Hi guys. I let the other thread die because I figured you all got kind of tired of hearing me whine, and not be able to track down the problem. But, I had not quit working on it. I just quit whining about it. If you read the previous thread recently you know it got to where it would not start, and I had to increase the tach interrupt filter value to 50 for it to start. However, it would runwith a lower setting if I turned the tach interrupt value back down after it started up. It became obvious that the starter motor was the culprit. Noise from the starter was being interpreted as tach input signals causing the RPM at start up to shoot up to 3K thus flooding it, and no start.

In talking to Peter earlier he said it was important to get the ECU inputs isolated from chassis ground noise. So, I insulated the HEI module from the chassis and ran the ground wire for it to a sensor return tap on the MS relay box.

VOILA! No more start up noise. I turned the tach interrupt filter down to 5% and it still starts right up with no false tach reads. I have not taken it on the road for weather reasons. But I strongly suspect that the tach interrupt filter was what was causing the misfire under acceleration. All the noise was coming back through a ground connection at the block. I know many of you were telling me there was a ground problem, but I could just not track it down.

So this is huge. This is important information for anyone using Mega Squirt. Be sure to take all ECU input signals through the megasquirt ground, and not to the engine block or chassis.

In the mean time, I have installed a trigger wheel, and hall effect pickup, and will eventually use that to trigger the ignition. But now there is no hurry to get to that.

I will let you know how the test drive goes. It will be Thursday before I get to that.

Sam

_________________

_________________
I have no idea what I am talking about.

Cool beans! Keep it up!

During a break in the weather last evening I took the car out and drove it. The miss that has plagued it from the start is gone. It now breaks down at 4K, which is a huge improvement. It just tops out and stops pulling, and feels like it is getting too lean maybe, but the AF ratio reads 11.2:1, so suspect that is not the case. I lost some coolant while R&R ing the radiator, and because I am using waterless coolant, I cannot just run down to the auto store and get more, so had to stop testing when it got hot. Got home and checked under the hood, and several of the intercooler hoses were not tight, so maybe that was the culprit. Anyway, things are considerably better. I still have the tach interrupt filter on, and set at 5%. I will turn it all the way off today and test it again.

Also, the other sensors are still grounded at the block. There are not enough taps on the relay contact strip for all of them now. Does anybody know if there is a problem with taking all the sensor grounds to a common terminal strip and then connecting that block to the ECU ground through the relay box?

I suspect the 4K limit will not be as elusive as the 3K one was. I could drive it forever in normal driving conditions and never need to go over 4K. The turbo is a fabulous addition for the road. When I took it to Mason Dixon last month it went over the mountains just like a V-8 as if they were not even there. And this was with the ignition not working right because of the grounding situation. I can hardly wait to take it on the road again. I think tuning will clear up things from here.

Bottom line is it runs much better under all conditions. It idles better,and it is much smoother cruising. I suspect the HEI was getting all kinds of small misreads through the block ground wire. There must be a lot of EFI crap piggy backed onto that ground plane. Being a piano tuner I understand all too well frequencies that piggy back onto a carrier of some kind. I just did not understand the grounding concerns in an auto electrical system. I suppose the other sensors still could be picking up stray noise and feeding that to the ECU, which would compromise the integrity of the fueling and timing strategy. I must deal with those soon.

All you guys who offered your advice 9 months ago when I was trying to track this thing down: Thanks! You were right. We did find the problem. And, as many of you suspected, it was a grounding issue.

So, there is another problem which is of a mechanical nature. I will start a thread on that in the engine section.

Sam

_________________