Snagging List of Modified 3.2 JTS,

[Ascari32]

 I seem to get a few weeks where I am pretty happy with the cars performance, then niggling issues creep into dampen the mood.

The latest was a failed M.O.T. and the expense of new plugs and coil packs. Even so, the garage managed to get a pass on retest, but said, despite all the effort and nearly £300 in garage costs there were still occasional misfires.

This coincided with a spell where I was topping up with E10, RON 95.

I have reverted back to E5, RON 99 which brought back the sparkle of the engine, with no misfire issues, even under the hardest of acceleration. Moral - "This engine is 11.2:1 compression ratio so demands the highest octane rating, particularly as when she was rebuilt, both the heads and the block were skimmed, increasing the C/R by a fraction more.

I will not make that mistake again.

MAF issues have dogged the performance at either the bottom end, with the top end sublime, or poor top end, yet extremely good tractability at the low end.

The MAF was modified by removing the slatted screen which greatly improved matters. However, testing with four new MAF sensors never seemed to suit the engine entirely and doubts began to creep in as to what the problem may be. Yes, I have avoided remapping, yet she has recorded 291PS on the Dynamometer, so she isn't that far away from what should be possible. And I would not entertain remapping until I was happy with the mechanical alignment regardless.

Having removed the slatted screen, I wondered if the venturi tube into which the MAF elements fit, may be having some additional effect on the performance. Still needing to fit a normal screen across the venturi, where the slatted screen had been, I resumed testing. In this instance, from the Bosch data sheets, I decided to try elements - 280 - 218 - 019, from a 2.0 GTV, 280 - 217 - 531, from a 3.0 GTV and a 280 - 002-421, from a Porsche 3.8/4.0L Turbo.

The Porsche element struggled at the bottom end, but above 3000 rpm is absolutely glorious. Whereas, whilst the 3.0 GTV element proved sharper than the 2.0, both where relatively dull compared with the Porsche element. All three however, seemed to be a bit raucous/edgy at the bottom end. The correct element for the 3.2 JTS is better in this regard but it quite clearly runs the engine on the rich side and the noise iis excessive.

Given, I still had to fit a normal screen across the venturi tube, to protect the sensor, I decided to "Bin" the internals altogether. So both the venturi tube and the remaining screen to which it is attached have been stripped out and the internal surface smoothed over to remove any sign that the body was anything other than like either the Porsche, 2.0 or 3.0 GTV bodies. I then fitted the screen from a 2.0GTV MAF body, to the now naked 3.2 JTS MAF body front face, fixing it with Epoxy Resin.

The newly modified MAF body has been re - installed and the ECU picked up on it immediately. Currently the 2.0 GTV element is fitted and progression is really good, with excellent torque from below 1000 rpm. and sharp throttle response. Despite the lower rating of this element, I would say its performance is still considerably better than the standard JTS engine. I will have a few days testing with it first, before seeing how well the 3.0 element performs. If it is anything like the difference between them, when fitted to the 3.0 Busso, I think there will be a marked improvement.

Again, I shall have a few days testing before finally testing the Porsche Element again. I have high hopes for that. With improved airflow through the MAF Body, and the MAF Element appearing less of an obstruction than hitherto when inside the Venturi Tube, then it's performance at low revs could be sufficient to warrant it being permanently fitted.         

[Ascari32]

And in conclusion, having completely "Gutted" the 3.2 JTS MAF Body, 280-218-182, it now represents a 3.0 GTV/3.2 GTA version of Bosch's MAF, the exception being, the body is a greater diameter.

Finally, I decided to "Bite the Bullet" and remove my 3.0 GTV's MAF Sensor which has less than 1000 miles of usage on it.  Clearly the original 3.2 JTS Bosch Sensor will not work unless it is fitted within the Venturi Tube of the original body. Bosch clearly state the Slatted Screen enables the sensor to function by measuring a very small portion of the airflow and I assume this is because of the approach they took with regard to "Emission Control"

It has not been easy, not having the resources to develop/research, how best to optimise airflow and fuel rates appropriate for getting the best performance from my modified engine. It still rankles me that after all my modifications, the engine still was sub 300PS. And every change/modification comes a a financial cost, given to get from A to B invariably means modifying and original component, which can never be returned to its original state. Thus one is faced with having to purchasing a replacement, in the event the modification doesn't work.

Historically, that was never a problem for me - I was paid to make prototypes work and any cost was borne by my employer/customer. Now, that cost is wholly down to me and being retired, there are many other things which have a prior claim on my resources. So, a considerable part of the process in deciding whether to modify or not is spent  researching, analysing and investigating other issues not immediately obviously connected with what one wants to achieve.

In most cases, there is "No Way Back". So, expense is never far from my mind and being acutely aware of the cost is something that greatly tempers one's enthusiasm. But there comes a point, where if conviction overrides all other consideration - one has to go with what your logic leads you to believe is correct.

Stripping all else away, the issue of AFR has long been a thorn in the side of this project. I still do not fully understand why it was after my first Dynamometer Run, I felt the results were lying to me. The Torque and AFR response curves just did not stack up, although I have my suspicions as to why, I still cannot explain it adequately. None the less, subsequent plots have lent support to my suspicions but still they are not sufficient to draw any positive conclusion. That may change when I take my car for the next Dynamometer Session.

My GTV is currently being prepared for sale, with the front bumper still needing refitting after respray and a new exhaust system in my garage, awaiting installation. So I decided to "Borrow, the MAF sensor from it, to test my theory.

The original 3.2 JTS - Brand New - will not work unless it is housed within an unmodified MAF body.

The Porsche 3.8/4.0 Turbo sensor, although having a brilliant response, will not function as it's output level at tick-over is too low. The consequence of this is, fuel gauge falling to zero, Hill- hold failure, ASR/VDC failure and an unwillingness to idle. I have discounted any further mods on that despite missing its glorious response ar high RPM.

A GTV 2.0 MAF Sensor works well with excellent performance at low and mid range rpm - better than a standard 3.2 JTS. However, it runs out of steam due to it's max 480 Qm/kg/h rating. It may appear to be sufficient for the 3.2 JTS, but it is the Dy/dt response that is important - ie, the high rate of change is needed to get maximum response by the engine to throttle inputs.

However, the figures were not too far away from one and the other, the 2.0 GTV and 3.0 GTV. The only real concern was the lowest output level of the 3.0 version compared to the 2.0. If the output of the 280-217-531 was above 1.2390 volt at tick-over, as is the 280-218-019 at tick-over, then it should be ok. That was my only concern, given the max Q/kg/h of the 3.0 GTV is given as 640 should be more than enough!

And so, after stripping out my GTV 3.0 MAF Sensor and replacing the 2.0 Element in my 159 with it, it was time for a test drive.

Plenty of after market suppliers claim that engines with performance enhancements should not sound too different from their standard counterparts. Bitter experience has taught many an enthusiast that it just is not the case. But it is all too simplistic to think one can change elements involved with performance and not have considerable effect on how the engine sounds or performs. Everything is interrelated.

Colombo Bariani, Supersprint and Autodelta and others said the characteristic changes to sound should not be outlandishly loud. But after quite some years of experimentation, this has been proven to be true, in my case. And throughout the offending element has been essentially Mass Air Flow Metering.

Essentially, I have ended up with and established MAF Element, as installed on the 3.0/3.2 GTV/3.2GTA, the exception being the body is slightly wider in diameter - in all other regards it is the same.

If one now looks at where I have ended up, I have a 3.2 JTS equivalent to the 3.2 Busso, the only difference being the JTS has VVT and the Busso has not.

Which is the greater engine - The Busso clearly for its history.

Which is the better engine - The 3.2 JTS clearly for its strength, direct injection and it's variable valve timing. I personally think, it is a Masterpiece which Alfa failed to recognise. And consigned it to the basement as a failure. It is not the 3.2 JTS that is a failure. It is Alfa Romeo who failed their History, by adopting a "Marks and Spencer" philosophy when acquiring this engine, as if it was an "Off - The - Peg" engine, just as M&S's clothing is!

My job is done now, all attention now turns to my GTV, and buying a MAF Sensor to replace the one now fitted to my 3.2 JTS.

However, I always believed in the 3.2 JTS engine. Just to look at and examine the quality of the block, its inherent strength and the superb attention to detail which has gone into every aspect of it is remarkable. From the way the crankshaft is embedded within, to the cooling of the piston crown and lubrication of the bores by spray jets, whose other function is to eliminate entrained air within the oil such that the main bearings and journals are protected from the risk cavitation is just mind blowing for what is essentially a a mass produced engine. Porsche still hold patents on some elements incorporated in the design.

It is a fabulous engine, more powerful than the 3.2 GTA Busso and strong as an ox in a way the Busso never was - I can say that having owned three V6 Busso's. I suppose, the only way to compare their performance would be for a 3.2 JTS to be installed in a 156 GTA, or a 3.2 GTA engine installed in a 159. Which ever way one chose to go, the GTA would come a poor second to the LLT Holden derived 3.2 JTS engine.

[Ascari32]

Hello Richard,

Weather held for our visit to Yorkshire, which just encouraged an idiot like me to think it was high summer and thus sat out without either a coat or a hat on, drinking some local ales. Result – a stinking cold.

Car went really well with an average of 28 mpg at 60+ mph in both directions – about 220 miles each way. That was rather pleasing as it is early days as far as my latest modifications to the MAF body and replacement sensor. I had previously removed the screen from the Venturi, but further tests led me to conclude both the Venturi and the MAF element were causing Airflow restriction at high rpm. And so, the entire body of the MAF has been gutted and nothing remains of any of the original internals. Indeed it looks like this now: -

   

Left in both pictures being a Porsche 911T 3.8/4.0L, right being that of the 3.2 JTS. The only difference being the diameter of the body, the Alfa being that bit smaller. The other difference is the MAF sensor in now that from a 2.0L GTV, although the 3.0L version from my 916 GTV also works well.

My replacement thermostat housing (82deg.) is the 3rd picture and shows the new position of the NTC sensor. I need to drive for ½ - ¾ of an hour before the core temp of the block is sufficiently warm enough to reduce/stop the NTC from provoking the ECU into dosing extra fuel into the manifolds, which now of course, being Autodeltas do not need.

So by moving the NTC to the Thermostat housing, to where the mean temperature should be higher; given it is where both front and rear bank coolant combine, the error between what the original NTC indicated and that which the ECU expects, should be that much less. What error there may be means I should be able to make small adjustments to the value as seen by the ECU to reduce, if not entirely eliminate any dosing of fuel.

This will be done on the 2nd of May after which I will give it a few more days of testing/fine adjustment before taking her for another Dynamometer run.

Then on the 4th June, there is an All Italian meeting at Sywell Aerodrome in Northants and there may be a chance to have a couple of timed runs on the track/runway – that should be fun. I hope someone has a GTA they want to try against my 159, because I am certain I have the beating of it in my sights. I calculated the Q4 needed 280-ish to match the power to weight ratio of a 156 GTA, so 291 PS should be enough. However, planting the power via the Q4's superb 4wd should give it a distinct advantage. Plus as Carroll Shelby said, "It is Torque that wins races", my 159 has more torque at 7000rpm than the GTA has at peak.

How's the car coming on? I do hope you decide not to sell it – it won't be too long before everyone hankers after one. It still; for me, is the finest Alfa since the old 75 & GTV6's. The Giulia of course is a great car, but not unless it is the QF. Not a fan of 4 pots anymore – not once one has experienced a V6 Alfa.

Keep in touch.

Kind regards,

Brian.