c4boom

first off how do you determine the duty cycle of your injectors. i would like to know this so that i dont go static on my wot tuning and how do you add fuel for wot conditions only

RBob

iTrader: (1)

Duty cycle is calculated from RPM and injector PW:

DC% = PWmSec / (1 / (RPM / 300) )


During WOT operation there is an AFR vs. RPM table.

RBob.

JPrevost

Just thought I'd tell anybody else that's reading this that this duty cycle calculation is ONLY for the TBI firing strategy. Port injection has twice as long to fire and sequential has double that.
Example; TBI injectors at 6000rpm have only 5ms of time to open.
DC%=5ms/(1/(6000rpm/300))=100%
For batch fire port injection (TPI);
DC%=10ms/(1/(6000rpm/600))=100%
SEFI like most LT1s and newer cars;
DC%=20ms/(1/(6000rpm/1200))=100%
Also a DC% of 100 is worthless because you've now lost complete control over the fuel delivery of the engine. Not even a normal decrease in volumetric efficiency will let you regain control at higher rpms.

c4boom

ok how do you determin the PW by looking at the maps in tuner pro

dimented24x7

iTrader: (2)

You dont... The actual pulsewidth is the result of using the VE, modified MAP term, etc in a calc to determine teh fuel flow for each cylinder. Either on a bench or in the car is teh only place you can really detirmine what teh PW truely is.

BMmonteSS

You can output your PW into the aldl stream pretty easily, you can use JPrevost neat little aldl .ecu file to change what gets sent out the data stream in what place. I just spit my PW out the prom ID stream, then converted it in excell.

c4boom

thats cool where is a good place to get this file

dimented24x7

iTrader: (2)

EDIT: Unghhh... I really feel stupid now

c4boom

say what? i think i am in over my head again some one please throw me some floties

JPrevost

http://www.moates.net/files/3)%20Bin...DL%20patch.zip
That's the stuff I put together. It has a pretty good readme file that should make sence (has examples).

c4boom

thank you
it seams like i keep reading but my brain dose not suck it up i have been working on my truck for months and have just now started to get some good results

i spent some time looking at the txt files i did not see a address for the DC would i have been under some thing that i did not see or is it part of a calculation that i am going to have to work off of

Dominic Sorresso

c4boom,

You won't see anything for DC. You need to feed the PW through the ALDL stream. Then that raw data is converted through a formula to arrive at Duty Cycle. Check Jon's "Read Me" for the correct address on PW.

c4boom

i was looking for some thing but ill i can find is bpw

dimented24x7

iTrader: (2)

I just realized that I messed up with the addresses. Its AA and AB. I misread them.

Ok, what you do is you type in '00AB' for the first byte of the PROM ID. Then you type in '00AC' for the second byte. When you go to datalog, there will now be the pulsewidth in raw form in place of the prom ID. This is in place of the origional adress', which where D002 and D003. The 0xAA and 0xAB are the two addresses that are the 16 bit synchronous base pulse width in the memory, or the time that each injector is supposed to be open.

In your datalog, if you want to know what your pulswidth is in milliseconds, you multiply the number you see as the prom ID by .01526. That will give you the pulsewidth in milliseconds instead of counts. Thats pretty much it.

EDIT: made correction.

dimented24x7

iTrader: (2)

Dont know if JP has the formula commented in his ALDL stuff but the DC isnt really too terribly hard, either.

Each injector fires twice per engine revolution. That means that it only has half the time it takes for a revolution to fire. The duty cycle itself is the ammount of time that the injector is open divided by the total time that the injector can be open. Lets say it takes 20 milliseconds for the engine to complete one revolution at 3000 rpm. Since each injector will have to fire twice, that gives a maximum time of 10 milliseconds that the injector can fire. After 10 milliseconds, the injector will have to fire again.

Ok, so we have a time of 10 milliseconds that the injector has to fire. Say the pulsewidth from the datalog is 5 milliseconds. 5/10 is .5, or 50%. That means that the injector is open for 50% of the time. The other 50% of the time, its closed. If the pulsewidth was 8 milliseconds, then it would be open for 80% of the time. If it was 12 milliseconds, then it would be open all the time, since 12 milliseconds is longer then the 10 milliseconds that the injector has to fire. After 10 milliseconds, its time for the injector to fire again.

If I didnt make any math mistakes, the duty cycle as a percentage can be found by the following formula:

(raw PW x RPMs)/19660.

The raw pulsewidth is the number that is spit out in the datalog. The RPMS are the engines speed that the pulsewidth was recorded at, in revs. per min. of course.

DieKnow

I just measure it at the injector. Use any meter or scope that has a duty cycle option and connect to injector wire.

Grumpy

If your broke or code challenged, a dwell meter will work in a pinch. *Dwell* is the amount of time the points are closed, expressed in degrees. Soo, on an application where you see ~23d dwell that would correspond to 50% DC, and 45 would be 100%.

All of this is much easier to work out on an ecm bench rather then beating the engine to death just gatering sampling data to get a rough idea of where you are.

Saturn5 is going to do another batch of boards, and they pay for themselves just in gas saved.

Dominic Sorresso

c4,

This is how you make the change in the ALDL stream to cappture PW in the PROM ID slot of the data stream. With this raw data you then convert it over to PW and then Duty Cycle. dim, I think the correct address for PW in a 7747 is AB, AC.
http://home.comcast.net/~surrexi2/ALDL_Mod.bmp

Dominic Sorresso

c4,

This is how you make the change in the ALDL stream to capture PW in the PROM ID slot of the data stream. With this raw data you then convert it over to PW and then Duty Cycle. dim, I think the correct address for PW in a 7747 is AB, AC.
http://home.comcast.net/~surrexi2/ALDL_Mod.bmp

dimented24x7

iTrader: (2)

I thought he was talking about a 8746... I think thats what I put the first time for the addresses but then I saw jons post and thought 8746 and forgot the thread title. I really had a serious brain fart there.