3.0 16V engine mods, etc

[Thom]

I said I would not touch the turbo but I could not resist and replaced the 0.82 A/R turbine housing with a 1.03 unit, again from Tial. The tongue area on the larger housing is a fair bit wider than on the smaller unit and "of course" interferes with the turbo mount on our cars considering how the housing must be rotated, and the mount needed quite a bit of grinding to "slide" in between the cold and hot sides. It was a tight fit, but a tight fit is still a fit. Well pleased with the results as there were moments I thought I would either have to refit the smaller housing or go with a Tial mount.

_20240520_121852.JPG (914.94 KiB) Viewed 1352 times

_20240520_121920.JPG (1.23 MiB) Viewed 1352 times

_20240520_121830.JPG (1.53 MiB) Viewed 1352 times

How does it drive then? Well, I had to add between 5 to 10% of fuel pretty much everywhere and lower the boost from 1.6 to 1.4 bar due to traction issues. Boost seems to build a little later but my understanding is that the net result of increasing flow with the larger housing is increased torque with less boost and the engine feels clearly faster with less of the frantic boost response allowed by the smaller housing. The performance increase through flow rather than boost makes for a more "refined" display of power, as if the engine was larger and normally-aspirated (notwithstanding "torque lag"). It is not easy to describe but not too dissimilar from what I experienced when upgrading from a GTX3076R 1.06 to the GTX3576R 0.82 on the previous, 3L 8v engine.
EGTs are also lower, even when cruising down low where the engine reacts more smoothly to light throttle inputs, which is great for cruising between 2000 and 2500 rpm in sixth gear.

_20240520_122547.JPG (2 MiB) Viewed 1352 times

As said, traction is even more of an issue as when torque kicks in at ~4200 rpm the rear tyres give up as can be seen in this log from the sudden rpm increase. It is fun but slightly overwhelming and requires even more respect and focus. The OS Giken LSD keeps doing a great job as the car remains absolutely straight and makes the car feel like the clutch is slipping when it's "only" the rear losing traction, even at speeds above 120 mph when flooring the throttle in 5th gear. It would be "interesting" to experience what the rear end would do with a stock type ZF LSD using a fixed locking rate... or not.

I will go back to the dyno in the coming weeks but I would not be surprised if the larger turbine housing provided the cranking power required by the compressor to max out the flow potential of the stock 968 top end, and the lower EGTs with improved smoothness even at the lowest loads suggest the engine simply runs more efficiently altogether, also suggested by another detail such as wastegate duty cycle which needed to be reduced by ~15% to reach the same boost targets, there again suggesting a higher mass flow rate.

Anyway, the display of power with such a strong top end and freely-revving engine reminds me of vids of 1000+hp Toyota Supras revving beyond 10000 rpm and losing traction at insane speeds, of which my 944 turbo may now be a scaled down version. What's not to like...

[blade7]

My motor -- which if I recall is a near clone of your old one -- is prone to knocking too. I've managed to keep the head on with less timing, e85 or high octane, water/meth injection, strong spark, etc. but that's a lot of work, so lately have been driving it around at 16-17psi. I have a catch can, which helps, but probably not as much as having it open to the atmosphere. The oil can't help for sure, but I've always figured the pistons were the root of the knocking issue. Despite advertising and labeling, they work out to nearly 9:1 compression, which doesn't sound like much these days, but is pretty high for that motor and engine management. The early 930's had a 6.5:1 CR for comparison...

I thought a benefit of running the sealed breather system was that it draws some pressure from the crankcase. Yes there's oil contamination, but as you mentioned a catch can helps.

[Thom]

On the topic of crankcase ventilation I had noticed whilst taking the turbo off that oil had leaked from the CHRA onto the mount. Since there had never been any leak with the catch can connected to the compressor inlet I just reconnected it. I will see if the air filter is now large enough to avoid excessive oil vapour suckage into the intake, but at least after yesterday's drive with the new turbine housing and more than a few full load runs I cannot spot any leakage on the turbo mount. Touch wood.

[Thom]

Here is a picture showing how close the 1.03 housing gets to the modified stock turbo mount. Spent some time today preparing the car for the biennial inspection coming soon and had a look at every corner for possible leaks/issue. All looking good.

[Boostfeen]

Wow it works though great pics and write up on the comparisons for turbo and flow rates/vs power

[Thom]

I did not have enough time for more tuning yet but simply driving to and back from the technical inspection today brought further to the light a totally unexpected virtue of the larger housing - as said earlier the engine is significantly smoother down low and slowly squeezing the throttle open allows increasing speed in top gear with the engine barely breaking a sweat while staying in a high-ish vacuum range. It was already possible with the smaller housing but nowhere near as smoothly and the engine didn't "like" it, whilst it now feels good doing it and enjoyable to the point I think it's worth reporting, again as if engine size had increased slightly. This improvement suggests to me that the whole set up has never been so well-balanced in terms of its various bottlenecks of lack thereof, and this puts to evidence the virtues of a free-flowing set up that does not necessarily needs to run under boost to shine.
Opening the throttle progressively certainly allows air to build velocity smoothly whereas going full load abruptly down low would force the turbo to build boost in an RPM range where it is not best reponsive. In fact at this stage I would say that the larger housing has brought more torque everywhere but perhaps only lacks in transitional loading phase, during boost build up when going WOT, where the smaller housing shone but elsewhere prevented maximising flow under steady state conditions, either under a vacuum or boost. The larger housing however, despite a slightly slower spool, allows for less "torque lag" as even during the transitional phase of building boost under full load the engine still makes more torque thanks to the increased flow - smoother but stronger torque build up, again as with a larger, normally-aspirated engine, only without the instant torque response typical of a high compression engine.

Anyway, the extra smoothness down low allowed by the larger turbine housing is a good surprise and makes sense for a street car that could run all day long at highway speeds in top gear with a very small throttle opening and minimal fuel consumption. The general drawback of "turbo lag" becomes I think a non-issue with that extra type of performance down low, at least when it only takes a shift down to turn the flying carpet into a missile. Flow is the real thing, and it seems to me that a good turbo engine should also need to perform "well enough" under vacuum.

[Bergerac]

Do you think the extra low down torque work with the long 6th in the diesel 01E? One of my concerns about that transaxle is it being sluggish cruising at 70mph @2200rpm (along with the 1st/2nd gap)

[Thom]

For me it has always worked, though never quite as well as it works now. I would say that 70 mph is the lowest speed the engine tolerates in 6th gear, below which it is asking too much from it. It will "work" as the turbo will build boost but it does not feel right to apply that much load down low. 6th gear is mostly to be considered as an overdrive anyway.
The gap between 1st and 2nd gear is fine with a 968 gearbox.

[333pg333]

It's amazing how far you've come Thom. Great capturing of your journey too. Thanks.

[Thom]

Thanks Patrick. At this stage it is probably just a matter of luck. I noticed oil leaking from the turbo mount after getting back from the technical inspection and thought I'd have to remove the turbo again to investigate but luckily I only needed to tighten both mounting bolts to fix the issue. Touch wood.

Fine tuning boost control has allowed improving traction, though it is not quite as fun as feeling the rear end get loose at the end of the rev range, so I settled on settings agressive just enough to stay right on the "edge". It is a very juvenile satisfaction to experience the engine hover between 5500 and 6000 rpm in 3rd gear as the rear tyres spin and struggle to regain traction, and start again in 4th at silly speeds. As observed earlier the figures suggest the engine runs cooler, with lower EGTs and all, and it seems there may be some headroom left for even more ignition timing, but I'm not sure what I would do with it except waste it into further rear wheel spin.

log.jpg (113.83 KiB) Viewed 857 times