His trap speed is 106mph. You can't drive your way out of that. That's the indicator of power.

 

and we already talked about that earlier in this very same thread.

People: Stop worrying about and commenting on the ET. The ET don't matter and R/T matters even less. It's ALL about the trap, and we have that info. The OP is wondering about his power. His trap & weight give us his power, which is somewhere just over 300 NET hp. There are already two, pretty definitive things that should raise concern in this thread, for the OP...and they ain't ET or R/T. They are:
104-106 trap speeds which =~300-320 NET CHP. That is too low for that combo.
This quote from post #1: looks like it should be pulling strong to 5500. Instead, it feels like it really falls off before 5000

Falls off before 5000? Even with the mild cam, those good heads should support power higher than that. The exhaust concern is aligned with the perception that top end power isn't there as well as the weak trap speed.

 

You're right, I had that backward. That's a good idea!

Wow, very very interesting! Thanks for running that! I'm actually kinda shocked that those two agree so closely. I guess it's definitely time to figure out if I have a backpressure issue.

Thanks guys. That's exactly how I felt - I'm glad that I'm not crazy in thinking that my engine should be capable of more than that.

Unfortunately, fixing the header gasket hasn't been the magic bullet I hoped. It seemed to pull a little better up top, but everyone knows the butt dyno can't be trusted. The datalogs tell a different story - looks like it was running right around the commanded AFR at WOT, which means the volumetric efficiency hasn't appreciably changed. I'm surprised - I thought I'd see at least a few percent higher VE. However, if the exhaust is choked up at the cat like you guys think, I think that would diminish the effect of the bad gasket (higher pressure -> higher density -> lower velocity -> lower frictional backpressure).

I'll order stuff to check the pressure and update when I have results.

 

Do you have sensor bungs in both headers... if so with the engine warm pull one or both and go around the block. It won't be as huge a difference as dropping the exhaust but if your having a back pressure issue that drastic, it should be noticeable. Wide open will be okay as you're open loop anyway. If its that bad you will feel it.

Also, most autoparts stores have the back pressure testers on loan. Pay test and return for nothing more than a hold on your card until you bring it back. May be an easier route than buying.

 

You think it would be noticeable even at part throttle/low RPM? I'll keep that as a last resort so my neighbors don't hate me. I'm sure it'd make quite a racket with the bungs open.
Very good to know about those pressure test kits - definitely going to look into that first thing tomorrow. That would be nice not to have to fab something up. Thanks for the suggestion.

 

I had a similar problem with emission testing in Mass, and had to do this ...

 

^Nice work!

 

Sorry, the cam is over a decade old before Crane reorganized and I didn't see it listed in their catalogue. The heads are about 2 decades old. Holley worked with Brodix and sold a number of 18 degree heads for a short while. These are also pretty rare. Good luck.

 

Clever, I like it! I appreciate having the numbers for PSI with the single 3", the modified single 3", and the true duals. Do you have a crossover pipe on the duals? Doesn't look like it but I can't quite tell.

No worries, thanks for the info.


I have the pressure tester on order, and it should be here tomorrow afternoon.
Just for fun I ran the numbers for exhaust backpressure effect on power. I assumed 4" bore, 3.75" stroke, 6000 RPM, and 10 psig. Multiply pressure by piston area to get the force on the piston (126lb), and multiply that by mean velocity (RPM / 2 * stroke * 8 * (unit conversion factors) = 125 ft/s) to get power in ft-lb/s, then divide by 550 to get 29hp. And that's a lower limit, because the backpressure will undoubtedly be reducing cylinder fill. Maybe not a super useful calculation at the moment, but I thought you guys might be interested.

 

Alright guys, in concert with testing the pressure I'm continuing to datalog and tweak the tune. I had pulled some timing at the top end, but I'm starting to walk that back in now that the AFRs are mostly dialed. It's probably just psychological, but it seems like it gained a bit in the top end with a couple more degrees of timing. It seems like I'm giving it a ton of timing, but it's not getting any knock retard (I know it's possible to give too much timing without getting knock, which is partially why I had pulled the timing initially). The timing curve I just ran (getting no KR) is below. I've added 1° at 4000+ in the next tune. That brings it to 38° at 5200+, which I know is not likely to make peak power. However, I'm just kind of curious if I'll feel it drop off and/or what it will take to start getting some KR up there.

I wish I could ramp up the timing a bit faster, but it is just not having any extra timing at 2800 and 3200. I keep trying to sneak a degree back in there, and every time I get KR. Now that my WOT VEs are pretty dialed-in, I might see if giving it more fuel there lets me give some timing.



Another possible issue is that AFR jumps around a lot during a single pull. Ignoring some outliers, it might be +/- 5% within a given pull. I've noticed my tach signal is pretty noisy in the datalogs. For example, I see two adjacent cells in a WOT pull where MPH went from 68 to 70, but RPM actually decreased from 5700 to 5650. In another, MPH goes from 66 to 67, but both cells report 5250 RPM and the same injector PW. Because PW is the same, it shows that the ECM is using this RPM to reference the VE value. I'd think it would use the time between dist ref pulses, but I guess not. If the ECM isn't seeing the right RPM for looking up the VE tables, my AFR is never going to be consistent. And +/- 5% is pretty significant: at a target lambda of 0.85, that's 0.81 to 0.89. Assuming stoich of 14.73, that would be a target of 12.5 and ranging from 11.9 to 13.1

Does this just come with the territory of old EFI cars, or is something up with mine? I've replaced the tach filter because the wire broke on the old one.

 

38 degrees isn't horrible. I agree its beyond what should make power but not horrible. Who knows.

For the AFRs... again, seems a little much. I could see when targeting 12.5 12.3 to 12.8 but not that much I feel...

 

Don't you think a bad injector might have something to do with the A/F jumping around? You know cyl #1 temp is off.

Too much timing advance does nothing good. Your lower compression may tolerate more with pre-ignition but it still does no good. Your not on a dyno so you can't see power taper off but once power peaks there is no need to add more timing advance.

 

^^^ What he said. ^^^ You need to get your engine running perfectly in its current tune before you try to change anything. Otherwise you'll never get it figured out.

 

Thanks again for all the suggestions and help guys. I should have some good concrete data this evening if the pressure gauge comes in on time.

Yeah, that's kind of why I keep adding in timing. Once I hit knock at least I'll have an idea of a safe range within which to experiment. I've read these heads like ~34 degrees, but you never know... maybe something screwy is going on with my ECM causing it to not give as much timing as it should. I've learned not to immediately rule out the implausible.

Yes, but the wideband is on the passenger side bank. I don't see how the #1 cylinder would be affecting that. It could cause the narrowband (driver side) to read lean, causing the wideband to show rich across the board, but I don't think it would cause AFR to bounce around. I've also observed the ECM giving the same fuel 1/4 second apart in the datalog because it's incorrectly reading the same RPM twice in a row, and I've seen it jump a huge amount from one point to the next for the opposite reason. I'm not certain the #1 cylinder isn't an issue, but I'm pretty sure it's not this issue.

Like I said, I know 38° is past what's likely to make peak power. I'm just experimenting. It's currently fairly rich up there (lambda 0.83), which slows down the burn and requires more timing. After logging this tune, I'll pull back to a more reasonable advance (35° maybe?) and start pulling fuel. I think I've read these heads typically like lambda 0.85-0.88. I agree that a dyno is necessary to really dial in timing and mixture, but experimenting on the street is a lot cheaper than a dyno session.

How will tweaking the tune affect my ability to troubleshoot this? Sure, it has to relearn BLMs after a new tune is flashed, but that's just a matter of a quick heat cycle to re-learn.


As for diagnosis, I have a few ideas that don't involve pulling the intake. Obviously, a leakdown test would be good. Also, if I run it with the #1 injector disconnected and then with another injector on that bank (say #3) disconnected, the wideband should read the same in both cases once BLMs stabilize in closed loop. Oh, I just remembered - I pulled the driver side plugs a couple months ago and took pics. They look pretty crusty, but I'm not experienced enough to make any conclusions - I think maybe a hotter plug would be good, but that's not the current issue. What do you guys think?

 

When I said current tune, I meant with the current cam, exhaust, etc. as well as the timing and stuff. Looking at the plugs in your pictures, why is #5 so different from the others? You're like, "I think I need hotter plugs..." No, you need to figure out why #1 is running so cool and why #5 looks like it's running pig rich. Or is that oil, I can't tell from the pic.

 

I took the pic of #5 at a different time and in different lighting - that might be it. When I do the leakdown test, I'll get new pics of the plugs (except #3, which is a bear to get out). I think new pics and leakdown data should be informative.

 

OK, you're in Lynchburg, right? You've got one of the greatest engine builders in the country right there - Sonny's Automotive. I doubt they'd want to mess with your car (and you probably couldn't afford them even if they did, LOL), but they'd sure know a good local guy to take it to. Also, you're not that far from Natural Bridge Raceway. It's an eighth mile strip, but you could do a full throttle run and then immediately pull the plugs so you could get an accurate reading on them. You could also pull your exhaust there to see if your cat really is holding you back.

I ran my impala there one night while we were on vacation at Natural Bridge. Nice laid back track.

 

That is a good idea, but I think you'd need more than 9 seconds or so? To get a good temp/reading on the plug?

 

Sure, the longer the better, but I think the closest quarter mile strips are all the way over in Richmond. Doable, but probably an overnight trip.

 

Plugs 1 and 3 look nasty. How many miles/run time do you think you have on them?

It reads like you expect the computer/ECM to react a certain way. With bad/out of range inputs it gets stupid. I mean it maybe trying to chase it's tail and cycling inject pulse width and all injects on that bank. Only your data log knows how it reacts. Heck it may even change PW on the opposite bank too - it's just a machine that does what it's programmed to do.