weird boost issue

[ehsnils]

B.t.w. - have you matched the manifold pressure value with what your boost gauge tells you? They shall be very close in value. If they aren't then you can look at the MAP sensor.

[cypher0117]

Gotcha iNVAR. That makes more sense. So there really shouldn't be any way to overrun your injectors until you literally have them completely open 100% of the time.

 

ehsnils. My boost gauge is fairly close from what I get from the manifold pressure in logs. I don't stare at the thing while doing pulls but I have the alarm set to a half tick over my peak boost target. So when it goes over I hear it audibly, then take a quick glance.

[cypher0117]

Another question. When I am in open loop, does the ecu just "guess" how much fuel based on maf readings by dumping fuel? Or does it actually look at the learned a/f ratios

[iNVAR]

The former.

[cypher0117]

Here is another wrench to throw into the fire......

[iNVAR]

Your fuel trims still indicate a boost leak... You are subtracting a fair amount of fuel in your upper ranges, which would explain the knock. The CEL is what happens when you overboost. You hit the boost limit and tripped the fuel cut, and you surely would've felt it if you were accelerating at the time. The entire car would've lurched. Do you remember exactly what you were doing? Gear? Uphill? Shifting? RPM? Speed? Better yet, do you have a log of it?

 

Go check to see what your boost limit is under the boost control limits. Post them here.

[PtPixel]

Also post your WOT initial wastegate duties. Poor tuning can cause the boost control to act too slowly, causing overboost

[cypher0117]

It felt like boost cut. I was passing a semi on the freeway. I did this twice last night and both times I hit ~20psi in the logs. There is a blank line to separate out both passes in the log. The second time is where the boost cut hit.

 

I also uploaded a picture of my boost limits. All four tables.

boost cut log.csv

[cypher0117]

PtPixel. That's what I think might be causing the overboost. Putting the full exhaust on dropped the back pressure. So the turbo can spin up that much faster. Turbo is spooling faster than what was tuned and the gs ebcs isn't compensating fast enough. I had this same tune on the stock exhaust last winter, with colder temps and never saw anything above 18psi

[iNVAR]

No. It's a partial throttle boost tuning problem. Infamous, and many other tuners, never bother to tune the partial boost, and as a result, there are certain conditions such as this which can cause a spike and boost cut. You see how your Turbo Dynamics Integral is way up there? It's because at 20-30% throttle, your boost target is set too high, which results in a high Boost Error. So TDI keeps accumulating. When you suddenly apply WOT, that extra TDI gets added on top of your normal WGDC which causes you to briefly overshoot to 20+, until TDI winds back down.

 

The way to combat this is by 1) Lowering boost targets and 2) Increasing the boost target activation threshold to 7 or 8 PSI.

 

(1) In effect prevents your Boost Error from being so high. You're simply not going to achieve 5 or 6 PSI at 20-30% throttle no matter what gear you're in, so why are the targets set so high? Most tuners don't touch those from stock, and on stock, it wouldn't be a problem, but on a Stage 2 setup, as you surmised, boost comes up a lot quicker so you overshoot by a lot if the TDI is wound up

 

(2) Will prevent TD from even donig anything until you are targetting a boost of 7/8. Your boost control system is useless for anything below 7 or 8 because your WG can't even open until you develop 7 or 8 psi.

 

I'll help you out later when I'm at home and have my map in front of me.

 

In the meantime, post up your Target Boost table.

[cypher0117]

Target Boost Table A and B are identical. Only A is shown

[cypher0117]

Also food for thought. I have been looking through my past LV I've taken. I have uploaded my stage 1 LV. Some from summer and some from cool november weather. My fuel trims are very similar to how they are now. So I guess I've had this issue for more than 2 years and with 2 different incomplete tunes. I also find it curious as to why the knock that I'm seeing now isn't apparent during summer but is during cooler weather. I also found my LV from when I discovered my inlet leak. Idle trim is stupid high, but the rest seem to be in check.

 

Maybe I'm just comparing apples to oranges at this point..

[iNVAR]

See how at 30% throttle it calls for 13.3 psi by 2800? Good luck with that ever happening... I'm sure Infamous had his reasons for doing that, but I don't know what they are exactly...

 

Anyway, backup your map somewhere.

 

Under Boost Control - Turbo Dynamics->

TD Activation Threshold (Target Boost), put in 6.97 and 7.50 for Disable Below and Enable Above

 

Save and reflash. Let's see if that helps a bit. I don't want to touch your WGDC and Target Boost tables unless I have to because then it starts getting really tedious.

 

As far as your questions, the knock you saw in this most recent log is a result of the spike. Remember what I said about a small spike being ok IF you don't knock? Well, you're knocking, so that's not good.

 

And your fuel trims, not sure... Check the scaling between the two maps and see if they're the same. Also compare your scaling to a stock map.

[cypher0117]

made the modification, wrote it. Seemed to be better for a few miles. Boost built slower but never went over. Then got on the highway and just did some quick pulls like I did last night. Still over boosting, but as soon as I heard my gauge alarm I let off. Still overboosting and still knocking...

stilloverboost.csv

[iNVAR]

Did you ever pull any timing like I suggested back on page 3?

 

Also, you're overboosting because you are flooring it in 5th at 2600. If you want to accelerate, downshift to 4th. If you absolutely insist on doing that, then a LOT of work has to be done to compensate for that sort of condition without sacrificing too much in boost spool. It involves pulling WGDC and readjusting your TD tables a lot. And it will really, really require you to understand how to tune your own car. I don't mind explaining stuff to you, but you have to understand it really is a lot of work involved.

[cypher0117]

I have no problem devoting some time to changing things, but I don't want to just be changing numbers not knowing what it's doing.

 

I'm also still a bit confused as to why I didn't have this problem last winter

[iNVAR]

You're reducing timing. If you make changes gradually in small steps, you can't do much, if any, damage, especially not by retarding timing.

 

As far as this problem occurring last winter, some people have lucked out before without this sort of problem. It's a combination of temperature, humidity, uphill, weight in the car, etc. and sometimes it's just the right condition to push you over the edge and trip the CEL.

 

Again, don't floor it in 5th at 2500-2600 RPM and expect it to behave 100% correctly without huge adjustments. Infamous and most tuners gear their tunes towards 3rd/4th behavior and quick spool, which requires having WGDC higher between 2500-3000. If you want to be able to floor it in 5th at 2500, you have to reduce WGDC in 2500-3000 and also make adjustments to your TD tables. It's not a small undertaking.

[iNVAR]

I'd be surprised if it was your catback that's doing it.... You had a full DP, not a shorty one, right?

[cypher0117]

cnt catted dp

[cypher0117]

Also, I just need to pull the timing on the base timing right?

[iNVAR]

Correct. Try to do it smoothly. In general, the timing table should increase going downwards, and decrease going rightwards. You need to learn how to work with the airboy spreadsheet because what matters is TOTAL timing, and in order to get total timing, you do Base + Advance. The advance table isn't the same value in all the cells (as you already know) so it may look smooth in the Base table but if the advance table is doing 8 in one cell and the next cell is only doing 5, your resultant timing won't be smooth.

 

http://www.romraider.com/forum/topic957.html

 

You have to get both your base and advance there, then have it calculate your total timing. Make adjustments in the spreadsheet for total timing, then have it recalculate your base timing (by subtracting Advance from Total). Then use the new base table.

[cypher0117]

seems as though my car is just starting to fall apart.......

 

I replaced the front o2 sensor, and now my fuel trims look a ton better. A, B, and D are all withing 2% of 0. C is 7%.

 

But last night just cruising on the freeway my IAM dropped to 0 again. I was not in boost, I was not accelerating. 5th gear and 16-18% throttle and suddenly IAM dropped to 0. I can post a log when I get home.

 

probably related: Had a coolant leak pop up about a week ago. Took a while to chase down. It was leaking out of both hoses of my turbo coolant reservoir. That is fixed and no longer leaking.

 

New issue: As of yesterday, when the turbo starts to spool I hear what sounds like a small supercharger. I hope it's just the fact that I somehow effed up an o2 sensor install and it's just leaking, but with everything else going on I feel like my turbo MIGHT be ready to explode, and maybe it has already been sending fragments into my engine. My oil pressure is higher than normal (30psi warm at idle).

[cypher0117]

In the matter of minutes my knock sensor literally went nuts. It's rolled over the knock count 3 times just cruising on the highway. I pulled my dp off to check my turbo and shaft play is near non-existant. Though I did find some interesting colors. Plugs look alright, but the turbine and front o2 sensor are white. It would not rub off either.

lots-o-knock.csv

[ehsnils]

Looks like a very lean condition.

[iNVAR]

Hmm that doesn't look good. Not really sure here... White usually means lean if I'm not mistaken. Maybe your MAF *is* going bad. That would explain the AF trims. Or maybe your O2 sensor. Hmm, very very interesting.

 

Take it very easy on the car until you figure this out.... Very lean condition usually = kaboom.