legacycentral bbs

Well, considering we should be pushing almost 9...

1psi difference is enough to warrant a cause for concern. Fluctuations mean there's a problem with wastegate control, the wastegate itself, or possibly compressor surge. Could be a problem with the turbo, or any number of things.

I wouldn't say that 6 pounds is a gift over "low-boost" of 5 pounds. It's STILL almost 3 pounds less than stock.

I'll install my MBC tonight and see how it works out for holding boost.

So I cleaned the controller with the car running...some goop came out, but I've still got the same problem. Just to make sure it's not my wastegate, I unhooked the control hose. Holy goddamn! Sure spools fast. I tapped the gas and hit 12 before I could even react.

I'm guessing it's a lot cheaper to just go buy a MBC than a boost control solinoid anyways, so I'll just pick one of those up.

Yea, you can even make an MBC for like $3 in parts at Home Depot.

I, personally, have the VRG3-MBC. I can't wait to install it.

aspect - Yeah, isn't it crazy how fast the turbo spools? You can't understand it until you see it; the stock boost control doesn't spool up as fast as it could. A ball-and-spring MBC like the one I made Phil gets you close to that kind of spoolup.

I ended up having to do about 20 minutes of driving today with the ECU in reset-codes mode. Lo and behold, I got 9 psi!!

I again had 9 psi for about 20 minutes of driving after the car was parked most of the day, then it went back to 5ish. Suddenly it came back, but now its gone again. So my ECU does know there's a boost control solenoid there, its just VERY tempermental when it comes to using it.

So now I'm wondering... The ECU just pulls timing when it sees knock, right? It wouldn't go to wastegate boost as well, would it? I can't see the car knocking on Sunoco 92, but maybe I should try a couple of tanks of 93 or 94?

I'm going to try doing a 45 minute drive tomorrow with it in reset-codes mode, and see what happens.

Hopefully I can also figure out why I had intermittent valve lash adjuster knocking today too.

Hm.

It's very plausible that it would reduce boost when knock is detected. I've never seen that mentioned in any literature I've ever read about our ECUs, but many other factory turbo ECUs do it.

If you're willing to risk it, you could try running with the knock sensor disconnected after resetting the ECU. The ECU will keep timing conservative but it won't be able to detect knock.

I could fill it with 94 and try it. There shouldn't be any knock then, right? Unless I've got some nasty azz carbon deposits causing detonation.

The other thing to do would be to leave the scan tool connected, and get someone to stare at it continuously. That Pinging variable indicates when the ECU detects knock, right?

There shouldn't be knock, yes.

Yes, the pinging variable is when the ECU is detecting knock happening. I haven't actually ever observed it "ON" myself, but I haven't continuously watched the display since I haven't tested the scan tool in the car with a co-driver.

I should grab your dumps of the ECU's code Vikash, and try to start learning what's going on (after I learn how to read it). I'm starting to think that's going to be the only way we're ever going to get this stuff figured out.

I can email you the ROM dump if you like, Mike.

I totally expect to have the code all figured out by the end of this calendar year... Right now I just can't put time or effort into it.

A reduction in boost would be reasonable if the car was detecting knock.

But that wouldn't make the turbo spool slowly.

When in "low boost mode" it will take longer to get to 5psi than "normal mode" takes it to get to ~8.

Plus it's so sporadic, that it's unlikely. I'll be "low-boosting" in 3rd, and full-boosting in 4th.

Depending on how it's done, it would indeed make the turbo spool slowly. If the ECU just scraps its boost control map and simply stops using the boost control solenoid, then it will no longer accelerate spoolup like it does when running normally, right?

I noticed that too.... My boost gauge seemed more lively when I was getting the proper 9psi.

How big is the dump Vikash? I may need to give you an alternate email address. I honestly wouldn't mind looking at it.

About 20 kilobytes gzip'd.

That's it? Send away!

Done and done.

k, back to full boost again for no reason. Has been steady for a day or two.

Anyways, it WOULD explain the slow spool. I tried connecting the compressor outlet pressure hose directly to the wastegate, and got almost the exact same result. As far as I can tell, the boost solinoid is just set to "open" all the time when in this mode.

I went out to home depot in the hopes I could build a ball-spring MBC but ended up with a needle valve instead. Havn't hooked it up yet as the boost seems to work properly.

Vrg, did you actually build the mbc from scratch? or is there somewhere you can get a ball-spring valve that would be suitable?


edit: some lady just walked into the store and asked me if I could recycle her old 386 laptop for her. I gladly agreed. Scantool, here we come =D

I built it from scratch. I went to the hardware store and bought:
- A 1/4" female pipe tee
- A 1/4" male pipe to 3/16" hose barb adapter
- A 1/4" male pipe to 1/4" hose barb adapter
- A 35mm long M10x1.0mm bolt
- Two M10x1.0mm nuts
- Two 5/16" ball bearings
- A spring that nicely gripped the ball bearing and wouldn't bind on the end of the bolt and was about the right length to fit inside the tee without bowing

I drilled a 1mm hole at the base of the 1/4" barb.

I put one ball bearing into the 3/16" barb from the threaded side, and hammered it down a little with a drift, and then discarded the ball bearing.

I threaded the 3/16" barb into one side of the tee and the 1/4" barb into the middle of the tee, and soldered them in place.

Then I soldered one of the nuts to the remaining open side of the tee.

Then I put the remaining ball bearing in through the nut, then the spring, and then threaded the bolt through the remaining nut and then the nut soldered onto the tee.

The bolt adjusts the pretension on the spring. The extra nut acts as a locknut to hold a setting. The 3/16" barb is the inlet and seals against the ball bearing because it is deformed to fit correctly. The 1/4" barb is the outlet; the 1mm hole relieves pressure trapped between the valve and the wastegate.

I hope that makes some sense... Maybe Phil could take some pictures to make it clearer.

Yea, I can get some pictures tonight or tomorrow.

Just a quick update. I filled the tank with Pioneer 93 (up from Sunoco 92), drove a bit to clear the codes / run diagnostics, and got my boost back. I've done a couple of local trips (nothing on the highway so far), and every trip into boost land has netted me around 8-9psi, vice the 5-6psi from before. I'm going to be hitting the highway this weekend, but provided nothing explodes, I think I've discovered that the ECU does pull boost when it detects a knock event. Or more likely, several knock events.

So... Would that be an overly sensitive new knock sensor, or carbon deposits? I can't think of anything else that would cause the engine to knock on higher than required octane. Oh wait... Would oil in the intake track cause that? I can't remember. I'm suddenly down a quart, and it looks like its all over the PCV lines.

It's the exhaust and new o2 sensor....

Actually, I find that my car runs best on pioneer/sunoco - so maybe yours does too.

I once again was only getting 5-6psi, so I did a trip on the highway with the car in clear memory mode. This means I've finally followed the clear memory procedure properly, as per Josh's site. It looks like getting the car into 4th gear is critical.

Previous runs had just been on side streets, so while I'd get into 3rd, and be getting the proper boost readings, trips onto the highway meant I went back to the 5-6psi.

It looks like, based on what I've seen my car do, that the ECU takes a look at everything that happens as you go through the gears, and decides whether to allow full boost or not, based on the gear. I guess because I was only getting into 3rd that it didn't do anything with whatever variable it sets for 4th. The trip back today on the highway was fine in all gears, running Sunoco 92 again.

I still want to get rid of any carbon that may be hanging around, but I'm no longer positive knock causes a boost reduction.

Yeah, tearing down an 83k mi turbo engine, the only sign of use was heavy carbon build up on the valves and exhaust passages. A real pain to clean off, but at least I could work at the valves a little.

Steve

Mike - Huh, that's interesting... I wonder what would happen if you disconnected the VSS?

Steve - Was the buildup in the combustion chamber too?

Combustion chamber was fine - just normal varnish-color. Cleaned up easy. Build up was on the backs of the valves.

Steve