Limp mode under high load

[rzyao]

I've been having a strange issue where if I ask the engine for too much load (say WOT in gears 3-5) my boost cuts entirely and either comes back after cruising for a little while or doesn't come back until I get a code reader and tell it to clear the codes. Unfortunately, I don't have vag-com working on my computer so I can't check any other codes than the basic CEL ones.

I'm assuming the car is hitting limp mode for some reason and the N75 is left open by the computer, though I don't know what could be causing it as it pulls hard until the boost cuts out. I've tried cleaning the MAF to no avail but I suspect it could be a boost leak, has anyone else had this problem and know of any good places to start looking?


---
EDIT: After installing a new N75 valve, the problem went away and I now hold a steady 16-17 psi. For anyone else having this issue who finds this thread, this is what solved it for me.

[Cybersombosis]

Do a boost leak test. You are most likely getting a lean code and the computer puts the car into limp mode. Get VCDS working and scan the car for codes ASAP and report back.

[Blazius]

Is the car 'tuned'?

[rzyao]

Do a boost leak test. You are most likely getting a lean code and the computer puts the car into limp mode. Get VCDS working and scan the car for codes ASAP and report back.

I'll have to order an OBD cable, but I'll see if there are any historic codes I can find when I get the chance. Is this a common symptom of boost leaks? I do plan on building a smoke tester to take a look as well.

Sent from my SM-A536V using Audizine Forum mobile app

[rzyao]

Is the car 'tuned'?

The car has a GIAC chip and boosts to about 18 psi

Sent from my SM-A536V using Audizine Forum mobile app

[walky_talky20]

Boost leaks generally cause a rich condition. You are losing air that the MAF already counted and added fuel for, so you end up rich.

To get a "lean in boost" situation, the leak would need to be in between the MAF and the turbo inlet (air getting sucked in, but NOT getting pulled through the Air Filter / MAF). For example a tear in the Turbo inlet pipe could do this.

Homework:
- We need to know what Year/Engine your car is. The wideband cars will behave much differently with regard to lean/overboost issues.
- Take a look at your fuel trims in Measuring Block 032. This will give a lot of insight into any air metering issues.

Extra Credit:
- Data log of Actual vs Target AFR (if wideband)
- Data Log of MAF
- Obviously these logs on a 3rd gear WOT pull or wherever you can best reproduce the issue.

[rzyao]

I've been having a (maybe not?) unrelated issue where my diverter valve will get stuck closed and I get flutter, so if the issue is something to do with my TIP replacing my DV could end up solving the issue. Should I start looking for possible boost leaks around the intake pre-turbo? It sounds like limp mode is only triggered by a lean condition.

The car is a 2000 with ATW, so I believe it doesn't have a wideband ECU or sensor.

When I get a chance (likely in a week or so), I'll take a log of everything I can.

[walky_talky20]

A week is enough time to eBay yourself a hex v2 22.3

Then you can see all the things.

[Cybersombosis]

Does it run better with the MAF disconnected? It should run rich if it doesn’t have a MAF signal.

[rzyao]

I made myself a smoke tester today, and I wasn't actually able to find any leaks. I think soon my cable will come in and I'll be able to check the codes, but for now I suspect the diverter valve is causing a leak of some kind.

[QuattroBucc]

I really really doubt theres absolutely zero leaks.... probably a faulty test. boost leak test it up to the psi you run. Most people think they have no vacuum/boost leaks but when they pressure test it up to 20 psi they find a small/big leak at almost every hose clamp connection etc

[rzyao]

It was a smoke test and I think I'll run it again but the system was full of smoke and I didn't see any coming out of anywhere

[Cybersombosis]

You need to pressurize the system. This in combination with smoke may be easier to pinpoint a leak. Get a 3” ABS end cap and cut a 1/2 inch hole in the end. Get a tire valve and pull it through the end. Undo your hose clamp to your air box and shove the end cap in the end of the TIP and tighten the clamp. Take a wine cork and plug the PCV pipe hole in the TIP. Get a hand pump or a compressor and a tire inflator valve and pressurize the system and listen for leaks.

8BE0A270-8FF8-4712-807C-3ED905EB405F.jpg

[rzyao]

Well, I checked again and as it turns out one of my PCV hoses is just entirely disconnected (and torn). I don't know why I didn't catch that last night, but it definitely gives me a lead on where to start looking.

[QuattroBucc]

completely disconnected pcv hose and no smoke? very faulty test then. You need to pressure test it

[rzyao]

No, it was absolutely pouring smoke after I redid the test haha that's how I found it

[rzyao]

Just installed a new diverter valve and still hit limp mode even with my PCV system blocked off, I'm getting a P1555 Charge Pressure Maximum Limit Exceeded, and I don't know what could be causing it.

[Blazius]

Boost regulation not working well due to many reasons, bad tune etc.

[walky_talky20]

Faulty N75 or faulty software map.

You can probably fix it by adding a ball/spring type manual boost controller in parallel with the N75 which will set an upper boost limit mechanically without affecting the boost curve lower in the map.

[Blazius]

Ditch the mbcs. If you need an mbc something is not working right. Should never ever be an mbc plumbed in, there is a reason for the boost PID.

[rzyao]

I suspect it's an issue with the tune as my wastegate still opens when in limp mode and I never boost past 20 psi, which the car sustains briefly before shutting off all boost. I'm going to try to get my PCV leaks all sorted out but until then I'm not sure if there's a way to see my requested vs actual boost levels while logging, which might be useful to figure out exactly what's happening.

[walky_talky20]

That's a bit of boost.

One of the failure modes of the N75 is getting "lazy" and allowing boost to go higher than it should. This is probably more true when it's both 22 years old AND taking double the stock boost pressure *right in the face*. If you've already checked for leaks between the compressor housing and the wastegate actuator, a fresh "tight" N75 might be worth a shot.

Note: "Lazy" probably isn't the right word. More like "loose" or "easy", as you get more N75 activity (more control pressure dumped to the TIP) when energized at a given PWM than it ought to give.

If hardware is fine and changing the software is not practical, a manual boost controller in parallel may the only reasonable option. Parallel installation leaves the ability for limp mode to still cut the boost as normal. You definitely don't want to eliminate the N75 or go in series with it as you want to retain the ECU's boost-cut capability.