axl911

Interesting....

Post it right here. It would be very interesting to learn & know more.

NineMeister

Sorry to take so long to get back to reporting on the outcome of this project, but it ended up being a bit of a saga (which I possibly could have seen from the get-go).

To summarise the story so far:
1. car booked in for Motec only
2. dyno test as-arrived shows odd fuelling & apparent resonance flap issue
3. resonance flap repaired
4. Motec fitted
5. Car back on dyno

At this point it should have been plain sailing, however reality showed an outcome that neither I nor the customer expected. As arrived the engine did not sound happy and certainly did not feel responsive, but the power we achieved seemed reasonable enough. However, once I had fitted the Motec we immediately ran into an engine misfire problem at idle which was not down to tuning and after intial power runs I really struggled to make an impact to the power numbers even though the engine was noticeably more responsive.

These were the power/torque graphs which show the engine as arrived (green) the very first run of the engine on the 9m Motec start map (red) and the last run following a couple of hours fine tuning on the dyno (blue).

The Lambda trace was interesting as well. Note how the initial motec start map was seriously rich (although level) and how the last motec lambda trace was much improved after pulling around 15% from the fuel table.

NineMeister

Obviously I had for the first time fallen well short of our expected performance result, so I started to look more closely at the engine for answers. In the end I suspected that we may be dealing with a bent valve, so I pulled the car back off the dyno, onto a ramp, removed the lower spark plugs and did a leak down test on the engine. The result of the test showed huge leakage on one cylinder into the exhaust system and all cylinders were heard to be leaking into the crankcase by taking off the oil tank filler cap & listening to the breather pipe; from memory the average leakdown from 5 cylinders was around 15% and the worst (cylinder #3) had 100% leakage.

"That would be worn cylinders & rings and a bent exhaust valve then, sir."

I broke the bad news to Kenn and he graciously accepted his engine's fate and authorised us to take it out and do the necessary repairs.

Unfortunately I did not have the chance to take any photos of the build because I was too busy working on other jobs, but suffice to say that we found plenty of faults to fix. Firstly the cam timing had been incorrectly set, the left bank (123)was 0.9mm retarded, the right bank (456) was 1.2mm retarded. On the 123 bank all the exhaust valves had connected with the pistons, cylinder 3 was bent badly and cylinder 1 had another exhaust valve that was tweeked slightly. The engine was an early 964 motor with the spigoted cylinders, hence no surprise either that the head to barrel joints had all been leaking and that the heads would need refacing. The rings were on their wear limit & the exhaust valve guides were also worn badly.

The budget for the engine rebuild was limited due to the committed spend on the Motec, however we did our best for Kenn and rebuilt it with fully refurbished heads (i.e. re-faced, new guides, recut seats), six straight exhaust valves, fresh rings and a good set of used 993 cylinders. I also took the time to make an additional piece to fill the gap between the SC engine mount and the 3.2 Carrera rear tinware piece which would also prevent any hot air from the exhaust from upcycling through the engine fan & causing unneccesary heat load on the engine.

Time for "take 2".....

NineMeister

So, back on the dyno I re-loaded the standard 9m Motec start map and started again from scratch. A quick break in proceedure for the rings and after quite a few full load runs I managed to dial in the fuel requirements to get the results below. I spent quite a bit of additional time playing around with injector timing as well as mixture because I wanted to achieve the best from the package of a 964 engine on SSI exhausts as this was the first time I had tuned this combination. Suffice to say I was happy with the end result: 333hp & 307lbft from 278bhp & 246lbft starting point.

The first graph shows the result of the engine rebuild work with before & after on Motec.
The second graph shows the final result with rebuild & Motec vs as arrived with Motronic.

When Kenn collected the car we spent some time out on the road to do a little fine tuning in traffic and the engine proved to be extremely flexible with the SSI's, even pulling from 1000rpm in 5th when the need arose. This was a sweet package that made the little ST replica an extremely quick car in any gear at any revs.

Wachuko

I want, I want, I need, I need, gimme, gimme...

NineMeister

Finally, because I ended up mapping this car the day before collection I decided to do a lot more work on the dyno working on the part throttle mixture settings because I knew that I would not have time to do this on the road. At the end of the part throttle session, I put together a series of dyno pulls from 50% throttle to 100% throttle in 10% increments (just for the fun of it). Hence this is what is actually happening at the flywheel when you lift off slightly to correct that unexpected oversteer moment......

(PS La table is not flat deliberately)

NineMeister

So, not much wrong with SSI's on the 3.6 from what I can tell? I especially like the flat torque band between 4500 & 5700rpm.

944CS

how did the engine make 278 hp with all the problems you mentioned?

NineMeister

It's a good question. The engine had been remapped, so I would have expected it to make 300hp all day long, add into the equation SSI's, the free flow back box & retarded cams the best I could see it achieving was maybe 310hp which is the best I've ever seen from a 964 on a stock Motronic installation. Note that the resonance flap would not have affected peak power because at that point it would be open anyway.

Hence 310hp ~ 52hp per cylinder. The cylinder with the bent valve also had a badly worn guide (indicating that the valve had been bent for a while), so at high speed there would have been sufficient mechanical distortion to allow it to make a seal on the seat and more importantly gas flow dynamics would increase the natural volumetric (cylinder filling) efficiency, so if we assume that the worn rings would drop power by 5% (15hp) and the valve by 35% on one cylinder (17hp) then the result would be 310hp -15hp -17hp =278hp. Obviously a simplistic approach, but it has some relevance.

More interesting to me was the complete lack of torque.

sergiDA

Hi Colin:

I have got a 964 c2 with cup pipe, Cat bypass, LFW, Steve Wong Chip, KN filter, and full rebuild 3000 kmts ago: What would be the cost of a conversion such as the one you have made to this Yellow 911 and who could do the installation right here in Barcelona?

May be you would be happy to spend a few days here in the sunny Catalan coast?

vanhauen

Colin. As I have already told you via e-mail, the car is simply fantastic! The 250 mile journey back was a real treat. With the weather being good last weekend, I had the opportunity to test it properly; the engine is sooo responsive and so strong from lower revs. I am very pleased indeed with the outcome! In fact, I will go as far as to say it is the best 911 I have ever driven and this includes the 997 4S 3.8L PDK version I had for a week earlier this year...yes, I wouldn't do a straight swap

crg53

Colin, those are impressive #'s, like you said, those SSI exchangers seem to work very well, perfect torque/hp curves for a street driven car.
What sort of muffler are you using with the SSI's ?

NineMeister

SergiDA,
Coming from the North West of England you should already guess that I'd never turn down the chance of a few days in the sunshine! We do a fixed price on the 9m Motec conversion of £3495 +VAT, just add onto this an allowance for travelling & other incidental expenses. The full conversion includes Motec M48, wiring harness, sensors, injectors, air intake & tps conversion, to do the conversion at your place I would need access to a working area with basic tools and equipment.

Provided your engine is in good condition you should see similar results to Kenn's car as long as you are running 98 octane super unleaded. The start maps are now very clost to optimum, so although I would prefer to tune the car on a dyno it is not essential and we could achieve very close to optimum by driving the car on a suitable road or a track.

NineMeister

Kenn's car is fitted with a conventional looking mild steel Dansk twin inlet silencer that has had the left side outlet plugged and is fitted with twin "pea-shooter" outlets at the rear.

Mind you, when I say pea-shooter I think they could actually fire fairly large potatoes....