Injector duty cycle ?

[Razor]

Ok..what determines injector duty cycle? Is it a combination of load, AFR, o2's? Commanded enrichment within the eprom?

I have seen a car run identical times with two different chips. The difference was how direct scan displayed the injector duty cycles. If the O2's at the end of the 1/4 mile are 780 and one chip has 97 percent and the other has 72 percent, where is the skew in the way DS shows the IDC?

Is direct scan wrong in its computations... and cant be trusted for these type of measurements?

Anybody else seen this condition. Maybe if the fueling on the PE tables is cranked up, and the BLM's locked it may display wrong/different IDC numbers?

First I saw this was a year ago whereas I believe Two Lane had posted a DS recording of a high 11 second run with 50's at close to 100 percent DC on the BMCS DS board. two lane if it wasnt you..sorry..working off of memory. 50's should be good for 10's..

Thx..i'm missing something..

 

[BM Computer Src]

I have seen the inj DC vary based on the chip PE fuel trim settings and fuel pressure.

IE: TE-44, 50's, running 11.70's @ 115 MPH @ 25 PSI boost. The injector DC should be at 80-85%, roughly. However, I have seen this same scenario when the chip was programmed with too much fuel and the Fuel pressure was too low. The injectors showed close to 100%.

The fix was to increase base FP, and reduce the fuel commanded in the chip. The injector DC returned to the about normal 80-85%.

Hope this helps,
Bob

 

[Razor]

How bout if you lower the injector constant to lets say 40 lbs for a 50 lb chip. And dropped enrichment across the board...Wouldnt the chip command less fuel? And thus the lower IDC number?

Yet the injector is delivering the same fuel with two different IDC numbers displayed by DS?

Car in question fuel pressure wasnt changed between both chips. Different vendors.. same o2's and EGT's.

 

[BM Computer Src]

You will see the DC change in [just chip changes] as well, if they have different injector constants and PE fuel trim tables.

The injector constant change will act the same way as the Fuel pressure change I described earlier.

 

[bruce]

Originally posted by Razor


50's should be good for 10's..

If you want to start a headgasket collection.

Running an injector at over 90% is putting you into terriotry where the injector starts getting erratic as it goes from running normally to where it's static. As it makes this transistion it's also happening fast enough that the sample rate of DS, Turbolink, Diacom, etc, can miss the O2 drop, and the O2 if at all old ain't gonna react quick enough to catch what happened.

I had a lab guality scope on my ecm bench and an EE with a good scope look at things, and the numbers matched for PW and DC.

The cal needs to go something like this.
Set the MAF tables and scalers to get 14.7 when not in PE. So set the TPS PE enable up and work out the MAF tables.

Then set the TPS back down, and set the PE to some reasonable figure. And verify that commanded AFR is an actual AFR, yes, means using a WB. Then trim you PE vs RPM tables for best performance.

Least that's the way I do it.
Seems to work for me.

YMMV
Not to be attempted by those that are in a hurry, or without adult supervision, etc, etc

 

[Razor]

Bruce.. I understand what your saying, but the issue I'm unclear about is the way direct scan calculates injector duty cycle. I know that looking at a waveform on an oscilloscope duty cycle is the amount of "On Time" the coil is being activated to fire the selenoid(injector) (BTW I have a 5 trace w/storage)

Could the possibility of having a car run identical times and horsepower yet direct scan read over 20 percent difference between two chips?

But the IDC is whats puzzling? How does direct scan as a program measure the duty cycle? It would have to go back to the micro and with an algorithm measure the??? I know it doesnt take the info from the drivers themselves..

I'm going to look at those runs and post back..

 

[BoostKillsStres]

Razor, here's my shot at it for ya.
2 chips could be off that 20% you mention and still run close. A good example is a aftermarket stock injector chip for 93 octane. when I first started modding my car I bought DS and a chip from a very well known vendor known for not being too rich and it commanded 120%+ DC at WOT, almost all stock inj chips do go over 100%.

I'd say a car 'should' run the same at 100% DC as it would at 120% since your not going to get any more fuel in. Now weather or not a car would run the same at good IDC like 65% vs. 85% would be hard to guess unless like BM Comp said it could be because of a big difference in fuel pressure/delivery and PE table differences. Its possible to set the FP at 30 psi and command a boat load of PE vs. RPM and get close to another chip w/ 40 psi. I bet as long as you were doing tuning with a wideband you could get very close although the other driveability would take a bit of work to get the same but WOT would be pretty easy.

For the DS percentage, the formula is looking at what time is available for the inj to be on and calculating the percentage over the entire RPM range since you have less time available as RPM rises to fire the injector. Thats what the other floating dial gauge line is on the dial guage display screen for IDC. It floats and shows you where 100% is so you can look at that gauge and see when you approach or cross 100% then look at the text display page and see the actual numeric percentage.

HTH

 

[Razor]

Ok..I looked over the runs again and it looks like after 4000 RPM's one chip commands more fuel and increases the DC to close to a 100%. The AFR's according to DS also drop to mid 8's as well. when the AFR is at 10 the IDC is 70 percent. When it drops to 8.4 its like 100 percent.

Whats funny is the O2's instead of climbing due to the extra fuel stay the same.

Its like I would expect to see O2's in the 800's with 50's at 100%.

Not 760's at 100 percent one chip, and 760's at 70's with another chip.

If you drop fuel in the PE tables, you also lean the car out. So dropping fuel on the chip thats at 100 will lean it out even more? And not have 760's.

This is why the confusion.. I guess dropping injector constant and dropping PE fuel trim tables above 4000 will bring down the IDC?

Ok?

 

[bruce]

Originally posted by Razor
Ok..I looked over the runs again and it looks like after 4000 RPM's one chip commands more fuel and increases the DC to close to a 100%. The AFR's according to DS also drop to mid 8's as well. when the AFR is at 10 the IDC is 70 percent. When it drops to 8.4 its like 100 percent.

Whats funny is the O2's instead of climbing due to the extra fuel stay the same.

Its like I would expect to see O2's in the 800's with 50's at 100%.

Not 760's at 100 percent one chip, and 760's at 70's with another chip.

If you drop fuel in the PE tables, you also lean the car out. So dropping fuel on the chip thats at 100 will lean it out even more? And not have 760's.

This is why the confusion.. I guess dropping injector constant and dropping PE fuel trim tables above 4000 will bring down the IDC?

Ok?

Don't guage the engine's fueling by what the O2 is doing.
They are EGT, and back pressure sensitive.

Use the injector constant that's correct for the injectors your running. The bandaid technic of using the wrong size IC rather then changing the MAF tables, is just compounding misunderstanding what the ecm is doing. If you have rich and lean spots, then correct the MAF tables and scaler. Ya, it's more work. Ya, the car might run about the same, but the PE AFR will be anyone's guess about what's really going on.

And like I said as the injector starts to go static, it's erratic, ie you can't tune an injector that is erratic.

If your looking for 10s, and not having headaches get some bigger injectors, or add water injection.

 

[BoostKillsStres]

Basically, yes, if you either make the injector constant leaner or lower PE vs. RPM then the IDC will be lower. Trying to make a graph or comparison using the narrow band O2 is worthless, not saying that to be hostile but its the case. If you find the perfect chip tune then "MAYBE" you can use the o2 numbers for comparison but if you ever change that o2 sensor then you could be way off again.

Wideband is the only way to go. When I got my WB and did some fun stuff to set it up so I could log it into DS, then I saw the light. My chip was commanding something like a DS A/F ratio or 9.8:1 which I of course thought was way too rich however based on my FP and inj const settings I was seeing a WB reading of about 10.5:1 so I was way rich in my programming and was seeing 780 on the stock o2's. Just goes to show ya. I'm now commanding around 11.2:1 in DS and I'm seeing about 11.5:1 on the WB with my chip tuning and my current stock o2 sensor is reporting o2's in the 680's. Now I imagine it may have a problem but its never seen race gas and has great cross counts and gets good mileage and the BLM's all look fine so its stays in there.

Now I forget the formula to calculate the ms time available to fire the injectors depending on RPM but with some work you can figure it out. Certain RPM, 720 degrees of crank rotation, 6 cylinders, only firing on the intake srtoke and so on where you can simply change the RPM and then figure the total time in milliseconds available.

 

[BoostKillsStres]

Actually here's the simple formula I found from JDEstill:

Max time available for the injector to fire in milliseconds = 120,000/rpm

So if you have inj firing at 18.7ms at 5500 RPM then the percentage is 18.7/21.82=85.7%

Thats for a sequential fire by the way. Batch would be 1/2. or 18.7/2.

 

[Razor]

Originally posted by BoostKillsStres
Basically, yes, if you either make the injector constant leaner or lower PE vs. RPM then the IDC will be lower. Trying to make a graph or comparison using the narrow band O2 is worthless, not saying that to be hostile but its the case. If you find the perfect chip tune then "MAYBE" you can use the o2 numbers for comparison but if you ever change that o2 sensor then you could be way off again.

Yes to the wideband..especially for tuning.
This I will not argue

But the same o2 is being used for both runs. Done back to back. I only use the numbers as reference since its whats available to me. Chips in question are a red armstrong for 50's and a thrasher for 50's. EGTs, o2's are same with difference being IDC. No changes to fuel pressure.

Anyone want the files..i'll send em your way

BTW..Mark..ya going to Reynolds this weekend?

 

[BoostKillsStres]

I'll be glad to take a look at the files if you want to send them to my profile email, not sure how much I can help but I'll take a peek. Not sure yet if I can make reynolds, if I do make it I'll probably only spectate I have a bad tendency to lean on my car when I go to the track the very little bit that I do so its best I do it closer to home as the tow bill will be less plus I'm still chip tuning and haven't yet hooked the alky back up and tuned a chip for it yet so there's no way I would be ready to run what I want to run so I'll just save it for later.

 

[Razor]

Come and spectate.. its great.. I wont be racing either, but watching people break stuff

Got to make a showing of support.. Be glad this is a lot closer than Bowling Green, Bristol, Vegas, etc...

Next year..Southern Nationals in my back yard..yay..ya bet i'll be racing..for sure.

I'll send ya the files latter today.. theyre not of my car, but was asked to look at them.

 

[ijames]

The basic warmed up closed loop fueling equation is pw=inj const * maf * fuel-air-ratio * (1/rpm) * blm * int, times a few trims for battery voltage and such, times a calibration constant for the maf (air meter base constant). When in pe mode this gets multiplied by (pe-rpm trim + pe-tps trim) * time-in-pe, and the fuel-air-ratio comes from a table. The ecm doesn't calculate duty cycle, it just calculates pw, and if the pw is longer than the available time then the injector driver just stays on (note I said driver not injector; I'm ignoring fine points of real world injector behavior). Direct scan reads the memory location where the ecm stores the final calculated pw, and its reporting of pw has been verified by a few people with oscilloscopes and data loggers. DS then takes the pw and rpm and calculates duty cycle using dc=pw*rpm/120000*100%, and again, so far as I know it reports this correctly. The air-fuel ratio that ds reports is calculated by ds and is not the reciprocal of the fuel-air-ratio used by the ecm in the fueling equation.

Since ds does not report all the values that go into the fueling equation you cannot calculate what the pw should be under some set of circumstances just from ds data. You have to read the chip itself and manually check all the fueling parameters to compare the pw that two chips will produce under some given set of inputs. Oh, in closed loop the ecm uses the O2 reading to vary the INT to cause the engine to continuously alternate between rich and lean so it will average stoichiometric because that is what makes the cat work best.

So that is how the pw gets set. As for the O2 volts, as others have said the O2 sensor is not accurate out at air-fuel ratios of 12-ish, with varying EGTs and exhaust manifold pressures. It is very useful for maintaining a tune, and with other inputs like looking at plugs and egt can guide the process of getting a good WOT tune, but just because two cars show 780 mV does not mean that they are running the same air-fuel ratio.