anesthes

iTrader: (12)

It appears the general consensus of people who *don’t* burn their own chips is: its hard, its black magic, leave it up to the professionals, etc. Obviously I don't preach it being the easiest thing in the world. Well, I guess easy isn't really the right word. It is complex, can be difficult at times, but far from impossible.

One thing said I am NOT the best prom tuner! I'm not even that good. I ask for help frequently, and often get stuck. You will too - and you will get frustrated. But believe me, the results are rewarding.



So, you've read Traxion's introduction to prom tuning, and you read Grumpy's "programming 101" and you're still lost, or scared? I understand. It's a large step.

Traxion goes into great detail on chips, and programmers. I'm assuming at this point you already have a pocket programmer, some flash chips, etc.


A lot of people recommend different programming software. I use Tunercat mainly. It costs money, but seems to work the best. TunerPRO seems to be getting a lot of fame as of late, and it has a built in ALDL scanner.

Speaking about scanner, you're going to need one. So start building an interface cable. Without a scanner, diy-prom is useless. I use datamaster, moates, and sometimes TunerPRO. For most people, moates and tunerpro should be fine for scanning.


What the heck is a bin anyway?

Lets explain this in simple terms. A lot of people I talk to who are getting into prom tuning are really confused about the whole thing.

First you have the ECM - the computer - the brain. This is what runs your car. If you have an 86-89 MAF TPI car, you have a 1227165 ECM. If you have a 90-92
SPEED DENSITY TPI car, you have a 1227730. Traxion explains all this in his article, so if you don't know what ECM you have - you’re in trouble!

So what is a mask? The mask is the program ID. It's the name of the program. $8D is the mask for the program that runs in a 1227730 ECM for speed density cars.

Then there is the broadcast code. What is this? Consider this the version. Gm wrote this program called $8D for speed density cars, but there are obviously numerous different models (some with manuals, some with 5.0 liter, some with 5.7, etc) So GM had to make various versions of the $8D program. Each unique version has a broadcast code.

When selecting a version (broadcast code), pick the one that matches your car! Obviously if you have an automatic 350, select the broadcast code that was offered for a 350 automatic. This will provide a close baseline, and there will be less to have to modify initially.

So now you know what program your ECM uses, and you know to get the version (broadcast code) that as close-as-possible resembles your setup. Now what?


Get used to the software!

At this point I recommend you get familiar with the tuning software. Open your new bin up, and start looking at all the options. Don't dive in too deeply! Lets start at
Constants. Injector flow rate - does it match yours? If you bought some larger injectors, you should modify this to match. Fan enable temps? Why buy one of those $40.00
Jet fan adapters, when you can just change the on/off temps here! Initial spark advance - some people like to set their initial dist advance to 10d. Make sure you
set this value to whatever you set your dist. You can even disable EGR, CCP, AIR INJECTION, etc by changing these constants.

Ok we got past phase one. Burn the chip! Go try it in the car. Did it work out ok? Great! If not, start over again. Keep a notebook handy of your changes. Document each change, and its effect.


Aldl logging, your new best friend.

If your new fresh bin is able to idle, and drive the car, now its time to bring your partner along. Your partner is going to be a cheap laptop, and an ALDL cable. You're going to want to let the car idle up to closed loop (aldl software will display this), then start driving. Try to use every gear, every throttle position, etc. accelerate, decelerate, stop, everything. Get a good long scanner log.


Beginning VE / LOAD tuning:

Once back from your drive, you'll have plenty of data to start diving into the fueling part of the prom. What I like to do is export my CSV data to a excel spreadsheet.
I label the columns; hide the junk I don’t want to see, and then freeze the label windowpanes in excel.

What we want is to get your bin as close to a baseline as possible for part throttle and cruise driving. You'll notice you have a column referred to as BLM, or l-term. This column is the long-term fuel correction that the ECM is providing. In most documentation I’ve read, the BLM directly affects the fuel injector pulse width (how long it opens for). It appears as such, the inj PW uses the BLM as a multiplier. When the BLM is 128, its "dead on". If the BLM is 120, 118, 108, or anything less than 128 that means the car is programmed rich, and the ECM is trying to reduce fuel. (Obviously, multiplying by a lesser number will decrease).

If the BLM reads a number higher than 128, then the car's programming is lean and the ECM is trying to add fuel.

There are numerous methods of getting your LOAD or VE tables in check. I recommend reading some of the tuning articles specific to your ECM. Generally a lower number in the VE or LOAD table means less fuel. A higher number means more.


Advanced VE / LOAD tuning:

This is an ongoing process. Making changes, driving the car, making more changes, etc. The idea is to get the BLM as close to 128 as possible. Now you'll notice I said earlier to wait to drive your car until its in closed loop. The reason is if your tables are WAAY far off, I want the ECM to make fuel corrections so you either don't get stranded, or don't blow something up by running too lean.

Once you have everything close to 128 in closed loop, its time to make sure your calibration is actually correct. The way I do this is I turn off closed loop mode. Most any ECM program ($8d, $58, $6E, $32B) will have a constant for closed-loop enable. Usually a coolant temp setting. Set this to some crazy high number, so it never enters closed loop. Then go for your ride. You'll probably notice some of your BLMS that used to be 128 or close, are now different. That’s because the ECM is not correcting anymore, therefore not skewing the numbers any. Now what you want to do is get your BLMS as close to 128 as possible again. Once you've done that, you have mastered your part throttle / cruise fueling.


What about spark advance?

Spark advance comes after, and only after you have your fueling right. Unless your combo has super-high compression, and you had to lower your timing in some areas to get the car drivable, then NOW is the time to tinker here.

Everyone has different opinions on your spark table. Though it appears fairly split. Most people like to get the spark advance as high as possible everywhere, without generating knock. Remember a few things however: As you increase spark advance, you'll increase fuel consumption. Often you need to then add more fuel to compliment
Your spark. So you maybe jumping back and forth between spark and part throttle / cruise fueling.


Ok man, I got the car DRIVABLE but what about performance?

This is another topic that has been greatly covered and will be specific to your ECM. But in any event, most everything has a PE (Power Enrichment) mode. Generally as long as the car is warm, you enter PE mode when throttle reaches a threshold. (Programmed by a constant usually). In $8D I think its 70% throttle.

PE mode is going to do two things: Add tons of fuel, effectively changing your AFR, and adding tons of spark advance.

Two things to keep in mind are:

1) PE switches the car to open loop. The BLM will stay at 128 static, so please for the love of *** - ignore this value in your PE log. Instead look at the 02 voltages.
.900volt is typically (but not always) 13:1, so use this as a baseline for PE fuel. If its way less than .900, add fuel.

2) If you get knock in PE, reduce timing by what PE reduced it, and then take out 1extra degree. You can ramp your timing back up, but lets be safe. Pistons
Are not a renewable resource.

Again, please read one of the well-documented articles on PE specific to your ECM / program.



Obviously there is hundreds of other options I left out. This is going to be a learning curve. However, reading this should help (some) people get a better idea of
What’s going on. Its a lot to get involved with I know. But once you get used to scanning the car, making fueling and timing changes, you'll be ready to go onto the more
advanced things. Again, the results will be rewarding. Stick with it!

vernw

iTrader: (2)

Excellent article/post! From a newbie just buying his programmer today, THANKS!!!!


- Vern

11sORbust

You should note if the blms are above 128. There is fuel correction if they are over 128.....

Mc Firebird

I am quite new to this so can you please expand on the ADVANCED VE LOAD TUNING part.

There seems to be a few different suggestions for fine tuning VE tables. narrowing BLM ranges / using INT values / TV cable disconnecting etc etc can they all be lumped into one SUPER-VE-TUNE method.....

anesthes

iTrader: (12)

In pe? Not in any of my logs.. I've leaned it WAY out in PE (boosted app) we're talking to the point of massive detonation, and it did not correct BPW

-- Joe

anesthes

iTrader: (12)

Theres plenty of methods. I just posted mine. I get it close in closed loop, then force open loop and try to get it dead on.

I'm not a professional. I'm sure there are better ways. If you've gotten to the point that you want to fine tune it even moreso, then searching will find some of Grumpy and RRobs methods.

-- Joe

Ronny

iTrader: (1)

where was this post when i started tuning 6/2003?
TY anyway good stuff.
i have been watching my 02 volts in AE and PE but it seems so eratic and difficult to track for me. also a bit dangerous without a passanger. it does seem to move based upon AE or PE but the AE i am tuning based upon drivability rather than sensor activity .previous a cough(lean?) on AE now a gurgle(rich). so i am backing off on my prior massive PS. PE seemed to take care of itself as commanded 12/1 turned out to be 12/1 on dyno so changed commanded to 12.5/1 and left alone. the bunbg for WB02 is in so now need to wire it up. that will be a big help.

11sORbust

There is some wot correction in the ecm. You did a good job with this thread. No sense getting off topic though. I don't want to take away from what you are trying to do.. You can search on the subject and will find the exact parameters and calc.

No sense in confusing the new guys....

Ronny

iTrader: (1)

we were confused long b4 this.

anesthes

iTrader: (12)

Well. Hopefully I helped a little.

MTPFI-MAF

I think this thread deserves to be put in to the
Important/Useful: Tuning Articles broken down by ECM on the main DIY Prom Page.

Very helpful :hail:

TRAXION

Definitely STICKY this post in BOTH the tuning articles AND the How To sticky at the very top.

Woot! Nice writeup.

Tim

TRAXION

These statements are wrong.

1) PE Mode will not only add fuel ... it can be used to subtract fuel. That is very important.

2) I have already DEBUNKED the OLD thought that PE = open loop. This is not true. The ECM is still in closed loop. It is just that learning has been disabled. HOWEVER, the BLMs can still be GREATER than 128 and the ECM can still ADD fuel in PE Mode. This is very important. I covered this too! Please remember what has been previously posted so that we can move forward instead of cornfusing the newbies. Also check into the S_AUJP binary where I disable this possible "greater than 128BLMs at WOT".

Tim

RednGold86Z

iTrader: (1)

Tim, are you meaning that at Wide Open Throttle aka Power Enrichment aka PE, the ECU is not only dumping in extra fuel, but is also making Closed Loop corrections (to target 14.7:1 A/F)?

RednGold86Z

iTrader: (1)

The way block learn learns at 100 kPa or full load is that it can reach 100 kPa while the throttle is below the throttle position threshold for Power Enrichment, then once the throttle opens above the threshold, closed loop is turned off and PE fuel is added to get the A/F down to 12 or less (from the factory), but since you are still in the same block learn cell (100 kPa, whatever RPM) that Block learn factor is applied if it is greater that 128. It's a warranty issue to have learning remove fuel, but adding it is A-OK. There is a curve that can be found where at each RPM there is a throttle position that corresponds to 100 kPa, where at 900 RPM, a low TPS can achieve 100 kPa easily, but 3000 RPM, maybe 60% is needed. The block for 100 kPa extends down to as low as stated in the calibration, be it 80 kPa or whatever, it is easy to learn in that block while in normal closed loop mode, but closed loop should definitely be disabled if the throttle position is greater than the PE tps threshold.

MTPFI-MAF

Opps wrong window
Sorry!!!

SLP-GTA

iTrader: (1)

thanks for the write up, I like to see a similar write up on what steps to take after a big cam install

Grumpy

No Tim, it's not closed loop.
It's a special condition. In SOME masks, the BL correction is applied to the PE calculation, but not in all.

Closed loop here means ACTIVELY using the O2. Which it's not. If it's not using the O2, it's not closed loop.

The ecm MAY be using a correction learned from while in closed loop, but that doesn't mean it's in closed loop.

TRAXION

I beg to differ Bruce. I think you are wrong. There is open loop and there is closed loop. HOWEVER, closed loop can have different states: Learning and non learning. When in PE Mode the ECM is in closed loop but is in non learning mode and the ECM bypasses certain corrections. Go through the source code. The ECM flag for closed loop remains in effect! Yes, the flag for closed loop remains on. It is not turned off. Thus, the ECM is still in closed loop. HOWEVER, the learn mode flag is turned off. Furthermore, the code will check for the PE bit and bypass certain other things. It is still in closed loop though. To prove me wrong please locate the section of code in the ANHT_HAC which shows that the closed loop flag is disabled when PE is hit. You can't show it to me because it doesn't happen. The ECM stays in closed loop.

Tim

RednGold86Z

iTrader: (1)

It could be that since the commanded A/F is NOT 14.6, that closed loop disables itself, without forcing it "disabled" with a flag. From the P4 Section 9 document, it goes into an INTEGRATOR RESET mode.
http://www.sunflower.com/~leroy/P4/section9.htm
That IS Open loop by resetting the Integrator, but not by "disabling" or turning closed loop off.

Grumpy

I said it was a special condition.

And once again, not all masks use the BL correction, when in PE.

But in ALL cases, PE is a S-P-E-C-I-A-L C-O-N-D-I-T-I-O-N, in the 3rd gens.

Nice try with the ANHT hac comment, but that doesn't apply to all the masks used in 3rd Gens..

anesthes

iTrader: (12)

Why dont one of you guys load up AUJP in a ecm bench, pin tps and set map to like 100kpa or something, and then play with the o2. See if it changes the bpw.

-- Joe

funstick

common guys play nice. im writing up some stuff for this post becuase it needs it. on the whole bruce is right. on the narrow he is wrong in the sense of anht. it depends. yeah ive seen the flag that gets set for closed loop in anht but not in ajup. i think thats why they went to ajup. now on the newer obd2 car they are always approaching closed loop operation at almost every load and wow are those scew calcs screwylooking. there reading catalyst efficiency etc to determine the scew from stioch. its wierd to say the least.

JP84Z430HP

The open loop / closed loop question isn't much to worry about, however, what happens in PE in YOUR bin IS what's important.

As mentioned, PE is a SPECIAL CONDITION, and you need to know what is going on in PE, whether or not BLM correction is being used is very important. Whether the flag for closed loop drops, is not, at least to me..... From what I can tell, learning is turned off in PE, always, and in some codes the nearest cell's BLM is used.

Does this simplify things?

TRAXION

You are holding on to old thoughts for some reason that I just don't understand.

Closed Loop to me means that the Closed Loop bit is set. Remember, we are simply talking about binary code here. If the binary code as the Closed Loop flag set then, according to the binary code, the ECM is in closed loop. I don't see why you guys are saying that it is no longer closed loop. The flag is still set. The ECM will still register as Closed Loop. The ALDL stream still shows closed loop. In PE the closed loop bit is never UN-set. It is still set. It's just that learn control is disabled.

For me it is Closed Loop with Learn Control Disabled. Technically speaking - this is 100% correct AND it explains everything.
For you guys it is something else that has no meaning in terms of the actual assembly language.

I guess I can cause more trouble now that I don't moderate ... or is it that I am more thought provoking?

Again - somebody please show me the code where Closed Loop is set to False when PE Mode is enabled. We've already been through this before. It doesn't happen. It remains in closed loop. I am speaking of ANHT and AUJP. AUJP is the most recommended binary so it is valid for me to bring this up.

Tim

TRAXION

Says Who? I disagree. It is "Learn Control" that means the O2 is actively being used. The ECM can be in Closed Loop and not using the O2. This is true.

I respectfully disagree. I don't mean to be a butthead. I just think that everyone needs to get on the same page. Since we have already covered this before it would be nice not to take a step backwards and to keep moving forwards. Closed Loop can still be enabled with Learn Control off in PE. Granted - you are absolutely right that this may not be true for all BCC's / Masks. However, for the most recommended BCC (AUJP) it is true. Thus, it is rather important. It also helps people to remember that BLM corrections CAN still happen at WOT (>128)

Tim

JP84Z430HP

What I'm trying to point out is that it really isn't of much importance, compared to the PE flag being set, and knowing what the code dose in that case.

I'm trying to figure out where it may be of importance for m,e to know that it is still in Closed loop! From what I can tell, the closed loop flag just means that all the conditions are correct for closed loop, even if it is in a state where it ignores the O2.

The only TRUE closed loop, purely by definition, not by how the ECM set's it's flag, is when the O2 is being used for fueling corrections. Is this not correct?

Not trying to step on any toes here, just trying to clarify the point! I know that the parties discussing this are FAR more experienced than I, and I'm just speaking based on what I have learned from them, and my own work.....

Maybe we should be discussing this in a "When is closed loop NOT closed loop" thread?????? This is a great thread, and I want it to grow...... No be a grounds for old arguments.....

Grumpy

As long as you ignore the GM document and basic electronic principals, you'll be on the wrong page.

If all your looking at is the Flag being set, then your ignoring what's actually going on.

As I mentioned what about DFCO?. Flag is set, and there is NO FUEL delivery. The flag being set is only part of what's going on.

Your the one that's tied his ego to this deal, with your claim of having debunked something, and when you do that you've taken a position, and it's doubtful you'll be open enough to see another view. Sorry, that's just human nature. If you want to actually discuss things then you need to use rational points, and discuss them.

The Flags being set are only one part of the equation, and that's just fact.

A quick goggle turned up this,

http://www.netrino.com/Publications/Glossary/PID.html

Or is this something else that we're just supposted to ignore?. Yes, let's move on.......

Grumpy

The world of electronics and code revolves around definitions. You can't just call things what ever you want.

A flag being set is not the final say on what's going on. Just because the flag is set doesn't mean anything then the conditions are meet. If something else is going on the overules that, then it's what the output is doing is telling what the truth of the matter is.

PE, DFCO, soft fail mode, as well as limp home mode are special conditions, or operating modes.

TRAXION

Wrong again. I haven't ignored anything. Here's the problem Brucey ... The definitions of closed loop were created before we know what we know now. Everyone is holding on to the older definitions when they are just not appropriate anymore. The ECM is in closed loop. Again, show me the code that proves I am wrong.

And yet again you put words into my mouth that I never said. Why do you continually do this? I never said that. In fact, I talked about learn control which is definitely not ignoring what is going on.

Ummmm - learn control?

Ummm - no. You're the one being closed minded by not being open enough to actually hear what I am saying. What you SHOULD do is actually listen to what I am saying and try to prove me wrong.

And yet again Bruce makes comments that makes no sense in an attempt to TRY to make somebody else look horrible. WTF are you talking about? Rational? Rational is looking the code and not seeing the Closed Loop status reset to False on Power Enrichment. Not being rational is you. Why? You are holding on to some definition that somebody else developed without actually being rational and looking at the code.

Ummm - yea. Learn Control.

Bruce I have read more books on ECMs than what you are giving me credit for. I know what the general popular definitions are. I grew up on them. However, if you never question the general consensus then you'll never grow and move forward. You stay right were you are if it makes you feel better. However - stop using your words to try to make me look like I don't know what I am talking about. I know about everything you said .... I knew about all of that years ago. I have chosen to shed new light on stuff. Again - prove to me that the ECM is no longer is Closed Loop.

Tim

TRAXION

Yea, don't forget that there are various modes to Closed Loop like Learn Control being on and learn control being off.

Tim

11sORbust

you are both wrong,lol. The ecm is not going to switch to open loop just for PE. But at the same time it does turn off learn control. So closed/open loop is not the issue. PE is PE, not open or closed loop. PE will override both conditions..


I'm starting to get confused. Is this thread about teaching noobs or debating over *************? Sorry to be so blunt. BUT, I have really learned alot from getting banned. One thing I was guilty of was hyjacking threads to create a debate. I was never aware that I was doing it. The original author of the thread is the one getting the shaft. I'm aware of this and now tend to start a new thread instead... Not saying that's what happening here. Just some thing to consider.

tpi_roc

I agree on the point that to effectivly communicate with a group of people on a specific topic you have to accept the existing termonology.

Taken from the website posted and the deffinition I've used in my ventures:

Closing the loop

Systems that utilize feedback are called closed-loop control systems. The feedback is used to make decisions about changes to the control signal that drives the plant. An open-loop control system doesn't have or doesn't use feedback.


I would like to see an ecm bench add PW in PE based on 02 input for thirdgen ecms.

3.8TransAM

great thread guys:-)

Lets just play nice and keep it clean with no slinging:-)

Want to keep this one going, I have found quite a difference in terminology from book to book and even tuning program to tuning program. Lets use this to find out the who/what/why on the way things are names, vs. who is right and wrong. More than likely we can all pick something up from this thread

later and thanks
the mods

JPrevost

I'd also like to see this because if that's the case, then closed loop is applicable in PE. Learn mode and closed loop seem to be confusing to most.
If the code forgets to turn off the closed loop flag but the code ignores the o2 feedback then the system is in open loop even though it's telling you different.
I forget where I read this or if I even did but wasn't there a mask out there that used closed loop without learn where the BLM was locked (no learn) but the INT was still moving. If this is incorrect or another "special condition" correct me.
As for the article... looks good from the bits I read.

Grumpy

OK, so what do you call it?.
I post where PID, closed loop is explained, and you just continually ignore the material. I thought I'd connected the dots well enough that with the mention of DFCO, you'd grasp that a flag being set doesn't mean something is final determination about what's in fact going on.

vernw

iTrader: (2)

Guys, I admire you both and what you've accomplished and done for all us neophytes, but please don't tarnish this thread with your detailed discussions and disagreements on terms, etc.

At the risk of getting flamed, would you mind if I asked you to please take that elsewhere so we can preserve the "magic" of this thread.

Have a Great Day!!!


- Vern

B4Ctom1

Im confused a bit about the terminology, I don't understand this "$8D" I don't have "$8D" all I have is AUJP.

3.8TransAM

The $8D is simply the mask for a given vehicle type/line and its ecm.

$8D takes care of 90-92 SD f body(730 ecm)

like $58 works for the Sy/Ty apps(749ecm)

later
Jeremy

RBob

iTrader: (1)

The $8D is a maskID. It is used to define a calibration layout. The $8D cal is used for the SD TPI F and Y body's.

AUJP is a $8D mask.

RBob.

locked. . .