PDA

View Full Version : B4005 Small Pulse Adjust How does GM create this table?



onfire
July 17th, 2007, 02:48 AM
I've looked at a lot of GM files...B4005 has many differnt "looks" depending on what car /injectors it's in. It's not a standard slope.

I've had to change mine bigtime to get 86lb low imp injectors to work in COS5. I've watched some of the changes by using RR in realtime....

I'm getting close but would like to understand the small pulse adjustments better. The nice thing about the table is that it will command an adder where you want it....plus it can be a negative adder which is nice with big injectors.

It and the voltage oddsets are the keys to big injector tuning.

SSpdDmon
July 17th, 2007, 08:16 AM
I've looked at a lot of GM files...B4005 has many differnt "looks" depending on what car /injectors it's in. It's not a standard slope.

I've had to change mine bigtime to get 86lb low imp injectors to work in COS5. I've watched some of the changes by using RR in realtime....

I'm getting close but would like to understand the small pulse adjustments better. The nice thing about the table is that it will command an adder where you want it....plus it can be a negative adder which is nice with big injectors.

It and the voltage oddsets are the keys to big injector tuning.
Please elaborate more on tuning for offsets and pulse adjusts....

onfire
July 17th, 2007, 02:21 PM
Only time I needed to adjust the spa,vof,etc was for the 86lb injectors. The 42lb svo's and the 60lb Mototron worked well with a proper VE table and a proper IFR table. The 86lb low imp injectors require more tuning. Basically they are so large that if you only adjust the VE and the IFR the minimum pulse at idle is around 1.8ms.....this is too much for a 14.7:1 idle....I was getting 11.5:1 too rich.

Did not matter if I set the Minimum pw to 0.6 and the Default pw to 0.6...I still could not get the ecm to go lower than 1.8pw at idle.

So you are forced into altering the offsets. Basically they are adders. You start with a base pulse width and you add to that with the voltage offset and the spa offset....together it makes the minimum pw around 1.8 So you have to lower the offsets to get the minimum pw lower. By lowering both table I now can get a 1.38pw at idle and can go as lean as 16 :1 if I want it.

I'm just curious if anyone knew how GM calculated their spa curve. I had to do mine with trial and error based on RR feedback in realtime and a wideband.

SSpdDmon
July 17th, 2007, 03:28 PM
I asked because I noticed the same problem with my SVO's during decel prior to DFCO kicking in (and they're only 30's). No matter how much I lopped off the VE/IFR/MAF tables, I'd get a rich condition (13.5:1). Min. PW was in the 1.8ms range too. I think it may also be contributing to a rich condition below 1600rpms. I'd be very curious to know how as well how GM got their numbers...

2002_z28_six_speed
July 18th, 2007, 03:20 AM
hmmmm this is VERY interesting!!

Listening in....

From the sound of what has been stated it almost seems that the SPA is being used to patchwork anamolies in the injector.

In onfire's case, it sounds like he is watching what frequencies are hitting the injector [miliseconds/pwm] and adjusting for how the frequencies [miliseconds/pwm] are affecting the injector. IS THAT WHAT THIS IS???

Adjusting for the frequency of the signal?

Thank you.

onfire
July 18th, 2007, 03:32 AM
hmmmm this is VERY interesting!!

Listening in....


How tight is the variance on this aftermarket injectors? If you put in one set would the table numbers work for the next? I am sure they bench flowtest them. Just asking.

Thank you.

I had mine flowmatched at Racetronix so they would be within 1%.

2002_z28_six_speed
July 18th, 2007, 03:42 AM
Changed my post a little bit after I stared at the table for awhile. Sorry.

redhardsupra
July 18th, 2007, 04:54 AM
flow if the injector is not linear, it has at least 4 distinct stages with 4 different fueling rates. first is the offset stage, when the injector is building up working against the spring, so there's no flow yet. second is the opening stage, at which the injector is changing the flowrate. third is the 'proper' fueling at a rate dictated by the rated flow, operating fuel pressure and manifold vacuum (if not referenced). fourth is the closing stage, when the injector goes from full flow rate to zero.

http://www.marcintology.com/tuning/INJECTOR_nonlinear.png

the actual fuel mass injected is the 'area under the curve' of the various fuel rates in time of injection (integrate from 0 to IPW across the four different functions for IFR).

if you look at ford tuning, they treat it as two stages, they actually have two flow rate numbers, and a 'breakpoint' which is the time the first fuel rate is used for, and everything after that is calculated with the second fuel rate. of course ford tuning people with their natural tendency of 'screw the numbers, right is whatever works' attitude tweak the numbers randomly until they get it to the point where the fueling is acceptable.

GM did it differently. Once the airmass is obtained (SD calcs or MAF lookup) the fuel mass needed is airmass*AFR. the PCM knows the IFR value at the current MANVAC so it uses that to get the pulsewidth it thinks it should use (IPW=fuelmass/IFR). however, this is wrong, because that's assuming the IFR at the various stages of the time of injection is the same, and it isn't (paragraph #1). the design people knew that, thus created a 'lets make up for the discrepancy with another table' table ;) this is the short pulse adder table.

short pulse adder is the time required (at the official IFR) to make up for the difference in the expected fuel mass, and the fuel mass that actually goes in. at long pulse widths, the areas with the non-official IFR account for a small portion of the total fuel mass, thus are just ignored. however at short pulse width they make up a significantly larger portion of the fuel mass, thus cannot be ignored. That's why we have the short pulse adder.

there's also the offset table, that seems to be mostly dependent on the physical characteristics of the injector (resistance?), operating MANVAC and voltage. offset seems to always get added to the pulsewidth, long or short.

i'm working on a spreadsheet that will convert injector data into short pulse adder table. this is the best explanation i've been able to come up with. of course it might be all horsepoo, so please let me know if i'm full of it. if you got better explanations/data please let me know so we can get this short pulse adder spreadsheet done quicker, we need the fueling to be more precise and this should help a lot.

SSpdDmon
July 18th, 2007, 05:56 AM
What I think I know -

When it comes to injectors, there's 3 main factors to pay attention too:

1) When the injector opens (which I thought was offset)
2) How long it stays open (IPW)
3) How much fuel is dumped (IFR)

From what I've read, the software allows us to alter flow rate based on injector performance and 'offset vs. ECT' in the timing tables. Then there's the SPA and the IPW voltage adjustments are simply adder tables to the base IPW, which is calculated using other sensors/tables.

I haven't read anything about adjusting the timing of the injector vs. the ECT. All I've seen is people posting some offset vs. voltage tables for various injectors, which typically commands a bigger pulse width. My problem is, I can't shake the rich decel condition I have from my SVO 30's even with the stock voltage correction table.

I'm losing train of thought....I'll be back later. We definitely need to crack this puzzle though. :lol:

joecar
July 18th, 2007, 06:07 AM
So is B4005 the area under the sloped parts of that curve...?

redhardsupra
July 18th, 2007, 06:32 AM
no, area is fuelmass, b4005 is time.

if the pcm tells the injectors to dump let's say 2.5ms of fuel, it will get less than that, because the injector at first doesnt flow at all, and then it flows less then the official 'static' flow rate. short pulse adder is the amount of time that's needed to make up for that by artificially telling the injectors to open longer than they would have to open for if they provided a perfectly linear flowrate.

ok, here's a pic, it should be easier to visualize. area under the red curve is going to be bigger than the blue curve. thus we must make the red curve longer to effectively make up for what the blue is missing.

http://www.marcintology.com/tuning/INJECTOR_nonlinear2.png

2002_z28_six_speed
July 18th, 2007, 07:21 AM
This is helping tons! I think I am starting to understand.

joecar
July 18th, 2007, 07:32 AM
Ah, I see.

joecar
July 18th, 2007, 10:09 AM
Is the red curve supposed to include the offset at the start...?

redhardsupra
July 18th, 2007, 10:19 AM
yes, that's what it would be if the injector would start instantly, hold perfect and closed instantly. that's why it's called theory, and why it's only an estimator for the real thing.

SSpdDmon
July 18th, 2007, 02:18 PM
If the voltage correction is vacuum referenced kind of like the VE, why do we need it since it's only an adder?

Also, why does the adder increase with vacuum? I would think more vacuum sucking the fuel out of the injector during decel would require less voltage correction than WOT where manifold vacuum is minimal.

2002_z28_six_speed
July 18th, 2007, 02:29 PM
If the voltage correction is vacuum referenced kind of like the VE, why do we need it since it's only an adder?

Also, why does the adder increase with vacuum? I would think more vacuum sucking the fuel out of the injector during decel would require less voltage correction than WOT where manifold vacuum is minimal.


That is a good observation about {B3701}.

5.7ute
July 18th, 2007, 03:05 PM
Possibly the vaccuum holds injector closed & you need more correction.
BTW great discussion you have going on here.

onfire
July 19th, 2007, 04:42 AM
If the voltage correction is vacuum referenced kind of like the VE, why do we need it since it's only an adder?

Also, why does the adder increase with vacuum? I would think more vacuum sucking the fuel out of the injector during decel would require less voltage correction than WOT where manifold vacuum is minimal.


It's because the injectors (if wide open) would flow much more with a negative pressure on it pulling the fuel out, but the above graph shows how the start up and shut down reduces that flow....so the voltage/kPa table adds it back .

onfire
July 19th, 2007, 04:48 AM
Another thing that really skews the SPA table is adding low imp injectors. Low imp injectors open and close much faster than stock high imp injectors and don't need as much offset to get the same area under the curve.

What gives me heartburn is I've reduced the Minimum Pulse Width to 0.3 and the Default Pulse width to 0.3 and it does not lower the idle pw under the magic 1.8. The SPA and the Voltage Offset tables are adding to something else to create that minimum pw number.

SSpdDmon
July 19th, 2007, 05:01 AM
That's my point. If it's a vacuum issue, why not just deal with it in the flow rate table.

I too have a mad hatrid for the number 1.8 as I hit it all of the time during decel. On one hand, I want to learn why it's bottoming out there and figure out how to fix it the right way. On the other though, I just want the problem gone and am two steps away from the trial and error mentality.

redhardsupra
July 19th, 2007, 05:16 AM
Another thing that really skews the SPA table is adding low imp injectors. Low imp injectors open and close much faster than stock high imp injectors and don't need as much offset to get the same area under the curve.

What gives me heartburn is I've reduced the Minimum Pulse Width to 0.3 and the Default Pulse width to 0.3 and it does not lower the idle pw under the magic 1.8. The SPA and the Voltage Offset tables are adding to something else to create that minimum pw number.

well yea, this pcm is not exactly designed to work with low impedance injectors, so the fact that it works at all amazes me ;)

you can tell the computer whatever you want about the minimum pulse width, but ultimately it's gonna to perform something it cannot physically execute. that's why it's important to find a way to convert the injector data into something that the PCM can understand. post all the data you got on your injectors, i'm working on a spreadsheet, might as well work on some real numbers.

SSpdDmon
July 19th, 2007, 05:29 AM
well yea, this pcm is not exactly designed to work with low impedance injectors, so the fact that it works at all amazes me ;)

you can tell the computer whatever you want about the minimum pulse width, but ultimately it's gonna to perform something it cannot physically execute. that's why it's important to find a way to convert the injector data into something that the PCM can understand. post all the data you got on your injectors, i'm working on a spreadsheet, might as well work on some real numbers.
Any ideas on SVO 30's while you're at it???

redhardsupra
July 19th, 2007, 05:34 AM
gimme data and i'll do it

jfpilla
July 19th, 2007, 06:20 AM
I've looked at a lot of GM files...B4005 has many differnt "looks" depending on what car /injectors it's in. It's not a standard slope.

I've had to change mine bigtime to get 86lb low imp injectors to work in COS5. I've watched some of the changes by using RR in realtime....

I'm getting close but would like to understand the small pulse adjustments better. The nice thing about the table is that it will command an adder where you want it....plus it can be a negative adder which is nice with big injectors.

It and the voltage oddsets are the keys to big injector tuning.

onfire,
This is an interesting subject and I know nothing about it. It's hard to find answers. You might want to look at an LS7 file. The table values related to injectors are very different from LS1's. I have a crate LS7 in a C5. The injectors calculate to 41.8#s. I was having some surging problems and low rpms at hot startup. After putting in all the LS7 tables related to fueling ie: pw, default pw, fl.rate, min. pw etc. All the issues went away. I should log pw and see if it's as low as in the min. pw tables. It didn't occur to me until I read this.

SSpdDmon,
Am I not understanding? As manifold vacuum decreases pulse width increases and visversa in my tables.

onfire
July 19th, 2007, 07:19 AM
well yea, this pcm is not exactly designed to work with low impedance injectors, so the fact that it works at all amazes me ;)

you can tell the computer whatever you want about the minimum pulse width, but ultimately it's gonna to perform something it cannot physically execute. that's why it's important to find a way to convert the injector data into something that the PCM can understand. post all the data you got on your injectors, i'm working on a spreadsheet, might as well work on some real numbers.


Here's where I'm going to start a firestorm...lol..

With 86lb injectors or larger....the "IFR calculator" really jacks the VE table up wayyyy over 100% on cruise, etc....so that's an issue I'm still playing with by lowering the IFR slope by different percentages.....

I'll get you some data tonight...would have it now but my RR skewed all my tuning data and I had to start over today without it.

onfire
July 19th, 2007, 07:20 AM
I've looked for a LS7 file but did not find one at the Holden site. Do you know where one is? TIA.




onfire,
This is an interesting subject and I know nothing about it. It's hard to find answers. You might want to look at an LS7 file. The table values related to injectors are very different from LS1's. I have a crate LS7 in a C5. The injectors calculate to 41.8#s. I was having some surging problems and low rpms at hot startup. After putting in all the LS7 tables related to fueling ie: pw, default pw, fl.rate, min. pw etc. All the issues went away. I should log pw and see if it's as low as in the min. pw tables. It didn't occur to me until I read this.

SSpdDmon,
Am I not understanding? As manifold vacuum decreases pulse width increases and visversa in my tables.

SSpdDmon
July 19th, 2007, 07:21 AM
onfire,
This is an interesting subject and I know nothing about it. It's hard to find answers. You might want to look at an LS7 file. The table values related to injectors are very different from LS1's. I have a crate LS7 in a C5. The injectors calculate to 41.8#s. I was having some surging problems and low rpms at hot startup. After putting in all the LS7 tables related to fueling ie: pw, default pw, fl.rate, min. pw etc. All the issues went away. I should log pw and see if it's as low as in the min. pw tables. It didn't occur to me until I read this.

SSpdDmon,
Am I not understanding? As manifold vacuum decreases pulse width increases and visversa in my tables.
Right...during decel....high man. vacuum...my pulse widths are too big. No matter what I do to the IFR, MAF, VE, or min. PW tables, it wants a 1.8ms pulse width and nothing smaller. I'm about to just lop it off of the voltage table where I see the problem (Man. vacuum >74kPa).

jfpilla
July 19th, 2007, 08:25 AM
I've looked for a LS7 file but did not find one at the Holden site. Do you know where one is? TIA.

You will get one soon.

I logged coming home and the pulse width at idle and decel are a little higher. So PW was not the issue. It could be the Commanded fuel when cranking values. They are much lower for the LS7.Or, maybe it's just a better balanced tune as I worked on it.
It's raining so I could not do a WOT decel to observe min PW. I will, since I'm now curious.
http://forum.efilive.com/images/statusicon/user_invisible.gif

TAQuickness
July 19th, 2007, 10:05 AM
Right...during decel....high man. vacuum...my pulse widths are too big. No matter what I do to the IFR, MAF, VE, or min. PW tables, it wants a 1.8ms pulse width and nothing smaller. I'm about to just lop it off of the voltage table where I see the problem (Man. vacuum >74kPa).

It's been a while since I tuned my SVO 42's, but I was getting stuck on the 1.8 ms PW as well. I was eventually able to get the commanded PW down to 1.5, but then I discovered that anything less than a 1.78ms PW and the injector wouldn't fire (AFR's in the 20's).

I ended up tweaking the tune so I would see no less than a 1.79ms PW and made my DFCO a bit more agressive.

onfire
July 19th, 2007, 12:29 PM
Red,

Here's the lastest 86lb low imp tune.

SSpdDmon
July 19th, 2007, 01:41 PM
It's been a while since I tuned my SVO 42's, but I was getting stuck on the 1.8 ms PW as well. I was eventually able to get the commanded PW down to 1.5, but then I discovered that anything less than a 1.78ms PW and the injector wouldn't fire (AFR's in the 20's).

I ended up tweaking the tune so I would see no less than a 1.79ms PW and made my DFCO a bit more agressive.
I found this out tonight as well. I got the PW's down to 1.5mS and saw high teens and low twenties on the WB for decel. I bumped it back up to the mid 1.7's now and it's better. Going to target a 1.7ms PW and see what that does.

redhardsupra
July 20th, 2007, 06:03 AM
http://www.picotech.com/auto/tutorials/fuel-injection.html

take a look at this. injector voltage and current graphed as a function of time on a working injector. this would give us the offset/open/close times, and probably gives us an idea on the shape of the IFR curve for the closing phase.
anyone's got a scope to try this out?

oh this is even better, they actually chop up the fueling stages just like I did, maybe I'm onto something here:
http://www.lindertech.com/newsletters/nwk7~03.pdf (http://www.lindertech.com/newsletters/nwk7%7E03.pdf)


and another one:
http://www.automotivetestsolutions.com/images/escope/fuelingectortutorial/FuelInjectorWaveforms.htm

Ira
July 20th, 2007, 07:31 AM
anyone's got a scope to try this out?


I have a scope and if anyone want's to show up at my place in West LA, we can have a look and post the results.

Ira

redhardsupra
July 20th, 2007, 07:37 AM
can't you just hook it up to your car and see what the waveforms look like? it would be really nice to have few different kinds to compare, like lucas, ford/bosch, gm stockers, holly...

redhardsupra
July 20th, 2007, 09:08 AM
Red,

Here's the lastest 86lb low imp tune.

do you have some 'official' data on the injectors? are they the racetronix ones or something else (rceng?)

onfire
July 20th, 2007, 01:28 PM
http://www.racetronix.com/3105FM.html

redhardsupra
July 20th, 2007, 01:30 PM
what fuel pressure do you run them at? do you have a manifold referenced regulator? does your fuel pressure drop off at higher boost/rpm?

onfire
July 20th, 2007, 01:54 PM
48 base psi. Boost referenced only 1:1 . Twin Bosch 044's so there's zero fuel pressure drop...when I was running Mototron 60's I had the base pressure at 74psi ramping up to 91psi....the pumps are now laughing at a base pressure of 48...

Chalky
July 21st, 2007, 02:19 AM
I found this out tonight as well. I got the PW's down to 1.5mS and saw high teens and low twenties on the WB for decel. I bumped it back up to the mid 1.7's now and it's better. Going to target a 1.7ms PW and see what that does.

Jeff:

Any update? For those of us running the SVO 30's, I wonder if another injector would not work better to address idle issues. Maybe something with lighter internal parts that would respond quicker to minimum IPW's.

SSpdDmon
July 21st, 2007, 03:46 AM
I gave up... :lol:

Seriously, I accepted the fact that I wasn't working with any sort of data and was only making trial/error type adjustments. I don't like working like that.

I just decided to turn the DFCO delay down to 0.1 seconds. That way, I'm not bottoming out the IPW's. The easy thing then is to increase the throttle cracker a little so the car doesn't engine brake too hard, yet stays in DFCO. :)

redhardsupra
July 30th, 2007, 02:58 AM
jeff, since you're mr experimental, why dont you try this:
calculate the real ipw from dyncylair/(AFRwb*IFR), then subtract the appopriate offset from it, and the difference should be the the short pulse adder (i'm not 100% on this but it's an idea).

SSpdDmon
July 30th, 2007, 03:54 AM
jeff, since you're mr experimental, why dont you try this:
calculate the real ipw from dyncylair/(AFRwb*IFR), then subtract the appopriate offset from it, and the difference should be the the short pulse adder (i'm not 100% on this but it's an idea).
I'll give it a try next week or so. I have bigger fish to fry (or master cylinders at least) before I get to the track this Friday. Thanks for the idea none the less! :D

BTW - is this based on the assumption that I used the injector calc. to determine IFR values? Right now, I have everything running off of a backwards mapped IFR table. In other words, I did an IFR tune on a pure stock MAF calibration to determine my IFR line. I am currently in SD with that IFR line and theoretically, I could switch back over to the stock MAF curve and all would be well.

mr.prick
December 2nd, 2008, 12:17 PM
flow if the injector is not linear, it has at least 4 distinct stages with 4 different fueling rates. first is the offset stage, when the injector is building up working against the spring, so there's no flow yet. second is the opening stage, at which the injector is changing the flowrate. third is the 'proper' fueling at a rate dictated by the rated flow, operating fuel pressure and manifold vacuum (if not referenced). fourth is the closing stage, when the injector goes from full flow rate to zero.

http://www.marcintology.com/tuning/INJECTOR_nonlinear.png

the actual fuel mass injected is the 'area under the curve' of the various fuel rates in time of injection (integrate from 0 to IPW across the four different functions for IFR).

if you look at ford tuning, they treat it as two stages, they actually have two flow rate numbers, and a 'breakpoint' which is the time the first fuel rate is used for, and everything after that is calculated with the second fuel rate. of course ford tuning people with their natural tendency of 'screw the numbers, right is whatever works' attitude tweak the numbers randomly until they get it to the point where the fueling is acceptable.

GM did it differently. Once the airmass is obtained (SD calcs or MAF lookup) the fuel mass needed is airmass*AFR. the PCM knows the IFR value at the current MANVAC so it uses that to get the pulsewidth it thinks it should use (IPW=fuelmass/IFR). however, this is wrong, because that's assuming the IFR at the various stages of the time of injection is the same, and it isn't (paragraph #1). the design people knew that, thus created a 'lets make up for the discrepancy with another table' table ;) this is the short pulse adder table.

short pulse adder is the time required (at the official IFR) to make up for the difference in the expected fuel mass, and the fuel mass that actually goes in. at long pulse widths, the areas with the non-official IFR account for a small portion of the total fuel mass, thus are just ignored. however at short pulse width they make up a significantly larger portion of the fuel mass, thus cannot be ignored. That's why we have the short pulse adder.

there's also the offset table, that seems to be mostly dependent on the physical characteristics of the injector (resistance?), operating MANVAC and voltage. offset seems to always get added to the pulsewidth, long or short.

i'm working on a spreadsheet that will convert injector data into short pulse adder table. this is the best explanation i've been able to come up with. of course it might be all horsepoo, so please let me know if i'm full of it. if you got better explanations/data please let me know so we can get this short pulse adder spreadsheet done quicker, we need the fueling to be more precise and this should help a lot.

did you ever make the spreadsheet?

redhardsupra
December 2nd, 2008, 12:39 PM
yes, i made like 8 spreadsheets, but none of them are agreeing with anything else, so something's wrong. Greg Banish claims to have something, might wanna look into taking his class.