PDA

View Full Version : ls1 with ls7 maf, sanity check, input appreciated



rosey
October 31st, 2009, 03:26 PM
Motor is a 2000 camaro ls1/t56, ported 5.3 heads, 227/239 113+2 cam, bolt ons, in a 89 rx7 with a 3.55 final drive.

So I made a new intake several months ago, wanted to try a credit card style maf(ls3/ls7 style). I bought the maf new from gm and a flange to mount it, had it welded into a 4" aluminum pipe as shown.
http://img.photobucket.com/albums/v331/rosey769/SNC00309.jpg
http://img.photobucket.com/albums/v331/rosey769/SNC00308.jpg
I then inserted it into an intake made of pipes and couplers, and a LS2 C6 blackwing intake, excuse the dirtyness...
http://img.photobucket.com/albums/v331/rosey769/SNC00035.jpg

Having been through the autoVE and autoMAF processes many times, I figured it would be a piece of cake to redo the maf calibration table. I started with the table bruce melton and lingenfelter both give out as a starting point for an 100mm maf on an ls1...
http://powrmax.com/PowrMAF100%20Tables.html

Set the b0120 value to 400rpm and logged mile after mile, applying BEN corrections as I go, and while the idle and WOT stuff wasn't too bad, steady cruising and low throttle would cause very jumpy fueling, nothing seemed to help. The logged maf frequency seemed to jump around a lot, so I decided to put the stock maf in place of the 4" ls7 maf pipe and compare. The stock maf worked great, and the results are precisely what I expected to see, but unfortunately I don't have much of an explanation for it.

First the ls7 maf, this screen shot is just steady ~70mph crusing. the purple line on the bottom is maf freq on a 3000-4000hz scale. Notice the yellow AFR line in the second chart, it's all over the place, remember, this is after many long driving cycles and BEN correction factors having been applied...
http://img.photobucket.com/albums/v331/rosey769/ls7maf.jpg

Next is the ls1 maf, which is a stock 00 3 wire 78mm maf that has been descreened but is otherwise stock. This was with a slightly tweaked stock maf calibration table, but it could still use some tweaking. Even in its slightly rough state of tune, notice how much better actual AFR follows commanded. The purple line at the bottom is maf freq on a 4000-5000hz scale. The drop about half way through is just from a slight backing off of the throttle, but it is pretty obvious how much more steady the signal is...
http://img.photobucket.com/albums/v331/rosey769/ls1maf.jpg

Has anyone else had this problem, or is there an explanation for this that I may be overlooking? I would love to figure this out and be able to go back to the ls7 style maf, seeing as the stock one is a slight restriction, but more importantly I will max it out when spraying(dry nitrous).

Thanks for reading...

mr.prick
October 31st, 2009, 04:22 PM
Both MAF Hz are shaky, they should not be with TP% steady. :confused:
My ported MAF is steadier than both of yours, yet my AFR jumps a bit. (CL)
http://forum.efilive.com/attachment.php?attachmentid=6563&d=1257049146
Is there a difference in LTFTs between the MAFs?
What does the MAF table look like?

rosey
October 31st, 2009, 04:36 PM
Both MAF Hz are shaky, they should not be with TP% steady. :confused:
My ported MAF is steadier than both of yours, yet my AFR jumps a bit. (CL)
http://forum.efilive.com/attachment.php?attachmentid=6563&d=1257049146
Is there a difference in LTFTs between the MAFs?
What does the MAF table look like?


Its hard to compare on different scales, if you look at the maf frequency on the 0-12000hz(green line) scale it looks fairly steady and about the same as yours dones. I purposely made the bottom purple lines for both mafs on a scale that only covers 1000hz to exaggerate any unsteady-ness.

It is running in OL, so no LTFTs either way. The maf table for the ls1 maf looks like a stock one, basically, with a few minor tweaks. The ls7 maf table looks like the ls7 table I started with, but because it is so unsteady, it is very hard to get accurate correction factors and the values tend to jump around and require a lot of smoothing to make reasonable.

Highlander
October 31st, 2009, 04:47 PM
THis could very well be related to the size of the LS7 Maf tube that you have. The bigger the tube the less "sensitive" it will be to a given airflow. The calibration of Hz vs flow will be very different. Aside to that, if your MAF tube for the LS7 is of different size it will create a turbulence that will make the MAF a lot more uneven at low to idle flows.

Bruce Melton
November 1st, 2009, 02:12 AM
You are well into a major science project. There a many factors involved in getting a smooth laminar airflow. Without a degree in the subject, it will be a frustrating trial and error process.

First, the GM Hitachi element is far superior to anything previous so you are off to a good start. You can try to picture in your mind what the airflow looks like across the tube. You want that sensor opening in a non turbulent, smooth stable, location in the airflow path. This is VERY sensitive and minor changes make a huge difference.

From here it looks like you need to increase the the thickness of your element mount to cause the element tip move closer to the inner wall of the tube.

Another factor is the transitions you have just before and after the MAF. The air is being steered and the air velocity across the tube is changing. Ideally, all the pieces should be the same size and shape so as to have the same effective square inch area.

Jim Hall of Halltech, and I are months into developing an CAI intake with MAF for the new Camaro and we have made dozens of changes trying to achieve a smooth, even airflow. Even an eighth inch diameter change will cause a dramatic change in airflow. Very minor changes can net positive or negative results on the dyno. Often a minor change will net lower dyno numbers than even the stock intake. Bigger is often, not better.

Watch your fuel trims at a constant low rpm to see the difference between different configurations.

Bruce
the PowrMAF guy
http://dotster@lithomax.com/BCM/IMG_3545.JPGhttp://dotster@lithomax.com/BCM/IMG_3560.JPG

Highlander
November 1st, 2009, 07:17 AM
You are well into a major science project. There a many factors involved in getting a smooth laminar airflow. Without a degree in the subject, it will be a frustrating trial and error process.

First, the GM Hitachi element is far superior to anything pervious so you are off to a good start. You can try to picture in your mind what the airflow looks like across the tube. You want that sensor opening in a non turbulant, smooth stable, location in the airflow path. This is VERY sensitive and minor changes make a huge difference.

From here it looks like you need to increase the the thickness of your elemnt mount to cause the element tip move closer to the inner wall of the tube.

Another factor is the transitions you have just before and after the MAF. The air is being steered and the air velocity across the tube is changing. Ideally, all the pieces should be the same size and shape so as to have the same effective square inch area.

Jim Hall of Halltech, and I are months into developing an CAI intake with MAF for the new Camaro and we have made dozens of changes trying to achieve a smooth, even airflow. Even an eighth inch diameter change will cause a dramatic change in airflow. Very minor changes can net positve or negative results on the dyno. Often a minor change will net lower dyno numbers than even the stock intake. Bigger is often, not better.

Watch your fuel trims at a constant low rpm to see the difference between different configurations.

Bruce
the PowrMAF guy
http://dotster@lithomax.com/BCM/IMG_3545.JPGhttp://dotster@lithomax.com/BCM/IMG_3560.JPG

On the dyno and a new intake, you have to watch for hot air getting into the engine. There is not sufficient flow to simulate 100mph on the street. That way you can make a better design out of it. But....

I definitely agree on everything you said. MAF placement and achieving a laminar flow is key to having a stock driving vehicle.

joecar
November 1st, 2009, 08:53 AM
Very interesting... as Bruce said, a major science project.

My simplistic understanding:
- place MAF sensor a further distance from the bend (if it's after the bend),
- clock MAF sensor so that it sees the air coming from the outside of the bend, since the air from the inside would be turbulent,
- avoid sharp bends (easier to say than to do).

Bruce Melton
November 1st, 2009, 08:57 AM
Camaro> We are waiting for the plastic heat shield pieces for the box that will surround the air filter. There is an outside air intake low, beneath the filter.

rosey
November 1st, 2009, 01:51 PM
Looks like I have some experimenting to do. I'll have to try moving the element around until I hopefully find some laminar flow. Someone else also suggested adding a screen like the stock maf used, which I may play around with too.

Thanks for the words of wisdom :cheers: