V8Soarer.com - #1 site for V8 Soarer mods

As you saw in Part 1 there are a lot of other components to a working ITB system than just the throttle bodies. Fortunately I had a spare lower intake manifold to test how everything would fit together on the bench before final fitting on my car. However many of the smaller components have changed between initial conception and what's now running, such as the vacuum collector for the MAP sensor.

I also faced many and varied problems which challenged my own understanding of intake systems (which is fairly general) and which also tested my troubleshooting skills (which are pretty good). Like any new modification it helps if you know what you should get as an end result and most of my time was wasted try to solve "problems" which were just quirks of ITB's... but more on those later.

Fitment was actually very straight forward, and as most things are with Soarers you just follow your nose and you can't go too wrong. From the time I parked my car until I first turned the key was 2 days at a leisurely pace BUT it was another 3 months of one or both days on the weekend working out the details until I had the car on the road then another month of fine tuning.

If you're using 40mm throttlebodies you can stop after you remove the plenum but for 45's you'll have to continue and remove the intake runners to enlarge the opening where the throttle bodies bolt on so there is no sudden step.

With all the intake removed and the fuel rails out of the way there doesn't seem to be much engine left. Something I did notice was how much crud had accumulated in the vee of the block. This shot shows why changing the starter motor is such a PITA... that's it tucked right in the middle of the valley.

Refitting the manifold is easy enough but be careful with the injectors to be sure that the rubber seals are seated properly and aren't getting squashed.

Then you can bolt the wiring harness back in place in the centre and for the waterline that was going through the throttlebody and ISCV I used some fuel injection hose through the middle here:

This is the tube for the vacuum lines to the MAP sensor which is 10mm OD aluminium with pneumatic connectors from Festo which are also used in the throttlebodies. The line to the MAP sensor & fuel pressure regulator is in the middle underneath.

When bolting on the throttlebodies I had to take extra care to ensure the adjoining shafts were totally parallel - if they met at the slightest angle they would bind and not fully close. I found it easiest to tighten the joiner before tightening the bolts to the manifold to ensure everything's square (indicated below).

I made gaskets from 1mm gasket paper which sit on the throttlebodies and the top of the throttle cable bracket lines up with the top of the gaskets. When the plate is bolted down, a countersunk hex bolt screws into the cable bracket to brace it.

The plate bolts to the throttlebodies with countersunk hex bolts then the trumpets fit over the top.

I ordered a custom air filter from UniFilter which fits over the lot until I get around to having a cold air intake system made with remote air filters behind the headlights.

In the above shot you can just see the oil catch can in the right side of the shot which just has a mini air filter on it for now, but this will be plumbed into a source of vacuum soon. This really does need vacuum to suck the blow-by and other contaminants out of the crankcase to stop it fouling the oil, not to mention the emissions concerns.

I'm using a vacuum pump to provide vacuum for the brake booster and I fitted this along with a vacuum tank from a Saab in front of the driver side front wheel. I used the existing wiring for the old air suspension compressor to power it.

The white object is the vac tank and the pump is next to the chassis rail - the photo was taken from the wheel arch looking forward.


1. Throttle pulley 2. Cruise control/return spring bracket 3. Return spring 4. Throttle stops 5. Joiners	6. Vacuum lines 7. Throttle Position Sensor 8. Diagnostics port 9. Fuel Pressure Regulator

Sealing
The main problem I faced from the beginning was getting a good seal when the butterflies were closed. The stock cams only have a small overlap so the vacuum at idle is very strong and 8 small air leaks from each throttle body quickly turn into 1 large air leak resulting in a high idle speed around 1500rpm.

I sent the throttlebodies to EFI Hardware to have new butterflies machined up and they did a good job but the air was also leaking around the throttle shafts as the holes in the casting were too large. I tried various ways to get them to seal better but what finally worked was 4 coats of Loctite Moly Dry Film Lubricant spray inside the bore and around the butterflies. An hour after each coat I cracked open the butterflies, closed them and did another coat.

This brought the idle speed from 1500rpm+ down to 600rpm cold and 800rpm warm which is perfect. The seal around the butterflies is airtight so it's actually idling on the air leaks around the throttle shafts. I may re-employ the Idle Speed Control Valve so the ECU can raise the idle speed for cold starts to avoid stalling.

Sticking Butterflies
Getting the butterflies to fully close when gradually leaning on and off the throttle was also a problem. The main cause of this was that the shafts weren't exactly in line and met at a very slight angle. With the joiner tightened up it limited the free rotation of the shafts so I had to very carefully check and double check the alignment as I tightened the bolts connecting the throttle bodies to the manifold.

With the throttles moving freely I still needed a relatively strong return spring so when the butterflies are closed slowly they wouldn't have any friction problems. I needed quite a long spring so it would provide a fairly constant force over about 10cm of movement.

Vacuum
The first thing I noticed when I first started the car (apart from the high idle) was how low the vacuum appeared to be. With the stock intake it was around 25-27 inHg at idle but had dropped to about 8 with the ITB's which meant I'd lost about 2/3 of my resolution in the fuel map. What I didn't know was that the vacuum signal seen by the MAP sensor is inherently lower with an ITB setup but even 8 inHg was too low. This was because my first design used a collection tank for all the vacuum lines then everything that needed vacuum fed into that. The problem there was the PCV and charcoal canister both let air/vapour into the tank which lowered the vac signal.

After separating the MAP sensor and Fuel Pressure Regulator to their own vacuum lines the vacuum at idle went up to about 15 inHg which was much better so I asked for a custom firmware for the MAP-ECU which went from 15inches to 0 so I could have the full resolution of the table. This worked much better until I was cruising on the open road and it would hesitate when opening the throttle from closed. After checking on the laptop it was evident that the vacuum when cruising went as high as 21 inHg so the flat spot was a result of the MAP-ECU waiting for the vacuum to go from 21 to 15 before it would register.

In the end I settled on a firmware which goes from 23 inches to 0 which covers the full range but still gives small increments for fine tuning, but had I known what to expect it would have saved a lot of time.

I did most of the fuel tuning myself as I have a wideband AFR meter which I covered in the MAP-ECU section here but because the shape of the fuel table had changed it took some time. Something else which hindered my efforts was the stock ECU when running in closed loop so I had to disconnect the oxygen sensors in the extractors to stop it interfering. Before I did that I couldn't work out why I was getting an AFR of about 20:1 even though I was adding heaps of fuel. It seems that when the ECU saw it was far too rich, it compensated too much and pulled out way too much fuel. Once I started removing fuel, it started to register as rich once it got back into a range the ECU deemed acceptable.... not exactly intuitive.

*UPDATE* - I've recently found out why the wideband was seeing a lean mixture even though it was too rich. When it gets overly rich it will misfire so when this occurs in a cylinder it is mainly clean air flowing out of the exhaust port - this sends the reading way off the lean end of the scale.

When the idle was high it was running OK, but not as well as it should and sometimes it felt like the ECU was cutting fuel when the car was stationary in an attempt to bring the idle down as the exhaust would pop a bit as though it were too lean, the revs would drop, then they'd come back up and on it would go. I took it to the dyno but they couldn't tune it as the ECU was pulling timing and adding fuel all over the place.

After I sorted out the idling problem and the idle came down to the normal range of 600-800rpm it instantly felt much smoother and more willing to rev. I took it back to Unique Autosports for another run on the dyno and it ran smoothly and without any issues which allowed Ben to finish tuning the fuel map. He was initially skeptical that it'd be any different to the week before when it was being a pig, but even though it was his day off he gave it a go and we were both pleasantly surprised to see it reel off 178kw at the wheels by the time he was done. John at UAS was who put me onto the moly spray to help seal the butterflies so they've been extremely helpful.

While I'm happy with the peak power, what I've really noticed is the mid-range punch that it didn't have before. When changing from 2nd to 3rd at WOT it felt like it had to climb up the torque curve again as it lacked that real shove in the back. Now, however it feels like it has gobs of torque from about 2000rpm onwards.

There is a bit of a flat spot at around 3700 on the graph but looking back at an old dyno sheet when it had a totally stock intake that was always there so it's probably a characteristic of the engine but it's not noticeable on the road.

If you compare the output of stock V8 Soarer which is about 125-130rwkw with the graph above, you can see that I get the same power at about 4000rpm... not bad! Now we just need to extend that power curve up to about 6500 or 7000rpm which a nice set of custom cams will do. Well, that'll be in a few months as I just want to enjoy what I have for now :-)