SCCA Camaro

Since I had two weeks before my next SCCA Solo II Autocross event in the 68 Camaro after United Airlines left me in Austin, TX and I missed my July 13th event, I decided to put a digital fuel pressure gauge in the Camaro so I could keep an eye on the fuel pressure generated by the Walbro 255 inline pump and the Holley fuel pressure regulator.  I wanted a digital gauge in red to match the Dakota Digital dash and the Holley Wide Band Air/Fuel Gauge I had already installed.  The best bang for the buck seemed to be Cyberdyne; their digital fuel pressure gauge for EFI reads from 20 to 100 PSI, comes with the sending unit and wiring, and can be found for under $100.  I needed an adaptor in the fuel line somewhere to tap the sending unit into. 

The sending unit has 1/8th NPT threads and you can buy a pressure gauge adapter with either barbs or AN -6 fittings on either end but I decided to machine my own from brass with barbs to go inline in the 3/8ths rubber fuel-injection rated hose (SAE J30).  I put a ½ diameter brass rod in the lathe and bored the center out to 0.330” before turning down the ends to 0.375” on either end.  I left the center at 0.500” and also left the barbs at 0.440” with sixty degrees backcut and forty five degrees on the end going into the hose.  I radiused all of the cuts and put the tube in the milling machine to drill and tap the hole for the 1/8th NPT threads.  I also picked up a couple of stainless steel worm clamps to put the adapter in-line in the fuel hose.  I installed the gauge on the passenger side firewall with a plumbing clamp I found at the local hardware store and re-bent to fit securely over the unit with a single screw.  This is essential for proper gauge performance because the sending unit needs to be grounded, which the rubber fuel hose obviously doesn’t accomplish.  For the gauge mounting I simply drilled an additional hole in the 20 gauge sheet metal panel I had made to mount my Holley Digital Wideband Air Fuel gauge in the dash.  I also made an equally-space dimple in the center of the panel so the addition of a third gauge in the future will be easy.

Wiring the gauge was straightforward: I tapped into the twelve volt supply for the Air/Fuel Gauge as well as the ground while routing the sending unit wire through the firewall grommet with the wiring harness from the Holley Commander 950.  The gauge worked well and confirmed fuel pressure at about 41-42 PSI upon putting the key in the ignition position.  The fuel pressure regulator is vacuum referenced and with the car running at idle fuel pressure was around 35 PSI.  The gauge fit well into my red LED scheme and allowed me to monitor the fuel delivery system for problems which is critical when retrofitting the 68 Camaro with EFI so I was happy.

Upon pulling out of the carport I realized the Camaro was not going to be as docile as I hoped when I spec’ed out that mild 112/118 duration cam!  It was very torquey and surged when attempting to accelerate and torque braked hard as soon as you let off the gas.  This was noticeable even in third and fourth gear.  This made it almost impossible to log useful data in traffic as the only way to drive it without the car violently bucking around was “flat out”.  My first test drive also involved topping off the gas tank and I found out that I had a leak somewhere high up on my tank (it was not leaking when not full but leaked when full).  This wouldn’t be that big of a deal except I still didn’t have a reliable reading from my 68 Camaro’s 40-year-old gas gauge sending unit so I was guessing about how much fuel it had at any moment—making it difficult to fill it up three quarters of the way.

I put the following on the www.pro-touring.com forum board hoping for some help:

When taking off in first, i find it very difficult to not spin the tires. If you are going balls out on a track it might be OK until you have to come off the gas–but in street traffic when taking off at a stop light and then trying to feather back off the gas so as not to rear end the vehicle in front of you, the car surges forward, then if you let off the gas it jerks or bucks very hard due to the torque braking. I have the Commander 950 software running and am looking at my throttle position sensor when i am doing this, and i have a relatively light foot (TPS is 1 closed and 172 for WOT, i am applying values of ~10 when this is happenning). I have found for the street the only way to drive it is to let traffic leave me to create a gap, then take off aggressively in first, then go to second almost immediately, then go to third and essentially allow the car to idle along in traffic (idle is 850 RPM). In third, the ammount of torque on tap is diminished enough relative to the load that i can come on and off the gas (again, i’m talking 3-5% of WOT) with controllable surges rather than violent bucking. Traffic around me still think i’m an idiot because the car bucks with even the slightest progressive addition of gas (i looked at my datalogs and i’m talking TPS values of around 5–roughly 2-3% of WOT).

There were a lot of thoughtful responses, and three things I could implement pretty quickly: double checking throttle linkage and spring tension, moving the idle up higher in the RPM band, and setting the shocks softer.  I disconnected the spring from the throttle linage but didn’t see any other way to make it more progressive in the short term.  I also set the DSE-sourced Koni adjustable shocks at one click out from full soft in the front (they had been set on one click from full hard) and to ½ turn from full soft in the rear.  Then I raised the idle RPM to 950 in the Holley Commander 950 software and adjusted the throttle return set screw to get the idle correct so the idle air control solenoid was in control of the idle.  I figured I would deal with the gas tank leak later since it only leaked with a full tank of gas.  Upon taking it for a test ride, it was still very rough but seemed somewhat improved mainly by the throttle spring and idle speed changes.  Unfortunately upon adjusting the front suspension I was now bottoming out on bumps at speed.  Although I had the 2001 Trans Am as a back-up, I decided to race the Camaro on Sunday and see from there what to do next.  Unfortunately I was travelling on the mainland for work and United Airlines stranded me there on July 12^th, the night before the race I had been waiting 3 months for!  The only good news was that there was another SCCA Solo II race two weeks later, so I wouldn’t have to wait too long to test out my modifications.

 Upon starting my 68 Camaro successfully after the Cam swap and installation of the Holley Stealth Ram and all of the attendant plumbing, wiring, and sensors, as well as the installation of the TPI serpentine belt set-up, I was ready for a little tuning.  With the engine running I kept a very close eye on the Holley wide band air fuel ratio gauge on my dash to make sure I wasn’t running too lean.  The Holley Commander 950 base map I had chosen (WBmp3008) seemed to be close enough—a little lean but still safe at around 15:1.  One thing I had to figure out was the idle speed.  I had set it in the software at 800 once the car warmed up (it starts cold around 1100 RPM).  This was down about 50 RPM from the WBmp3008 base map since I had considerably less overlap than the cam used for the basemap and figured it would be easier to get a nice idle.  But the Camaro was actually idling about 1100.  It turned out I needed to adjust the idle air screw on the side of the throttle body as well as set the actual timing on the car to match that which the Holley.  A few turns of the set screw closed the throttle blades enough to get the idle down where I wanted it.  I had to purchase a timing light with a dial back feature in order to match the timing on the distributor with the 23 degrees the Commander 950 was giving the Camaro at idle, but it was straightforward with the correct tool.

I also noticed a water leak from around the water pump, which meant the block had to be mostly drained of coolant and the TPI serpentine accessory drive had to come mostly off the Camaro so I could pull the long style reverse rotation water pump and try again with a different set of gaskets.  The ones that came with the water pump were sort of a cork-like material and I used the Chevy specification for the bolt torque as well as a thin layer of RTV sealant so I was a little surprised by the leak but I used a FelPro composite-type gasket set re-prep’ed the block and didn’t have any more problems.

 
The Camaro was running lean in the lower RPMs due to the higher volumetric efficiency of the shorter-duration cam I was using relative to the one used for the base map so I added ten percent to the pulse widths in the first three columns of the fuel based map and tapered them off to the rest of the map where the map stopped being lean.  This meant at the higher RPMs the Air/Fuel ratio was a bit rich which I wasn’t too worried about so I left that part of the program alone.  When the dual Spal fans kicked on the idle would start to surge back and forth 50 to 100 RPMs around the set point so I played with that quite a bit.  The Commander 950 Software has two features to keep the idle steady, one involving fuel trimming and one for adjusting timing to keep the idle steady.  I enabled both and still found the idle to hunt a little.  Good enough for a test drive though.