dimented24x7

iTrader: (2)

See the attachment before reading on...

I did some datalogging with a wideband and Im seeing some funny results. It looks like there is a kink in the table for some reason. Its almost as if its two MAF curves superimposed over eachother. I dont think its anything on the computers side as it appears to be producing the correct output for a given input.

I suspected that there would be a point in the cars operating range where the flow through a big MAF would go from laminar to turbulent. It seems logical as, from the reynold's number, it should start off laminar and go turbulent at some point at higher flowrates. Anyone think this could be the possibility? Maybe the MAF is hosed? Its definatly the right direction as the car is wicked lean in that area with the stock LS1 MAF table. With that table, its fairly close for the most part. The intake track itself is basically a large cone air filter, MAF, and then the carb hat hooked up to the TBI. I dont know, maybe there really is such a thing as 'too big'.

Fast355

iTrader: (2)

I am not seeing the same problem as you though.

This is what my MAF table looks like

It holds between 12.4 and 12.7 on the Wideband across the RPM band.

I think it is that large straight piece of pipe that you are using for an intake.

My intake setup has a couple of large sweeping turns too it.

RednGold86Z

iTrader: (1)

This is TBI? Like standard 454 TB with big injectors? Or some other setup?

If it has something like dead-headed injectors, and a solid fuel system, I could imagine a harmonic resonance in the fuel system, but that would show a strong RPM correlation, not a MAF problem. Could add an accumulator (basically a fuel pressure regulator with the outlet blocked).

I'd try a longer straight intake pipe between MAF and air filter, and then try a longer pipe between MAF and hat, and see if either makes a difference.

Fast355

iTrader: (2)

This is with TBI.

Dimented one other thing, you might be setting up a standing wave in that particular area. My suggestion is to put a 45 or 90* corner in the intake tract and see if this anomoly goes away.

dimented24x7

iTrader: (2)

The weird thing is that I dont have any piping. The maf, cone element, and carb hat are all hooked to each other. If it was resonance, id expect to see it be more a function of RPM. Id also think fuel, but that would show with RPMs as well. This seems to be global and based on MAF flow more then anything else. I was also thinking of trying the old setup with piping, but I recall having standing wave issues with that. Seems like it needs to be more non-uniform like the later intake systems. Ill have to keep looking...

dimented24x7

iTrader: (2)

I lied. Im still stumped...

dimented24x7

iTrader: (2)

I did reconfigure the MAF ducting. It did help, and the car does go 0-60 in 5 seconds or so, but its still lean down low.

Fast355

iTrader: (2)

While my Van does run well with the MAF setup, I have not been able to achieve the same WOT power that a well tuned EBL setup would give. Then again I haven't even made any 8192 logs, just used my scan tool, a scratch pad.

On a side note, what engine are you running now Dimented.

Is that the deep dished piston HD 20270 crate 350 TBI with Vortec heads and 204/214 "RV" cam combo that you have been running? If so, I might need to pull out my Vortecs, blend the bowls a little, port the exhaust, cut them for 2.02/1.60s and swap them on.

Grumpy

Can you post a pic of the intake tract?.

Might be your MAF table is reflecting something *reversion* wise going on. The air flow has to match what the engine's *demand* is. ie If the engine has some point where the mechanical VE goes high or takes a dip, the fuel *demand* will make the same change.

dimented24x7

iTrader: (2)

I think I figured it out. After some more tuning, I got a standard MAF table with no kinks in it. Like you implied, the problem only happens below 1800 RPM. I think the issue is backflow from cylinder pulses in the intake. Easy way to fix it is not to operate the engine at low RPMs/high load, or run a baffle style MAF. Nothings perfect, I guess. Thats probably why GM has the dynamic fueling.

Fast, its a 350 w/ vortec heads, edelbrock intake and performer plus 'RV' cam. Has 8.7:1 or so CR. I finally got the timing fully dialed in (needs LOTS of it), and it is SICK. Doesnt pull real high, but has cummins turbo diesel like torque through 3600 RPM. Have you done any wideband datalogs? How about the timing? From tonight, the MAF is very sensitive to the intake configuration. Incidetally, at WOT, SD works fine with TBI as theres not much fuel dynamics at high MAPs. Its only at part throttle that I had all the problems.

dimented24x7

iTrader: (2)

Its definatly a reversion issue. Once the throttle passes a certain point, it bogs down at low RPMs. Let out a little bit and the AFRs immediatly come right back up. Its not a huge issue, but a little annoying with a stick. With the auto, it was never an issue as it was impossible to go below 1500 RPMs.

Ive seen that companies make probe MAFs specifically designed to handle reversion. Ill have to try one one of these days.

Fast355

iTrader: (2)

I have noticed that as well ocassionally. My Van will ocassionally go lean as I open the throttle in OD with the converter locked. At 45-50 MPH it is only turning like 1,100-1,200 RPM. After it hits about 1,500-1,600 it is fine, just down low such as pulling a hill at 45-50 MPH it goes lean and starts bucking lightly.

dimented24x7

iTrader: (2)

Yeah, its similar with mine. Goes lean at very low RPMs and high throttle. Other then that, it runs well. Guess Ill have to take the good with the bad...

Grumpy

Or, go MAP. <G>

Fast355

iTrader: (2)

Funny you mention that. It did it with the MAP to, except worse. I think mine is more an overall combination problem.

dimented24x7

iTrader: (2)

If I had MPFI or TPI I would... I have a CAI and its bad enough having the AE be a moving target. The MAF is also nice because its easy to tune.

Fast, have you looked at the timing in that area? Might also be a good idea to verify it with a WB before coming to a conclusion, if you havnt already. It could also be that the lower flows in the MAF table are off. My table needed alot of tweaking as the actual flow and the reported differed alot.

Grumpy

Count the total number of data points, that have to be tuned, and there's just not that much of a difference. There are a large number of table entries that the engine can never acheive.
Hmm, easy to tune, and here you are having reversion/ tuning issues.... <G>

Grumpy

You're right, nothing will cure a bad combination.

So far thou, I haven't seen a combo., that didn't run it's best on MAP. With MAF your always, always having to fight the misreporting error of manifold filling. Well, as far as with a hi-po combination, goes. The transistional AE will always be something you have to chase around with a MAF. Long small runners, and a small plenum are all crutches to not needing much AE, interestingly enough that discribes TP to a *T*. But, even with those items, GM still went to MAP in the later TP's.

dimented24x7

iTrader: (2)

With TBI, MAP isnt as easy to tune. It can be purely empirical based on the setup (standard carb intake and CAI). If you look at the PCMs, the IAT isnt used at all, even though there is sophisticated routine in there to blend the temps and calculate the true inverse density term. With TBI, the intake charge temp goes all over creation. Basically its like injecting R-134A into the intake randomly and chilling it. Really has the potential to throw the fueling into a top hat. With MAF, the fueling is fairly consistent with TBI.

As for the tuning, its one small area that normally would never be realized with an automatic. With 2.77's and a TKO, its real easy to duck into low RPMs, and even easier to dump down below 600 RPM where the timing is reduced to prevent the engine from kicking the starter clear off the block. Other then that, its quite drivable, and runs as well as an old **** heap can be expected to. 0-60 in 5 seconds with good low end torque aint bad for an old junker

We could go on and re-ignite the whole MAF vs. SD arguments, but in some apps, one system will work better while some apps the other will work better. Despite the whole low end bog at max throttle, I definatly prefer MAF with the TBI setup. It decouples the fueling from the engine and the fuel dynamics. If I had MPFI, Id probably run SD. The MAF has very good resolution, and is more versital then SD, but lacks the dynamic range. Conversly, with SD, the fueling and engine are closely coupled. Neither sysem is perfect... Such is life.

dimented24x7

iTrader: (2)

They do make MAFs that can handle the full range of inputs, take a look at fords MAFs with the baffles to prevent low RPM cylinder pulse reversion at high loads. Another likely reason they dumped it is that bosch MAFs where complete garbage, not to mention the poor resolution the 8 bit systems had. +/-4% varience in the idle fueling is terrible. The later hybrid systems on the LS1s are definatly the shizzle. Very smooth running setup.

To some extent, the AE is also there to cover up delays as well as enrich the acceleration. One issue Ive had with both MAF and MAP is the need for a short pulse of AE to cover for lag in the system. The AFRs will go dead lean for a split second without some form of a short, quick pulse to cover the transition.

dimented24x7

iTrader: (2)

I took another look at the datalogs from this. It doesnt look like an issue with the MAF at all. The MAF is reporting around 100-120 gms/sec at 1800-2200 RPM. This equates to VEs of around 95%. The PWs too are in line with the airflow. The PCM and MAF are doing what theyre supposed to be doing, but still the AFRs where dipping down to 20:1 at WOT until the RPMs come up a bit. Also not giving full throttle to the motor will bypass the issue to a good extent. It will also not happen at higher RPMs. the lower the RPM, the worse it is.

It could be a fuel issue, but the pressure is steady at 18 PSI, so the pump is ok.

Could this instead be some cam timing issue, or maybe the fuel pulses being scavanged out of the cylinder by the cam overlap? something wacky with the injectors? This is definatly one wierd and very strange problem. Even at full injector duty cycle at low RPMs it still leans out a little. When it does have fuel, it goes like hell. It also has a slightly lopey idle, which is strange for a cam with only 204 degrees of intake duration. In addition, it sounds like theres ball bearings bouncing around in my y-pipe when the engines under load. Sounds like fuel burning off in the exhaust.

jwscab

Sounds an awful lot like where the cam starts it's effective range. You start scavenging more effectively, and the engine VE is going to be high. You might be stripping the fuel off of the intake walls, and causing a momentary lean spot. Could be that on the throttle transition, you need to kick out some extra fuel.

dimented24x7

iTrader: (2)

Aye, captain, shes giving us everything shes got...

Ive gotten to the point where Im at 100%+ DC with the added MAP and TPS AE during transition. Fast355 reports his engine is pig rich with these settings during AE, so its definatly me. In third or fourth with sudden WOT, it starts off dead lean and gradually gets richer and richer until it comes up to the desired AFR at 3000 RPM. The MAF and PW readings look good. Sometimes its ok, but alot of times it just goes dead lean and doesn make much power. My theory is that maybe the fuel pulse is being sucked clear out into the exhaust.

dimented24x7

iTrader: (2)

I wonder if slewing the injector firing timing would help? Maybe delay the firings enough to give the exhaust valves time to seat on the cylinders that will be on their intake cycles.

Fast355

iTrader: (2)

Dimented,

How much fuel are you running on your combination?

I have been tuning Kurt's (OldRed95 on FSC) 1995 C1500s TBI 350 and have found that I had to peak out his VE table around 2,000-2,800 rpm, despite running 61 lb/hr injectors at 25 PSI. His engine is basically the same as yours. The more fuel we keep giving it, the more it responds with that fuel.

Dominic Sorresso

dim,

You think your experience is analogous to what I have been discussing in this thread?
https://www.thirdgen.org/forums/diy-...n-pattern.html

dimented24x7

iTrader: (2)

It could be, but this seems to be a high MAP problem on my car. At low kPa's, its fine. At higher kPa's is when it starts to lean out and make that annoying bouncing ball bearings sound in the y-pipe.

It definatly responds to more fuel, but the calculated VEs are around 150%, which doesnt make much sense. Once the RPMs build, the AFRs come back in line. Due to the fact that it can fuel it even at increasing PWs and speeds leads me to believe around 160 PPH should be enough to feed it down low, unless there is something really FUBAR. I donno...

My best guess is that the exhaust pulses are vacuuming out the manifold at higher vacuum and low RPMs. Its really bad at around 400-600 RPM. Starts reading in % oxygen on the wideband. It was also an issue with the old ECM and ignition system. Its a damn strange problem. Ill have to keep looking, maybe theres some more logical reason for it.

Fast355

iTrader: (2)

I had to ditch the cone style filter on mine. Put in an Astro style air box with matching filter. The cone filter was disrupting the MAF sensor readings.

I honestly still have not touched the stock 5.3 MAF tables, except to put 5 gms/sec in the lowest cell.

dimented24x7

iTrader: (2)

Just remembered this... Know what else is interesting, they force async fire at 80 Hz above 1300 RPM and around 55 kPa in these PCMs. Basically Im in async whenever Im at higher loads. I wonder if that is the issue? It seems plausable as the large, sudden intake pulses could be left high and dry with the injectors dribbling out fuel at 80 Hz rather then delivering it all in one shot in sync fire. With my luck itll agrivate the problem more being full sync, but if it cures the lean issue, Im going to drop a nuke on its *** and blow the manditory async code out of my bin. Im praying for thick clouds of black smoke out the tail pipe next time I hit the gas.

Dominic Sorresso

Dim,

Interesting that you bring up asynch fueling. I have asynch disabled in my bin.

Fast355

iTrader: (2)

I've had the manditory Asynch disabled to 4,000 RPM in mine. The stock Van bin is disabled to 4,000 rpm then switches to Asynch. I have had a high rpm miss for sometime now. Going to push the manditory asynch up to 6,400 rpm and see how about 4,000 rpm runs now.

dimented24x7

iTrader: (2)

The RPM thresholds dont actually do anything as the MAP thresholds are checked FIRST. If these are met, the PCM enters async. Ive had the same issue, a strange, high frequency shudder at high RPMs. Hopefully deleting the code will kill two birds with one stone.

Fast355

iTrader: (2)

The MAP thresholds are checked first, but the RPM to enable still has to be met.

I am going to bump the MAP up as high as it will go and do the same with the RPM. Will let you know how it goes. At WOT, it definately waits until 4,000 rpm to enter asynch though.

dimented24x7

iTrader: (2)

Oops... I stand corrected. Went back and checked and the RPM threshold will disable it.

dimented24x7

iTrader: (2)

Man, got my hopes up too that I might have found the problem, lol...

I dont know, then. Maybe async is the answer. It does seem to be ok and starts recovering once its gotten up past 1500 RPM. Ill have to tinker with it and see if it makes any difference.

RBob

iTrader: (1)

I've found that to run async for the main fueling that the injector duty cycle has to be high. Greater then 90% is best. Otherwise it just doen't run right. I figured it had to do with fuel distribution. Some cylinders will be lean with others rich. All depends upon when the injectors fire and which cylinders are inhaling at the time.

RBob.

dimented24x7

iTrader: (2)

Ive found that out as well. At lower DCs, the engine acts odd at higher loads and speeds.

Whats driving me crazy, though, is that at low speed/high loads its going really lean for no real reason at all. The fuel should be getting in there. At part throttle its fine, but once the loads build, it leans way out at low speed. I wonder if delivering the fuel as shorter pulses at a high frequency would help when its under low RPM, high load?

Fast355

iTrader: (2)

Here is what happened at WOT with stock 68# injector running asynch on my stockish 1991 G20. Notice the graph at WOT and high RPM.



I turned the asynch off and redyno'd in hotter conditions. On a 15* hotter day with the clutch fan engaged, it lost 10 ft/lbs of torque, but held the same peak HP. It felt much stronger at high RPM when the fan was disengaged.

dimented24x7

iTrader: (2)

Thats exactly what mine does at high RPMs. The cylinder intake cycles are much faster then the injector firings at high RPMs in async and it gets a strange high frequency vibration from it. Probably the rich, lean, leaner, rich, lean, leaner... as the injector firings are out of sync with the intake cycles. Im going to disable it and see how it works then.

Im also wondering whether or not this is still an AE issue down low. How much fuel does an accelerator pump deliver over time in a carb? Im going to max everything out as a last ditch effort and see what happens.

Fast355

iTrader: (2)

Have you thought of lowering the TPS to enable PE down low and richening up the PE a/f ratio there as well?

Holley uses 30 and 50 cc accelerator pumps.

dimented24x7

iTrader: (2)

30-50 CCs is around 20-35 grams of fuel. Wonder how fast its delivered? Over 2-3 seconds? Thats actually a decent chunk of fuel.
----------
Ive thought of enriching down lower, but Id like to actually get it to target the correct AFR. It seems the MAF is reporting a reasonable ammount of flow for the load and engine speed, so it should be in the ball park AFR wise.

dimented24x7

iTrader: (2)

Im definatly thinking this is an AE issue. It seems like this manifold needs LOTS and LOTS of fuel during transition. Added alot more and it definatly helps. Now, though, the AFRs are pegged at full rich when the throttle closes after AE. Its really bad after cruising on the highway. Hit the gas and booooggggg. Ive felt the manifold and it gets really cold while driving the car. Im probably going to have to put an RTD right in the manifold wall so I can actually see what temperature it is. Right now the AE is all over the place when its based off of the cool temp as the manifold is at whatever temp if feels like being.

dimented24x7

iTrader: (2)

Think I figured out the other half of the problem. I didnt have any filtering set for the MAF at high loads, so the computer seemed to be fueling per the instantanious airflow, which is near zero when there arnt any intake cycles. No intake cycle when the MAF is read, no fuel... Set the filtering so Id get more of an integrated signal, and it seems to work much better. No bogging down low and it tolerates lugging the motor. This car likes to get my hopes up, so Ill wait and see how it works when I try it again with a little less AE.

Fast355

iTrader: (2)

I wonder if I just have never noticed it with my combo. The automatic with a 2,600 rpm converter just doesn't load the engine like a 5 spd can. If I hammer the gas it lays over for maybe 1/2 second than takes off like a unguided missle. I have always seen grey smoke out of the pipes though when I hammer it. I still have almost no AE to it. Atleast not compared to what you were running with your setup. Yours was so rich it would nearly flood the engine out on a quick throttle romp at idle.

dimented24x7

iTrader: (2)

I think it was a big mistake on my part to underestimate the speed of the sensor. In reality, its probably faster then greased lightning. It makes sense as the lower the RPMs, the more discrete the signal is, so the fueling would crater if the sensor reads and intake pulses where out of phase with eachother. Another strange problem... Im surprised it ocured to me in the first place to check it.

In the later MAFs, they may have put something in to integrate the signal internally. If its blowing black smoke and is rich on the wideband, then I wouldnt sweat it.

This is what I was using and it seems to help alot. Potentially itll never be perfect, though, as there will always be some aliasing in the signal at low RPMs.

dimented24x7

iTrader: (2)

Works ok now with the filtering to integrate the signal. Acts the same way it did when I put it back in speed density last night. Momentary stumble at low RPMs and then it goes fine. No stumble at all when warm, so it must be too much AE when cold. About time...

dimented24x7

iTrader: (2)

Did some real time high speed datalogs and the MAF output is a very dirty signal at WOT. Alot of it is from the engine intaking air, but a good chunk is also from the resolution limits of the hardware. Without the filtering, the PCM was tracking along the signal. Slide filtering helps alot and the car doesnt go dead lean for a long time like it did before. Leans a little, but recovers and goes just fine. Think adding a moving average filter would help on the frequency side?

The first graph is a 300 msec snap shot of the frequency output at higher RPMs. The second is another 300 msec snapshot of taking off in second gear from a dead stop.

HaulnA$$

Where are you picking off the data? Is that data taken directly off the MAF sensor itself through a F/V converter? What hardware are you saying lacks the resolution, the PCM or your logging hardware? A lack of resolution acts as a filter in and of itself. From the looks of the datalog and then looking at a MAF table from an LS1, you are varying about 200 Hz at 7.6 kHz. That is about an 8.8 gm/sec swing at 108 gm/sec. Right at 8%, a whole lot. A moving average filter will definetly help but I think the 20 point filter in the graph is too much. JMHO

Fast355

iTrader: (2)

I am curious about his wiring setup.

The setup has the frequency straight from the MAF going into the speed sensor circuitry in the ECM. Straight shot, NO converters of any kind.

I am running the stock 85mm MAF sensor on mine from a LS1/5.3 Vortec. I am running the stock MAF table from the newer LS1. Sits right at 125-131 BLM at all times, except at idle with the A/C on where it creeps up around 140 BLM.

Still don't fully understand his problem with the setup.

I think Dimented is talking about the rate at which the MAF signal is read by the code as well as possibly the datalogging equipment.

On the other hand, I have Monday off and am going to have to change yet another U-joint. HORRIBLE vibration again while accelerating around 2,500-3,500 rpm at part throttle. Checked and its the rear U-joint AGAIN. Crappy Taiwan U-joint. Anyone know of a good old quality AMERICAN piece to replace it with. Its taken a horrible beating lately.

dimented24x7

iTrader: (2)

Like fast said, it goes right into the PCM, no converters. The resolution is actually quite good. A 'good chunk' was bad wording on my part. Me not thinking before typing. I think the main problem is the fast response of the MAF sensor. There is a small ammount (<2%) error due to uncertainty in the counters as well as ~2% from the sensor itself, from what Ive read. This doesnt do much by itself, but it probably does factor in with all the other airflow noise in the intake. At light throttle and idle, the noise is just a percent or two. It was actually faster with the throttle partially closed at high loads without filtering. Probably because the restriction acted to filter out the intake pulses by itself. The 'noise' does seem to be actual airflow, so the average of all that should yeild the true airflow rate over time.

Fast, it shouldnt be, and never was, an issue with an auto as youll never actually get to low RPM, high load operation. At low RPMs, the instantanious airflow varies by +/- 20%. This causes the motor to run real erratic. When Im going around a corner in 2nd, without the filtering, the engine would plain crap out when I got back on the gas. Works ok with the filtering, but still goes somewhat lean at 800 RPM or so. Even that is more then tolerable, but Id like to see if using a rolling average filter on the frequency can help smooth things out even more. Ill have to tinker with it in excel and see what looks good. GM has one in these PCMs for filtering the derivative of the TIS input. That uses a full 50 samples as well as a convoluted method for weighting the start and end values in the sample set. Im going to take that over for the MAF and tweak it a bit. Maybe have the # of samples used be variable based on RPMs.