I'm stumped.

No load (sitting still out of gear. Acceleration through 2500 misses then good again at 3200 then again at 5100.

Wideband sensor (LM1) shows lean during miss (I think a symptom not cause?). If I pull the vacuum to the fuel control pressure regulator the miss at all 3 places is largely masked. In other words it seems to disappear; but completely - not sure. Fuel pressure actually tested OK at 29 idle and whatever spec is for engine off. Swapped pressure regulators anyway no change.

3 different AFM. 2 different DME. Several sensors. Plugs OK. Different cap (but not rotor - don't have one around). Swapped coils. No vacuum leaks.

Wish I had an ignition scope!?

Jon,

One of our guys just went through a very similar problem. If you've ruled out all the externals in fuel and electrical, it might be internal. His turned out to be a broken inner valve spring. My other thought was a lifter issue but they found the broken spring when they disassembled.

Here's his synopsis:

We changed fuel injectors, wires, plugs, cap, rotor, ECU, AFM. We swapped the injector wires and metered them. We metered the crank sensors, temperature sensors and spark plug wires. We ran a leak-down test and a compression test. Then at the track on Saturday Charles lent us a ECU and we tried swapping it again. We had Jerry and Tim Whitteridge over there trying to figure it out. We then did a running compression test. I had never done one of these before. My compression tester is old and we blew out the Schrader valves. Ron at AIM got us back up and running and we found that the Number 3 was different than the rest. We suspected mechanical failure in this cylinder. No one would believe that the symptoms we were experiencing were mechanical (except Jim Foxx). Everyone thought it must be either electrical or fuel related. (Our team certainly agreed that this was the most likely scenario. That is, of course, why we went through all the above diagnosis.) Our team concluded that it must be that either a lifter had gone bad or we agreed with Jim that it might be a broken inner valve spring.

Previous to Laguna Seca, the car would start fine and idle fine. Goose the engine and it would only run on 3 cylinders. The Number 3 cylinder would not run. Get the car out on track and the car would again run on 4 cylinders and seem to race fine. On Saturday at Laguna Seca, the car would idle fine, but then only run on 4 cylinders between 4500 RPM and 5600 RPM. Very hard to get a good lap in! Still, I was able to qualify at a 1:50.3. On Sunday, the car was worse. It no longer idled and it had a tapping sound. I ran the same lap times. All this made for a very frustrating weekend.

The car was hauled to the team mechanic, Art at Autopoint in Point Richmond. Art went through all the above diagnosis and finally took off the cam box. He replaced a lifter and a broken inner valve spring on the intake of the #3. The lifter was suspect, but the valve spring was conclusive

The fact it is happening at those RPMS tells me your flywheel triggers are the issue, why...because taking some RPM error on the tach and being hard to read exactly where it happens you are seeing steps in RPM. Either that or I am too into math.

At any rate check the gap on the triggers as well as the wave pattern on the sensors themselves. Look into the inspection hole at the TDC mark as the crank is slowly turned around to make sure your clutch is not coming apart (mine just did this and took out the triggers and sensors).

Another way to look at this is with a timing gun. Shoot the light at the TDC hole and look for a jumping signal at those RPMS, that is an easy one to see.

Compression 180 all 4. Timing marks all in order (crank, cam).

Thanks joe. Will recheck with a timing light today. The rpm is clockwork - and lasts for 150 rpm or so then smooths right out until the next step.

Had previously set sensor gap to .8mm. Then Swapped over to the late sensor bracket with the sleeve around the ref sensor. Noticed that with the associated space under the speed sensor the ref sensor is moved REAL close to the set screw. Bottom surface of ref sensor is scratched now.

Would love to look at the waveforms. As Click & Clack used to say 'mail me a Sun 9000 digital ignition analyzer ...'

Seriously. Why at this distinct rpm?? I hate old cars.