Load calculation/RPM reporting?

Anything about Ford EEC tuning. TwEECer and Moates questions dominate, but there's some SCT and OBD-II knowledge too.

Moderators: cgrey8, EDS50, Jon 94GT, 2Shaker

Post Reply
wickedsnake00
Gear Head
Posts: 22
Joined: Sat Apr 22, 2017 3:12 pm

Load calculation/RPM reporting?

Post by wickedsnake00 » Wed Aug 01, 2018 8:53 pm

Does anyone have a good grasp of how load is calculated, and how RPM is reported? I would assume that load is just a function of airflow vs. RPM like simple VE, and if anything would lag changes in RPM or airflow, if only by the time it takes to do that calculation. For RPM I would think reported RPM would be fairly close to actual and not lag too significantly. Is that accurate?

My concern comes from a stalling issue I'm trying to figure out. I've got it idling smoothly, but in the range of 90-120s after cold start it will seemingly randomly drop RPM fast, sometimes recovering, sometimes stalling. The only thing I noticed is that load and fueling jump up about 200ms before RPM drops off, even though airflow and RPM are still level at that point. I'm curious if there's something to explain this, or if its just the way they're calculated and broadcast.

Refer to the attached datalog for an example of what it looks like.
RPMdrop_datalog_080118.zip
(12.73 KiB) Downloaded 40 times

User avatar
cgrey8
Administrator
Posts: 10569
Joined: Fri Jun 24, 2005 5:54 am
Location: Acworth, Ga (Metro Atlanta)
Contact:

Re: Load calculation/RPM reporting?

Post by cgrey8 » Thu Aug 02, 2018 7:40 am

There are often two "Load" values that tunes have used and differentiated between. In the older EECs, they are referred to as Load and Load%. In newer EECs, the Load% I think is called PerLoad.

But basically Load is as you described however they also take into account the declared CID of the engine. And I honestly don't know the difference between Load and VE. They seem like the same thing to me. And Load is generally used by the Spark equations.

However Load% is a scaled load relative to the engine's max possible load...at least that's what it is supposed to represent when the scaling function is setup correctly for your application. In GUFx defs (and I'd assume in others), there's a function that accepts RPM & Load as inputs and outputs Load%. When this function is setup correctly, WOT at all RPMs should result in Load% being very near 100%. Why that's useful, I honestly don't know which is why I setup my tune to make Load% = Load and just tune that way to simplify my life.

As for your stalling issue, I don't have a suggestion there. Looking in your log, I didn't HEGOs or WB being logged to know if the drop is due to a sudden rich or lean condition. But that'd be what I'd want to know.
...Always Somethin'

89 Ranger Supercab, 331 w/GT40p heads, ported Explorer lower, Crane Powermax 2020 cam, 1.6RRs, FMS Explorer (GT40p) headers, Slot Style MAF, aftermarket T5 'Z-Spec', 8.8" rear w/3.27s, Powertrax Locker, Innovate LC-1, GUFB, Moates QuarterHorse tuned using BE&EA

Member V8-Ranger.com

User avatar
EDS50
Administrator
Posts: 3652
Joined: Sun Apr 05, 2009 9:17 am
Location: Tampa Bay, FL
Contact:

Re: Load calculation/RPM reporting?

Post by EDS50 » Thu Aug 02, 2018 10:04 am

cgrey8 wrote:
Thu Aug 02, 2018 7:40 am
There are often two "Load" values that tunes have used and differentiated between. In the older EECs, they are referred to as Load and Load%. In newer EECs, the Load% I think is called PerLoad.

But basically Load is as you described however they also take into account the declared CID of the engine. And I honestly don't know the difference between Load and VE. They seem like the same thing to me. And Load is generally used by the Spark equations.

However Load% is a scaled load relative to the engine's max possible load...at least that's what it is supposed to represent when the scaling function is setup correctly for your application. In GUFx defs (and I'd assume in others), there's a function that accepts RPM & Load as inputs and outputs Load%. When this function is setup correctly, WOT at all RPMs should result in Load% being very near 100%. Why that's useful, I honestly don't know which is why I setup my tune to make Load% = Load and just tune that way to simplify my life.

As for your stalling issue, I don't have a suggestion there. Looking in your log, I didn't HEGOs or WB being logged to know if the drop is due to a sudden rich or lean condition. But that'd be what I'd want to know.
I apologize for the thread jacking but I am also a huge supporter of setting prldsw to load%=load. My question to you is do you also set your load values for fn072a to the same values as fn071 or do you leave fn072a at default value? Just for reference purpose I set my fn072a to match fn071.


As for the OP... Looking at your log your your idle air integrator shits itself @ 120 sec but your afr stays pretty consistent. Timing seems to start moving around after the integrator gets out of range. My guess at this would be you need to add more air to your startup isc multiplier or look into your iac going bad.
1992 LX - 25.1c Chassis, Vortech Blown Dart 332, Lentech Trans, TRZ Backhalf, A9L, Moates QH/SL v1.9, BE, EA, TunerView

User avatar
cgrey8
Administrator
Posts: 10569
Joined: Fri Jun 24, 2005 5:54 am
Location: Acworth, Ga (Metro Atlanta)
Contact:

Re: Load calculation/RPM reporting?

Post by cgrey8 » Thu Aug 02, 2018 11:10 am

EDS50 wrote:
Thu Aug 02, 2018 10:04 am
...do you also set your load values for fn072a to the same values as fn071 or do you leave fn072a at default value? ...
For those not familiar, those are the Load/Load% scaling functions for Spark and Fuel. And yes, mine are ALMOST identical. The only difference between them is the lowest Load row value. One is set to 18 and the other is 16. I cannot now remember why they are different or if that was intentional. My guess is I could go back and make them the same and from a driveability standpoint, I'd never notice the difference.
...Always Somethin'

89 Ranger Supercab, 331 w/GT40p heads, ported Explorer lower, Crane Powermax 2020 cam, 1.6RRs, FMS Explorer (GT40p) headers, Slot Style MAF, aftermarket T5 'Z-Spec', 8.8" rear w/3.27s, Powertrax Locker, Innovate LC-1, GUFB, Moates QuarterHorse tuned using BE&EA

Member V8-Ranger.com

User avatar
skunk
Tuning Addict
Posts: 391
Joined: Sat Jun 25, 2011 2:30 pm
Location: Newmarket, ON

Re: Load calculation/RPM reporting?

Post by skunk » Thu Aug 02, 2018 12:01 pm

Since we are having this little sidebar, I have to add my 2 cents and add some clarity.
When this function is setup correctly, WOT at all RPMs should result in Load% being very near 100%. Why that's useful, I honestly don't know
It's useful in creating a third axis (RPM) for your ol fuel. This basically turns those 5 or 6 cells under your ECT operating temperature into a complete OL fuel table spanning your entire RPM range. Has little to do with WOT except that WOT occurs within OL.

My apologies to the OP as I have nothing to offer directly to your original post.

John
1987 Turbocoupe.
A3M1/A9L, BE/EA, Tweecer RT.
306,Single Turbo, Ported E7's, Ported Stock Intake, 42lb Matched Green Tops, PMAS 3" Blow-Thru Protube (42# supercharger calibrated),T5z, 3.73 gears
..... Nothin Fancy.......Just something to keep me from the honey-do list.........

User avatar
cgrey8
Administrator
Posts: 10569
Joined: Fri Jun 24, 2005 5:54 am
Location: Acworth, Ga (Metro Atlanta)
Contact:

Re: Load calculation/RPM reporting?

Post by cgrey8 » Thu Aug 02, 2018 12:10 pm

My translation of that is it allows more utilization of the cells in the OL fuel table. Just to exaggerate a little, if actual load ranges from say 1000RPM/70% then goes up to 4000RPM/150%, then there'd be cells above 70% for the low RPMs that are unattainable and therefore go to waste. However if you normalize Load to 0-100%, then you can make use of the entire table from idle to Redline. If that's the case, then I guess the Ford developers felt they needed the higher resolution for fuel but didn't need the resolution for Spark, EGR, and the other things that only reference Load, and not Load%.
...Always Somethin'

89 Ranger Supercab, 331 w/GT40p heads, ported Explorer lower, Crane Powermax 2020 cam, 1.6RRs, FMS Explorer (GT40p) headers, Slot Style MAF, aftermarket T5 'Z-Spec', 8.8" rear w/3.27s, Powertrax Locker, Innovate LC-1, GUFB, Moates QuarterHorse tuned using BE&EA

Member V8-Ranger.com

User avatar
EDS50
Administrator
Posts: 3652
Joined: Sun Apr 05, 2009 9:17 am
Location: Tampa Bay, FL
Contact:

Re: Load calculation/RPM reporting?

Post by EDS50 » Thu Aug 02, 2018 12:16 pm

I always had the feeling that Ford wanted the extra resolution for fuel so conditions for fuel were favored and in place before spark required it...if that makes sense...
1992 LX - 25.1c Chassis, Vortech Blown Dart 332, Lentech Trans, TRZ Backhalf, A9L, Moates QH/SL v1.9, BE, EA, TunerView

User avatar
skunk
Tuning Addict
Posts: 391
Joined: Sat Jun 25, 2011 2:30 pm
Location: Newmarket, ON

Re: Load calculation/RPM reporting?

Post by skunk » Thu Aug 02, 2018 1:12 pm

My translation of that is it allows more utilization of the cells in the OL fuel table.
Agreed.
, if actual load ranges from say 1000RPM/70% then goes up to 4000RPM/150%, then there'd be cells above 70% for the low RPMs that are unattainable
Not necessarily..... Sealevel Load Scaling allows you to determine the fuel commanded from the Fuel Table. If you want more fuel at a given rpm in OL you lower the sealevel scale..... Want less...then increase it. For example let's say at 1300 rpm I want to add fuel for lugging OL say at 40 load, ( I am making up numbers here) using Load=load% might keep me at 14.64 AFR. Utilizing load% and sealevel scaling allows me to command from the top of the fuel table say 11 afr if I so desire. Move now peak torque rpm.....I can choose like stock to keep things leaner longer setting sealevel scaling higher keeps me lower and leaner on the fuel table. This essentially makes all levels of the fuel table attainable.

I don't know that it actually allows for more resolution since it relies on calculating load% instead of using a look up table. I have no idea what's faster. My thoughts are it had more to do with space constraints.

John
1987 Turbocoupe.
A3M1/A9L, BE/EA, Tweecer RT.
306,Single Turbo, Ported E7's, Ported Stock Intake, 42lb Matched Green Tops, PMAS 3" Blow-Thru Protube (42# supercharger calibrated),T5z, 3.73 gears
..... Nothin Fancy.......Just something to keep me from the honey-do list.........

User avatar
skunk
Tuning Addict
Posts: 391
Joined: Sat Jun 25, 2011 2:30 pm
Location: Newmarket, ON

Re: Load calculation/RPM reporting?

Post by skunk » Thu Aug 02, 2018 1:31 pm

Sorry Chris....I posted without comprehending what you are saying. Lol....work got me distracted.

Yes I believe you surmised the point of load%.

John
1987 Turbocoupe.
A3M1/A9L, BE/EA, Tweecer RT.
306,Single Turbo, Ported E7's, Ported Stock Intake, 42lb Matched Green Tops, PMAS 3" Blow-Thru Protube (42# supercharger calibrated),T5z, 3.73 gears
..... Nothin Fancy.......Just something to keep me from the honey-do list.........

wickedsnake00
Gear Head
Posts: 22
Joined: Sat Apr 22, 2017 3:12 pm

Re: Load calculation/RPM reporting?

Post by wickedsnake00 » Thu Aug 02, 2018 2:36 pm

EDS50 wrote:
Thu Aug 02, 2018 10:04 am
cgrey8 wrote:
Thu Aug 02, 2018 7:40 am
There are often two "Load" values that tunes have used and differentiated between. In the older EECs, they are referred to as Load and Load%. In newer EECs, the Load% I think is called PerLoad.

But basically Load is as you described however they also take into account the declared CID of the engine. And I honestly don't know the difference between Load and VE. They seem like the same thing to me. And Load is generally used by the Spark equations.

However Load% is a scaled load relative to the engine's max possible load...at least that's what it is supposed to represent when the scaling function is setup correctly for your application. In GUFx defs (and I'd assume in others), there's a function that accepts RPM & Load as inputs and outputs Load%. When this function is setup correctly, WOT at all RPMs should result in Load% being very near 100%. Why that's useful, I honestly don't know which is why I setup my tune to make Load% = Load and just tune that way to simplify my life.

As for your stalling issue, I don't have a suggestion there. Looking in your log, I didn't HEGOs or WB being logged to know if the drop is due to a sudden rich or lean condition. But that'd be what I'd want to know.
I apologize for the thread jacking but I am also a huge supporter of setting prldsw to load%=load. My question to you is do you also set your load values for fn072a to the same values as fn071 or do you leave fn072a at default value? Just for reference purpose I set my fn072a to match fn071.


As for the OP... Looking at your log your your idle air integrator shits itself @ 120 sec but your afr stays pretty consistent. Timing seems to start moving around after the integrator gets out of range. My guess at this would be you need to add more air to your startup isc multiplier or look into your iac going bad.
Regarding load at least the one I'm logging with Livelink is the VE type, not load%. I know it goes into the 1.XX range when in boost.

I do have the wideband being logged under "Analog 1." I don't have the exact HEGO voltage values but I do have the STFT's being logged.

Regarding AFR you can see it's running nice and smooth right around 14, and goes rich after the engine dumps a bunch of fuel either in response to or causing that big dip around 125s. That's what has me scratching my head.

Now for the idle integrator, are you referring to it going negative at that range? I've dialed in the ISC tables and it's happy as can be hot, zero integrator. I've tried dialing in the ISC multiplier and have it turned up pretty well for colder temperatures, but regardless of what the value is, it always seems to do this. I started with it really low when it was stuck around +0.40 and kept creeping it up, here it's actually negative. The giant swing in it seems to be to try and save the giant RPM drop, which is also after the seemingly random spike in load and fueling.

User avatar
EDS50
Administrator
Posts: 3652
Joined: Sun Apr 05, 2009 9:17 am
Location: Tampa Bay, FL
Contact:

Re: Load calculation/RPM reporting?

Post by EDS50 » Thu Aug 02, 2018 3:26 pm

If you are happy with the isc function then I would start looking into maybe the timing starts swinging around causing everything to go out of whack. You definitely have something distinct going on which shows in the log you provided. We just need to determine what is the root cause to effect everything else. Does the transition from open loop to closed loop happen at this time interval?
1992 LX - 25.1c Chassis, Vortech Blown Dart 332, Lentech Trans, TRZ Backhalf, A9L, Moates QH/SL v1.9, BE, EA, TunerView

wickedsnake00
Gear Head
Posts: 22
Joined: Sat Apr 22, 2017 3:12 pm

Re: Load calculation/RPM reporting?

Post by wickedsnake00 » Thu Aug 02, 2018 8:48 pm

EDS50 wrote:
Thu Aug 02, 2018 3:26 pm
If you are happy with the isc function then I would start looking into maybe the timing starts swinging around causing everything to go out of whack. You definitely have something distinct going on which shows in the log you provided. We just need to determine what is the root cause to effect everything else. Does the transition from open loop to closed loop happen at this time interval?
That was one of my first thoughts about the switch to closed loop, but according to the flag it happens around 10-15 seconds after start.

Spark was pretty powerful. It used to surge, but I turned the idle spark gain down from 1.1 to 0.75 and RPM smoothed right out. I didn't log the spark source on this one, but on another log it looks like it's holding steady at 9 or 8, but when RPM has already dropped it's getting clipped by 1 or 2. It doesn't seem to ramp way up until RPM starts dropping, like fueling trying to save it.

The only thing out of the ordinary that seems to precede the RPM drop rather than react to it is the load/injector pulsewidth. Which really leaves me scratching my head given airflow and RPM don't appear to change.

wickedsnake00
Gear Head
Posts: 22
Joined: Sat Apr 22, 2017 3:12 pm

Re: Load calculation/RPM reporting?

Post by wickedsnake00 » Tue Aug 07, 2018 2:03 pm

Any other ideas on how airflow is calculated based on the transfer function? Does the EEC use simple linear interpolation, or is there more involved? I'm wondering if there's some sort of PID in the background that could cause issue.

And for the recovery, it seems like the engine dumps in a lot of fuel very quickly, but the IAC is slow to react to the RPM drop. Is there any way to tune that, or is that just normal behavior?

wickedsnake00
Gear Head
Posts: 22
Joined: Sat Apr 22, 2017 3:12 pm

Re: Load calculation/RPM reporting?

Post by wickedsnake00 » Wed Aug 22, 2018 12:34 pm

Car's been down with some other work so I haven't had a chance to mess with it much.

Here's a list of what I know it isn't:

Injector values - used latest 60lb/hr file

ISC neutral idle air - dialed in while hot, matches stock 03 Cobra values, same as the hardware I'm using.

ISC transfer function - using 03 Cobra values, same as the hardware I'm using

TB bypass air - calculated and populated at hot stable idle

ISC multiplier - dialed in, but does not affect stalling.

Cold start enleanment table - always occurs after this table has expired

Cold fuel table - always occurs after transition to base fuel table. Leaner values do not help.

Base fuel table - AFR matches the table, lambda of 1 in this region.

Fuel injector timing - optimized timing with cams. Runs smoother, but no impact on stall.

Idle spark gain - Big difference in idle stability, but no impact on stall.

Spark source - idle spark source is in control until RPM falls off.

Fuel source - 0, occurs during closed loop, not when transitioning states.

Coolant temp - does not occur at specific coolant temp, but only occurs before it's warmed up.

Voltage - voltage is a steady 13.7 when this occurs.

Fuel pressure - fuel pressure (drop) is a stable 39psi when this occurs

Idle speed - tried varying this in RPM and time. No impact.

MAF transfer/location - verified with a hot idle no load sweep. Also it doesn't occur at one specific spot.


So what has worked? On a whim I intentionally made the MAF transfer 15% lean in that region, and forced open loop to see what would happen. No stall. That would mean the variables of AFR and open/closed loop are in play.

I looked back in some very old log files to try and see if there was an issue with fuel pressure when the stall/RPM drop occurred. There wasn't. But I knew the file that old was still kind of lean in that region, and it still happened. That makes me think it may be more closed loop related. Next step will be taking it as is and forcing open loop to see if it still occurs. If it doesn't, well, we actually have something to go on. If not? Well, I'm stumped.

wickedsnake00
Gear Head
Posts: 22
Joined: Sat Apr 22, 2017 3:12 pm

Re: Load calculation/RPM reporting?

Post by wickedsnake00 » Sat Aug 25, 2018 4:43 pm

Tried some back to back tests.

Open loop, accurate MAF curve - Ran rich, was a little better at saving itself, but the same result.

Closed loop, accurate MAF curve, Ran richer and worse than before. Stalled.

For a sanity check I reloaded the lean open loop curve. Ran rich and stalled, exactly the same as the others.

Ran with the evap/BOV systems disconnected and that port plugged. Same result.

Nothing makes sense.

I'm going to try smoke testing it and throwing a new set of plugs in it.

User avatar
EDS50
Administrator
Posts: 3652
Joined: Sun Apr 05, 2009 9:17 am
Location: Tampa Bay, FL
Contact:

Re: Load calculation/RPM reporting?

Post by EDS50 » Sat Aug 25, 2018 4:47 pm

Are you clearing the learned fuel trims before experimenting with the different tunes?
1992 LX - 25.1c Chassis, Vortech Blown Dart 332, Lentech Trans, TRZ Backhalf, A9L, Moates QH/SL v1.9, BE, EA, TunerView

Post Reply

Who is online

Users browsing this forum: No registered users and 26 guests