Odd boost/fuel problem on a "customer" car

[chaloux]

I'll pose the question first: what am I dealing with here? I believe it's either an ignition or fuel supply problem. Logs attached.

Car is an S4 with a 30r, VEMS, 1000cc injectors, welded ebay header, big FMIC

Finally got around to re-tuning it. Previous tuner said it was running out of fuel, we found a variety of odd little problems which were all fixed, and just had it out this morning. Everything looks pretty good as far as ignition and VE table, but boost PID was on and the refDC table was pretty wonky. So I set PID to 0 and started with refDC. We maxd out at about 200kpa with the stock wastegate untouched, so we drilled it and ended up screwing it down all the way. He'll probably need a bigger spring because N75 DC is still quite high to achieve <200kpa

Anyway, once we hit about 255-260kpa the car starts bucking, somewhere inbetween a misfire and fuelcut. Fuel looks good and nothing else really looks suspect. What I'm thinking is that the fuel pump is just maxing out and the fuel disappears which is why it doesn't really show on the log. Lambda is great until the problem happens, and when it does, requested shoots up so it's even technically overfuelling. At ~240kpa everything is hunkydory. But more boost and it is not having a good time. Dad will be putting new plugs in as well as a fuel pressure gauge which should help us diagnose. I assumed that I would see the car leaning out if the fuel pump was unable to supply the fuel needed, but perhaps it just hits a wall where fuel pressure goes to near 0.

Any insight? Thanks!

[loxxrider]

Matt, have you tried leaning it out? That's an awful lot of fuel for that much boost.

[chaloux]

Thanks Chris. Wow... I didn't even think to check if they modified the lambda target table and it's been a little bit since I've had to redo everything from scratch so I wouldn't have seen the boost/corresponding lambda connection. They modified the lambda table in a big way! I just assumed it was the same. I should know by now never to assume anything!

That makes a lot of sense that it would be rich misfire. That's sort of what it felt like, I felt it on the 4k. I'll have to import a normal table and roll with that. I bet it just couldn't light the mixture with that much air and fuel. Too dense.

Will report back! Thanks for that buddy.

[bradyzq]

Hmm, that's weird. I'd have thought that with a cranked down 33mm wastegate and a 30R would be a recipe for boost creep, yet you're losing boost.

Check your battery voltage too.

[chaloux]

Quick update. Imported my lambda table from the 4k, took out some fuel, then added some fuel up top. Doesn't seem like the pump is maxed out yet. But there's definitely a slight misfire. Going to check plug gap tomorrow and go from there. I tried adding some more dwell which seemed to help a little, but didn't totally eliminate the misfire.

The engine in this car was essentially left untouched, it got a TDI tranny install, and a huge laundry list of items addressed (eBay manifold welding and machined flat, tons of random suspension issues and small electrical issues etc). And now a retune.

[SEStone]

I can't look at the logs, but I've gotten misfires like that from running too little timing. Obviously use experience and be super careful going the other direction, but I've done it a few times.

Sam

[chaloux]

Yes me as well, when I was overly cautious in the 4k after the 2nd rebuild.

Figured a few things out. We had multiple things going on.

1) rich misfire. This was combined with
2) huge plug gap @.036 on ngk bkr7e causing ignition misfires and
3) stock wastegate spring requiring huge n75 DC to make boost

Solutions were obviously to take out fuel, regap to .028, and put in a 1.9 or 2.2 bar spring dad had at the shop

Without the misfires the car is absolutely out of fuel pump. Boost drops from 240kpa peak to 200kpa redline in order to keep the lambda happy.