Alfa Romeo Forum banner
1 - 20 of 20 Posts

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #1 ·
In this thread I'll be documenting the changes that I'm making to the ECU, fuelling and engine wiring of my 1996 Ph1 GTV.

Why are you doing this? My car has a bit of a running issue, it stumbles badly under anything more than light acceleration. I probably could have fixed this with replacing distributor bits, wires and cleaned injectors but then it would only be as good as the original 1980's designed ECU could make it.

Work started yesterday on ripping out the wiring harness. Alfa makes this relatively easy, the complete ECU / engine harness is a standalone item. There are three connectors joining it to the car (Alfa code/tach/fuel pump/AC/Ignition) signal plus three wires that run to the trunk for power and vapour solenoid. I'll need the vapour solenoid so those got cut and re-used rather than being removed from under the seats etc.

Next up, fuel pump and injectors.
 

Attachments

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #2 ·
GTV TB uses two ECU: One controls most of the spark and some of the fuel. The other, most of the fuel and some of the spark. They talk to each other by serial connection. The injectors are all batch fired - yes, all six at the same time. The new set up will have sequential injectors and sequential coil on plug ignition. For that to happen the stock injectors have to be replaced with a higher capacity.

The fuel pump has just enough capacity but is 23 years old and noisy. It is replaced with a new, higher capacity Deutsche Werks pump. During the replacement of which I think we found part of the car's poor running problem. The insulating mount for the pump has mostly disintegrated and is returning to the earth from which it came. There are parts of it everywhere. And, of course, no more mount for the new pump - a little ingenuity with some zip ties and filter location solved that issue. I replaced the main fuel filter and will replace it again after a few hundred miles - you should see the crap that came out the intake side of the old filter - yuck! I'm hoping that none of this will get into the new injectors....

Next up, Manifold pressure and Intake Air Temperature sensors.
 

Attachments

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #3 ·
Three of the most hated words in the English language: Drill and Tap

The existing two ECU have no knowledge of boost pressure - they use a bunch of inputs, airflow, Throttle position, rpm, amongst others to calculate boost pressure. The new system will read this directly from a Manifold Air Pressure sensor. There is no suitable place to screw this MAP sensor but the intake casting has an unused cast-in boss above the cylinder one runner. I also need an Intake Air Temperature sensor and there's another suitable boss cast into the intake tube that runs from the turbo to the throttle body, just upstream of the throttle butterfly.

With this setup there will not be any plug wires hence, first cylinder detection using an inductive coil won't be possible. This cam position looked like it would be the hardest to overcome until we discovered that the mounting and drive arrangement for the TB distributor is exactly the same as that for an '89 VW GTI that uses a Hall sensor. Once three of the four 'windows' were cut from the rotor we now had a single cam sync for the Hall, perfect!

The spark will come from motorbike coils. These do not need an ignition module (ignitor) because the new ECU has high output coil drivers internally.

Next up, a whole lot of wires.
 

Attachments

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #4 ·
For something that's supposed to take hundreds or thousands of years to decay, plastic sure does become brittle awfully fast.

Initial investigations of the feasibility of this project led to several connectors cracking and clips breaking and that's what prompted replacing the harness rather than augmenting with needed additions. It makes for a whole lot of work though.

Another of the items I was dreading - removing the Lambda sensor turned out to be no problem - these usually "weld" themselves into place with heat and time. I will be using an AEM wideband to send exhaust O2 signals to the new ECU.

Speaking of ECU, it got mounted into the bottom of the passenger side A pillar (note: left hand drive car) where one of the previous pair of ECU used to live. The wideband controller is mounted nearby on the back of the glovebox frame.

Next up, relays, fuses, wiring the ECU and initial power up.
 

Attachments

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #5 ·
It really is a lot of wiring.

Removing the two ECU freed up space under the centre dash behind the radio. Just exactly the right amount of space to mount a box containing Main and Fuel relays and associated fuses.
The two speed radiator fans are normally controlled by a temperature sensor build into the radiator. The new ECU allows them to be under ECU and hence programmable control. At a later time I'll be adding a small fan on the intercooler which will also be under ECU control and respond to both coolant and intake air sensors.
The ECU allows more than one tuning map which is changed with a simple switch - duly mounted.

Today, the initial power up of the electrical systems - no Lucas smoke, phew and then start 'er up.

It needs a whole bunch of tuning but it starts and runs!!!!

Things to be fixed - I ran out of time today:
- The throttle position only knows 0% and 100% so either the wiring diagram is wrong or I pinned the plug wrong.
- There's fly back power from somewhere keeping the main relay on after key off. Probably the idle valve.
- The MAF removal requires the intake to be reconnected to the air filter - parts required.
- Lots of testing, tuning and then button up the harness properly.

No more updates for a week, I'll be away.
 

Attachments

  • Like
Reactions: bathas

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #6 ·
Did you miss me? My one week turned to two - destination wedding, summer cold and overseas guests will do that to you.

Got back to sorting some of those outstanding issues:
- The throttle position sensor pins are labelled 1 thru' 6. The connector for it is also labelled 1 thru' 6. But, the order is reversed. Inside the TPS is a on/off switch for off-idle indication and a potentiometer for throttle position. The reversed connector pins were picking up the of-idle switch. The new setup does not need this switch, only the potentiometer.
- Fixed the flyback power to main relay, by using the same negative switch as the ECU relay uses, to control the main relay. The instructions say to use the ECU relay output to control the main relay, however this created a loop inside the ECU of some kind. The loop voltage (10.6V) was not enough to pull in the main relay but once latched was enough to keep it pulled.
- Decided to add a fuel pressure sensor so waiting on parts and added the necessary wires.
- Five spare wires (supplied with the ECU) got laid into the harness for future uses.
- Started to loom all the wires.

Up next: More, lots more, looming and fitting the awaited parts.
 

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #7 ·
Looming all but finished. Still awaiting the fuel pressure sensor and MAF delete hose parts.
 

Attachments

  • Like
Reactions: bathas

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #8 ·
It lives!
The fuel pressure sensor isn't in yet but everything else is. A couple of hours of tuning over three days and it starts, idles and runs very nicely. Some occasional hot return-to-idle stalling to fix - the idle control valve is a pain, it is an electromagnet and spring balance thing. The factory ecu likely had a battery voltage compensation which the AEM Infinity does not and the Bosch alternator is very lazy at low engine rpm allowing the voltage to drop to battery. I may end up switching to a stepper motor idle control valve.
It is currently running on wastegate boost, next week it is off to the local dyno to tune the boost control and see what we can safely achieve.
 

·
Registered
Gtv 2.0 v6 tb `95
Joined
·
175 Posts
Respect for the job! I have the same ***** model. I thought about transplanting installations with Alfa 166 2.0 v6 tb. But somehow it passed me. Greetings.
 

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #10 ·
Respect for the job! I have the same ***** model. I thought about transplanting installations with Alfa 166 2.0 v6 tb. But somehow it passed me. Greetings.
Thanks for your kind comment.
While doing this job I discovered the biggest reason that it had not been running well was that the fuel pump support piece in the tank was in an advanced state of decay. The decay takes the form of something like sand and it was present all over the tank bottom, in the pump, clogging the filter and injectors.
If you are having any issues with the engine running badly then I encourage you to check your fuel tank. If it is like mine then I imagine cleaning the mess, replacing the fuel pump, fuel filter and having the injectors cleaned will make a huge difference.
 

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #11 ·
Dyno results.

Discovered a fault in the ECU whereby the boost controller would not respond to incremental control. It became an on-off switch, the 'on' part being discovered when suddenly the full boost of the turbo became apparent. Logs showed it hit 21psi (1.5 bar) and over 240 whp (243 wps) and the power inrush was almost instant - yikes.

There was no time to find and fix the problem at the dyno but once the ECU issue was found later and fixed (boost control moved to a spare channel), we street tuned the boost control settling on a maximum boost of 15 psi (1 bar). We found that anything over 1 bar makes the turbo efficiency drop off markedly. We'll head back to the dyno to dial in the settings later, for now the engine feels very strong likely maxing at a little over 200hp at the crank - similar to original factory numbers.

The idle issue has been significantly improved - the ECU was trying to control idle at anytime engine rpm was below 2000. While slowing in gear, the idle valve is driven to fully closed trying to get the engine to 900rpm. Applying the clutch at this point allowed the revs to drop, occasionally to stall, because the idle valve could not recover fast enough. The issue is made worse when the radiator fan is running at second speed. Modifying the idle control rpm from 2000 down to 1400rpm reduces frequency of this issue since the clutch is usually applied before getting down to 1500. This is still a work in progress.
 
  • Like
Reactions: bathas

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #12 ·
Here's an unexpected side benefit of this new brain. All the main relays and fuses from the firewall were relocated. The V6 TB washer bottle (bag) is located in the boot because the air filter is where the washer bottle sits for every other GTV. It's a long way from the boot to the wipers, sooooo. I got a bottle from a Cappuccino, made a bracket from strip of aluminium and a heat shield from some aluminium sheet et voila!
 

Attachments

·
Registered
Joined
·
434 Posts
I like the project and the work you do :)

Please tell me, will you be using the original turbo or will there be an upgrade somewhere in time?

Is there a targeted power you are aiming or just running for optimization?

PS Your last few photos are small and can't see the work you do.
 

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #14 ·
I like the project and the work you do :)

Please tell me, will you be using the original turbo or will there be an upgrade somewhere in time?

Is there a targeted power you are aiming or just running for optimization?

PS Your last few photos are small and can't see the work you do.
There is a power target but there is a path to get there. Next up will be a timing belt/water pump change since I don't know when it was done before (there are no records from the import from Japan).

The brakes are the base sliding caliper - those will be replaced by 4 piston Wilwoods.

The suspension appears to be OEM and may be the original from 1996 - Not sure what is a available other than the Koni so I'll likely end up with something custom made.

The differential is original, there will be a Quaiffe going in along with a stronger clutch.

Once these are done the turbo will be replaced with an EFR, size to be determined, to get somewhere around a reliable 240hp (243ps) wheel horsepower. The existing turbo is capable of that now but it would not survive very long.

Sorry about the picture size - I put them in as an idea of what I'm doing - I didn't think anyone would want to see detail. LOL
 

·
Registered
Joined
·
434 Posts
There is a power target but there is a path to get there. Next up will be a timing belt/water pump change since I don't know when it was done before (there are no records from the import from Japan).

The brakes are the base sliding caliper - those will be replaced by 4 piston Wilwoods.

The suspension appears to be OEM and may be the original from 1996 - Not sure what is a available other than the Koni so I'll likely end up with something custom made.

The differential is original, there will be a Quaiffe going in along with a stronger clutch.

Once these are done the turbo will be replaced with an EFR, size to be determined, to get somewhere around a reliable 240hp (243ps) wheel horsepower. The existing turbo is capable of that now but it would not survive very long.

Sorry about the picture size - I put them in as an idea of what I'm doing - I didn't think anyone would want to see detail. LOL
The car needs stronger brakes for sure, 4 pot calipers are the way to go, at least 310mm up to 330mm so you can fit 17" wheels.

Koni is a great shock no doubt. But for the money of new shocks + springs, I would invest a bit more and buy KW coilovers, V1 or even better V2.

Quaife is also a good way to go and I would say mandatory.

With the smallest EFR you will easily make 250 HP, and best to pick the smallest one for response and spool time.
 

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #16 ·
The idle issue has been significantly improved - the ECU was trying to control idle at anytime engine rpm was below 2000. While slowing in gear, the idle valve is driven to fully closed trying to get the engine to 900rpm. Applying the clutch at this point allowed the revs to drop, occasionally to stall, because the idle valve could not recover fast enough. The issue is made worse when the radiator fan is running at second speed. Modifying the idle control rpm from 2000 down to 1400rpm reduces frequency of this issue since the clutch is usually applied before getting down to 1500. This is still a work in progress.
We may have found the cause of the idle problem. I removed the idle control valve and gave the rotor and passages a complete clean to remove all traces of oil and dirt. After reinstalling I connected the AEM software to check its function. I noticed that the throttle position closed percentage kept changing, ranging from 0.2% to 1.4%. The idle active threshold is set to 1% throttle. Altering the filtering of the signal did not reduce the percentage changes. Messing around with the throttle position sensor I found that the closed position is electrically 'noisy'. Rotating the TPS so that the closed position is moved reduces the percentage change when closed to 0.2% maximum.

I haven't had change to drive around much but can report that the short test drive yielded no rpm drops or stalls.

Why this may be the problem:
If the closed TPS has a percentage above the set point of 1%, the idle control will not function.
 

·
Registered
Joined
·
5 Posts
Wow. Just...wow.

I wish I had the know how to do this on my car...very well done. I wish you had continued updating this, but it is great as it is. Thanks.
 

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #18 ·
Not a whole lot of updates to add at this point. I ended up replacing the throttle position sensor with one from a Kia. The sensor required a custom bracket which I 3D printed in ABS. The AEM ECU does not need the idle position switch from the standard TPS which made sourcing a replacement much easier.
I played around with some of the more advanced settings of the AEM and ended up with a rock solid idle at 950rpm and 1100rpm with air conditioning running.

I have acquired but not yet fitted the brakes. I went for 6 piston Wilwoods because, well, why not? I'm waiting for the custom water jet cut brackets which have been delayed by this pandemic.

The interior got a bit of attention, in Canada it is almost obligatory to have cup holders...... I grabbed a centre console from a Scion TC (a Toyota off brand in north America). It is the right width to fit between the rear seat cushions and extends forward over the tray behind the parking brake. Some fettling gave me two cupholders, a phone storage, a lidded storage bin and three power outlets (USB and cigarette lighter style). My wife often works from the car while I drive so there's also a power outlet and inverter for her laptop. All the power things are ignition controlled.

Still a whole bunch to do but it is a long term project for me.
 
  • Like
Reactions: IgoRR

·
Registered
Joined
·
195 Posts
Brakes already fitted? :cool:

I like those ignition coils: looks like they are made for it!

Nice project!!
 

·
Registered
GTV, '96, 2.0V6 TB
Joined
·
306 Posts
Discussion Starter · #20 ·
Tire Wheel Automotive parking light Car Vehicle


Brakes finally fitted on Friday!
Those are 6 piston radial mount Wilwood calipers on 310mm, two-piece Wilwood rotors.
Notice the unusual position of the wheel balance weights - my aftermarket no-name brand wheels are 16" and there is not enough room to fit balance weights between the wheel and caliper.
 
1 - 20 of 20 Posts
Top