New Rev G DV failed within 50 miles?

[pudding]

Probably why the K04s were more expensive - to recoup the tooling costs to make the DV parts 

The DV was moved back onto the turbo again for the EA888 engines, both on the BogWiener and IHI turbos, so it's clearly VAGs preferred dumping method.

[Hypertuned]

Probably why the K04s were more expensive - to recoup the tooling costs to make the DV parts 

The DV was moved back onto the turbo again for the EA888 engines, both on the BogWiener and IHI turbos, so it's clearly VAGs preferred dumping method.

Was that on the lower end models? I know the MK5 Edition 30, MK6 Golf R & S3 all had their DV moved to near the throttle.

Sent from my SM-N9750 using Tapatalk

[pudding]

Every car except the 6R, S3, ED35, TT-S, Scirocco R and maybe the K1 Cupra R.  They all basically had a derivative of the EA113 CDL engine....BYD, CDL, CDLA, et al.  All E888 engines have the DV on the turbo, even the high power ones.

The K04 EA113 was a bit of a mongrel engine, both in terms of it's long block and it's ancillaries.

[Hypertuned]

Every car except the 6R, S3, ED35, TT-S, Scirocco R and maybe the K1 Cupra R.  They all basically had a derivative of the EA113 CDL engine....BYD, CDL, CDLA, et al.  All E888 engines have the DV on the turbo, even the high power ones.

The K04 EA113 was a bit of a mongrel engine, both in terms of it's long block and it's ancillaries.

Mongel..haha! But a good engine with a lot more character than the EA888.

Sent from my SM-N9750 using Tapatalk

[breeze]

I just replaced the o-rings on the intercooler hose and went for a drive.

No change at all.

DV looks to still be fitted properly, next job is to remove and inspect.

Question: Should I be considering a revision D valve? Not keen on swapping as they break (if I find that the new G is split). Aware of the pros and cons and that the D valves can leak from the factory. I should say that VW did try and sell me the D and did say that it was recommended for the car.

[Hypertuned]

Have you considered one of these?

Sent from my SM-N9750 using Tapatalk

[breeze]

I have seen the GFB kit but honestly I’m not that keen. The factory D part should be good enough and I have read mixed reviews of the GFB solution.

I still need to confirm that it is the DV although that is looking most likely. I feel like I’ve read every single forum post involving GTI and ‘boost leak’ and it does point towards the DV or something big like a hole in the intercooler when the drop is as significant as it is.

Going back over the timeline, it looks like I had about an hour at the high boost before something failed. Expensive. 

[pudding]

Unlikely to be a hole in the intercooler. The A/C condenser takes the Lion's share of the stone chip abuse.

What are the other O rings like?  Oil patches around the two intercooler connections? Assuming you still have the standard GTI one fitted of course.

Rev D works fine.  The people saying they 'leak from the factory' are usually tuners running Stage 2+ boost levels.  It's a plastic piston at the end of the day and has it's limitations. I've used it on Stage 1 tunes without any problems.  Boost holds just fine.

I don't like the GFB either.  I'm on my 3rd.  They do work well but they need regular-ish maintenance. The fact the throttle response off boost is so much better with the GFB is the only reason I persevere with them.

[breeze]

So… I just took the new Rev G out. It looks fine. Then I compared it to the old Rev C. And that looked fine as well. I am wondering whether both valves are actually OK. Aware that the internals can fail to hold boost, but certainly there was no gaping tear in the rubber.

There is a chance that what is causing the problem is electrical. I disconnected the battery in order to do the engine mounts, so I am thinking that could have reset something.

No fault codes prior to replacing the C or after fitting the G.

I had a good look around the turbo outlet pipe and it looked fine. Nothing obvious.

Stumped. I think the next port of call is probably something along the lines of a physical boost leak test. Possibly N75 or wastegate actuator.

I am very tempted to buy a cheap piston style DV just to completely rule it out.

[pudding]

Log the actual and requested boost channels, along with the N75 duty cycle.

A sharp drop off in boost along with a bigger than usual N75 duty usually means wastegate trouble.

Probably also worth logging the fuel pressure (LPFP) and actual & requested rail pressures (HPFP) just to make sure fuel starvation isn't causing a boost cut.

[breeze]

VCDS ordered. Will do some logging.

In the meantime, is there any kind of ‘soft limp mode’ that the car will go into if there are issues? Something that is reset when the battery is disconnected?

I’ve now pulled apart both the C and G DVs and the pistons and rubber seals both look fine. That doesn’t mean that they are fine, but does reduce the chance of issues. I think the swap of the DVs and the battery disconnect is what allowed me to hit full boost, but that is no longer available, hence the ‘soft limp mode’ thinking. Car still revs to the redline.

[pudding]

Good stuff. VCDS is invaluable for troubleshooting stuff that doesn't light up the CEL.

There certainly is.  Mine is currently flopping into and out of limp mode frequently these days.  Some days it's a tarmac shredder, and other days it struggles to keep up with an Hyundai i20.  As usual, zero fault codes.  Just the ECU being all grumpy cat on me.

When you pull the battery, the adaptations and learned values are wiped from the ECU memory, so you get your peak performance back again.  Then over some miles (it can take up to 250 miles to relearn) it gradually pulls the fun back. It's why most tuned GTIs feel epic after a remap, and pulls some mega hp/tq numbers on the first few cold dyno runs, and then a month later......meh.
Obviously if the car is a daily, this behaviour is far more noticeable than a weekend toy.

Anyway.....the only 'hard' limp modes I'm aware of are from throttle and crank sensor failures.  The latter won't allow more than 3000rpm and the former won't allow any more than a 10mph crawl.  Everything else (inc the cam and MAF sensors), the car drives relatively normally - hence why minor problems are so hard to track down.  The ECU is too good at band aiding problems to get people home.

Keep us posted!

[breeze]

Do I need to subtract atmospheric pressure from VCDS logs for actual and specified boost?

[breeze]

The conclusion is that all is well…

Long story short, the app (Car Scanner) that I was using was showing corrected peak boost pressures. In other words it was correctly deducting atmospheric. I thought it wasn’t, so was subtracting 14.5 PSI (ish).

The reason why I thought it wasn’t deducting atmospheric:
- With a brand new Rev G fitted, it showed ~34 PSI peak (and was briefly crazy fast). I read this as 20 PSI.
- After the first drive it started showing ~24 PSI peak (and was much slower). I read this as 10 PSI.

What I think happened is that the boost spiked massively after fitting the new Rev G diverter valve. This was a one-off. Maybe because the battery had been disconnected and all of the adaptations had been lost. That was enough to register on the app. It settled down to a much more normal level for a Stage 1 car after a little while.

I had VCDS delivered yesterday. With logging it was boosting to 20 PSI, with specified and actual matching very closely. In other words, there are (now) no issues at all.

Lessons learned:
- My old Rev C was probably fine.
- You may get a one-off spike with a new diverter valve/battery reset. If you do, it will be fun.
- Car Scanner reports corrected values. Not with perfect accuracy, but for £5 or so quite useful. Probably good enough to tell you if you have a significant boost leak (and whether you need to replace your DV).
- VCDS not offering a mobile app for £225 is ridiculous (sorry, it’s 2022… a laptop on the passenger seat makes no sense). And no, a WiFi web interface for £430 does not count…

Hopefully this is useful to someone. Got there in the end.

Thanks for all the help.

[pudding]

Ah yes, the mild annoyance of fitting the MAP sensor in the boost pipe instead of the intake manifold.  Glad you figured that one out.

Yep, the wheels of progress turn very slowly at Ross-Tech.  I have VCDS 'mobile', but it's basically just a sh1tty little XML thing that runs in a web browser.  It can do 'Gauges' but only 4 at once from memory, and setting that up is buggy as hell.  And of course, running that on a phone means you have to switch off auto screen lock before using it, or setup and automation.....but iOS doesn't seem to have an automation to disable screen screen lock on a per app basis. 

So yeah, VCDS mobile is completely shat really so I use the laptop.  Ross-Tech have a forum and if you ever mention how terrible it is, they just churn out the same indifference.  "Oh it's a beta and not supported".  Yeah, it's been in beta for years.

Until a rival can figure out to read and translate the VAG specific PIDs like they have, we are stuck with it.