I haven't taken the time to play with my car much since last fall, but looking back at my datalogs I noticed that airflow comes to a head around 55-5700rpm and almost flat lines from there and I'm not really sure what to make of it. You can see the motor in my sig. It should have alot more top end than it does and I feel like I'm down about 20hp from where I should be... don't think it's valvefloat because airflow doesn't taper off it just flat lines... heads and induction are healthy shouldn't be choking up... doesn't lean out and timing is where it should be. The headers overlap the exhaust ports a smidge because they're d port heads dumping into round tubes but I ground on the header flanges as far as I felt was safe to get them close to matching so I don't think that's a problem... ideas? All I can think of is POSSIBLY the lifters pumping up and hanging the valves open a tad at those speeds.

 

Hows your air filter setup? High flow filter?

 

Breathing through a 70mm maf and a tpi air box with standard filters.

 

You've put together a lot of parts from different vintage 60* engines there. Not really sure where to start other than to ask how you know you're not running lean (wideband?) and if the injectors are showing a "flat" pulsewidth to match the "flat" airflow over 5500.

If you're floating the valves at higher RPMs it usually shows up in some very erratic O2 readings above that point. I doubt an ignition (or timing retard) problem because that wouldn't cause airflow to slump, it would just eat power.

Or maybe you just have a very well matched combo that peaks out at 5700. Just because you hit peak power around 5500-5700 (which is still pretty high for a 2 valve pushrod motor) doesn't mean that's where you shift. I bet it pulls OK to well past 6000, right?

Having played aruond with mildly modified 3.4 TDC motors I can tell you that even with all that 4-valve airflow goodness they don't peak out much higher than yours. Used to think we were spinning one of them to 7000 until we bought an aftermarket tach with shift light and found it was actually only about 6400-6500. Bummer. But that was reality. Still pushed an old Fiero through the traps just a scosh under 13.00.

 

If you have a flexible air-duct in front of your throttle-body, check to see if it could be collapsing due to vacuum created at high RPM.
I would also check vacuum in your plenum during WOT runs to see what's going on there.

Have you datalogged while this is going on?

 

@ Damon, using a wideband and pulsewidth levels out too and no erratic afr's. Continues to hold just fine to 7k and above if I want to go there.

@ 305sbc, there's a tpi accordian-style duct ahead of the throttle but I don't think it's collapsing. Vac readings are normal all through wot and I've datalogged it.

 

When you say it's "flat" above 5700 just how "flat" are we talking about? And at how many g/sec (i.e. have you maxed out the MAF)?

If you're talking about absolutely DEAD flat (like "Doctor, the patient has died!" kinda flat) then I'd start suspecting some kind of ECM/programming limitation.

Also, if I may make an observation..... you definitely get the gold star on your lunchbox for going WAY further with a pushrod 60* V6 than anyone I've ever known. Where did you learn how to match parts from different generations like that?

 

Have you tried setting your lifter preload to as close to zero as possible to see if it has any effect?
If not that, then I think you have an airflow stall happening in your intake ports. You'll have to go through and look for any choke points or sharp turns, - things that would cause turbulence.
Have you had your heads tested on a flow-bench?

 

@ Damon, not dead flat but a very slow rise. I wish I deserved all the credit but most of it belongs to firstfirebird and ben over at 60degreev6.com. Converting the block to a roller cam is my innovation though. Check out what's on the v6 board, project89 has a destroked 3.4 around 600hp (though it's been touchey), Purple82RS has a 3500 with stacks in his camaro that I drool over and there's a few other fun projects going on.

@ 305sbc, Rockers are pedestal mounted so no adjustment to do, it's all in the pushrod length. Now that you mention it I do have formed steel gaskets on the manifold and when assembling it I noticed how closely the gasket matched the ports, maybe the gasket was mushed into the ports when torqued down. Need to borrow a scope so I can peer through the injector ports now...

 

How long are the intake runners from plenum to valve? I'm just wondering if you're suffering from them being too long to work at a high RPM range (like the limitation of a long-runner a TPI engine).

There's just nothing jumping out that says "ah hah, THAT's it!". It's got the feel of a combo that's just going as far as it can given the parts it's got. Like things are either too small to feed it at higher RPMs or the intake runners are just too long to work at a higher RPM range (intake resonance/harmonics).

On the 3.4L TDC motor we fabbed our own intake by basically mounting a plenum box on top of the stock lower intake. Never did get it sealed up properly or sort out the programming properly, but it showed promise. Low end torque took a big hit but from 5000-up it felt a lot stronger (and shift points moved up to an actual 7000 RPMs). Of course, the cam timing was changed quite a bit at the same time so I can't really say exactly what did what for sure. My belief was that the stock upper intake runners were teh most serious serious restriction not just becuase of their size but becuase the factory make them incredibly long to help bump up the midrange torque (to help haul a heavy Z34 Lumina out of the hole). I also think the stock plenum volume was is insufficient to work at elevated RPMs.

The math on stuff like this is, quite frankly, above my pay grade. But from what I learned from that experiment I took away that stock intakes are a serious compromise when you ask them to work 1000+ RPM over their stock range. They just aren't designed to work well over 6000.

 

can you post up your datalogs?


Sounds like it may just be a limitation in the bin file. If $A1 outputs calculated airflow like the older bins i'll post up what the cars i've tuned show.

 

that thing should feed well na ,though i do recall some negative things about the 3500 plenum from seeing one with the top cut off.its been a long time though since i saw what the inside of the 3500 plenum looked like.

i dont think it was bad enough to cause issues at such a low rpm though, hell with my custom upper plenum i was able to make iron headed motors rev to 6k n/a and 8+k turbo, so u shouldnt be having any issues with what u have

id be looking into an ecm lilmitation as stated above as well

 

Well, this could be a problem if you're getting lifter pump-up. I would check the amount of preload into the lifter.
If you have 0.050" compression of the lifter plunger, then that's how much clearance you can have when/if the lifter pumps up at high RPM. That will prevent the valve from seating by that clearance amount X your rocker ratio. That can be a LOT.

The available oil pressure, plus the spring in the lifter plunger tries to push the plunger up to the retaining clips on the lifter. If there is every any tiny clearance developed in your valvetrain during operation, due to slight loss of control by your springs, or the valves taking too long to seat because of a bad valvejob or worn guides, then the lifter plunger is designed to take up that slack. The problem is that once that slack is taken up it prevents the valves from seating until you slow the engine down and allow the lifters time to drain the extra oil back out.

Another issue is that the more preload you have on the lifter, the more that lifter plunger spring is compressed. If you go too far into the lifter with preload it can/will take away 10 lbs of seat pressure from your valvesprings. The lifter spring works directly against the valvespring.

That may not be your problem, but I think it's worth checking out.
Non-adjustable rockers are notorious for having a crazy amount of preload on the lifter.

 

Damon everything on the induction is pretty big, runners are only 8-9 inches if I recall and the plenum's huge. I don't think it's a breathing problem has to be something I've overlooked.

Project89 I think what you're talking about is the tube that runs the length of the plenum from the throttle and exits at the rear, I thought it was a funky design too.

305sbc I installed ls1 internal springs in the lifters because they're touted as pumpup resistant because of the higher spring rate, maybe that was a bad choice?

Purple82ta I'm still using the modified 6E code with a maf for airflow. Logs are on the laptop and I don't have a thumb drive handy but airflow looked something like the attached pic shows. Around 5500 to 5700 there's a somewhat sharp change in the curve.

 

...

 

Ok added a pic of the graph.

 

I may be wrong but I would look very close at the valvetrain, and I would try another set of lifters to see if any difference in results.

If a different lifter shows a better result, then I would try to get rid of almost all of the preload, - either with shorter pushrods, shims, or convert to adjustable.

 

Alright thanks.

 

shorty headers, an accordian "ribbed" shaped intake feed tube that is a restriction and a small cam, do you really feel that 5700 rpm is bad? I'd look into a smooth intake tube, long tube headers and then think about the cam being larger if you want it to pull higher rpms. What do you really want peak power to be at in a small engine? Do you have a 3600+ converter in the car?

 

not only the tube but iirc the runner entrances are off the floor of the plenum as well.

i highly doubt its anything in the valve train,many ppl have similar sized cams witht he same springs and rockers that turn 7K+ rpms and make power up that high.

just for the hell of it do u have a spare maf u can swap out, also take a vacum reading infront of and behind the maf sensor and see if theres a large difference

other then that only thing that comes to mind could be ignition starting to retard at rpm

 

heres the pic i was looking for , the runner entrances are on the floor but this shows the inside of the plenum



still have no clue why they designed it the way they did but the air has to come in then up and over the hump on the sides

 

xpndbl3, It has full length headers and I don't think the rubber tube is a restriction, if you look at the pic you can see it's sandwiched in there pretty good so there's no space from rib to rib. The change in the airflow curve @ 5700 is what caught my attention. Something funky's happening there because it shows up in every log... I'd like to make peak power around 500rpm higher to match my shift point a little better (7k). Converter's on the backburner 'till I have money but a 4k converter's the plan.

Project89, I have an LS6 maf but I need a translator for that, I'm starting to think you may be right about the programming. IDK what the ignition is really doing at that speed.

 

Combination of cam and intake design (and maybe the exhaust isn't quite up to snuff). I also now doubt my original suspicion of a programming restriction since all the readings are telling the same tale- good A/F ratio everywhere, power matches the airflow, there's just not enough of it above 5700. You are hitting the airflow wall.

I don't think you are going to find a single "ah hah!" fix for this. It's more systemic in nature than that. Like football is a game of inches, so is building power when you are well above 1 HP/ci like you are now.

Even in it's current form, restricions and all, I bet that's one f-ing fast N/A 3.4!

 

Lol it does alright and I still get 30 on the highway. So does the change in the curve @ 5700 seem natural to you? I would expect it to smoothly taper off instead of come to a sharp angle like that.

 

All it takes is one valve not seating correctly at high RPM to do what you describe.
Have you eliminated the possibility of one cylinder getting hot and going into mild detonation? You should see it on the sparkplug insulator as pepper-like black specs.

 

his car should be turning around mid to low 13,s id expect 11's out of it if he ever throws a turbo on that thing.

blc nextime u do a build i strongly recomend the destroked 3.4 route it deff makes a killer setup,just be sure to use a really good valvetain i had some issues with mine.

my TT iroc is almost done so soon ill be getting back to my v6 build again

 

305sbc, I haven't noticed any evidence of detonation. Chambers have been polished & smoothed and the c/r is mild so I wouldn't expect any, especially at this altitude.

Project89, everything loses about a second at my altitude even forced induction cars loose a good bit of power which really sucks, but it lets me push the edge a little too. I wanted to do a short stroke N/A 3.4 with 3500 heads and wring it out as far as it could go just for giggles, I might build a second and put it on alchol. I bet your 3.4 project whipped through the powerband with no trouble hu.

 

my destroked 3.4 -- i.e 3.0 build sucked down low off boost, i bet a stock 2.8 would take it off the line.

the motor has the power curve of the typical 1000hp supra no hp till 4k rpms but from 4-8k u better hold the hell on , but i was also spinning up a turbo with an 83mm turbine wheel and 1.15 housing. the bw turbo should spool much faster and make much better power.time will tell though

 

Yes, if you can explain it. No if you can't!!

You know what that looks like to me? Looks like the port is stalling. Not from a flow restriction but from being too small in the higher RPMs. Velocity is good but it has it's limits. Those limits happen around .5 Mach (half ths speed of sound). Now, usually that isn't a problem at only 5700 but I don't know how big the 3500 intake ports are (minimum cross sectional area). When you hit a velocity-induced stall in a port the flow curve hits a "knee" like that and just won't go any higher. Flowbenches can't simulate this becuase they run at low, constant pressure differentials. It's why aftermarket performance heads are sold in different port sizes. It's not just about higher flow it also about supporting that flow at elevated RPMs. At a certain point, size matters, not just flow numbers.

I will admit that this facet of engine design is well above my pay grade. And I am definitely not saying this IS the problem. There are any number of other things that could cause this from the cam to the intake design (which is still a good possibility). But you are definitely hitting an airflow wall. This should not be too surprising given the fact you are using mildly warmed-over stock parts. I'm just saying to stick this in the back of your brain in case you run out of other bogies to chase.

What I DON'T believe this is is your instrumentation lying to you. I believe it is exactly as your datalogs show. I also don't think you're floating the valves at that low an RPM given the upgrades you have made and the lack of any other symptoms that would indicate a loss of valve control. Those graphs and my gut are telling me it's an airflow issue.

 

the fwd heads have a really odd shaped port that i never really did like

they open up alot about an inch into the head. i never did pour a set to see what the port volume is though.
i have pics in my photobucket somewere that show the bowl area and how big it opens up and the shape of the port from the valve side ill have to dig those out

 

I have all that info stored somewhere on my computer from when I built the motor, I'll dig it up when I get a chance. So maybe my idea of the gasket intruding on the port wasn't so far off. Alot to look at, I'll let you guys know what I find.