When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
You need a better PCV or a check valve that keeps boost in your intake manifold from pressurizing your crankcase.
Pulled the PCV hose and capped the intake manifold. No more leaky. That’s a much better thing than air blowing past my piston rings that I feared; I have a healthy sense of paranoia that the shavings from drilling the oil pan are in there tearing up my engine, despite my extreme caution.
Able to pressurize the system up to 10psi now, with very slow air losses through the tip of the BOV screw; less than 1 psi/second.
Ordered the GTX PCV from Fab9. I’ve read like a hundred of these PCV/catch can threads, and they’re all pretty much inconclusive. Figured I’d start with easy/simple and work to difficult/complex as needed.
I get it- catch cans are easy. Two more hours of reading old threads. Problem is that I still require a PCV or check valve that doesn't suck, no matter which setup I go with. Fab9 is taking forever to ship the GTX one, it was a bad decision to order through them. Took a flyer on one from NAPA this morning, it sucks donkey dicks. Holds zero PSI. Somebody mentioned an Autozone one in a different thread, my store doesn't carry it.
Can somebody give me a part number for a decent PCV valve that will hold 10psi?
Went to NAPA and asked for the linked PCV. It has been replaced by a newer one that’s black and grey. It sucks as bad as the cheaper one.
Got the Autozone valve. It also sucks.
Went back to the valve that was in there in the first place. It seals up tight, but only sometimes. Now that I know what my issue is, I took it for a drive tonight. It’s night/day when the valve is closing properly or not closing. I pulled it back out and washed it out with Naphtha... will see if it seals more consistenly tomorrow.
On a positive note, Fab9 decided to ship my GTX one. After charging me $9 for shipping, it took them almost a week and they shipped USPS flat rate. WTF.
On a positive note, Fab9 decided to ship my GTX one. After charging me $9 for shipping, it took them almost a week and they shipped USPS flat rate. WTF.
Thats terrible. I got one from mazda and had it the same week. I hope you didn't pay to much for the sloppy service. Flat rate is 3 business days.
This blows. I am going by Napa, pretty sure this is the main distribution warehouse near me, plus they know me because of my car. I bought 2 last time because they were so nice, but they are a little weaker and I broke one trying to pull it out. If they have them, and is says available in store, I will pick up a bunch and tell them about the difference. Not that this helps you any.
Of course, they get the good ones for a while and then stop carrying them.
Or you could just stop trying to run a PCV valve at all. They block half of the crankcase's ability to vent anytime you're in boost, which is the time you need it most.
^ I’m tempted. I‘ll see how the GTX one pans out first. That crankcase **** stinks, I hate venting it to atmosphere. Then again, that’s exactly what I’m doing right now.
After cleaning the one I started with (blue and cream color- maybe stock from 25 years ago?), I am 100% certain that the PCV is the culprit. It mirrors the behavior I saw when I was leak testing. When I jump on the throttle, it seals up tight and spools like lightning. If I roll on the throttle slowly, it doesn’t seal up and I hear air hissing out of the crankcase vent, pressure tops out at 9psi.
Now that I know how the car drives with the PCV sealing properly, I still have some boost creep. My overboost protection is set at 12psi and I haven’t hit into it yet, but I must have been awfully close because my boost guage was showing 12. More porting in order.
Finally got the GTX PCV valve. Nice quality part, and feels like it should block boost if you blow/suck into it.
Only problem is, it doesn't. My spool data is basically the same as with the stock PCV valve. This was a crappy 4th gear WOT pull, but you get the idea.
In addition, now that I know that PCV valves also block vacuum at full idle, I realize that it's not doing that job very well either. My idle is 300-400 rpm high, intermittently. It's basically acting as a vacuum leak.
Sixshooter is right. I'm done with PCV in my turbo car. The idea that it's vacuuming out all the bad juju from my crankcase is romantic, but it just doesn't appear to be true. When the valve is working properly, it's closed at idle. And it's closed at boost. It's only really open at part throttle, when there's very little vacuum anyway. And it's pressurizing my crankcase at other times.
I'm curious if there are differences in spool and top end with the intake manifold plugged properly. If it's leaking past there you should definitely see something change.
Plugged the IM, and threw a breather filter onto the PCV. New spool data:
This is a surprising outcome. The spool was actually better with the PCV valve connected to the IM than it was with the IM plugged and the PCV valve vented to atmosphere.
My train of thought at this point, with the data at hand:
- I have done a pressurized boost leak test, and there are no meaningful boost or vacuum leaks outside the PCV system.
- I had assumed that all the PCV valves are leaking boost, even the GTX valve. I still think that's true, but it's not the primary culprit for loss of spool. See next point.
- Something else is affecting my spool, unrelated to boost leaks. Hypothesis: the wastegate is opening prematurely. Looking at the shape of the orange plot above, it's cracking open around 4.5 psi and is mostly/fully open around 6.5psi.
- Now that I've capped the IM and I'm not losing boost through the PCV system, it may be providing a stronger signal to the wastegate actuator, which is picking up the wastegate sooner than before.
- I will try more preload on the actuator arm to see if this affects the outcome. If I see an improvement there, I can assume that I've gone as far as I can with mechanical solutions, and implement the EBC.
- Even though the PCV system isn't the main culprit for killing my spool, it's definitely leaking vacuum and messing with my idle. I'm also not convinced it's doing anything good to remove crankcase vapors. Therefore I am not planning to connect it back to the IM, I'm staying VTA. I will implement catch can and a more permanent installation.
Last edited by Schroedinger; 10-30-2017 at 10:55 AM.
Now we're getting somewhere. I added wastegate preload by tightening the actuator arm an additional two full turns.
Findings so far:
- the GTX PCV valve (blue) performed identically to the OEM PCV valve (yellow). No difference whatsoever.
- capping the intake manifold @ the PCV connection (orange) actually made spool worse, because it sent more boost to the WG actuator.
- Increasing preload on the actuator arm (grey) increased the spool by keeping the actuator closed longer, and made the system behave more linearly.
- I also logged a 1st gear WOT pull all the way to the rev limiter. Max boost is 170.6 kpi, exactly the same as it was with the arm looser. That tells me the wastegate is still opening fully.
I'm going to keep adding pre-tension 1/2 turn at a time until the wastegate doesn't open fully (max boost @ WOT, rev limiter exceeds 170.6), then go back 1/2 turn. Then install the EBC. I'm starting to get excited for how awesome the car will be with EBC. In theory, I could hit full boost around 3200-3400 rpm.
Last edited by Schroedinger; 10-30-2017 at 02:24 PM.
Ordered a catch can, keeping both crankcase vents VTA through breathers for now.
More experiments with wastegate actuator preload. In this graph, "neutral" was the hole in the actuator arm covering the pin on the flapper arm by 50% at rest. +1 is increasing the preload by 1 full tightening rotation of the actuator arm, and so on (+2, +3, etc). These are all 4th gear pulls; today I did the "+3" run.
Been better every time I tightened; today's +3 run was much better than yesterday's +2 run.
In all of the plots, the line is pretty straight after the WG opens. That tells me I've ported my WG well enough, because if it had creep then the line would curve upwards with increased flow. The slope of the line after WG opening is also the same in all plots; I'm assuming that's the spring constant of the WG spring. So by adjusting preload, I'm just changing the offset point, or the "knee" of the graph. I love it when experimental data matches theory.
The only problem is that the car is going mighty fast doing 4th gear pulls, and on local roads I can't run all the way to the rev limiter in 4th gear without going 50 mph over the speed limit. I've also been logging 2nd gear pulls all the way to rev limiter to see where boost peaks. Previously, it peaked around 170.6kpa (10.2 psi). Today's +3 run peaked at 180.7kpa (11.7 psi). I was going to dial it back, but I figured I'd plot the 2nd gear pull also.
Even though it's getting damn close to my overboost protection (12 psi), it appears to be hitting an asymptote just below 12 and going no higher. Basically, I've squeezed all the boost and spool that I need to out of WG preload. I don't think I've gotten to the point that the WG doesn't open fully yet, but I don't need any more boost than this.
My feelings on the turbo experience so far:
The car is pulling like a freight train, so much fun to drive. Going from supercharger to turbo was a big improvement. But going from poorly sorted turbo to well sorted turbo was an even bigger improvement. I have to think that people who don't fuss over the details of setup and tuning are leaving a TON of performance on the table. I'm going to drive it like this for a while, then put the EBC in when I'm ready to fuss with it again. The graphs suggest I could hit 10psi around 3200 rpm. Not too shabby for a 1.6L motor.
Last edited by Schroedinger; 10-31-2017 at 03:14 PM.
10-25-2017, 07:07 AM
0
