Mechanicals updated or repaired, new Datalog

Blown88GT

Founding Member
Nov 13, 1999
2,242
549
164
Palm Beach Gardens, FL
New crankshaft balancer.
New Ford Racing aluminum drive shaft.
New aluminum radiator, mounted differently from old 3-core copper brass.
Same Mark VIII fan & DCC controller, both remounted a bit differently.

Runs rough in the 1500-2000 rpm range, under load when accelerating.
Have not tried running VEAL again with new 47 lb injectors.
Running VE Analyze---Analysis from Megalog Viewer results in significant changes.
Should I try these?

For those that don't notice, am using a MAF for fueling & there is a supercharger which accounts for high VE values.
See anything that stands out in the datalog?
 

Attachments

  • 2019-03-07_09.48.38.msl
    4.5 MB · Views: 121
  • Sponsors (?)


New tune much better.
New datalog.


Well then... What this means to me is that it's time to stop spending zeroes and ones on data logs and put them to better use recording some video of this thing doing a little pavement damage. :chin

That's [my] theory anyway :shrug:

I've been patient in letting you guys handle the easy stuff. :jester:
 
  • Like
Reactions: 1 user
Well then... What this means to me is that it's time to stop spending zeroes and ones on data logs and put them to better use recording some video of this thing doing a little pavement damage. :chin

That's [my] theory anyway :shrug:

I've been patient in letting you guys handle the easy stuff. :jester:
I just got my 1st smartphone, so am trying to figure that out, too. Android, not Apple.
Car is old, like me, can't afford any pavement damage.
 
Last edited:
  • Haha
  • Like
Reactions: 1 users
I try not to step on other tuners toes.. but post the tune file that goes with the datalog and I will scrutinize it for you.
VE numbers increased a lot. Maybe you can tell me what this means?
I suspect it has something to do with the supercharger pumping air in the cylinders even when it's not in boost.
BOV closes at about 10 inHg.
Went from 30 lb injectors to 47 lb injectors; changed required fuel & dead-time to correspond to the LU-47's.
 

Attachments

  • 2019-03-10_20.38.49.msq
    120.1 KB · Views: 128
Last edited:
Nope... I will have to take a long look, the ve numbers should not have changed much at all if your deadtime changes were correct, assuming your original deadtime settings were correct.
Draw or blow through MAF? Got a picture of the engine?
 
Nope... I will have to take a long look, the ve numbers should not have changed much at all if your deadtime changes were correct, assuming your original deadtime settings were correct.
Draw or blow through MAF? Got a picture of the engine?
That's what I thought & why I am confused.
DT is correct; calculated same way for 30lb & 47lb, using seijirou's online spreadsheet for MS2.
For LU47: DT=0.901, BCV=0.129
Blow-thru MAF, nothing new about intake.
IMG_1295.JPG
 
lu47 frpp injectors from my datasheet and specd into the excel file.
Target Min PW (ms)0.885367077
Dead Time 0.896005678
Battery Correction 0.12830864

The MAF numbers will only change in a blow through application when the engine is seeing boost...
 
Now lets look at some information from your tune file and latest datalog.
Blown88GT MAF CURVE issue.PNG

In this histogram we are looking at your VE table populated with raw MAF values, notice how the maf values fall nicely into the cells. Now in order to fix the issue you are having we need to adjust the MAF curve by increasing the MAF curve value by the percentage that its corresponding VE table value has your VE table will start to fall back to 100%...
[as an alternative you could make this histogram as an raw adc7 count value on 10bit scale and you can then adjust the maf value as is correlates to the voltage input instead, same result from a different perspective]

Now I can do this for you if you like, I made some small adjustments to your idle strategy but have not touched the fuel curve as of yet... I figured i would give you the chance to sort it out first. you dont have to alter the MAF curve if you dont want to, thats what the VE1 trim table is essentially doing but it will help with the tuning process overall if you do. I did not touch the injector settings as the difference between out settings is non-consequential.
I did also notice in your datalog that the raw MAF value is not clean, meaning that the MAF is seeing turbulent air in its current position, placing it in a piece of straight pipe 10x its diameter in length will help, this is often not feasable so i suggest using a piece of metal mesh or even fine chicken wire and stretching it over the mouth of the MAF housing before putting the silicon coupler in place.. this is a cheap and easy way to help straighten the air flow out before the sensor.
 

Attachments

  • Blown88GT.REV.1.0.0.msq
    120.1 KB · Views: 124
lu47 frpp injectors from my datasheet and specd into the excel file.
Target Min PW (ms)0.885367077
Dead Time 0.896005678
Battery Correction 0.12830864

The MAF numbers will only change in a blow through application when the engine is seeing boost...
There is a slight discrepancy between the Google sheet & the local Excel file values.
MAF numbers should increase if BOV can't vent 100%.
 
the engine is going to consume the air the sensor is reading, unless the pressure exceeds atmosphere the MAF sensor cannot read pressures only air mass. I agree that the SC will increase VE of the engine but the sensor should read that air, just like a larger displacement engine.

IMHO you should fix the MAF curve as it applies to your installation, it is not uncommon for the lower end of the MAF curve to have to be skewed higher than expected to correct airflow VE. I can fix the MAF curve for you if you like.
 
Now lets look at some information from your tune file and latest datalog.
Blown88GT MAF CURVE issue.PNG

In this histogram we are looking at your VE table populated with raw MAF values, notice how the maf values fall nicely into the cells. Now in order to fix the issue you are having we need to adjust the MAF curve by increasing the MAF curve value by the percentage that its corresponding VE table value has your VE table will start to fall back to 100%...
[as an alternative you could make this histogram as an raw adc7 count value on 10bit scale and you can then adjust the maf value as is correlates to the voltage input instead, same result from a different perspective]

Now I can do this for you if you like, I made some small adjustments to your idle strategy but have not touched the fuel curve as of yet... I figured i would give you the chance to sort it out first. you dont have to alter the MAF curve if you dont want to, thats what the VE1 trim table is essentially doing but it will help with the tuning process overall if you do. I did not touch the injector settings as the difference between out settings is non-consequential.
I did also notice in your datalog that the raw MAF value is not clean, meaning that the MAF is seeing turbulent air in its current position, placing it in a piece of straight pipe 10x its diameter in length will help, this is often not feasable so i suggest using a piece of metal mesh or even fine chicken wire and stretching it over the mouth of the MAF housing before putting the silicon coupler in place.. this is a cheap and easy way to help straighten the air flow out before the sensor.
Thanks a lot Steve. You've been a great help.

I think it's easier to understand if MAF curve stays "raw" & VE curve is greater than 100%. I like to use VE1 as a trim table.
It is a C&L MAF with green calibration tube (42 lb). It has 10 inches of straight pipe in front of it. The outlet pipe from the I/C angles upward behind the battery & that's where the MAF is located. 10X diameter would be 30 inches, I think Lee Bender told me 3X, thus 9 inches.
I have some bronze screening left over from one of my boats. I could do as you suggest. It's a POA to get to the MAF, need to remove battery to get to it.

For some reason, I have a mental block when it comes to interpreting data logs. Still trying to figure out how you generate the histogram.
 
the engine is going to consume the air the sensor is reading, unless the pressure exceeds atmosphere the MAF sensor cannot read pressures only air mass. I agree that the SC will increase VE of the engine but the sensor should read that air, just like a larger displacement engine.

IMHO you should fix the MAF curve as it applies to your installation, it is not uncommon for the lower end of the MAF curve to have to be skewed higher than expected to correct airflow VE. I can fix the MAF curve for you if you like.
If you think that's the only way, fine. I tried this once before & just made it worse. Probably my bad.
 
No disrespect, but what kind of controls engineer would I be if I can't figure it out myself?
Maybe I should just burn my 2 degrees?
I was a rocket scientist & this really isn't rocket science (LOL).
No offence taken, just figured I would offer... Its just that i am very familiar with the software, if you spend a bit of time in it you will be as well.
 
I'm trying to learn how to use all the features of my first smartphone.
Windows I know, Windows 10 Insider for over 4 years.
I know PC's like you know MS.
I left Apple with the end of the original Macintosh.
Android is new to me.

BTW, I see you do document control with your tunes for others, i.e. revision control.
Finally, someone besides me who understand this is necessary.
I wore many hats at one job: team leader, design engineer, configuration manager for electronics, QA & QC for electronics.
Had to do it all for no extra pay.
 
Last edited:
You have given me 3 changes to try:
1. Add flow straightening screen to MAF.
2. Blown88GT.REV.1.0.0.msq - small adjustments to idle strategy.
3. Blown88GT.REV.MAF RE-CURVE TEST 1.0.0.msq - large changes to MAF flow curve

Some of the changes made in #2 are not made in #3.
Is this intentional?
Comparing MAF calibrations; how did you arrive at #3?
All cells have been changed, except 0.1, 0.2, & 5.0 volts.
The original flow curves did not have enough data points for MS & I had to interpolate.
Attached is the original flow curve from C&L.

I will do #1 first to see if it cleans up the MAF by reducing turbulent air.
Then get a datalog after #1 is completed.
Then try #2 & another datalog.
#3 will have to be tested later.
 

Attachments

  • 76mm 89-93 green tube.zip
    2.7 KB · Views: 80