I got it back this week. They did alot of stuff to it. Some of it was to clean up the original install, some to address the operational issues. The total bill was around $3K. Plus shipping each way. I'm in a little over $5K for this "tune up".
On a "worth it" scale of 1-10 we're about 8. There were lots of problems, and it's not perfect, but it is a significant improvement. I'm pretty sure I understand the final issue (which was probably there all along but now is obvious.) Read on...
The engine had never run right since I got it and I futzed around all last summer (with Roush on the other end of the phone) and it never got right. I'd started by flushing every fluid out of it, and it took about 11.5qts of
oil to fill. Way over what they said on their spec sheet. I went on line and was convinced that this was a 11qt Canton racing pan that had been changed. Well, it turns out that it was the original pan and that the dip stick was not calibrated. Poof - one big problem gone.
Oil was blowing by the breathers on the valve covers (even though they were baffled), and it would burn off the headers and run onto the floor. Plugs wires were soaked and the 90* plugs would vibrate off so I sent along a set of FRPP 45* boots that they installed. They also replaced the valve cover gaskets (you have to pull the driver side cover to get to the last plug). Problem 1 solved.
Issue 2 was that the engine was at a slight angle. The engine mounts aren't level so they shimmed. The driver side mount is still high but the pipe doesn't rub as much on the exit port. I still want to look more at that but it's OK for now.
They rewired the DFI controller harness and dressed up the install. There were huge wire looms where they took the cables and connected them to the other end at the front of the engine instead of running shortest route. Wires were merely twisted and black taped in the loom. I'd seen that and didn't want to tear into it myself lest I mis-wire something. I have a box of wires they trimmed off and sent back. I will post some pictures of before and after. At least now in future dealings I will not have to argue with them about the install job.
The third problem - rough running - is still there to some degree - and represents the remaining 2 of 10 points.
The original problem/symptom was that it would cut out (miss) at about 4000RPM under quick acceleration (like getting into an accel lane or off the block at the strip) When I got it home on Tuesday I took it for a spin and it ran a little rough on a quick accel, but the engine was cold. I let it warm up and ran a few runs into the accel lane I use for the test and it ran pretty well up to close to the red line. Then it started to spit rain.
Wednesday night I took it out again and it failed. I had just filled it up with gas. The only two changes from their tests in Detroit to here were gas and altitude. They had filled it with 92 (which is probably 95 here). I filled with 91 (which would be their 93+). But in case the pump isn't right I drove to town to get some octane booster. It rained on me on the way home so I didn't get a chance to run it hard. (There is a third change - the driver
)
I took it out yesterday for the same test and it still failed - rough running at 4000.
This is a DFI system that has altitude compensation (it reads the base from the MAP when the engine is not running). So last night I installed the Accel CalMap software and looked - the compensation table is null - there is no adjustment happening... So I'm waiting for Roush to tell me what they do next.
This afternoon I decided to take it for a test run and have my wife shoot a video on my crappy-cell-phone-video-camera to send to Roush. Guess what. It didn't fail. But it did rain on me. So there is a humidity/atmospheric pressure change (more humid, lower pressure) from the Thursday run (sunny, warmer, high pressure).
So now it's down to intermittent, but it still runs better than it did. I'd appreciate a pointer to anyone that has set up an Accel DFI Gen7 altitude compensation table.
Cheers!