TRAXION

P730: Part Throttle Fuel Tuning (no Power Enrichment)

Questions:
How is the fuel curve represented in the $8D calibration of the 1227730 ECM?
What are the areas (addresses) in the PROM that are used for the fuel curve?
What is meant by no PE (Power Enrichment)?
When should the fuel curve be modified?
How is the fuel curve tuned?

Answers:
Subsequent posts.

Comments:
In this discussion more questions will probably arise as compared to the ones that I have listed above. However, these are the first questions that have ‘come to mind’ when thinking of this topic. As more questions arise I will modify this post and paste the new questions into this original post. However - this post will only consist of subjects relating to Fuel Curve tuning for the 730 ECM while not in PE (not WOT).

Tim


------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

TRAXION

The fuel curve in the $8D calibration is represented in terms of Volumetric Efficiency. Think of the gasoline engine as a big air pump. It is pulling air in, doing something with it, and then exhausting the air. Air in – Air out. The more air the engine pulls in, the more fuel that is needed. Thus, we can talk about how ‘efficient’ the engine is at being an air pump. We can discuss how efficient our engine is at pulling air in and exhausting air out. The more efficient the engine is at being an air pump, the more fuel that is needed to keep a given Air/Fuel ratio intact. The $8D calibration has ‘Volumetric Efficiency’ tables (VE). These tables are three dimensional (3D) tables. They are separated into an upper VE table and a lower VE table. Both VE tables show how efficient the engine is at a given RPM and MAP. RPMs are the revolutions per minute the engine makes and MAP is the amount of vacuum the engine is pulling. I mentioned that the tables are 3D tables. They are according to VE vs. RPMs vs. MAP. The lower VE table extends from 400RPMs to 1600RPMs in 100RPM increments. The upper VE table extends from 1600 to 5600RPMs in 400RPM increments. The MAP readings for both tables are in kPa (kilopascal). kPa is a metric vacuum measurement whereas inches of Mercury would be a similar English measurement. One of the first questions about these tables is with regard to MAP. Why MAP? Everyone usually catches on to RPMs. But, what does MAP truly represent and why is it used in the VE tables? MAP signifies engine load. In most cases, the greater the engine load the higher the kPa value. Think about it this way. When you car is idling it is pulling a vacuum inside the plenum. As you open the throttle blades you are providing an inlet that will decrease the effective vacuum. Here is the often tricky part … LESS vacuum is represented by a HIGHER kPa value. Keep that in mind. Lets say we are driving uphill. You would consider driving uphill to be more of a load on the engine. If you are driving uphill you probably have your foot more on the gas. If you have the gas pedal depressed more then that means that the throttle blades are opened more. The throttle blades being opened more will result in LESS vacuum and HIGHER kPa value. So, MAP equals load where increasing load is represented by higher kPa values. When put in context of the VE tables, we can see that as the kPa values increase from left to right, this represents increasing load on the engine. When taken to the extreme we can view the highest kPa reading (100kPa) to be equal to Wide Open Throttle (WOT). WOT will put the highest load on the engine. 100kPa is atmospheric pressure ... no vacuum at all.

In conclusion, to make a long story short, if you raise a given value in one of the VE tables you will be increasing the amount of fuel the engine receives at that given RPM and that given MAP.

One question down. 4 more to go.

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

[This message has been edited by TRAXION (edited February 19, 2001).]

TRAXION

As mentioned previously, there are 2 tables: Lower VE table and Upper VE table.
The lower VE table begins at 6EE (hex address in the BIN). It has 13 columns and 13 rows. The end address is Hex 796. The Upper VE table begins almost directly after the Lower VE table. The Upper VE table begins at Hex 79A. It has 9 columns and 11 rows. The end address is Hex 7FC. How did I know this?

Go here:
http://www.geocities.com/ecmguy.geo/tuning/do_tune.html

Right-Click on the ‘$8D (1227727)’ link in the left pane and select ‘Save Target As…’. You will be presented with a dialog box that will allow you to save a file to your disk. This is the $8D HAC. It is entitled ‘anht_hac.pdf’. It is a PDF document and as such you will need Adobe Acrobat Reader in order to read this file. Adobe Acrobat Reader is free and you can obtain it from Adobe’s site. The $8D HAC is a complete breakdown of the code, tables, flags, and constants in the $8D calibration. A HUGE thanks to ECMGuy who made this HAC possible. I have referenced ECMGuy’s site in my article so, hopefully you have already found this HAC.

2 down ... 3 more to go

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

TRAXION

Power Enrichment is a plan in the ECM whereby other specific tables are utilized in the overall calculation for fuel, spark, and other specifics. This often means adding extra spark, extra fuel, not using AIR or EGR, etc for obtaining the most power possible (get it ... 'Power Enrichment'). For most of our discussions, Power Enrichment is Wide Open Throttle (WOT). However, it is important to note at this time that PE is engaged based on various throttle openings. It is not solely engaged at 100% throttle opening. There is actually a table that indicates when PE is entered based on a TPS threshold vs. RPMs. This table indicates the amount of TPS that is needed at particular RPMs in order to engage Power Enrichment mode. The table begins at Hex 603.

For our discussions in this post we will be focusing on tuning the VE tables while NOT in Power Enrichment. This means that this particular discussion is NOT concerned with WOT tuning. We will get to that at a later date.

Furthermore, discussing Power Enrichment and what it entails is a big discussion in and of itself. The answer that I have provided to this particular question is much larger when discussing the specifics of Power Enrichment. However, my goal in this post was not to discuss the full specifics of PE. It was to simply indicate that we will be focusing on tuning the VE tables while NOT in PE.

3 down ... 2 more to go

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

[This message has been edited by TRAXION (edited February 19, 2001).]

TRAXION

Anytime you change anything to the car that will change how efficient the engine is at being an air pump then you should consider changing the VE tables. Anytime that you change how your fuel system supplies fuel you should consider changing the VE tables. Anytime you change the point of combustion (spark) you should consider changing the VE tables.

The VE tables are a function of RPM (as well as load). This means that a small change may affect a particular RPM range (possibly even a very small RPM band!). Lets look at a few examples…..
1) Stock L98 F-Body with a brand new Edelbrock Base, AS&M runners, and ported plenum. This combination will allow the car to be much more efficient at air induction at certain RPMs and Loads (MAPs). Obviously this will have an effect on the VE curve. My experience has shown that this setup likes more fuel in the higher RPM, higher load areas. The L98 often struggles in the higher RPM region because of the 1.48” tubes. Using Large Tube Runners with the usual 1.660” tubes provides almost 21% more internal area! That is a large change and will definitely have an affect on the VE of the engine. Although your stock cam/head combo may not utilize all of this new found area the VE curve will still change. With this increase in area many things happen. I will not even attempt to cover everything that happens by converting to this setup. However, one of the things is the increase in RPM where max torque occurs. This will shift the shape of the VE curve somewhat higher.
2) Stock L98 F-Body with an LT4 ‘HOT’ cam (218/224 w/ 0.525”/0.525” on a 112LSA). We’ll assume that the heads have been modified to fit the LT4 springs and that this setup is running 1.6 rollers. Camshafts have the biggest affect on VE (as compared to other naturally aspirated modifications). The VE curves for this setup will change for several reasons. The obvious reason is that this cam’s higher duration numbers and lower LSA numbers will make this motor less efficient at the lower RPMs but more efficient at the higher RPMs (very general statement for the purposes of explanation). However, one of the biggest effects here is that bigger cams (especially those with tighter LSAs) will pull much less vacuum at idle (higher kPa). Less manifold vacuum means less vacuum to those devices that are connected to manifold vacuum. The fuel pressure regulator is controlled by manifold vacuum. More ‘pull’ (more vacuum, smaller kPa) on the regulator reduces the effective fuel pressure. Less ‘pull’ (less vacuum, higher kPa) at the regulator increases the effective fuel pressure. Thus, if we install a cam that pulls less vacuum at idle then the Fuel Pressure Regulator will see less vacuum and the overall fuel pressure will be greater. This results in a very rich condition at idle. The bigger the cam duration (with LSA and lift remaining the same), the less manifold vacuum at idle. The overall response here is to re-map the lower VE table to supply less fuel at those RPMs and kPa values where the car idles. We will have to decrease those VE values. This often is not too much of a problem once the car goes into closed loop because the ECM can adjust the amount of fuel by looking at the O2 sensor (REMEMBER we are discussing NON-PE mode which means that the car will most often be in Closed Loop). However, the car is in open loop when you first start it and for the amount of time it takes to get into closed loop you will be dumping A LOT of excess fuel into the engine. This was the biggest problem with my TPIS Level V PROM. They did not change the lower VE table even though I told them that I was running a 218/224 cam on a 112LSA . My car idled terribly until it went into closed loop. Once the ECM switched to closed loop the Block Learn Multiplier (BLMs) shot down to 108 and the integrator down to 112. If my cam had been much bigger then the ECM would not have even been able to compensate in closed loop. Remember, the ECM can only adjust the injector pulse width via readings from the O2 sensor when the car is in Closed Loop. Furthermore, it can only adjust the BLMs down to 108. Given everything that I have just said about the idle curve throw on top of that the affect the cam will have on the volumetric efficiency of the engine at higher RPMs and various engine loads. If you haven’t studied or researched cam dynamics and you are considering adding a cam then I suggest reviewing David Vizard’s book ‘Modifying Small Block Chevy Camshafts and valvetrains’. It will tell you more then you ever need to know.

I have only presented two small examples above. The biggest idea in my opinion to come out of this is that you NEED a good understanding of engine theory. You have to be able to understand how a modification could affect the VE curve. If you don’t understand how a cam operates then this will hinder you. Example 2 could be extended for days ….. seriously. I have only mentioned a couple things about #2 but it is so much more complicated than that.

The good stuff …. how to go about changing the VE curve is next.

4 down … 1 to go You guys can join in at any point now ....

Tim


------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

TRAXION

I will tell you how I tune my fuel curve. There will be those who agree and those who disagree and those who agree but have even better ideas. I am very much open to learning new ways of doing things. With that said, here is how I tune my fuel curves….

First I would like to say that I tune the VE tables through use of a Scan tool. I cannot perceive how one would be able to change these tables without a scan tool given their three dimensional nature. So, if someone has any suggestions please post. Anyhow, I always start with the lower VE table. Once you begin tuning the VE tables you will not want to change the car’s parameters. That means the following must now be static: fuel pressure, injectors, injector constant, thermostat, and the car is in an otherwise good state of tune (IAC and TPS have been set via the factory methods, the ignition system is in good condition, etc). The object is to achieve BLM values of 128 across the board at all RPMs and MAP readings. Readings of 128 indicate that the amount of fuel supplied by the injectors for a given RPM and MAP achieves a perfect 14.7 Air/Fuel Ratio. For an ‘ok’ reference on the BLM (Block Learn Multiplier) and INT (Integrator) check out Phil’s article here (http://www.racer-x.f2s.com/proc/blm_int.htm).

I make sure that I hook up Diacom (my scan tool of choice) and start the car. I allow the car to warm up to operating temp and enter Closed loop (using Diacom to register closed loop). I use the record feature of Diacom and record 30 seconds of data at various RPMs with the car in Park. I go from idle in 200RPM increments to 1600RPMs. This will yield 5 to 6 good data points. I record 30 seconds of data at each RPM because it takes about that long for me to get the RPMs rock solid and for the BLM to have adjusted and be rock solid. With this data I begin working on the Lower VE table. The Diacom recording will have RPMs vs MAP vs BLMs. Diacom shows the specific BLM used for each RPM and MAP. So, what do I do with this BLM? Let’s assume at 1000RPMs and 50kPa the stock VE value is 69. Lets also assume that Diacom is showing a BLM value of 108 for the same RPM and kPa. BLMs less than 128 indicate that the car is running rich and the ECM has had to decrease the injector pulse width to compensate. 108/128 = 0.844. We take the stock VE value of 69 and multiply by 0.844 to get 58.2. Change the VE table value from 69 to 58.2. Proceed with the other data points in the recorded Diacom session and adjust the VE table accordingly. This will only yield 5-6 data points. I have heard of others conducting this test using 100RPM increments which can be done and will yield twice the amount of data points. In any case, you have now adjusted a very small portion of your lower VE curve. At this point I will burn a new chip with the new VE values and run this chip in the car. I’ll check the new VE values by doing the same test in Closed Loop. The values should be very close to 128 if not on 128. I then get ready for a new Diacom recording. I again make sure the car is warmed up and in closed loop and record myself driving around town. I get 2-3 sessions of about 2 minutes of driving around. On this session I avoid roads with hills and I avoid pressing the throttle too far down so as to ensure I don’t engage Power Enrichment. With this data I then scan through and have many different RPM vs MAP readings. This will give me many data points for adjusting the lower VE table as well as data for adjusting the upper VE tables (when we get to them). After adjusting VE points in the lower table with this new information I use TunerCat’s graphing functionality and view how the lower VE curve looks. The points that have been adjusted will be obvious. I use these points as a guide for adjusting the other points in the curve to achieve a nice uniform even flowing curve. This may or may not be correct. Later readings from Diacom will show how ‘off’ my guestimates are. With the data collected I then begin working on the upper VE tables. Notice that the lower VE table has 1600RPM data points. The upper VE table also has 1600RPM data points. I transfer the 1600RPM data point values from my lower VE table to my upper VE table. I then use the Diacom data to calculate new values for various upper RPM VE values. Sometimes I use TunerCat’s graphing functionality here to start smoothing points that are obviously off. For example, if I have a new value for 2000RPMs and 40kpa (52%) and a new value for 2000RPMs and 60kPa (55%) and the old 50kPa value at 200RPMs is 65%, then I will adjust this value down to between 52% and 55% (so long as the trend is a slight increase along those kPa values). I then program a new chip with these new VE tables and go collect more data via Diacom. I conduct the same tests as before and see how well the new changes worked. This is a tedious process … collect data, evaluate data, adjust VEs, program new chip, collect data, evaluate data, adjust VEs, program new chip, repeat continuously. This process can take many days and many chips.

At this point in tuning the VE curves you will notice that you don’t have many of the higher kPa values. This is expected. In fact, I specifically stated that I avoid roads with hills. Roads with hills will allow one to tune their upper kPa VE curves. On straight roads the car lunges forward when you step on the gas (RPMs increase quickly). On roads with hills the RPMs don’t increase fast when you press on the gas. Thus, you can press on the gas on a steep hill (increasing your kPa because the throttle blades are opened more) and the RPMs will still be low. Simply put, driving uphill puts more load on the engine at various RPMs. Getting several scan tool recordings on various grades of hills is very useful for getting a handle on the upper kPa curves. However, don’t pick hills that are too steep or else you might have to press on the gas too much and thus, Power Enrichment will be engaged. Use these hill recordings to adjust the VE tables again. These recordings are only meant to get a handle on the upper kPa values. I actually use another method for fine tuning these values. The biggest problem is that you cannot press on the gas pedal too much or else Power Enrichment will be engaged. We don’t want to engage power enrichment. We want to operate solely off of the VE tables. It is extremely difficult to tune the upper kPa values in the VE tables because you have to push the accelerator pedal fairly far in order to get to the higher kPa values. This almost always engages PE. So, my method is disable PE mode in order to tune those higher values. First let me emphasize a warning here. It is very important in my method to conduct as many ‘hill tests’ as I can before proceeding with this step. You want to ‘have a handle’ on the upper kPa curve. The last thing that you would want is to disable PE and then find that you are lean in the upper kPa values of the VE tables. This can cause severe engine knock which can lead to engine damage. With that said, I don’t necessarily disable PE. I change the point at which PE is engaged. In many of the stock BINs PE is engaged anywhere from 50% to 70% throttle depending on RPM. I change these values to 85% across the board. The table that I am referring to is the ‘TPS Threshold for Power Enrichment vs. RPM’. It begins as Hex 603 and is 5 rows and 1 column (2D table). I change all table values to 85%, program a new PROM, and road test. Now it will take 85% throttle in order for PE mode to be engaged. That means that I can now test on the same inclined roads that I did previously and Power Enrichment will not be engaged. This allows me to begin to nail down the upper kPa values. I eventually change the PE tables to 90% and then collect more data, evaluate the data, and modify the BIN. I then change the values to 95% and then collect more data, evaluate the data, and modify the BIN. I then change the values to 99% and then collect more data, evaluate the data, and modify the BIN. You can see that I slowly narrow down the calibration. The one thing that you don’t want is to run lean in the upper kPa values. Patience is definitely a virtue here. When you are done testing for the day don’t forget to set your ‘TPS Threshold for Power Enrichment vs. RPM’ table back to stock settings!

You can easily see that this process is extremely iterative: road test and collect data, evaluate data, modify BIN, program PROM, … repeat. It can take many PROM revisions in order to ‘nail down’ your VE curves. You want to try to get the car to respond with 128 BLMs across the board.

NOTES:
1) Try to tune your VE curves on days of the same outside temperature. Temperature does play a role in VE calculation … especially if you have relocated your MAT. More on this issue in subsequent posts.
2) I have discussed getting your BLMs to 128. However, I didn’t mention the integrator. Why? More to follow.

5 down ... 0 to go (sort of) Any questions yet?

Tim


------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

[This message has been edited by TRAXION (edited February 20, 2001).]

GregWestphal

iTrader: (1)

Bravo! Very informative articles Tim, even though I don't have MAP car.

A couple corrections. The LT4 HOT cam has an exhaust duration of 228 degrees. Not a big change there. In the referenced article on BLM and INT, he made a huge mistake in stating that the INT needs to hit the maximum of 255 (or minimum of zero) before the BLM is changed. According to a couple hacks I've read through, once the INT is more than 5 units different than the BLM, the BLM is changed to bring the INT back towards 128, or at least w/in 5 units of the BLM value. I personally watched this on my scan tool yesterday to verify. In the BLM cell zero at idle in closed loop, I ended up with a BLM of 140 and an INT of around 136-138. Most of the other cells had BLM's around 128-130, and the BLM would actually start going back to 128 in cell zero if I increased the revs some, so I need to fine-tune my first MAF table some...

------------------
Greg Westphal
'87 IROC 305TPI/A4

TRAXION

Greg,

1) Correct on the HOT cam ... I know its 228 ... I must have been thinking about my cam (218/224).
2) Correct again on Phil's article. Notice my odd usage of the word 'ok' when I state where to find the information.

Let me take a moment to expand upon #2 a little bit more. The BLM has a max of 160 and a min of 108 and the INT has a max of 180 and a min of 80 with the stock 730 $8D calibration. As the INT adjusts to compensate for a rich or lean condition it may max out (or min out). If it does max out (or min out) then the BLM will correspond by increasing or decreasing one point. At this time the INT will reset to 128.

Lets give a good example.....
Car is running rich. INT adjusts down to compensate. Adjusting down to 80 does not compensate enough. The BLM drops to 127 and the INT resets to 128. The INT then continues to drop as long as the car is running rich. If it drops to 80 again and the car is still running rich then the BLM drops to 126. The process repeats over and over until the INT is between its max and min value signifying that the AFR is at 14.7 (as long as you have not changed the stoichiometric value for AFR in the PROM; stock value is 14.7).

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

[This message has been edited by TRAXION (edited February 20, 2001).]

[This message has been edited by TRAXION (edited February 20, 2001).]

branz28

Very very good, keep it up...there's very few people that are able to put things in terms that i can understand...

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Problems every other day with the car (probably not a mod, but to me it sure makes a difference)

james_fearn

How would you rate a stock PROM for a stock TPI engine? Did the engineers do a good job of maintaining the "128"?

I imagine the descepancy of needs between stock production engines are fairly common. So how perfect does it need to be? The stock PROM for a given year runs all the TPI engines for that year pretty well. Obviously you can fine tune a PROM for a particular engine and gain better driveability by doing so.

At which point do you reach diminishing returns? Are there diminishing returns? By diminishing returns I mean the more time you put into making the table perfect isn't well spent because the TPI sensoring equipment isn't precise/accurate enough for the new values to translate into meaningful performance gains.

James

------------------
1985 Z-28 305 TPI
221,000+ mi., Original Eng.

Neil

Tim,

I know your talking about the P730, but how different is it to program the 165ECM $32B calibration for part throttle fuel tuning?

Grumpy

I've talked with some of the guys that did the actual calibrations from 86-94. they put a tremendous amount of work in getting the calibration to work well under any condition you can imagine. ie to include driving with a flat tire.

Ever notice how oversized the oem injectors are?. This allows for small range of BL Int values to account for those conditions. On my GN I'm running 55#/hr injectors, I have no intenstion of running at 620 HP. But, I want to have 500ish available at all times. With the way I have things planned, I have enough resolution in my calibration, to twist the Boost from 15 PSI to 28 and have the proper fueling and timing for what I want to do.

Deminishing returns?.
Once you get into really seeing what you can do, it can be very addictive. Gets to be well I did this strategy, but so and so mentioned something else, and lets see where we go trying that. My first EFI attempt took 400 chips to get right. Knowing what I know now that would be probably 70.

Once you see and feel whats going on then you can really let your thoughts fly.

Or help a friend with his car and see how things vary from mask to mask. That is when things get real cool.

leirch

Brendan

Good job Tim.

branz28

James has a good question, could us maf guys use what you said here about the VE tables to tune our own cars?

------------------
89 IROC-Z 350 TPI

-Flowmaster Catback
-Performance Resource Chip
-700R4 (Rebuilt) Too much done to actually list
-K&N Airfilters
-Ported Plenum
-2.77 Gears (not much to brag about but eh, its there)
-MSD 8.5 mm plug wires
-Problems every other day with the car (probably not a mod, but to me it sure makes a difference)

David 91RS/Z28

Trax.....man this is awesome. Very, very informative!! I thought I had my part throttle tuned ok, but after scanning the car again several times the other day, it's still a little off. My question is what exactly is the BLM cell? How do I interpret this on my scanner?



------------------
David Tuschhoff
1991 RS Camaro M5
A SuperRammed, TPIS,
Ross, AFR, Scat, Crane,
CompCams induced 355!
My f-body page!

TRAXION

Guys,

The 165 is a different animal entirely. The 165 has no Volumetric Efficiency tables.

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

Pablo

iTrader: (1)

david, the way blm is stored in memory is by blm cells so that when you start your car or change throttle position(map)/rpm, the puter allready knows where to go in its fueling strategy rather than trying to adjust all over again.

BLM cells are much larger than regular VEcells rpm/kpa, I guess to save memory space? dunno
anyways, they take up the same area as the ve/map chart but only have like 12 cells if i remember correctly

The computer simply remembers what the last blm was while it was in a certain cell so that when it goes back to that cell it automatically moves to that blm
doesnt mean it doesnt change, there is a filtering to it to where the blm will eventually move to the new blm if there is a new blm.. but once out of that cell, the puter remembers what that blm was and will then automatically adjust to that blm once it gets back to that cell

the hack shows you exactly how the blm cell blocks are situated. You can adjust their borders i think. Ive been told that if you go with a bigger cam and a higher idle you wanna raise the bottom blm cell rpm some to include idle etc. I havent looked at it in awhile though.. havent messed with it

james_fearn

I realize the VE is to SD as the MAF sensor it to my MAF car ==> determining the amount of air to correspond to the correct amount of fuel. I'm just exploring the subject and trying to realize what is happening, whether or not it actually pertains to my particular TPI setup or not.


James

------------------
1985 Z-28 305 TPI
221,000+ mi., Original Eng.

FastBroker

Very good expalantion job, Tim. That's pretty much EXACLTY what I have been doing. Too, I believe that you cannot achieve true tune w/o a scantool...

I have found from experience and have read that some 7747 ECU's don't hold the learned BLM's when the engine is shut-off. Thus, a MOJOR reason that you HAVE to tune for 128's or you can have severe open-loop drivability problems becasue the ECU uses AFR ratios based upon the VE tables. I found this was the case in an early 7747 (not sure of the year) and the later 7747 ECU's were updated to hold the BLM's in memory upon shutdown.

88GTASTK

Great info Tim. Keep saving these posts and after a while you'll have a book to publish. One little thing that I can add is that when you get to the upper MAP areas at low RPM's standing on the brakes can get you to those points and give you the time to get good data. On automatic cars there are low RPM high load (MAP) points that you just can't get to because of the stall of the convertor.

Paul

Grim Reaper

Congratulations Tim on the article and work that you did on this first of, hopefully, many technical articles. This is a lot of effort on your part and I just wanted to mention it.

I would like to add that I too believe that for the VE Table that I cannot think of any way of tuning it without a scan tool. The importance of the table should not be underestimated. I found that for certain features like “Highway Mode” to work properly, that a good VE table is a must. Also, when you are in open loop (or drop into open loop due to a long deceleration period or prolonged idle), if this table is not correct you will either be too lean or too rich. Lastly, a good VE table sets you up nicely to get into the “Power Enrichment % Fuel Change”, which is another topic on its own. Also, tuning the MAT Delta Inverse Loop table requires a very good VE table to get that table to correctly adjust the injector pulse width for the MAT sensor; especially if you use a relocated MAT. I cannot stress the importance of this table for future eprom changes and why it is really the first table you must play with.

I have a slightly different approach to tuning the VE tables. Possibly it is because I will briefly touch on the “Spark Advance” which I suspect that you will be doing another article on. I found there is an interrelationship between both the VE table and Spark Advance. When I initially began tuning my eprom, I spent a great deal of time getting a very good VE table that was darn near 128/128 perfect. Then when I started to adjust my Spark Advance, all these values changed; increasing spark advance caused a leaning effect and retarding caused a richening effect. So I ended up having to “revisit” the VE table after my changes to the spark table.

While this might suggest that the Spark Tables should be hit first, I still believe the VE table should be done first because, because the amount of spark advance is also dependent on fuel. I believe the risk of running “lean” is a major concern, thus having a good VE table is important to avoid this. Plus, if you do the spark table and then the VE table you may encounter detonation if you were overly rich, and gave spark that the rich condition situation could support but the 128/128 could not.

I also use a slightly different method of than you mentioned, but I believe the two methods have harmony when used appropriately. You use a method which I will refer to as an “Open BLM” where you leave the min at 108 and the max at 160 as the stock eprom”. I use a method I will refer to as “Locke BLM” where I set the min and max both to 128. The biggest difference is that with an Open BLM, it takes a change of 5 in the INT to cause a change of the BLM (using the stock BLM settings, these are completely modifiable). With a Lock BLM, the BLM remains at 128 and only the INT moves to its max limits of 80 to 180.

A Locked BLM allows you to make “finer adjustments” to the VE tables to the point where you can get 128/128 perfect. With an Open BLM, you will get close but there still can be some inaccuracy and becomes difficult to attain 128/128 unless you consider an adjustment fact of the INT. I like to think of the two as this: adjusting with the Open BLM is like using a coarse grade of sand paper to get the VE table into the desired shape. The Locked BLM is like a fine grade of sandpaper to give it a smooth finish. They each have their advantages and disadvantages, and should be used at the appropriate time.

I like the idea of using the Open BLM as the “first pass” on the VE tables. It allows you quickly to get the “right” shape of the table which will be accurate enough to allow you to quickly move onto the Spark Advance, which is going to make you have to change the VE tables again due to the “leaning effect” of when you advance it. I will assume most will find that the stock eprom is far too conservative in the spark timing as I did.

Then when you are ready to make your second pass, after adjusting for the Spark Advance, you use the Locked BLM. I found that with a locked BLM that for every difference of INT from 128 that an adjustment of .2 to .3 got me “on the money”. I found for lower rpm/low load that .2 was the appropriate factor and that for high rpm/high load that .3 was more appropriate. I actually have a more sophisticated formula that I use with Excel tables that I put my initial VE table in, type in the INT number and it calculates a new VE table for me. But with that formula .2 - .3 was the general level of change for each change in INT. The simple formula is ((INT – 128) x .2) = Change to VE table value.

I will say that I do not recommend using the Locked BLM method for the first pass unless your VE tables are already fairly close. When you lock the BLM, it is very easy to set an SES code on the O2 sensor if the INT falls to 80 or rises to 180. Plus, this basically turns of Block Learning so ANY inaccuracies in your eprom will immediately show up in driveability. I will say that a locked BLM is a good test of how well your eprom is running, and when you Open the BLM, it makes if extremely smooth in removing “rough edges”.

Lastly, I also have a different method of using Diacom to help with the testing. I set it on its greatest frequency and I take a nice long drive, gathering as many captures as possible. How does 250,000 frames grab you? That is from a couple of road trips down to the “Coast”. I then took all this data and combined them on an Excel spreadsheet by converting the Diacom data to a DBF file and then having Excel read these DBF files.

I then read as many of these DBF files as I can and then sort them on RPM/MAP/INT. I then cut “groups” of RPMs and save those to a separate file. I take everything +/- 25 rpm of the intended RPM range I am interested in. I keep doing this, reading, sorting and cut and paste until I have a spreadsheet of all data for (say) 1975 to 2025 rpm in a separate spreadsheet. I then take that spreadsheet and resort it on MAP/INT. I then will take all the entries for a particular MAP range. For example, 40 kpa. Once it is sorted on MAP/INT, I will then calculate the average INT for 40 kpa. As a “reality check” I will also take the average from 39-41 kpa and 38-42 kpa and make sure that they all look reasonably close. When I believe I have a good average INT, I then use the forumula ((INT – 128) x .2) = Change to the VE table value.

I found this method gave me very accurate correction factors and virtually perfect VE tables to the point that I noticed the effects of air temperature, until I started to correct the MAT tables to get more accurate readings when the temperature changes.

Sorry for the long reply, but there is a lot of information here. I look forward to your next article Tim and I will see you all next week.

Bob Wooten

Tim,

great piece of work. thank you for the information & for the step by step. I look forward to the next chapter (i guess you would say chapter one, if you are going to cover idle & the affects of temp on VE).

a short note on myself. i have been watching this post for a few months now (found this site from DIY). you have a lot of great information here, 'specially for us new to EFI. if you would like to check out my car goto my WWW. Lots of torque & a blast to drive.

thanks for everything
BW

TRAXION

Bob,

Thanks for the encouragement! Appreciate it. What is your WWW address?

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

Bob Wooten

Oops sorry, i thought that when i put my info in when i got my password that it would show up @ the bottom of my reply, Oh well.

www.r71camaro.homestead.com check it out & let me know what you think.

again, thanks for the great info.
BW

------------------
Bob Wooten
r71chevy@earthlink.net
www.r71camaro.homestead.com

Bob Wooten

figured out the sig line. not really as dumb as i look, just sleep deprived. too many hours working on the car.

BW

------------------
Bob Wooten
r71chevy@earthlink.net
www.r71camaro.homestead.com

Yelofvr

TRAX:, Thank you very much for the great info! I have been very appreciative of the time you have spent contributing to the message board since I have been a member. It was thanks to you I started burning my own PROMS, and can't see how I got along before having this wonderful tool.
I did want to make a correction to one of your comments. When you were discussing the differences in a calibration for a motor with a large cam such as the LT4 Hot Cam, you made the following statement:
"Less manifold means less vacuum to those devices that are connected to manifold vacuum. The fuel pressure regulator is controlled by manifold vacuum. More "pull" (more vacuum, smaller kPa) on the regulator reduces the effective fuel pressure. Less 'pull' (less vacuum, higher kPA) at the regulator, increases the effective fuel pressure. Thus if we install a cam that pulls less vacuum at idle then the Fuel Pressure regulator will see less vacuum and the overall fuel pressure will be greater. This results in a very rich condition at idle."

My correction is that the fuel pressure regulator may "appear" to generate more pressure when the manifold pressure is lower, but it DOES NOT. The "gauge" pressure of the fuel rail will increase, but the pressure across the injector is ALWAYS constant, no matter what the manifold pressure is. Remember that the outlet of the fuel injector is at manifold pressure, a pressure gauge on the fuel rail is referenced to atmospheric pressure. The true pressure across the fuel injector is "fuel rail pressure" minus manifold pressure. When the injector opens, this is the pressure differential that squirts the fuel out. The whole reason that your FP regulator is referenced to manifold pressure is to keep the pressure across the injector constant so that the only thing that affects fuel delivery quantity is injector "on time". Here is a good illustration. Do you know why on TBI systems there is no manifold reference on the fuel pressure regulator? It is because the outlet of the injector is at atmospheric pressure no matter what load condition the engine sees. In order to have a constant pressure across the injector, the fuel pressure is held constant.

If I go back to the Fbody example with a big cam, true the fuel rail pressure will be higher, but the manifold pressure is also higher by the EXACT same amount, for this reason, the "pressure" pushing fuel out of the injector is exactly the same as the "stock cam" case. I hope my example makes sense to you, I know the first time I read about the operation of the FP regulator system, it took me awhile to "digest" it.

Thanks again for all your contributions to this forum. We all owe you a great deal of thanks for taking the time you do to share with us!!


------------------
Dave Zelinka

Grim Reaper

Yelofvr, that is an interesting point. However, can you explain why when you increase the fuel pressure the effect is far greater at high MAP (low vacuum) yet minimal at low MAP (high vacuum). If I measure the fuel pressure with the vacuum line connected and disconnected, the effects are reflected with the guage; disconnected the increased fuel pressure is reflected and connected the fuel pressure is marginally higher than prior to the adjustment.

This is further confirmed by the effects on the VE tables. When I initially setup my VE tables, my fuel pressure was 42 psi. When I later increased the fuel pressure to 48 psi, I had to significantly decrease the VE values at higher MAP readings but minimal (if any) at lower MAP readings.

If your point was correct, I should have noticed a 6 psi increase whether the vacuum line was connected and disconnected, and I should have had to decrease the VE values in a linear amount (or a constant %). Which in fact, I did not.

Just pointing out the difference between theory and actual results experienced.

SATURN5

"Less manifold means less vacuum to those devices that are connected to manifold vacuum. The fuel pressure regulator is controlled by manifold vacuum. More "pull" (more vacuum, smaller kPa) on the regulator reduces the effective fuel pressure. Less 'pull' (less vacuum, higher kPA) at the regulator, increases the effective fuel pressure. Thus if we install a cam that pulls less vacuum at idle then the Fuel Pressure regulator will see less vacuum and the overall fuel pressure will be greater. This results in a very rich condition at idle."
======================
I agree with this statement. With less vacuum at the regulator the less the fuel pressure is reduced from full static (zero vacuum hose off). And it would result in a rich condition at idle as the injector pulsewidths are calculated from the stock cam charactoristics. Hence that big black cloud you see as a 5.0L leaves the light. Bob

Yelofvr

Hi Glenn:

I fully understand your real world observations and the apparant discrepancy, I have seen it myself. Here is what I have concluded after playing with adj FP regulators. Lets take the stock case of about 42 psi. My engine pulls about 16" vacuum at 750 rpm. This is about 7psi below atmospheric pressure so theoretically when the ref port on my FP reg is connected, it should change the gauge reading by 7-8 psi. My pressure goes from 36 psi connected to 42 psi unconnected. This is about right and therefore, the fuel injector would have about the same delta pressure in both cases. Now, when I raised my Fuel pressure adjustment so that I got 48 psi open, when I connected it, I only got down to 43 psi. This is a delta of only 5psi. Now my injectors do not have a constant pressure across them in each case. My conclusion was that there are limitations with making a non adjustabe regulator adjustable. All I can figure is that when you push the spring harder to raise the pressure, the spring constant is higher and the ratio of big spring to small spring is not what the factory designed it at. The big spring is what responds to vacuum changes. In the end, this doesnt really matter once you dial in your VE tables, my point of the response was simply to point out the intent of the design. As a note, I have read two books who explain this incorrectly and two who explain it right. Thanks for your response, you are a great resource on the board and I've learned alot from you!

------------------
Dave Zelinka

F22Raptor

I swear to GOD that I am not trying to be a dik, but you guys who know about programming ECMs keep telling us to buy the equipment and burn our own PROMs and stop begging and blah blah, but then I see TRAXION's post where he states that "the 165 ECM is a whole different animal". I haven't read on these subjects for a very long time, but it is more than abvious to me that nobody here seems to know what to do with a MAF ECM.
So my point is, why bother and spend the money on software and tools if nobody (including yourselves) don't know what to do with it anyways?
I Hope you guys understand where I'm coming from. I'm not trying to put anyone down and TRAXION's work is very good, it just doesn't apply to me. The way I see it, there is no hope for us MAF guys.
Rick

------------------
'88 GTA 350 Full of Leprosy (rust)
Full exhaust, free mods, Xtreme energy cam, intake pieces and 2100 RPM TQ

13.39@102.50 MPH
1.90 Sec ----- Best 60FT
104.23 MPH -- Best Trap Speed
AIM HIGH!

[This message has been edited by F22Raptor (edited March 14, 2001).]

[This message has been edited by F22Raptor (edited March 14, 2001).]

F22Raptor

On top of that, I also see contradicting info in this post.
How can you guys tell if the BLM/INT are at 128 or whatnot? According to ya'll, the stock O2 sensor is not accurate and there can be big differences between its reading and what's really going on as far as A/F Ratio.
Someone please help me understand this!
Thanks, Rick

MR.HP

Hang in there Rick Tim is not talking about wide open PE. I'm in same boat I know nothing
about SD 730. I have work on Carbs all my life But I think the area is the same. When I do mind I will use 02 for cruise and what I have use for 30yrs on PE read the plugs I know some Tables are diff. But you can bet there is some one on hear that can tell you.
also cruise is drive feel whats good to you
so the stock 02 will back you up a little. It's better than not having one.Get your equipment play with it for a while on junk
just to make sure you are good with that.like
take eprom out at the wrong time and erase it
and not even knowing it.Keep study it will Come.
Dale

GregWestphal

iTrader: (1)

The main difference between the 165 and 730 ECM is the MAP vs. MAF stuff. In tuning the 730 ECM, you have a whole 3D table of volumetric efficiencies (VE) to tweak in order to get the block learns right. With the 165, there's just the MAF 2D tables that basically just calibrate the MAF sensor (to put it simply). Since the MAF sensor indirectly measures airflow (kudos to Grumpy for pointing out in another topic that it does have to calculate it), you don't have to tweak the VE any. Other than these differences, nearly everything is the same. The 730 has more things that can be changed, but most of the important tuning stuff is in the programs like TunerCat, GMEPro, and WinBin. I and quite a few other people have had great success tuning our engines and tweaking stuff with the 165 ECM. There is hope...

AFA the block learns and accuracy of the O2 sensor, these are 2 totally different subjects. The O2 sensor is accurate near the stoichiometric air/fuel ratio of 14.7:1, so that's how the computer uses it to adjust the block learns and integrators. Using that information from the computer, one can adjust the VE tables (730 ECM), MAF tables (165), and injector constant for global changes (affects all of the block learns cells and all other air/fuel calculations) to bring the block learns closer to 128.

The topic of O2 sensor accuracy centers around WOT readings. It is much less accurate in that area, but it still can be used for tuning purposes and tracking changes, although there is no direct correlation between O2 mvolts and air/fuel ratio in those ranges (like when you're in Power Enrichment mode).

------------------
Greg Westphal
'87 IROC 305TPI/A4

TRAXION

Rick,

Chill dude. No need to insinuate negativity ... it makes those who have knowledge not want to share.

Although the O2 sensor is a switching device it is very accurate at 14.7 AFR (which corresponds to 128/128). So, once you get your fuel curve nailed down to 128/128 then you can rest assured that it is accurate. At WOT the AFR is anywhere from 10.x to 13.x The stock 'switching' O2 sensor is not accurate in this range. If you need more info then do a search in the GMECM archives. Take a look at the response graph of the stock O2. This graph will visually explain the whole 'switching' behavior.

As far as the MAF system is concerned ... its no big deal ... slightly harder than SD and more limited. But, once you get the hang of it it is pretty easy and straightforward. In SD we have MAP vs. RPM and then change the VE with respect to these variables. In MAF you have the MAF tables (Counts vs Grams/Sec.). Changing the Grams/Second at various counts will modify the amount of fuel at those counts. Could I go into more detail? Yes. Am I going to? No. This is not the post for it.

Tim

------------------
TRAXION's 1990 IROC-Z
Best Time = 12.244 @ 112.51mph (1.778 60' / 7.819@88.32mph in the 1/8)
All Natural. No Force. No Drugs. Stock Bottom End. Stock Body Panels.
Gunning for NA 11's with bigger cam, bigger stall, and bigger exhaust.
-=ICON Motorsports=-
Moderator: PROM board at thirdgen.org

Grim Reaper

To further add to the posts above, at 14.7:1, the O2 sensor is "switching" around .450 mV where the O2 sensor is very accurate as pointed out above. It is when the O2 sensor gets further away from .450 that it accuracy diminisishes.

I recently have been testing Highway Mode and trying to see how lean I can go with it. I found at AF ratios leaner than 17.6:1 that the O2 reading got into single digits. What I noticed was my injector pulse width was actually higher with 18.0:1 than 17.6:1 on the same stretch of road that I have been using to test. Also, the O2 sensor seemed to take longer to get a "varying" O2 sensor reading. So, while the ecm can accept AF Ratios > 17.6:1, they become meaningless because the O2 sensor is not accurate enough to give a proper calculation for the injector pulse width.

WOT has exactly the same problem, except it is in the opposite "spectrum" of the O2's range.

F22Raptor

I'm not insinuating anything TRAXION, just trying to make sense of all this info.

F22Raptor

Just baer with me people. I work at night and sometimes my ECM goes haywire.

scanman717

Would it be safe to assume that this part throttle tunning is also going to yeild the best fuel economy??

Grumpy

The highway cruise mode AFRs are gnerally based on the 14.7 AFR learning. Soo, if the ecm hasn't learned in or the there is a problem in learning the 14.7 then it will "default" to a reicher AFR. Might also delay the *enable* time for highway cruise.
IMO, no need to run over 17:1. Gets to where you need alot of timing to realize that lan of AFR. Now in a real light car and lotza gear maybe. Also, at leaner then that a bad tank of gas might be really bad on the engine.

F22Raptor

TRAX,
If the above formula CAN’T be used on MAF cars, then I’m guessing the only way to adjust my AFR is to basically drive the car the way you stated (flat road, up hill, etc) and check the BLM/INT and then cut back or add up on the injector pulse width to achieve a perfect 128/128.
Am I missing anything? Please correct me if what I just stated is ridiculous.

Ps: I know this post is for the 730 ECM only, but if I post this separately, it won’t make any sense and I will be lost for sure.


Something else:
If I change the value of throttle opening needed to engage PE, say from 70% to 60% or 80%; will this be beneficial performance wise.
Thanks a LOT, Rick


[This message has been edited by F22Raptor (edited March 18, 2001).]

Grim Reaper

Back to the top. People seem to be asking this question a lot lately and can't find the post.

Tim Burgess

Very nice job, Tim. Great info!

Tim

Steves ZZ5

I agree!! Damn good article. I'm learning alot here guys, Thanks.

------------------
86 Trans Am WS6 Black with tinted windows, shaved door handles
ZZ4, Mini-ram, true duals, LT4 HotCam, AFR 195 heads.
***Too many other mods to mention***

"In the future, cities will become deserts,
roads will become battlefields, and the hope of mankind
will appear as a stranger" The Road Warrior

precision350

Thanks a lot Tim. Its a wonderful thing to have someone really experianced as yourself to help the people (like myself) understand the vast world eprom calibration. Thanks a lot and keep up the excellent work!

Grim Reaper

Back to the top for all to read. Seems to have a lot of questions that this post relates to.

Grim Reaper

To the top.

Grim Reaper

btt - people are asking about "tuning" their VE tables again.

doc

Rick (F22raptor),

I, like you, have a MAF car. Being an '87 165 ECM, my fuel map logic is a direct fuel injector pulse width method. The table is titled "BPW Fuel vs. Load". The vertical scale is engine rpm and the horizontal scale is air flow in gm/sec.

The values that populate the various cells on this fuel curve is the milliseconds that the injector is held open. So if you want to add fuel on the bottom end, you would increase the numbers by some factor that you have figured out or take a small incremental step.

Also, you can work with the MAF tables, there are six of them.

As with the SD MAP cars, you first set the fuel injector flow constant in the constants table.

I have read thru this post because I am thinking about switching to SD if I build my 396 stroker motor. Howerver, I think now that I should stick with my MAF system and see how well I can tune the engine.

I have been burning my owm Eproms for about 9 months and have learned alot on this board, so dont get fustrated. There are several prom burners on this board who have a MAF car.

Doc

super83Z

TTT I was about to post about the VE tables and I found this EXTREMELY helpful.

GO TIM!!!! Awesome article.

Synapsis

Thanks to Tim and Glenn for all the information they posted (a year ago). It's really going to help me get this thing tuned closer to where it should be.

I was wondering, were there any other articles posted like this in regards to spark tables or anything of that sort? Maybe we could put them together and make a link in Tech Articles to them?