Request: Chip Burning HowTo

[Tydriver]

For me personally it was a pros/cons list someone posted between the factory ECM and FAST sytem. Im not willing to give up some of the things the factory ECM does that FAST doesnt (TC lockup, etc).

Would FAST be simpler for someone that cant even figure out how to open a BIN file to look at like myself? Oh hell yeah. Am I gonna pay $2k for FAST/syty harness? Oh hell no... Not when I can struggle for a while but with some effort maybe learn how to burn my own stuff for a fraction of the cost.

Lastly.. If I can actually figure this stuff out its a personal accomplishment for someone whos not the sharpest tool in the shed.. and I may be able to help other ppl that honestly want to learn it too.

Couldnt agree with you more Myclone, also keep in mind, FAST and DFI are cool (I'd love to have that budget) but there's something cool about telling people you are modifying the STOCK ecm to be that fast, look at the canadians for example.. Also, I look at it this way, if I learn it on this vehicle what I learn can apply to many others that I may own in the future and dont want to equip with another stand-alone Engine Mgmt System.

Its gotta all be similar atleast in theory, once you learn the basics, they should apply for many years to come Now thats ULTRA COOL. 8)

 

[smeagol]

Modifying on the fly is cool.. but once you start datalogging and getting your act together, it's not that big a deal.

Ask Nolan what he spent the most time doing while tuning - I almost guarantee it is spent going over Datamaster runs, deciding what changes need to be made. Making changes on the fly promotes hasty decisions, not a good thing when it comes to blowing up motors


Also, making those on the fly changes is not the biggest aspect or feature of DFI or FAST. Once you compare the other features of the computers, things look different.

Wideband is H U G E. Being able to run a cheap, accurate wideband 02 sensor is awesome. Now that we have that option on the stock computer, it evens the playing field. Sure, FAST & the new DFI can do corrections, but to tell you the truth, I'm not that impressed with the correction feature yet. I did more datalogging, analysis, then made changes based on the run, than relying upon correction factors.

TCC lockup - not available in FAST, but is available in DFI.

Boost control - only available in stock syty ECM.

Idle control - much better options in stock ECM, the others don't have park/neutral input.

AC control - only new version of DFI has AC shut off at WOT.

timing adjustment tables - stock ECM has the best, better than FAST, maybe better than DFI.

Datalogging software - Datamaster is DAMN good, FAST windows software is a big improvement over the old DOS software, but is still only comparable with Datamaster IMO.

Nitrous control - only available in the aftermarket stuff, but most of us don't care about this.

Chips allow you to switch between programs quickly, if you use an Ultimate setup. For FAST, you must upload the program, which takes 30 seconds to a minute. So switching between a street program and a race program requires you having a laptop handy & connected.

Lots more, that's just off the top of my head.

 

[10secTy]

Modifying on the fly is cool.. but once you start datalogging and getting your act together, it's not that big a deal.


Yep I am tunning a 410 aluminum sprint car motor that is in a 1969 Camero. It has FAST. It is hard to tune on the fly, idle is easier but that is it.

Ask Nolan what he spent the most time doing while tuning - I almost guarantee it is spent going over Datamaster runs, deciding what changes need to be made. Making changes on the fly promotes hasty decisions, not a good thing when it comes to blowing up motors

Yep datalog for about 10 minutes then go home and go over the log for about a hour making changes. We are on chip number 54 now.
But I go slow and makes small changes.


Also, making those on the fly changes is not the biggest aspect or feature of DFI or FAST. Once you compare the other features of the computers, things look different.

Wideband is H U G E. Being able to run a cheap, accurate wideband 02 sensor is awesome. Now that we have that option on the stock computer, it evens the playing field. Sure, FAST & the new DFI can do corrections, but to tell you the truth, I'm not that impressed with the correction feature yet. I did more datalogging, analysis, then made changes based on the run, than relying upon correction factors.

I even turn the correction off as it can be confuseing sometimes. I work at getting the Ve's good with out haveing to rely on it.




TCC lockup - not available in FAST, but is available in DFI.

Boost control - only available in stock syty ECM.

Idle control - much better options in stock ECM, the others don't have park/neutral input.

AC control - only new version of DFI has AC shut off at WOT.

timing adjustment tables - stock ECM has the best, better than FAST, maybe better than DFI.

Datalogging software - Datamaster is DAMN good, FAST windows software is a big improvement over the old DOS software, but is still only comparable with Datamaster IMO.

Nitrous control - only available in the aftermarket stuff, but most of us don't care about this.

Chips allow you to switch between programs quickly, if you use an Ultimate setup. For FAST, you must upload the program, which takes 30 seconds to a minute. So switching between a street program and a race program requires you having a laptop handy & connected.

Lots more, that's just off the top of my head.

FAST is a good system but you still have to know what you are doing.
I feel personally that you will have better drivability and more features with the stock ECM. It takes some work to learn how all the tables work together but once you do its a great system. And best of all its cheap!

Nolan

 

[Blake]

I know this is an old topic but I just read through that thirdgen BB archive. OH MY GOD that was good!!They are sharing all kinds of stuff!!! If you read down close to the bottom they have stuff like:
Tuning VE curves(this is if you WANT 14.7 of course):
1) Discover your current BLM at a particular RPM and MAP using a scanner.
2) Divide that value by 128 (stoichiometric A/F for 14.7).
3) Multiply that result by the current VE value at the same RPM and MAP where you took the BLM reading.
4) Take that resulting value and put it into the cell with that RPM and MAP.

This stuff may seem simple to guys that have been doing it but it really clicked the lightbulb on for me.
Now I just need to apply this to the DIY WB o2 info that is hacked into the bin so it can be seen with Datamaster(gotta figure out what the numbers mean first) :roll:
There is also a reference to highway tuning for mileage...
Checking that out too
I'm with you, Myclone. Only I'm a retard and have to read stuff like 15 times before I can even ask a decent question, let alone understand it.

 

[myclone]

I know this is an old topic but I just read through that thirdgen BB archive. OH MY GOD that was good!!They are sharing all kinds of stuff!!! If you read down close to the bottom they have stuff like:
Tuning VE curves(this is if you WANT 14.7 of course):
1) Discover your current BLM at a particular RPM and MAP using a scanner.
2) Divide that value by 128 (stoichiometric A/F for 14.7).
3) Multiply that result by the current VE value at the same RPM and MAP where you took the BLM reading.
4) Take that resulting value and put it into the cell with that RPM and MAP.

This stuff may seem simple to guys that have been doing it but it really clicked the lightbulb on for me.
Now I just need to apply this to the DIY WB o2 info that is hacked into the bin so it can be seen with Datamaster(gotta figure out what the numbers mean first) :roll:
There is also a reference to highway tuning for mileage...
Checking that out too
I'm with you, Myclone. Only I'm a retard and have to read stuff like 15 times before I can even ask a decent question, let alone understand it.

Going there now.. Ive looked at one fbod site but didnt think to look for a thirdgen site.

I hear ya on having to do things multiple times to learn BUT.. I usually pick up something new/more useful each time. One reason I havent posted anything on here is Ive been working on it myself to get enough knowledge to even ask a somewhat intelligent question.. I know just enough to be dangerous right now IMO.

 

[Tydriver]

Tuning VE curves(this is if you WANT 14.7 of course):
1) Discover your current BLM at a particular RPM and MAP using a scanner.
2) Divide that value by 128 (stoichiometric A/F for 14.7).

FWIW, I can tell you for sure, that WOT BOOST stuff you dont want 14:7 ratio.. You are prolly well aware of this already..But Nolan's been schooling me a lil bit and says 11.8:1 to 12.1:1 at WOT with BOOST....

Just a noob's input that I am sure you already know..

 

[Blake]

That's what I'm learning. Of course WOT fueling won't be determined by BLM since that kinda goes out the window when in PE or WOT conditions.
Wish I could experiment with more WOT and PE stuff but I don't think my tired engine will stand for it
I need to play with the wideband and learn to see if this same theory applies to the numbers it spits out. I'm probably making it harder than it is.
Blake

 

[smeagol]

Datamaster has a nifty feature you can use for drivability tuning as well.

Make a closed loop program, set lock the BLM (set limits at 128 in the chip). Drive it when warmed up, being careful to run at *static* conditions - not heavy acceleration, deceleration, much driving sitting at a certain RPM/map level.

You can go into Datamaster and view a histogram of the run, and you will see what kind of corrections are being made at each RPM level. This will give you a good indication of what kind of fueling changes need to be made. The reason I say to limit the BLM is because that would introduce 2 variables into the equation, and would throw you off, and there are a couple other reasons I don't want to dwell on here.

I see drivability as 3 distinct areas - idle, regular cruising, and dynamic acceleration/deceleration conditions. The other 2 classifications I'd make are WOT and startup issues.

The idle stuff is *fairly* easy to dial in, try modding the F29 closed throttle table up and down. Your first few programs will get you real close, the final mods will just be fine fine tuning.

The regular cruising can be taken care of by the histogram method. You will still have to correlate the RPM vs integrator level to the necessary changes in the VE table, but that's the only difficulty. A few rounds of this will really start dialing the truck in.

The acceleration/deceleration can be a PITA, and can cause you to make changes to the VE map in error. I say to tune the VE map based on static conditions, and if you experience any other drivability errors (after VE tuning) they are almost certainly due to this area. AE stuff is difficult to tune. It is almost entirely qualitative, you can't take data on it most of the time. Basically trial and error if you are having big problems in this area. DE stuff - I've seen when DE is too strong, it will kill the truck on a deceleration. Too rich (DE is too weak) the truck will foul plugs, motor won't decelerate quickly (or at all), 02's will stay rich at closed throttle when you let off. DE is pretty easy to dial in IMO.


You have to remember that all the programs we are messing with, we have a starting point. So if I am experimenting, let's say with an injector that is fairly proven, but in a combination that isn't dialed in, and you are having, let's say, DE problems. I might make a few programs with stronger DE (goes leaner at closed throttle deceleration) and a few programs with weaker DE (stays richer at closed throttle deceleration) and try them all out. You may find a trend this way. As you start tuning the truck more and more, your tuning process gets more and more exacting, with more very fine tuning. Your notes may go from "Rich at 1000-2000 rpm acceleration" to "truck has slight stumble when you are at 1500 let off throttle 20% then stab it to 85% TPS". Solving problems at this stage can be a challenge, since you have dialed the truck in fairly well, you may screw up the tune more than fix it when you make changes at this stage. This is why you need to know the tables' role in the calibration, and when you make the changes.

 

[Blake]

Thanks for helping us with this, Brian.
It's an understatement to say how much we appreciate it.
Blake

 

[Sy769]

8) Brian You are much appreciated. A beer on me at the Nats


I am still working on the DIY burner. Been studying the tables and stuff in the meantime. I have a few peoples bins and starting to put together the larger injector stuff in my mind.
Some of the table entries for fueling seem odd to me. Is it normal for it to not be a smooth curve; or is what I'm seeing just an entry that worked so they haven't been changed when entries around them were updated?

Any other rules of thumb that you wouldn't mind sharing? In particular how much to change values. I am a little rich at this MAP and RPM so I need to change the value by X% then try again. Or do you have to do it a little more scientifically?

 

[kwick6]

Has anyone ever used this type of programmer...GTek Model 7228


I'm asking because it is all for sale together and I like the idea of getting everything I need at once.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1924313288


If the pocket programmer is much better than I might just go that route. Just wanted to see if this one will do the job just as good.



TIA

Scott

 

[keith455]

test

 

[kwick6]

test

 

[TheObiJuan]

hmmmmm, my email notification came through, however clicking on it yielded a message stating the post did not exist.

 

[Tydriver]

hmmmmm, my email notification came through, however clicking on it yielded a message stating the post did not exist.

Ditto, earlier in the day.......

 

[kwick6]

I know this is an old topic but I just read through that thirdgen BB archive. OH MY GOD that was good!!They are sharing all kinds of stuff!!! If you read down close to the bottom they have stuff like:
Tuning VE curves(this is if you WANT 14.7 of course):
1) Discover your current BLM at a particular RPM and MAP using a scanner.
2) Divide that value by 128 (stoichiometric A/F for 14.7).
3) Multiply that result by the current VE value at the same RPM and MAP where you took the BLM reading.
4) Take that resulting value and put it into the cell with that RPM and MAP.

This stuff may seem simple to guys that have been doing it but it really clicked the lightbulb on for me.
Now I just need to apply this to the DIY WB o2 info that is hacked into the bin so it can be seen with Datamaster(gotta figure out what the numbers mean first) :roll:
There is also a reference to highway tuning for mileage...
Checking that out too
I'm with you, Myclone. Only I'm a retard and have to read stuff like 15 times before I can even ask a decent question, let alone understand it.

This really made sense to me as well. I know my truck needs some fine tuning at different rpms. I'd much rather shoot for a real AFR than rely on the stock 02 sensor for the rich/lean conditions. It doesn't really narrow down the specific areas, more less just switches back and forth from 128 to 140 or whatever....need to have WB02.

This page is helpful as well...
http://www.geocities.com/ecmguy.geo/bruce/prog_101

Now if someone could explain this...
If you looked in the knock retard table, you might find the value $44 (68 decimal). What's is this in real world units? take the table value, multiply by 45 and divide by 256, which gives 11.9. That's almost 12 degrees of retard when knock is detected.

I understand the 68 decimal, becoming 11.9, but what the %&^* is "$44" and where did that come from?

Oh well, been reading for hours, need to get away.

later,

Scott

 

[kwick6]

Anyone using this adapter.... the $40 zif to ecm

http://moates.net/gmecm/hworder.html

he doesn't show the 7749 on the list that his stuff fits.
I also want something that will go right into the pocket programmer as well,
anyone got the best configuration please post, obviously trying to keep cost to a minimum. :wink:

TIA,

Scott

 

[PPI Typhoon]

Yeah, I'm using the P-3 adapter in my multi-chip 3-bar setup. Works great. I'm working with him right now to get it set up with a remote mounted EEPROM and controller. But the adapter he has right now works excellent. I have pictures of it in the ECM, if you'd like.

Let me know.

 

[BillC]

I understand the 68 decimal, becoming 11.9, but what the %&^* is "$44" and where did that come from?

The "$44" is representing a hexadecimal number (base 16), instead of a "normal" decimal (base 10) number. Hexadecimal (or "hex" for short) is commonly denoted with "0x" in front of the number, like "0x44", but there are lots of different conventions.

0x44 converts to 68 in decimal. If you can't (or don't like to) do the conversions in your head, the standard Windows calculator in "Scientific" mode can do the conversions.

Hex is useful because two hex digits completely represent one 8-bit byte, and is a closer representation of the actual bit structure within a byte. If you ever spend time working with the raw data in the chip, you'll have to get used to using hex.

 

[kwick6]

understand the 68 decimal, becoming 11.9, but what the %&^* is "$44" and where did that come from?

http://www.jaworski.com/htmlbook/dec-hex.htm

Just found this.

I don't think I'd want to do the conversions in my head.


I'm glad you mentioned the 0x representing the hex also, didn't know that.
I'm not quite sure how the two hex digits represent the 8 bit byte, but I was trying to figure this out...

Binary refers to base two or a two-state digit called a bit. A bit is either on or off - represented as a '1' for on (the set state) and '0' for off (the cleared state). Eight bits together form a byte and are written as 00110101b (or sometimes %00110101). The 'b' stands for binary, and lets you know that we aren't talking about 110,101, the decimal number. A byte or multiples of bytes set the register size for microcomputers. Hexadecimal (hex) is a base 16 way of representing one byte. Hexadecimal uses the digits 0,1,2,3,4,5,6,7,8,9,A, B, C, D, E, F. A byte requires just 2 hex digits. Thus, FFh is a two hex digit representation of a byte. FFh is the same as 11111111b (the little 'h' designates a hexadecimal value sometimes a preceding '$' is used as in $FF)

How is the FF the same as 11111111b?

Thanks,
Scott