0ski_dwn_it

Thought I would spend the time to share with you all the rehashing I am going through on my setup.

Brief review of why I am going through this. Many of you recall I just put together this setup. The setup is turning some very impressive ET/MPH while cruising etc is as nice as the stock setup was. I am very happy with the setup, but naturally want to get the absolute most out of it. My best ET/MPH YTD with only two track appearances are can be seen in my sig.

I toook the car to the dyno on May 21st I believe was the date and dynoed it. I was a little disappointed that the speed shop did not have the ability to log the AFR, but I was confident the Tune was pretty close with the help of my own WB02 (which co-incidentally I can't use when I drop my exhaust). Upon dynoing the car on the first run it made the following power:



I knew at that point from previous tuning that my AFR was at the upper range of safe and getting close to lean. The AFR at that dyno was 12.8. So I increased the PE % change to AFR vs RPM about 2% across the RPM range above my TC stall speed. I as expected I saw a gain, but only in the lower RPM range. Notice how the upper RPM range HP/TQ stay dead nuts on the same paths. That verified my suspicions that my injectors were going static from about 3600+RPM. Here is that graph for comparison purposes, notice the nice gain down low, but it quickly retreats back to the previous dyno numbers:



At first I was not all that alarmed but I made several more pulls adding fuel each time till the lower RPM range that I saw the gain in fell on its face, due to too much fuel, yet the upper RPM range stayed dead nuts on track even in lieu of the added fuel. What does this mean? Well that my injectors after about 3600RPM are staying open 100% of the time, and regardless of the tuning I give it they just can't supply any additional fuel. Lets take a minute to look at some math to see if this all makes sense:

For a batch fire system as the 165 ECM system is, here is the formula

Injector PW=600/RPM What all this means is as the RPM increases the available time for the injectors to fire become increasingly shorter and shorter.

This is an actual data log from Ease scanner software I took prior to the dyno runs. The information for actual PW is indicated on the graph along with other critical parameters.



Pulling the actual raw data for injector PW out of the program and comparing it to the actual calculated PW times, you can see that I am static at about the same time as the dyno sheet tell me I am also.
Actual Calculated
RPM PW TIME AVAIL
2200 11.67 27.3
3475 14.31 17.3
4150 13.85 14.5
4525 13.43 13.3
4900 12.74 12.2
5125 12.52 11.6
5075 12.48 11.8
4975 12.67 12.1
5075 12.41 11.8
5275 12.13 11.4
5300 11.87 11.3
5450 11.23 11.0
5825 11.03 10.3
5850 10.57 10.3
6075 10.42 9.9
6175 10.39 9.7
6300 10.24 9.5
6375 10.08 9.4
--------------------------------------------------------------------------------


When TIME AVAIL becomes less then BPW they are static, See how that nearly corresponds exactly with the dyno sheets.

Now the big question do I need larger injectors. After some hashing around the conclusion was NO, what you might ask. You have another option first. You can increase the fuel pressure to make the injectors act larger than they are. Since I was not off much with the injectors already, I felt confident this would work. So last night I increased the pressure from 43psi with vac off, to 55 psi vac off. With the same tune as before I set out to get some logs. I could immediately tell the car was way out of tune, as the fuel was just dumping in, and causing the BLMs to be way off at cruise, but that was expected. The real test was going to be at WOT, with the increased pressure, would the injector constants come down to a under the calculated values. I don't have my laptop with me here, to pull the data, but the final outcome was yes they did. My WB)2 ranged from 12.2 in the lower RPM range to 11.4, this is a good sign that the injectors are able to fatten up the F/A micture once again, but how much room was left in the injectors. Upon
analysis of the PWs taken during the runs at 5750 RPM the pulse widths were about 8.83ms, well below the time available, that puts my now at about 83% duty cycle, and most likely it will get better as I pull fuel out to return the afr to up around 12.5-13.1 AFR for best power.

I do not have the logs here in front of me, so I can't share them with you. Sorry.

From this bit of information you can see that you need to use all available tools to hash through a setup to try to optimize it best as possible. I am hoping that now with more injector available I can squeek out a few more ft-lbs and some more horses out of it to continue what apprears to be a nice hump that was starting to occure on the dyno sheets. This added power I think will afford me the needed power to catapult me into the 10s hopefully. That is if everything else can take the power ( :smash: on wood).

I need to now start from scratch on the tune and hammer out the part throttle stuff along with idle again, then move onto smoothing out WOT, then to a dyno for the final WOT tweaking to optimize each RPM range. Then to the track for a final testing. :cheers:

Hope there is some information here in this post that can help others find where they rest with their setups, and some useful information on diagnosing a setup.

Talk to you all later....


PS for you sequential fire guys, your Pulse width equation is much different.

Here it is: PW = 1200/RPM what this does is afford you much more time to fire the injectors, at a higher RPM. So you can get away with less injector, at the same RPM demand.

Here is the math worked out side by side:

RPM Batch Fire Sequential Fire
2000 30.000 60.00
2250 26.667 53.33
2500 24.000 48.00
2750 21.818 43.64
3000 20.000 40.00
3250 18.462 36.92
3500 17.143 34.29
3750 16.000 32.00
4000 15.000 30.00
4250 14.118 28.24
4500 13.333 26.67
4750 12.632 25.26
5000 12.000 24.00
5250 11.429 22.86
5500 10.909 21.82
5750 10.435 20.87
6000 10.000 20.00
6250 9.600 19.20
6500 9.231 18.46
6750 8.889 17.78
7000 8.571 17.14

Here is the new one. I did notice that the system said closed loop, even while at WOT, and the blms did go to 128 during the runs, so I am assuming that is OK. PLease tell me if that is not.. I am starting to doubt everything I once knew.



Here is the data captured during that run:

Hope it comes through ok. It didn't here is a snaphot of it...




You can clearly see a definate decrease in PW from one log to the other. Really quite crazy. Maybe I am missing something. but everything is identical, with the exception of the FP.

What do you guys think this is. I hope this post makes sense, I got interrupted about 10x, and had to keep coming back to when I left off.

I thought you guys would think this was interesting. I hope I am not missing something completely obvious LOL. I wanted to examine them last night before posting, but got tied up with some plumbing maintenance at one of my apartments. Oh well. Curious to what you guys have to say.

Thanks!

[Modified by ski_dwn_it, 3:17 PM 6/11/2003]


[Modified by ski_dwn_it, 3:18 PM 6/11/2003]

John Row

I say use bigger injectors. I'm a big "systems" guy. I've always found that a system works best with all the right parts working correctly. Running injectors at above the design pressure might be squirting enough fuel (theoretically), but is it in a pattern that can be efficiently used.

It could be that in a batch fire, the spray pattern is not as critical as the A/F mix sits static (relatively) before it's sucked into the cylinder. It could also be that the fuel just coats the inside of the intake and doesn't burn at all.

JMHO, and for the record, I've been wrong before (just don't tell my wife!).


[Modified by John Row, 9:30 AM 6/10/2003]

0ski_dwn_it

Actually John youe better off and typically the injectors spray a better pattern with everything I have read at higher pressures. Most injectors spray best at about 85%-90% duty cycles in most of the articles I read.

Running too large of injectors creates all sorts of control problems. For example, at idle the injectors are have such a short pulse that its hard to minimize fuel delivery, almost the opposite scenerio I had at WOT, but the exact opposite. What most people/tuners will do is they will reduce the pressure a bunch to make the injector look smaller, then they car open up the pws. Which gives a little better control.

I am very confident that the 24# injectors at this pressure will provide very adequate/efficient fuel delivery. The GN guys have been doing this for years. The only downfall is it puts added strain on your pump, but does it really? Who knows. If my stocker goes, then it'll just give me a reason to put in a walbro that can handle 80 psi :eek:

There is nothing hokie with raising FP to compensate a shortage of injector. As I said your better off doing that than going to larger injectors that cause the reverse effect, and then don't have adequate pressure behind them to create a good spray pattern. :cheers:

ol,RJ

ski,
you might want to hook up the fuel pressure gage and keep an eye on pressure during WOT. just to make sure the fuel pump is keeping up with the new pressure setting (at high RPM). if it is and your PW is no longer static then I think you'll be fine.
BTW, good detective work.

John Row

I love this stuff. :cool:

If you are concerned about low RPM performance, then the bigger injectors make sense as well. The vacuum line on the pressure regulator reduces the pressure at high vacuum levels. If you are running higher pressure you will be running rich at low RPM.

Again, you need the system to behave. You would need a different pressure regulator that would reduce the pressure more at low RPM.

If it is possible to shorten the pulse width at lower RPM, you might be able to compensate somewhat.

0ski_dwn_it

OlRJ,

Yes the pressure is keeping up. Tested it last night with the a few WOT passes. So yes its going to be a continual monitoring process to ensure it keeps up the performance.

John,

I low RPM is not a concern, as I stated before the PW was well in range in the lower RPMs, but about 3600 RPM the PW got longer than the time available, hence the reason for the need to increase the FP to bring the PW into a controllable time frame.

Now the entire RPM range is running pig rich. I will use the prom tuning to bring the AFR back to a leaner state, decreasing the PW even more. Yes, in the lower RPM range I will need to pull even more out. But I am going to let the dyno results dictate to me Where and how much to pull. Follow my train of thought?

On a side note, I just bought a new car hauler so Dad and I can both take our cars to the track. Looks like I can try it out this weekend, but don't expect to see very good results just yet until I get the tune back on. But I can always run it where I was at with maybe a few more pound to richen it up a tad, and maybe get lucky enought to touch out a 10sec pass ;)

Your right John, I love this stuff too! :cheers:

ralph

That's great work Jesse. however i have a question for you. If the system goes into open loop at WOT, how did the ECM know to reduce the pulse widths for the higher pressure????

0ski_dwn_it

Ralph that is a very interesting question, and one that we over at the thirdgen have argued over for quite some time.

Here is the mystery. Or the controversy. Does the does the ECM still somehow in all its trickery, still take reading even in PE/WOT/open loop? This as many have argued supports that theory. Some of the hacker gurus argue that the voltage from the MAF still is read and therefore it actually can be used in a sense past 255 g/sec.

I was going to post this information on the TGO to see what some of the gurus over there had to say about the information. But yes in a sense you are correct, and to be 100% honest with you, I don't know the answer.

I will try to find out why this is happening.

:cheers:

scorp508

To the best of my knowledge the only sensor ignored at WOT is the Oxygen sensor. All other sensors keep working.

ralph

Right Scorp, but which sensor would have picked up the additional FP if not the O2?

Here's my take on this, but i have no real facts to back it up. I think the ECM learns based on O2 readings over time and adjusts the fuel table accordingly, so even tho it goes into open loop, it knows based on the long term accumulator that there is more fuel present and had to cut back pulse widths while in closed loop. So assuming your injectors were adequate, raising/lowering your FP may not have much of an effect over a certain period of time, because the ECM will learn and adjust.

The only evidence i have of this is from my car. When i first got the chip burnt, it was pretty good right outa the box, but after further modications, it's a little off now. So if i disconnect the battery and reset the ECM, it will run like crap until it goes into closed loop. However, after driving it a while, it will run fine in open loop. So i believe the ECM does learn over time.

Jesse, how long did you drive the car with the FP at 55? In your case the ECM couldn't keep up with the FP at 40, but apparently it can at 55. Let us know what you learn elsewhere.....This is very interesting.

scorp508

Gotcha, hmm.... I have that "corvette fuel injection" book, I wonder if there is any mention of these things in it.

ralph

Scorp, just copied you and Jesse on an email about the electric water pump for our cars. Looking for opinions.

scorp508

Thanks, Ralph, I'll look for it. Did you use scorp508@attbi.com ?

ralph

yep

0ski_dwn_it

Guys,

As promised I posted the new logs above in the original post for you to compare easier, but you will have to clip on the images to make them clearer.

This has me baffled. :confused:

HWoods

Bigger injectors +electronic fuel pump controler? Maybe that would work better? Aeromotive offers one, I think MSD makes one

scorp508

Jesse I was thinking about this a lot today. We're talking about 8 bit ECMs right? It just isn't possible. Perhaps the MAF wires have the ability to sense past 255, but the ECM can't interpret it no matter what right?

8-bits = 2 to the 8th power

2^8=256
256-1 = 255 (counting starts at 0)

0ski_dwn_it

Yeah scorp,

But there is a whole lot of stuff, if you go to the thirdgen and see my post there, first you will see the rashing I get there when I post :) LOL. I have no love loss with some of the members, but people like Funstick and a few other that know the code better than any others agree with me there are some funky things that happen with even the MAF systems and their 255 g/sec limits.

scorp508

I'll have to check it out, but I just don't understand how it works. Maybe GM coded galactic worm holes into the MAF systems. :D

NanoBrain

Granted the following may be incorrect:bb, but what I’ve gathered of the 165 ECM was that the “Limp-Home” mode & cold start parameters are non- adaptive. Meaning what the factory Calpac had, is what you get even with engine mods. Now, when in closed loop the ECM uses the “Integrator” as the dynamic sliding scale for fuel, and the “Block Learn” address cell as the long term base for the “Integrator”. I think the PE value is non-adaptive also?

As the engine is mod’ed the ECM, when in Closed Loop will compensate with the Integrator. (Speculation here)- after a predefined sample rate/oscillator clock count the ECM uses the Integrator values to set the Block Learn cell values. At WOT, from what I’ve read, the 165 ECM ignores a few sensors (maybe even the MAF) and relies on the TPS, IAT, CTS, EST, & the HEI pulse signal. From these and the stored BL cell values, I speculate the ECM creates an interpolated set of parameters for fuel and spark tables @ WOT.

I think this is why, large CI SBC, with MAF’s don’t freak out in the upper RPM range @ WOT.:skep:

With all the people out there burning their own EEPROM’s, & flashing their PCM’s, someone must have the definitive answer to these questions? I realize the source code is very cryptic, but the 165 ECM & it’s Mircocontroler have less computational power than most of the new auto stereo Head-Decks out there now:rolleyes:!

BTW, ski_dwn_it: How do you keep from spinning tire all the time at stoplights with all that low end Tq? As a matter of fact, all of y'all that have >500Ft/Lbs in the low RPM range! It must be like driving in the rain - all the time?

:crazy:NanoBrain:crazy: