SOHC V6 tuning

[RangerOZ]

The other error potential is that even if the stock MAF sensor has the capability of outputting a voltage greater than 5 or 6 volts without performing a bench flow test the transfer function is not known. Most bench flow tests are terminated at slightly above 5 volts output. Years ago I attached my stock MAF sensor to my electric leaf blower for some testing. With the blower on high speed (120 mph) the MAF sensor output voltage was 5.25 volts. I suspect the MAF sensor sensitivity decreases rapidly above 5.0 volts. The corresponding voltage output for my LMAF 90 was 3.2 volts. According to SCT's value file for the LMAF 90, 3.2 volts is equivalent to about 19 lbs/min.

On Ford MAFs, which are manufactured by Hitachi, the MAF will output voltage to battery voltage. However, the ECU is set up to only acknowledge a maximum of about 4.7 volts.
from
http://www.lasotaracing.com/techsctmaf.html

So do I have an Hitachi ?
Now I have to find out how to check.
Is the 4.7 magic and I should stay under 4.7 with or with out scaling?


https://www.diablosport.com/products/mafia.html

 

[2000StreetRod]

PCM max input vs MAF output

My stock 55 mm MAF sensor that output 5.3 volts with the leaf blower has an XF2F-12B579-AA part number. Since the PCM does not extrapolate if the MAF sensor output voltage exceeds the max in the MTF table the PCM uses the top entry of lbs/min. That will result in a lean condition since there is more air flow than what the PCM is using. Don LaSota's comment means that increasing the max value in the MTF table will increase the range of the MAF sensor up to 5.0 volts. Your stock MTF table had the same air flow for 4.7 and 5.0 volts. Your modified table has a greater value for 5.0 volts than 4.7 volts. Your PCM will use the larger air flow associated with the limit of 5.0 volts so that's not a problem. As long as your datalogged MAF AD count is less than 1024 your MAF sensor has not "pegged". However, the air flow in the MTF table associated with 1024/5.0 volts may not be accurate. That will affect load and your AFR. Even if you had bench test air flow data for 1024/5.0 volts and entered that in your MTF table the AFR still might not be correct. That's why you need to measure it with a wideband AFR meter.

You can change the AFR by adjusting the slopes for your injectors which will not "corrupt" the load calculation. However, that is difficult and takes a lot more time. I was not able to achieve my desired AFR using injector slopes with my 60 lbs/hr injectors. I was forced to corrupt (deviate from bench test flow data) my MTF table to achieve the desired AFR. Corrupting the MTF table corrupts the load calculation so all tables with a load index have to be modified.

One approach is to change the engine displacement to make the FI load comparable to the NA load. This works best with positive displacement blowers (M90) because boost is produced over a broad RPM range. It would not work as well with turbochargers or centrifugal superchargers. Since changing load affects spark advance care is needed to avoid engine damage. I changed my engine displacement to decrease load and forgot to put back in my stock spark table. I had massive WOT detonation due to large spark advance with the low load. I immediately recognized the detonation and aborted the "pull" in time to avoid damage.

 

[2000StreetRod]

WOT upshift criteria?

I'm attempting to determine what the criteria is for upshifting when the accelerator is at WOT. I calculated the N/V ratio (engine revs/vehicle speed) to be 43.4 and changed it in a test tune from the stock value of 43 to 43.5 (closest I could set).

My vehicle speed is computed by the 4WABS module and I've noticed that the datalog value is about 10% different than my computed value based on engine speed, gearing and tire size. So I changed the Max Learned N/V from 1.25 to 1.01 and the Min Learned N/V from 0.7 to 0.99 (0.9899) to increase the accuracy. I also decreased the Trans Shift Schedule 12 to 35 mph at TP = 400 and greater.

Edit: I then went for a short drive and data logged the upshift. The transmission upshifted at 33.5 mph (instead of 35) and 4255 rpm instead of the Trans WOT Shift RPM 12 of 6150.

The larger rpm may be due to the torque converter not being locked. The smaller vehicle speed may be due to the 4WABS being inaccurate.

 

[2000StreetRod]

TP for WOT

My problem is that during earlier dyno testing upshifts occurred early in both RPM and mph.

Some have determined for their particular PCM strategy upshift is more accurate when relying on the Trans Shift Schedule mph value and setting the Trans WOT Shift RPM higher than the corresponding mph. Others rely on the Trans WOT Shift RPM and set the corresponding Trans Shift Schedule mph higher. General speculation is that the PCM will upshift whenever one of the criteria is satisfied. My simple test described in the previous post confirms that.

To make it easier for testing and to avoid tickets I reduced the TP for WOT from 700 to 400 for all engine speeds.

I also decreased the Trans WOT Shift 12 from 6150 to 3500 rpm which is less than the PCM computed engine speed (3767 rpm) for the 35 mph Trans Shift Schedule 12 speed. Based on past experience I predict the upshift will occur when the engine speed is about 3450 rpm.

 

[2000StreetRod]

surprising results

. . . Based on past experience I predict the upshift will occur when the engine speed is about 3450 rpm.

Well, I'm not going to make any more predictions because the test results were quite unusual. I kept the throttle at WOT (TP = 755) in first gear until I thought the upshift occurred at 3378 rpm and 36 mph. I think the drop in engine speed at time = 13 seconds was due to the torque converter changing ratios (decreasing) faster than the engine speed was increasing.

The surprising thing was when I left of the throttle the engine speed increased to 4100 rpm (44 mph) and stayed there until an obvious upshift at 47 mph. It was as if the PCM got confused about which gear to be in (or my transmission slipped excessively).

Anyway, I've proven that the upshift was based on engine speed (but 122 rpm less than specified) instead of mph.

 

[4pointslow]

Datalog

Did you datalog desired and commanded gear?
I am not quite sure what you mean about the torque converter changing ratios, they are either locked or not locked.
Did you change the TCC lock schedule along with the shift schedule?
Upshifts with downshifts making sure they don't cross?
I thought there was a "switch" in Advantage3 that could turn the trans from RPM to VSS shifting too.

Curious about this topic, that's why all the questions. There seems to be more engine performance tuning information available than trans tuning.

 

[2000StreetRod]

torque converter ratio

Did you datalog desired and commanded gear?

No, at WOT I didn't anticipate anything other than an upshift from 1st to 2nd.

I am not quite sure what you mean about the torque converter changing ratios, they are either locked or not locked.

Here's a fairly good explanation of how a torque converter works: Torque Converter
I copied the graph below from the website.

I couldn't find any specifications on my torque converter but I suspect it has comparable characteristics to the example. The graph explains why the slope of the plot of WOT engine speed vs vehicle speed decreases around 3800 rpm, levels off and then increases even though vehicle speed continues to increase (3rd speed shown).

Did you change the TCC lock schedule along with the shift schedule? Upshifts with downshifts making sure they don't cross?

There is no capability to lock the torque converter is 1st gear in my strategy. Only 2nd thru 5th speeds.

I thought there was a "switch" in Advantage3 that could turn the trans from RPM to VSS shifting too.

Not that I have found in my strategy.

Curious about this topic, that's why all the questions. There seems to be more engine performance tuning information available than trans tuning.

I agree. The Mustang owners have spent a lot of time trying to get their upshift points where they want them. I've about decided to use rpm instead of mph and just increase it about 150 rpm greater than what I want. I'll make sure that the corresponding vehicle speed for each gear is considerably greater (except for the 4 to 5 shift because max vehicle speed is only 127.5 mph).

 

[2000StreetRod]

Spark advance .5 degrees high

At 14.5 seconds:
IAT = 100 deg
ECT = 192 deg
rpm = 3800
load = .93
spark source = borderline knock table
knock sensor retard = -.25 deg
spark = 12.75 deg

According to my borderline knock table spark advance should be 14.0 deg
The correction for IAT should be -1.6 deg
The correction for ECT should be 0
So the corrected spark should be 14.0 - (1.6 + .25) = 12.15 deg
The LiveLink resolution for spark is .25 deg so the value should be 12.25 deg

.5 degrees of unplanned advance could be significant if my spark tuning gets aggressive.

 

[4pointslow]

Torque converter

I see what you mean now,
I had to go and look at some of my datalogs real close. Around 3,300 RPM's you can see the increase in rpm start to slow as the torque converter seems to "catch" (Torque ratio changing).
I guess it just looks a lot different when you have the shifts happening a few rpms after it. Makes it look like two shifts, or the TCC functioning. If you had not posted it I would have thought it was two shifts.
I will have to look at a few 5.0 trucks that have lower stall speeds, just for comparison.

Our torque converters are supposed to be 2900 rpm stall speeds, the vehicle moving would account for the RPM being around 3300 instead of 2900 when it "catches".

 

[4pointslow]

Switch

I thought this was the switch, but last time I tested it the results said otherwise.
I may have loaded the wrong tune though. I will have to retest again when the temperatures outside cool down some.

 

[2000StreetRod]

turned on

My first vehicle was a 1950 Olds with fluid coupling - no torque converting or lock capability but was indestructible.

I have that switch set to 1 (same as stock) because I thought it applied to normal driving and that the Trans WOT Shift RPM value would apply during WOT shifting.

What are you using to control your upshift points during WOT and how accurate are the upshifts?

 

[4pointslow]

Shifts

I am using MPH to shift. How accurate it is I couldn't tell you.
It started for me when I thought I was hitting the rev limiter but couldn't figure out how because It was set to 6250 and I was experiencing it at 6050. So I was informed to turn off rpm based shifting and try mph shifting. I found that switch and turned it off (set to 0).
In the Ultimate EFI Tuning Guide by Don Lasota, he states that cruise shift points are determined by MPH and TP position. He goes on to say that WOT shifts have both MPH and RPM shift points. So I was thinking that when you turn off RPM anticipated shifts that it was only for WOT (or what the pcm see's as WOT. (70% or more throttle in a stock tune)).

Now in my stock tune from SCT, the MPH shift points did not look like a reality at all. They had some shift points at 7,000 something rpm's. I can't help but wonder if that was to keep the transmission shifting off RPM only at WOT.
So after I turned the rpm anticipated shifts off in my custom tune I then changed the shift points to a much lower number to make it shift way too soon. It worked and I was getting shifts at 4,000 something. So then I just raised the mph until I got the shifts I wanted.

Keep in mind I was trying to lower the rpms when it shifted to stop what I thought was the rev limiter. Later I found out it was spark blowout from a large spark plug gap and high boost/compression/turbulence. I lowered my gap and have been able to creep the shift points back up. I still have not finished tuning it completely but hope to this year.

At some point I had adjusted the N/V number and the shift MPH numbers resembled something believable. I can not remember when I figured that out. Keep in mind that even though you set a shift point to a certain MPH or RPM, that those are only trigger points and the shift takes time(filling clutches) so you end up having to set the point earlier than when you really want the shift to occur.

Also I have to add that since I had changed the mph shift points at the same time I turned that switch off(RPM anticipated shifts ) I just assumed it had worked. Now I am not so sure and would have to do some more testing to find out.

 

[2000StreetRod]

WOT upshifts: stock, Henson & mine

Trans Shift
Schedule . . . Stock . Henson .. Mine
. 1 > 2 . . . 66/7021 55/5851 66/7103
. 2 > 3 . . . 78/6221 73/5822 78/6293
. 3 > 4 . . . 91/5768 93/5895 101/6476
. 4 > 5 .. 115/4945 125/5375 127.5/5546

Trans WOT
Shift . . . . . . Stock . Henson .. Mine Desired
. 1 > 2 . . .. . 5500 .. 5850 . . . 6150 6000
. 2 > 3 . . .. . 5500 .. 5800 . . . 6050 5900
. 3 > 4 . . .. . 5800 .. 5900 . . . 6350 6200
. 4 > 5 . . .. . 5800 .. 5850 . . . 6150 6000

It is obvious from the above that the stock strategy was using rpm for the upshifts except for 3 > 4. However, James was using mph except for 2 > 3. I've decided to use rpm because the PCM computed mph vs rpm is not accurate for my gearing and tire radius. My desired shift points are based on when the product of the gear ratio times the rwtq is less for the existing gear compared to the upshifted gear. I've added 150 rpm to the desired because that's approximately how early the upshift occurs. Obviously, I will never reach the desired 4 > 5 upshift because the greatest vehicle speed (127.5 mph) will always occur before the upshift rpm. The PCM number of bits for vehicle speed limit the value to 127.5. I noticed that on one of my dyno runs. I don't know how the Mustang owners get around this if any go that fast. Anyway, at some point I'll run thru the shifts on the dyno to determine when they actually occur and adjust the added 150 rpm up or down as appropriate.

On my first dyno run with my own custom tune I thought I hit detonation in 4th due to excess spark advance after changing the displacement which changed the calculated load. I realized a few days ago that what actually happened was I hit the vehicle speed limiter (103 mph) which I had not defeated from the stock tune.

Do you leave rpm anticipated shifts turned off for normal driving? If so, does it have a noticeable affect? Leaving mine on may be why my upshifts are always early.

 

[4pointslow]

rpm anticipated shifts

Edit: The switch for RPM anticipated shifting does not make the car shift off mph only. It is only for allowing the PCM to see a rapid rise in rpm as an excuse to start the shift sooner to prevent the rpm from going to high before the shift actually happens.

I have left it off for a couple of years now.
I thought it was making the transmission shift off of MPH only, but I don't know anymore. After thinking about how at WOT the transmission will shift off of MPH or RPM (which ever it hits first), I remembered how I dropped the MPH lower than it was to get the trans to shift sooner. So what I thought was the switch working was probably just the MPH being set lower.

None of the MPH numbers ever made any sense until after I switched from 4.10's to 4.88's. That's when I realized I had to change the N/V.
I remember now that it was then that I changed the N/V number and the MPH and shift RPM's started to look believable. I also had to change the final drive ratio under Axle in my tune. My final drive ratio doesn't effect the speedometer on my truck since the rear differential speed sensor goes to the ABS module and then to every other module in the truck. So my speedometer is calibrated in my ABS module (I can do it with the NGS scan tool). I never had to calibrate it though, since I never changed the tire size or tone ring on the differential carrier.

After changing N/V and final drive ratio I scaled the shift MPH to make up for the gear change. Then through trial and error I have been constantly adjusting the MPH numbers to achieve the shifts I wanted. Just recently I noticed that first gear was hanging way too long under light throttle so I changed the shift point to make it happen a little earlier. I also softened up the shift from 1-2 under light load because it was banging with the 4.88's in there.

 

[2000StreetRod]

Performance factor?

One performance factor is horsepower - one horsepower is the power needed to move 550 pounds (249kg) one foot in one second. I think it was defined by James Watts in the late 1700s when comparing his steam engines to draft horses. Another factor is torque (force x lever arm). A dynamometer measures torque and then computes horsepower: HP = TORQUE x RPM ÷ 5252. It is inconvenient (and a little expensive) to strap my vehicle to a dyno any time I want to evaluate a potential performance improvement so for some time I've been looking for an alternative. Yesterday I realized that the PCM must compute acceleration in order to determine shift points. This morning I searched the available pids and found Vehicle Acceleration Rate. I added it to my datalog configuration file and went for a drive. I am pleased to report that the pid works and is computed and reported frequently. The units appear to be in vehicle speed (mph) per second.

red = vehicle acceleration rate
blue = vehicle speed (0 to 53.5)
green = engine rpm (950 to 3710)
purple = relative throttle position (35 to 319)

I determined the change in vehicle speed in a one second interval and it is comparable to the acceleration value displayed. Note that the acceleration starts decreasing as the rpm increases from 3240 to 3710 just before the 1 > 2 upshift. The next time my Sport is on the dyno I plan to record the acceleration in each gear to evaluate my other method of optimizing my shift points.

Note: The vehicle acceleration rate is not accurate if the tires break loose from the pavement.

 

[CDW6212R]

One performance factor is horsepower - one horsepower is the power needed to move 550 pounds (249kg) one foot in one second. I think it was defined by James Watts in the late 1700s when comparing his steam engines to draft horses. Another factor is torque (force x lever arm). A dynamometer measures torque and then computes horsepower: HP = TORQUE x RPM ÷ 5252. It is inconvenient (and a little expensive) to strap my vehicle to a dyno any time I want to evaluate a potential performance improvement so for some time I've been looking for an alternative. Yesterday I realized that the PCM must compute acceleration in order to determine shift points. This morning I searched the available pids and found Vehicle Acceleration Rate. I added it to my datalog configuration file and went for a drive. I am pleased to report that the pid works and is computed and reported frequently. The units appear to be in vehicle speed (mph) per second.
View attachment 85546
red = vehicle acceleration rate
blue = vehicle speed (0 to 53.5)
green = engine rpm (950 to 3710)
purple = relative throttle position (35 to 319)

I determined the change in vehicle speed in a one second interval and it is comparable to the acceleration value displayed. Note that the acceleration starts decreasing as the rpm increases from 3240 to 3710 just before the 1 > 2 upshift. The next time my Sport is on the dyno I plan to record the acceleration in each gear to evaluate my other method of optimizing my shift points.

Note: The vehicle acceleration rate is not accurate if the tires break loose from the pavement.

Excellent post Dale, that discovery may be one of the most valuable tools there is for computer tuning.

Most people do not realize that HP is not the goal, acceleration is. Most people think that a dyno result is the ultimate test, it's not, a timed acceleration test is the best. That is why serious performance people do dyno testing, and then fine tune the vehicle at a track with acceleration tests.

If that built into the PCM ability is valid, that can be so much more valuable to the tuner and owner than any dyno. You can adjust parameters to obtain the quickest acceleration results, and then you may be very close to results from tuning at a track. That would be awesome.

 

[4pointslow]

Vehicle acceleration rate

Here is what mine looks like at WOT.
My 3-4 shift really slows down the acceleration.

 

[2000StreetRod]

excellent values!

John, thanks for the plot. Obviously, you have been logging acceleration long before I discovered it. Those are excellent values reflecting your large boost and axle ratio advantage. Your 1 > 2 shift point looks good since there is no drop (early upshift) or increase (late upshift) in acceleration. The 2 > 3 shift looks early since there is a noticeable drop. With a fairly flat torque curve the wider the gear spacing the higher the needed upshift point to maintain overall torque (acceleration).

 

[4pointslow]

Vehicle acceleration rate

I have been logging it for a little while, but not really using it.
A member on here mentioned it in a post and I thought it would be a great way to try to fine tune shift points. I think it was Jakee.

The drop in acceleration rate on the 3-4 shift was compounded by a little detonation which led to timing being pulled. That along with timing already being pulled to protect the transmission during shifting 3-4.

The 2-3 shift was at 6188 RPM, don't know if it is safe to go higher in this stock engine. lol.
I also question the accuracy of the vehicle acceleration rate pid(and others) with 20 "PIDs" being datalogged, there is definitely a resolution concern when data logging lots of sensors etc.
For what ever it is worth, I also recall the distance traveled only changing in 1/4 mile increments, it was when I was trying to compare the time from a 1/4 mile time slip to the datalog from the same run. There is some inaccuracy there.
The PCM gets more information on distance traveled than it shows in live link though, since the vss signal has a high resolution so maybe the veh acc rate is more accurate then I previously thought.

 

[2000StreetRod]

tip-in spark retard eliminated!

Since installing my M90 I have been plagued with unwanted tip-in spark retard. I prefer to have spark advance according to the spark tables based on calculated load and engine speed and then corrected for ACT and ECT. Yesterday, I loaded ten 40 lb bags of topsoil in the back and then did a datalog of acceleration on the way home. Last night while looking thru the log I noted that even at light throttle acceleration (with the added load) there was significant tip in spark retard. So far I've set the
Min Vehicle Speed for Tip-in Torque Control to 127
Amount to Reduce Spark for Tip-in Control to 0
Max Spark Retard for Tip-in Based on ACT to 0
Max Spark Retard for Tip-in Based on RPM to 0
Tip-in Spark Retard for Change in Air Mass to 0

Spark source = 3 (torque control)
Torque source = 5 (tip-in shock control)
Relative TP = 55
RPM = 830
Load = .42

Borderline knock table = 16
ACT correction = -1.6
ECT correction = 0
Knock sensor retard = 0
Overall spark should be 14.4 instead of -2.0

So this morning I decided to change the Threshold to Identify Tip-in from the stock value of .000065 to the maximum value I can enter (.0155) making it impossible for the PCM to recognize for spark that tip-in has occurred. The results (still with bags of topsoil) are very encouraging.

Spark source stays = 16 (borderline knock table)
Spark stays about = 12.5
Torque source stays = 0 (torque control off)
Relative TP starts at 40 increasing to 756
RPM starts at 880 increasing to 6172 (1 > 2 upshift)
Load starts at .36 increasing to .92

Borderline knock table for 3500 rpm = 14
ACT correction = -1.6
ECT correction = 0
Knock sensor retard = 0
Overall spark should be 12.4 compared to actual of 12.5 which is correct since the LiveLink recording increment is 0.25

I was concerned that disabling tip-in for spark may impact tip-in for fuel but it appears the transition is still the same as what I had previously achieved (about .2 seconds).

I was surprised at the 1 > 2 upshift rpm of 6172 and 53.5 mph vs the set values of 6150 and 66 mph. Now PCM is shifting a little late instead of 150 rpm early. The acceleration vs rpm verifies that the shift is late because the acceleration value increases slightly after the upshift.

I'll change the rpm upshift values back to the desired instead of the desired + 150.