92 zzz28

Okay, after reading lots of posts about the CCP disable feature, I decided to see what would happen if I set the constant to 0%. Well, I think it makes the engine run somewhat better. My BLMs never went below 120 and never went higher than 133 during a test drive this afternoon. How ever I did not an unusally high amount of knock count during this test drive. I registered 37 counts in about 15 min. Before this, on the last couple of days, I was getting them down to 10-15 knock counts in a similar length of test drive, driving in the same manner.

What does that mean? Am I running too much advance now that the BLMs have stabilized to a more ideal range?

Another thing I am very cloudy about is the fuel trim cell that DM has. What are these? I no longer idle in cell 4, which I read should happen. If I understand it correctly, with the CCP disabled I am disabling HWY mode. However, I still read high SA. As high as 56-57 degrees advance while cruising at about 60mph with a KPa of about 45-55.

Any ideas of why the SA is so high? I am currently using the stock ANHT stark tables and have my max SA constant set to 38.4*. BTW, 7730 ECM, latest TDF $8D.

What do you all think?

Thanks in advance.

Ryan

11sORbust

Sounds like you are on the right track setting the forced idle constant to 0. You should now idle in the cell that your map and kpa is at on that handwritten grid in the other thread. I now idle in cell 9 because my idle speed is 850rpms and the map reading is 65-75kpa. The grid says that is correct. Using more blms seems to be very good. I was going to just adjust the cell boundaries but noticed the ccp activity in the log data. AT this point I have the forced idle cell under xx% DUTY CYCLE set at zero. The CCP is still functioning so I'll get highway mode. Nobody has proven ill effects by keeping ccp on (in the code).

Your s/a should be that high under some conditions. I use the aujp because the spark tables are more mild.If your max s/a (part throttle)is at 38* then you should be in the safe zone.

I'll give you my opinion on knock counts. You should ignore them, in general. When you see knock counts then check for timing retard. If there is no retard then you are ok. Knock counts can be generated by anything. That means loud valvetrain,header clearance issues,etc. The ecm CAN tell the diff between false knock/true detonation. That is why I said to check for timing retard. I think knock sensors are good for nothing. But then again, my engine is too noisy for a k/s to even work. The funny thing is that my ks lead is grounded with a resistor and disabled in the chip but I still generate knock counts. No timing retard though...

11sORbust

forgot to mention something. I will be remapping the cell boundaries. It'll be just like how GM did it. There will be a dedicated idle cell, etc..................

RBob

iTrader: (1)

Wow, which ECM is this, I need to get one of these!

Thanks, RBob.

11sORbust

They are standard option on all thirdgen F-bodies

Grumpy

Nope, it's just looking for a particular frequency.

And pre-ignition can be silent, so the Knock Sensor is completely blind to that.

RBob

iTrader: (1)

And where is the false knock vs. real knock in the pretty pictures? Seems to me that they are showing the difference between no knock and knock as picked up by the sensor, whether false knock (such as the exhaust hitting the frame) or true knock.

RBob.

11sORbust

what is your point?

11sORbust

I agree and thanks for the correction. I was just trying to tell 92 that knock counts doesn't always mean detonation. THat sometimes there will be knock counts for false reasons.Not to get hung up on a few knock counts......

RBob

iTrader: (1)

You made this statement:

The ecm CAN tell the diff between false knock/true detonation.


And we all want to know which ECM can do this and how?

RBob.

11sORbust

That was a general statement and have been corrected,technically. BUT the point is that knock counts don't always mean detonation. THe part you seem to overlook is that the ecm can determine false knock(to a degree). The sensor cannot and I'm aware of that. It might be a crude method but GM made it possible to detect false knocks,GENERALLY. What do you think stuff like knock counts are for? And why do you persist in picking at me?.....

RBob

iTrader: (1)

The point is you made a statement and are unable to back it up. Which is not surprising figuring that statement is false. Here it is again in case you have forgotten:

The ecm CAN tell the diff between false knock/true detonation.


The fact is the ECM can not tell the difference between false knock and true detonation. Knock counts are knock counts.

Now you make this statement:

THe part you seem to overlook is that the ecm can determine false knock(to a degree).

Again I ask, how? Pretend I am from Missouri and show me.

RBob.

11sORbust

The answer is when you get knock counts and no timing retard.......

RBob

iTrader: (1)

OK, and how does the ECM know that these knock counts are from false knock while other knock counts are not?

RBob.

11sORbust

I don't know the ecm code so it's hard to say. I can give you something else to support my "false statement". If the ecm notices detonation(based one set parameters,maybe constant), it'll retard the timing, as we all know. If there is a FALSE knock then the ecm will not retard the timing.

Btw, I think you are acting a little harsh. What do you have to prove. I'm not a bad guy... I wouldn't consider the quote you took as a statement,"The ecm CAN tell the diff between false knock/true detonation." If you didn't skim past my opening statement you would have read that the following post on knock counts was my opinion.

So can I be taken from the cross? If I'm wrong then prove it and I'll retract my statement that you claim to be false. No need to get emotional......

RBob

iTrader: (1)

First off if you'd like to stay around here I suggest that you knock off the accusations. . .

Back to the subject on hand, you say you can't prove your statement because you don't know the ECM code. Yet I do know the ECM code and claim that your statement is false. False based on how the code operates.

Then this statement:

If the ecm notices detonation(based one set parameters,maybe constant), it'll retard the timing, as we all know. If there is a FALSE knock then the ecm will not retard the timing.

The ECM doesn't know detonation from a pothole. All it knows is knock counts. Simple as that. If you are looking at a scan tool and aren't catching the knock retard, maybe it is because of the slow frame rate relative to the ECM loop rate.

If you can give a little more to go on, just maybe I can explain why you are seeing what you are.

RBob.

11sORbust

I think we should just call it a day. I would like to stay around here so I guess forget it. I didn't make any accusations, just said no need to get emotional. I also didn't make any statements. I just typed a paragraph about knock counts in which I said is my OPINION. I am really sorry that this has gotten ugly. No need to go on, lets just let this one sink to the bottom.......please....

92 zzz28

WOW, I sure missed alot yesterday. Hate when that happens.

Okay, back to my original questions.

First off, Thank you 11s for pointing me to a very good post about BLM cells, I learned quit a few things from that post. Good info. BTW, when I was searching I searched for about everything I could think of that was related to that topic, but never actually thought to try "blm cell".

I get knock counts, I don't hear anything, which I know doesn't mean anything because knock can be silent, and my engine is kinda loud anyway, plus I usually am running the AC. I do get spark retard. That is what I don't like. I scan with Datamaster and I always use the spark retard warning thing. You know that bar graph that shows a huge purple bar as you get spark retard. Anyway, what do you all think I can try to get that to go away?

I did notice that once I turned the CCP constant to 0%, I am running with BLM and INT #s closer to where they may need to be. I have read much on the subject of tuning here and elsewhere and I am completlely aware that 128/128 may not be where the engine wants to be. Anyway, I want to start from 128/128. And I am using the stock ANHT spark tables as per Trax's advice. I noticed a subtle difference in acceleration. Probably as he warned, too much SA for what I am doing, but better than stock AUJP SA for me.

Should I set the HWY SA table to 0s to get my SA down to a more reasonable #? When I added my SA in the main table to the SA in HWY SA, I get what DM is showing in the logs. I thought by setting the CCP constant to 0% that would eliminate HWY mode? WHat gives here? If I am right and I should not be getting any HWY SA, then what should I look at next?

Thanks a bunch all of you.

BTW, Rbob and Grumpy, I always appreciate anything you guys post because not only is it usually humorous, but always very informative.

Sorry for being so long, I guess I just have too many questions...

RBob

iTrader: (1)

When I start out with a new setup I disable as many extraneous items as possible. Set EGR enable temperature at the max, air pump temp to the max, disable CCP, set hiway mode MPH enable to 255, etc. This allows the tuner to concentrate on just those items still alive.

If you are getting knock consistently in an area then remove SA from before that area and from that area. The before part is more so if the engine is accelerating. If you are getting knock more or less all over pull timing from the entire table and try again. Also be sure that the ECM initial SA value (cal term) matches your distributors base setting (as set with a timing light).

In the $8D mask there is also a forced knock test to test the knock detection system along with a low octane knock retard. Best to disable both of these for initial tuning.

One problem with BLM cells is that they can cover a large area of the VE table. If the VE value in one corner of a cell is rich then BLM will drop, if the VE value in the opposite corner of the same cell is lean, then going from the first mentioned corner to the lean corner will cause an even leaner AFR. Just something to be aware of.

RBob.

92 zzz28

Cool, thanks for the advice.

I understand all of what you are suggesting and telling me to try out except for the below statement. Can you kind of simpliy that for me?

I am hazy when you are talking about the VE value of a cell having a certain characteristic and in the same cell at a different corner, having a different characteristic. What/How does this work? I am just now, the passed few days, starting to learn about the fuel cells. I guess a quick turtorial is in order for me.

Thanks,
Ryan

RBob

iTrader: (1)

This characteristic comes into play during the initial tuning phase. Once the VE table is properly calibrated it is much less of a problem. I figured since you are starting out with a whole different combo (and looks like a nice one at that) that you should be aware of this.

I'll use cell 14 as defined in AXCN ($8D bin). Cell 14 is bounded by 1600 RPM through 2400 RPM, and from 44 KPa through 100 KPa. So it covers quite a bit of the VE table.

Here is the VE table within cell 14:

Every one of these points in the VE matrix is affected by the value in BLM cell 14. As an example lets say you are crusing along at 45 KPA, 1600 RPM, and that entry of the VE table is too high. In response cell 14 will drop below 128 to remove fuel. Cell 14 then stablizes at the value of 110.

Comes time to pick up a little speed and you press on the go pedal a tad. The MAP goes to 100 KPA as the RPMs climb toward 2400 RPM. The cell 14 value of 110 is still in affect, removing fuel the entire time. If the VE point of 2400 RPM and 100 Kpa is currently a little low (we are still in the tuning phase) then it will be that much lower due to the BLM being at 110. The engine can stumble, cut out, or ping, all dependent upon how lean it goes and the current SA value.

At the same time lets say that the VE point of 2400 RPM and 100 Kpa is dead on. With the BLM at 110 that is removing 14% of the fuel. Then afterwards you are looking at your data logs and see cell 14 at 110. So where do you really make the corrections? (Eventually cell 14 will rise again if held at 2400 RPM & 100 KPa long enough, but that is not likely as the engine would be accelerating).

Early on it can help to lock the BLM's at 128 and just use the INTegrator for corrections to the VE table.

RBob.

P.S. I edited the {code} tags so that they are correct.

92 zzz28

Rbob,

Thanks, great description of how the BLM cells work. I know understand a little more!!