Power drops significantly at 4500rpm

[archangel62]

Hey all, so I'm thinking of addressing a long-term issue of my car. I'd been ignoring it because I intend to go aftermarket EFI, but if it's a relatively simple fix, I'd love to get it sorted. The car has always done this, since I first put it together. Note that I bought the thing without an engine, so figuring out the complex engine bay has been an uphill battle, and I may have got something wrong.

Here's what my engine does, and always has done:
- Hesitates off idle with hard throttle application
- Feels intermittently sluggish until 3000rpm, but not lag-related - it feels more like fuel or ignition is holding it back
- Runs out of power very suddenly around 4500rpm - and runs marginally better with less throttle beyond this point, ie 2/3 throttle goes marginally faster than flat out

Observations:
- Airflow meter is very clean and can move freely throughout its range
- If timing is advanced any further it pings, if retarded it misses
- Vac advance is connected
- The car seems to overboost marginally in high rev range, perhaps around 10-11psi if pushed beyond 5k, but no more - suspect wasegate creep from poor flow

Everything's "stock" - well, forged pistons, ARP head studs, Top Gun leads and Iridium splugs, but stock compression ratio, ECU, sensors, injectors, etc.

My crazy theories:
- Timing wrong somehow, ie the wrong type of 4Z*1 dizzy, or a dizzy failure
- Fuel pressure not correct, perhaps from how I've hooked up the red and green pod things (or black switch box) that hook up to the vac line to the fuel pressure reg, or maybe blocked fuel filter
- Knock sensor freaking out?
- Iridium plugs producing poorer-than-factory spark

Any thoughts? I don't suppose anyone could put up some pictures, diagrams or info about the stuff connecting to the fuel pressure reg, to make sure I got it right?

[IZU069]

I'd check or substitute the O2 sensor.

The 4ZC1-t dizzy should be readily available and its ignitor identifiable due to its extra lead(s) (if my memory is correct...).
But also check that neither if its springs are broken.

With the few 4ZC1-ts I tuned, I found that retarding the dizzy has the usual retard slow down, and then advancing had the usual advance speed-up until a point of no change was reached - ie, the knock mechanism kicks in.
Not that I advanced further to obtain pinging, but maybe it is possible to advance beyond the point of compensation - ie, many anti-knocks operate by retarding a set number of degrees (or multiples thereof) and maybe its programming has some upper limit. Not that those systems occasionally remove that retard to recheck for knocking.
But IMO checking for similar behaviour by retarding & advancing the dizzy at idle should confirm that the anti-knock is at least working - at least to a basic extent. High-speed or full operation may be compromised due to circuit breakdown (as I recall, some of that circuitry is analog rather than a pure digital (EMS) solution).

[archangel62]

Ahh yes, cheers for the reply! I'll add o2 sensor to the list of things to check! I'm almost tempted to disconnect the knock sensor and tentatively run it and see if it changes anything - if it pings, I can back off straight away, but if it's the same then I can rule it out.

Are o2 sensors fairly generic? Or would the Piazza need something very specific?

[IZU069]

I think O2 sensors are reasonably generic, especially for single wire jobs.

[archangel62]

Slight update of sorts, the crappy stock turbo manifold has cracked again. I've welded it up but the flange/gasket area still has a massive ridge in it from the crack, it's so uneven that it'd need to be removed, studs taken out, machined and re-fitted to have a hope of sealing - and then the manifold could just crack again and repeat the issue.

So basically I'm not going to bother. I've started talking to a workshop about fitting my GT2860RS and aftermarket manifold, as well as my Microtech LT8s and a tune. I'm having some 380cc Merc hose-tail injectors flushed and then I'll fit them into my spare inlet manifold. Then I'll swap manifolds and wire the new ECU, and somehow get the car down to the workshop. With a bit of luck I should see 160-190rwkw with good response, and do away with both the flat spot and the massive exhaust leak!

[Piazza_man]

Which exhaust manifold did you go for?

[archangel62]

FIXED IT!

Ahahaha ooohhh dear... So of the three fuel pressure regs (AFAIK), one of of them had the vac line hooked up wrong. The lower of the two regs on the driver's side shock tower, I had the vac line hooked up to the tiny black cube near the cruise control etc. This is how I'd seen it on other Piazzas. Anyway, I saw a picture where they'd connected it directly to the inlet plenum, the hard metal line under where it says "turbo", towards the right - which goes to the throttle body. As soon as I did this, bam! Good power to 6000rpm+.

Pity the standard exhaust manifold is cracked to shit now and I'm about to rip off the standard turbo and ECU and upgrade it all.. Oh well!

[Bugle]

The tiny black cube thing is correct, it's a vacuum switching valve controlled by the ECU, maybe the vacuum source to that is wrong?

[Piazza_man]

Well that's finally some good news for you.

[archangel62]

Bugle - the vac source was good, it's now re-routed into the pressure reg and runs well. I'm guessing whatever switches it (ie ECU) isn't right or isn't behaving itself. Regardless, this will do the job until the car gets its big overhaul and new turbo. As it is, I'm using exhaust concrete (eww) to band-aid the cracked and warped turbo manifold until I can bring it in to get the overhaul... so that's going to be a fairly finite length of time (read: until I can fix the R31 daily and have it behave for more than a week).

Definitely good news though. It's nice to feel what a Piazza is meant to be like, before modifying it. I'm a big fan of that, hence why I went to such extraordinary lengths to get it running with standard bolt-ons in the first place, instead of just going aftermarket from the word go.