lighter lines) retrofit form.

3rd gear pull. The Venturi is able to suck the crankcase back down to -0.7psi WELL before the turbo is even fully spooled. The Basic retrofit gets the crankcase down to that level at 4500rpms.

2nd gear pull. You can see the Venturi is able to suck down the crankcase back to -0.7psi or more vacuum almost immediately as the turbo spools, while the Basic retrofit form isn't able to get the crankcase back down to it's target vacuum level until nearly redline.

ANY time that the crankcase isn't at least -0.7 to -0.8psi, the diaphragm is OPEN. This means a vacuum source is pulling a BIG volume of air on the bottom end in an attempt to replenish that vacuum level.

2nd gear is a LOT harder for the PCV to keep up because it runs through the RPMs a lot faster plus load is probably a bit less (and airflow across the TIP is less also).



I took an old phone and put it to use (since I can't find my damn Gopro)... That hood vent is coming in handy.
Ran out of daylight to test with Venturi... but this is coming up soon and will show exactly what I'm talking about with regards to why the full retrofit is the only way to fix the OEM system's shortfall...

I am not 100% certain I'm going to be running the crankcase sensor on TRACK, but 100% will be doing at an autocross with and without Venturi. That won't be until April though. Depending on how the day/weather goes in a few weeks, I MAY try it though. Just for grins. Won't be with my phone stuck to the hood though

 

[PerceivedShift]

Great stuff haze, the switchpatch is a gamechanger! Looking forward to making the uhhhh "switch" next year. It's definitely looking like if you run an aftermarket TIP, its highly recommended to go with the full retrofit without testing for that specific TIP performance.

I'll be ordering the parts and attaching the venturi to the EQT charge pipe hose as @yakev724 pointed out.

Edit: The bung can be purchased on ebay with free shipping.