bradyzq

We the Hybridz collective have talked about this (airflow limits of sidedrafts vs ITB EFI) often, but when doing the math, your 235whp at 6500RPM requires 51ish cfm per cylinder, which, in a 36mm unimpeded veturi, would travel in the neighbourhood of 53mph. Not very high or demanding numbers, are they? At 4500RPM, the numbers are less than 75% of that. I wonder if there is something else at play here. Maybe an overly strong aux venturi? We don't hear them mentioned often. They must have an effect on fueling, though, or they wouldn't be there.

Looking at the dyno graphs' shapes rather than the actual numbers, it seems that on your recent build, you maintained 90% or more of your max torque between 4200 and 6700RPM or so, which is slightly better than your 2008 engine which held 90% or more between about 3900-6200. I seem to remember that you have a badass ignition system (Electromotive?) If I'm wrong and you still have a single coil and dsitributor, that can make the difference between lighting off a mixture that's way lean or rich, or getting a misfire/stumble.

Is the boost reference hose to the regulator feeding anything else?

You have my condolences. This is major suck. My guess is valvetrain. Unless I'm missing something (entirely possible...) for a sparkplug to get hit, you pretty much have to have a broken valve stem and a valve head bouncing around in the combustion chamber.

Oh, I would suggest clean spark plugs too. If the plug is fouled, it becomes more difficult for the timing light to pick up the spark. On a related note, you can try getting the timing close at cranking speed with the injectors electrically unplugged. That way you won't foul the plugs if your timing is off.

Have you tested to see if there actually is a spark at a spark plug, not just through the plug wire with the timing light pickup? If so, and there is none, have you tested that the coils have power DURING CRANKING? Sometimes, a seemingly good candidate source for 12V sneakily gets shut down by the load reduction relay while cranking.

iVTEC would be awesome on this head. The lift portion would make for more expensive cam design among other things, but what about the continuously adjustable cam timing? Is there room for a tuna can on the end of at least the intake cam? It's amazing how much broader a torque curve you can achieve with even one cam being adjustable on the fly! I am not able to offer another solution, but the cam profiles (presumably low cam) of the K20A2 cams were arrived at assuming that the cam timing would be variable. I wonder if the "ideal" cam profiles would be different in a fixed cam timing scenario. Maybe more aggressive?

Hi All, I thought it was time to shout out a big "Thank you!" to the hybridz community for keeping the Z flame burning in me for the past (not so) few years since I last drove my Z. It seems that other priorities in life have jumped the Z and moved to the front of the line. This isn't permanent, but in the meantime, it sure helps to read all these threads and occasionally be able to help out and/or learn something along the way. So.... Thanks!

@ BLOZ UP, I love DIYAutotune, Megasquirt and Tunerstudio. The only thing lacking is phone support, and with so many users, and so many homebuilt kits of unknown build quality and installation, I can understand why they prefer to support (which they do very well) by email. I also really like the Haltech stuff from Platinum 1000 (PS1000) and newer, software included. And the guys in tech support have been great there too, including by phone. BUUUUT, when you mentioned buying an old used Haltech E6 as an option to a new MS, I laughed. That software is horrible, and those ECUs have been outclassed over time. You'll be miles ahead with a new MS2 or MS3X.

Another thing to consider is that the fuel pump is not getting good voltage. Check this while the engine is running, and preferably while it is exhibiting the symtoms you describe. IF original, the wiring in these cars is OLD.

Wow, you didn't mess around! Looks great. You can use Megasquirt maps if you really want to use someone else's maps. Just download the free version of Tunerstudio so you can look at them. But, honestly, since you already have the car running, your base map did its job. Work with your own maps. That way you don't run the risk of inheriting someone else's bad tuning to build on. Infinity is nice to use, but a bit of a pain to configure displays, especially if your screen is small.

Have you mechanically checked that your timing marks are correct? When you bring piston #1 to TDC, does the pointer aim at the 0 mark? Next, I would ignore the forums saying that "correct" offset is 60 degrees or whatever. There are so many ways and positions to mount trigger wheels and sensors. Just measure it by eye first, then with the timing light once it's running. When the engine is actually at TDC on cylinder 1, not just by reading what the pointer is aiming at, you should be able to see what the offset is from missing tooth to sensor within 5 degrees at worst. While the engine is still at TDC on cylinder 1, have a look to see where the distributor rotor is pointing. Hopefully it's at post number 1 on the cap... You can tweak it a bit once you know what your full range of ignition advance will be, so that you're not ever jumping the spark from the corner of the rotor. Last, there must be a spec for the sensor gap. It won't be "as close as possible without touching the trigger wheel." Find it and set it. Getting these things right tends to eliminate most sync issues.

Do you have Dave Rebello's engine dyno graph of your engine? Though the numbers must be ignored, the shapes of the curves should be very similar on both chassis and engine dyno sheets, barring any huge non-proportional driveline losses that would be seen only on the Dynapack.

I think your bigger issue is that RX8 throttle bodies are drive-by-wire.

Have you adjusted your valves? You may have effectively different cam profiles for each cylinder if your valves are not well adjusted.

Looks great! You don't specifically mention it, but since I think I see some sanding marks on the port side of the TWM intake, I'll assume you matched the runners to the ports too. MANY moons ago, like 20 years ago, I had my TWM intake matched to the intake ports. Surprisingly, the runners didn't line up perfectly with the ports. It was as if the bolt holes were evenly off a bit. Did yours line up well?

I know you're not fixated on the numbers, but if the PO is to be believed, the closest match is that the car was run in 3rd, which would mean your torque is actually 2.3% higher than indicated and your engine RPM is 2.3% lower than indicated. Not a big deal. What do you mean by "didn't sound good?" Did it run on all 6 cylinders for the full run? If you lose ignition on a cylinder, that will introduce a lot of unburnt oxygen into the exhaust, and show lean on a wideband O2 sensor. Did the car smell? Any smoke? Do you have an electric fuel pump? If so, is it getting good voltage? Is the fuel filter clean?

I agree with Jon above. Don't give up on the cam til the car's running right. What transmission and diff do you have? And what gear did you run in on the dyno? I'm asking because I can't come up with 4.30 overall ratio no matter which diff gear and tranny I mix and match. The Dynapack dyno measures torque at the hubs then divides it by the overall gear ratio the operator types in, in order to come up with the torque on the graph. It also uses the same ratio that was typed in to calculate engine RPM. It multplies the measured hub RPM by the 4.3 factor. So, if the ratio is wrong, both RPM and torque numbers will be wrong. Peak power will be correct, but at the wrong RPM. I had an L28 on my dyno awhile ago that had a simillar problem with torque dropping off. But he had a stock cam and advanced cam timing, with the resulting pinging problem. Which makes me wonder: did you hear any pinging during the dyno pulls?

I would suggest boost signal from intake manifold, and feed it to the lower port only of the wastegate, with your boost controller inline. Leave the top port open to atmosphere for now. You should have no trouble hitting 20PSI like that.

Have you tried plumbing the wastegate boost reference to the intake manifold instead? If the boost comes back (maybe even a bit higher because your reference is now after the piping and intercooler pressure drops) and stays back, then you have a leak between the turbo and the intake manifold.

Oh, yeah, and power will continue to rise with RPM only if RPM is going up faster than torque is dropping. 100lb*ft at 5000RPM is 95.2hp. 83.33lb*ft (100 divided by 1.2) at 6000RPM (5000 multiplied by 1.2) is 95.2hp. So, above, torque is dropping but power is not.

I'm not sure if I mentioned this to you already, but a couple of things come to mind here as possible issues. On the AFR graph, you show AFRs of leaner than 21:1 up high on the unplugged tube curve. The car absolutely would not pull at all cleanly at those ratios. So the ratios are suspect in themselves. This can be caused by a dying sensor, and/or a partially plugged tailpipe probe. I too use an Innovate tailpipe probe on my dyno, and it occasionally requires cleaning inside. The other possibility is a dying sensor. Since you use a tailpipe probe, the exhaust is cooler and the sensor, even with its own heater, is more susceptible to damage from water. There is not much water in a cold exhaust up where an O2 sensor is usually mounted, but in our cases, any condensation in the entire exhaust goes past the sensor. This can kill them fast. To minimize this problem, I suggest thoroughly warming up the exhaust and drying it out before attaching the tailpipe probe and sensor. It may be too late for this sensor though. A dying sensor combined with a partially plugged tailpipe probe would show as spiking lean off the charts as a function of exhaust flow. In other words, it would seem normal up to a given HP level, then spike lean with no corresponding drop in power or smoothness. To do it right, I would suggest having a threaded bung welded in to your collector, and install your sensor there.

Fouled plugs can cause a timing light strobe to be erratic. BTDT many times!

It's a bit of a pain that you haven't got MAP on the x-axis in either of those tables. You have to use both tables to generate that useful number. I assume you got that because you included pics of both tables. Leave the on-board MAP sensor open to the air. Your compensation table will effectively multiply you fuel table by the barometric pressure measured in bar, assuming you live near sea level. For example, if you're at sea level (baro is 1 bar), you would have no compensation (multiplied by 1). However, if you drive up in altitude, the barometric pressure drops. If you drive to an altitude where baro is 0.85bar, then you would want your baro comp table to multiply your fuel table by 0.85.

Is it possible that the data really is for opening time, not deadtime (more or less opening time - closing time)? That might explain where your missing 0.35ish ms went, but it won't explain why the car won't run at lower pw. Do you have the small injector pulsewidth table active and zeroed out by any chance? 'Cuz that would impact 2ms and lower pulsewidths.