BPW and Inj. DC

[Maxtor]

Ok - it's a long day at the in-laws and I'm trying to understand a few things better.

When referring to Datamaster BPW - how does this correlate to injector duty cycle (%)? Is there a conversion formula? I read somewhere that at WOT 80% duty cycle equals a BPW of approximately 17ms, which is essentially the point where the injectors reach a static condition - is this right (stock injectors)?

I'm looking for the point(s) where the injectors are basically out of headroom at WOT as recorded by Datamaster.

 

[leroy]

At 5000 rpm, you are spinning 83.3 revolutions per second. Each revolution takes .012 sec or 12 msec.

Does that get you started?

Jim

 

[Maxtor]

Jim - how does the the time it takes the engine to make 1 revolution tie into BPW? Does it indicate you are static when BPW exceeds approximately 80% of the time it takes to make a revolution at that point?

For example:

Assume Datamaster records a BPW of 16.58mS at 3150RPM.
At 5000RPM the engine takes 12mS to make one revolution and makes 83.3 revolutions per second.
So - at 3150RPM, the engine is making 52.5 revolutions per second and takes 19mS to make one revolution.

So - at this single data point, Datamaster is telling me BPW is 16.58mS and one revolution takes 19mS. 16.58/19 = 87%.

Does this mean my injector DC is 87% at this point (essentially - saturated)?

 

[leroy]

Yes, you are really close. Remember that an intake cycle starts every other revolution though. At 5000 rpm (12msec per revolution), it's 24 msec between the start of each intake cycle.

At 6000 rpm we have 20 msec between intake strokes.....80% duty cycle would be 16 msec.

I would also consider static to be 100% duty cycle.

Jim

 

[Maxtor]

Thanks, Jim. Based on this and looking at my 7.7sec 1/8 mile pass, It looks like I have plenty of injector left through the 1/8 mile. I just wanted to ensure I understood how Datamaster calculated Inj. DC and what the corelation was with BPW.

Thanks again.

 

[kbbeckius]

Ok, I'm still confused. Datamaster shows Inj DC%. But it doesn't seem to calculate it properly, but according to some web sites it does.

Here's how some (including Datamaster) calculate Inj DC
%DC= [pulsewidth/period]*100
where period=1/(RPM/60)

from Datamaster (one of my 93's files)
RPM=4000
BPW=14.25
therefore period=15ms
and %DC=95
which is exactly what Datamaster shows

Where does this factor in?
"Remember that an intake cycle starts every other revolution though"
yes, 4 stroke engine, up(compression), down(power), up(exhaust), down(intake), so every two rotations the intake valve on a cylinder opens.

Our injection system isn't sequential though, it's batch fire.

So how many times do our injectors fire per revolution (or two)?

Obviously with batch fire, each cylinder is in a different position in the 4 stroke cycle, since we have an even fire V6 their position is seperated by 120 degrees. Most distributors run at half crank speed so in two revolutions all cylinders fire. Some web sites say that batch fire systems fire the injectors before the valve opens, and again when the valve is open, that would mean the injectors fire twice per two revolutions, or once per revolution. If all cylinders fire in two revolutions, do the injectors fire once per revolution or once per every two revolutions? Your statement leads be to think that you believe the injectors fire once every two revolutions. But from what I gathered about batch fire systems the injectors would fire every revolution.

Now I'm really confused, I just keep thinking in circles, like a dog chasing it's tail!!

I also noticed that in Datamaster the injDC% at times exceeds 100% which should be impossible, shouldn't it? Is it possible, the injectors are firing longer than the time it takes the engine to make one revolution, but then when are the injectors going to fire again? Since the injection event took longer than one revolution.

I would be inclined to believe that you are correct since I don't see how you could have more than 100% duty cycle. Why is the calculation wrong in Datamaster then?? An error in the program??? And why do others use the same calculation as Datamaster??? Different injector firing patterns?

 

[leroy]

Very good post kbbeckius.

I've seen it calculated both ways and it is confusing.

Here's how I think it works, and I am contradicting what I posted earlier:

Datamaster figures pulsewidth based on each engine revolution, as occurs in batch fire synchronous fueling. Synchronous fueling is timed with a distributor reference pulse. However, the injectors can be fired twice between intake strokes (each engine revolution), simulating sequential injection.

I think the second firing is the same as asynchronous fueling. At any rate, asynchronous fueling may account for duty cycles above 100%.

Maybe I should finish building the ecm test bench and check it out for sure.

Anyone else want to elaborate?

Jim

 

[BillC]

So how many times do our injectors fire per revolution (or two)?

Obviously with batch fire, each cylinder is in a different position in the 4 stroke cycle, since we have an even fire V6 their position is seperated by 120 degrees. Most distributors run at half crank speed so in two revolutions all cylinders fire. Some web sites say that batch fire systems fire the injectors before the valve opens, and again when the valve is open, that would mean the injectors fire twice per two revolutions, or once per revolution. If all cylinders fire in two revolutions, do the injectors fire once per revolution or once per every two revolutions? Your statement leads be to think that you believe the injectors fire once every two revolutions. But from what I gathered about batch fire systems the injectors would fire every revolution.

Actually, the ECM can be programmed to fire the injectors once per crank revolution or once per intake stroke (every other crank revolution). In terms of our ECM, once-per-revolution injection is called "Synchronous" fueling and every-other-revolution injection is call "Quasi-Asynchronous" fueling ("Quasi-Async", for short).

The stock chip, and (I believe) the Ultimate Chips for stock injectors, use synchronous fueling. The 50# programs use Quasi-Async, because it effectively doubles the "window" in which to inject fuel, so it makes it easier to control the amount of fuel injected at low RPMs (like at idle). There is actually an option in the chip to switch between Quasi-Async (at low RPMs) and Synchronous (at higher RPMs), but BG was never happy with the transition and decided to keep the 50# programs always Quasi-Async.

BTW, there is also "Asynchronous" fueling in our ECMs, but this is primarily used for AE (Accel Enrichment). When in Async fueling, the computer just injects extra fuel when it feels like, without regard for RPM.

I also noticed that in Datamaster the injDC% at times exceeds 100% which should be impossible, shouldn't it? Is it possible, the injectors are firing longer than the time it takes the engine to make one revolution, but then when are the injectors going to fire again? Since the injection event took longer than one revolution.

It is very possible for the ECM to command an injector duty cycle greater than 100%. All it means is that the ECM wants the injector to stay open longer than the amount of time physically available for a revolution. This is an indication that your injectors are maxed out.

 

[kbbeckius]

Ok, reading the Turbo P4 doc, helps explain fuel delivery modes, section 9.2, excellent reading!!

I was thinking that the injectors were always energized at the same time in relation to crankshaft position. I thought the injectors were triggered like, say every time cylinder 1 passes TDC, the injectors would open. I realize now that while that probably occurs most of the time with sequential fuel injection, in batch fire injector timing relative to crankshaft location is pretty much irrevalent.

I'm still a little confused about the need for Asynchronous Fuel Delivery. Couldn't the pulse width in synchronous mode be increased sufficiently to provide enough fuel for power enrichment?? That is "IF" the injectors were sized properly. It looks to me, like Asynchronous mode is just some scheme to make up for injectors sized too small. Since we shouldn't be driving the injectors at duty cycles higher than 100%, its hard on injectors, right?? Mabye not that bad, my near stock 93', shows duty cycles at and over 100%, at 125,000 miles the injectors seem to be working pretty well.

How does Datamaster come up with values for the BPW? From the tables in the chip? or from actual injector duty cycles? This would affect how it calculates duty cycle. Also does Datamaster take into consider which fueling mode the ECM is running? or does it matter? I guess it wouldn't matter since duty cycle is just the relationship between pulse width and the time for one engine revolution.

Thanks for helping me understand this better.

Brad

 

[BillC]

I'm still a little confused about the need for Asynchronous Fuel Delivery. Couldn't the pulse width in synchronous mode be increased sufficiently to provide enough fuel for power enrichment??

Don't confuse AE (Accel Enrichment) with PE (Power Enrichment), even though it's easy to do. AE takes care of transients, like when first pressing the throttle open. PE takes care of longer-term power fueling requirements.

While in PE, the computer does just increase the regular Synchronous or Quasi-Asynch pulse width as needed.

However, for some reason, GM decided to make the AE fueling Asynch. This lets the computer inject the extra fuel whenever it feels like. Their reason probably has to do more with the algorithms they chose for the chip, rather than any particular fuel/throttle response issue.

 

[kbbeckius]

Thanks for clarifing that. It would probably help me if I could get the Promgrammer program to work on my computer. I'd like to be able to look at the tables and compare the calibration data to what we are discussing.

Ok, I was thinking you could only have duty cycles over 100% in async mode, but the ecm must allow that in sync mode too.

Thanks again
Brad

 

[BillC]

It would probably help me if I could get the Promgrammer program to work on my computer.

What kind of problems are you having? Did you remember to install the JRE? Once that's installed, you should be able to run Promgrammer by just double-clicking the icon.

 

[kbbeckius]

Well, before when I clicked on the app, nothing happens. I finally got it working thought!!! I went through and erased anything java related, and then reinstalled it, and it works!! I think I had some old versions or parts of java software in various places on my drives.

Thx Brad