NefMoto

Hi folks.

I'm just updating my "Understanding TT 1.8T remapping guide" to reflect recent discussions on KRKTE & particularly KFMIRL/KFMIOP plus some minor editorial updates, corrections, etc. when I came across a possible issue/lack of understanding with respect to LDRXN & LDRXNZK. I'm looking for some assistance in getting to the bottom of it.

I compared LDRXN with LDRXNZK for both BAM & BFV engines and was expecting (perhaps naively) to find LDRXNZK significantly lower across the board. Indeed, a quick look at funktionsrahmen module LDRLMX 3.100 which I have yet to translate confirms: "LDRXNZK: about 15% less than LDRXN". However, this is not the case in either tune. See table below which probably won't render neatly, so jpeg screen capture attached. BAM (225 PS) on the left, BFV (240 PS) on the right.

RPM   LDRXNZK (%)   LDRXN (%)   LDRXNZK (%)   LDRXN (%)
1000   97.0   97.0   99.0   99.0
1720   128.1   128.1   129.8   129.8
2000   139.9   140.6   148.5   148.5
2100   143.6   147.4   154.5   154.5
2200   145.6   146.7   160.5   155.3
2520   148.5   140.8   176.3   166.5
3000   148.2   138.3   176.3   166.5
3520   146.5   140.9   176.3   166.5
4000   148.2   144.0   182.3   166.5
4520   159.4   151.8   188.3   177.0
5000   160.2   151.7   183.0   185.3
5520   157.7   157.7   168.8   171.8
5900   155.8   155.3   162.0   168.8
6000   153.2   153.0   153.8   160.5
6400   142.3   142.3   141.0   148.5

So they look to me to be about the same up to maybe 2,300 rpm, then LDRXNZK is bigger  :o until 5,300 rpm then smaller, but not by 15%. What is going on? Is it perhaps something simple such as the labelling information has been inadvertently or deliberately swapped around in the BAM engine DAMOS I've got, or is there a more subtle interpretation that I am missing?

Any thoughts, theories, information and even peripheral discussion would be greatly received.

Regards.

Doug

In the ECU LDRXN is used when the ECU doesn't detect knock, and LDRXNZK is used when it does detect knock. There aren't any scales or corrections applied to those maps individually, and the value that comes out of them is treated the same no matter which table it comes from.

Indeed. So I was expecting approximately 15% lower values in LDRXNZK since I had assumed a lower output would be desirable during continuous knock so LDRXNZK was part of a protection feature. However, that doesn't appear to be the case...?

Doug

Is it possible this engine is dealing with knock by adjusting the air fuel ratio instead of retarding ignition timing?

That crossed my mind, but I rejected that on two grounds: (a) that it would be more sensible to achieve that through lambda control not charge control and because the ignition timing route is present and valid for these engine variants. I.e. precursors to condition B_kfzk which controls switching between LDRXN and LDRXNZK appear to be normal. I briefly checked KFSWKFZK (ignition angle retardation threshold for switching between ignition angle maps) and found that it was -3.75 degrees crank in all addresses for BAM which appeared reasonable.

I checked the S4 and RS4 DAMOS files to find that LDRXNZK addresses are set to 100 across the board in both. Not something I was expecting. Is that correct? How have you modified yours in your Stage 3 tune Tony?

Since nobody else is jumping in on this thread with good words of wisdom (or even just the answer), I presume that the interrelationships between knock control and charge control are another aspect of ME7.x that is not that well understood? ;)

Doug

In the 180 ps AUQ file LDRXN is lower everywhere than LDRXNZK. Mayby because of the power loss of ignition retarding during knock activity it has more boost to regain that loss and the driver can't feel the difference in power.

In S4s, KFSWKFZK is -96.0 across the board...... Interesting find though. It would be a good idea to maybe think about using it to pull boost on too much timing retard. One approach to meth injection failure protection...

In my example stage 3 tune I set LDRXNZK to be lower than LDRXN. I did that because it made sense to me, and I have not verified the results on a dyno.

On a side note for a fail safe in my future meth injection install, I was planning on having a relay that cut power to the N75 valve. I was going to connect the relay to the meth injection controller and the meth fluid tank level sensor. That way if the meth controller isn't working or you are out of fluid, then you won't be able to build more than 7 psi of boost on the S4.

On a general note regarding knock protection... Decreasing the boost is the absolute slowest way to bring an engine away from knocking/pinging conditions as the regulation of boost (in general terms) is very very slow compared to ignition retardation that can be done from one engine revolution to another. Having a boost PID controller trying to adjust the boost level for this purpose in multiple steps takes ages compared to (a) ignition timing change and/or (b) injector timing change (enrichment) that is more of an instant solution to the problem.

In my previous experience (not for ME7 cars), ignition is used first as knocking/pinging protection, secondly injector timing is used and as a last resort the intake air mass is lowered due to the fact that the regulation of the air mass is so much slower than the other options.

Notwithstanding fredrik's comments, Giannis' theory looked plausible enough, but I thought that the onset of knock is not sufficiently predictable to counteract it by altering a boost profile hard limit alone.

What are the main influencing factors on when knock occurs? Combustion chamber temperature? Engine load?

Thanks for chipping in guys. It would be nice to think this is another Motronic subtlety which we can get to the bottom of rather than an error.

Doug

There is no better way to cure pinging/knocking than to pull timing, but what fredrik is over looking is that ME7's KR is *ALREADY* pulling timing!

The problem is, EGTs are going to go sky high when you start pulling too much timing. At that point, the best approach is to start adding fuel (see BTS), and third, reduce load (either pull boost, or close thottle).

But isn't too much fuel increasing the knock possibility also?

Adding fuel generally cools the intake charge (especially in forced induction engines)

I believe this is the case. Keep in mind that stock, the boost curve is very very conservative.

BTW stuff like this is why I laugh at people that try to save money by using low octane gas. Modern torque based engine management will just use more fuel to try to give the driver the same "feel".

Yes you are right about that. High performance car and low performance fuel can't go together. I hope someone has a differnt opinion to discuss about the subject, but i strongly beleive that it is all about keeping the same "feel" as you said.

I can laugh too now, but I had to do the logging runs to convince myself that there was a substantial drop in output under WOT with lower octane fuel... A lot bigger than I was expecting. I dropped to standard octane for a while because, sadly, 95% of my driving is steady cruise.  :(

Good discussion, but I'm still seeking that elusive answer as to why LDRXNZK is higher, not lower, in the 1.8T variants I've studied.

Doug

Err. I thought giannis explained it perfectly.

O.k., so the answer is that the knock control retards timing to deal with knock on the crank synchronous pathway as fredrik explains, but more boost is allowed to mitigate the loss of power by sparking later than would otherwise be required for peak cylinder pressure as Giannis suggests...

If that's the correct then I'm a happy man! ;D

If that's correct, why does the dumbass funktionsrahmen specify that LDRXNZK should be 15% lower than LDRXN?

Doug

LOL. Good point. No clue.

Just a thought; I've had my cars software updated twice by Audi. This was well before I knew shit about "tunes", "chips", logging etc (some may argue that I still don't). If they were in fact updating it, I could only assume it was to run more efficiently than before. If this was the case, maybe they found a more efficient way to manipulate the maps? Again, just a theory....and in my defense I'm falling asleep at the keyboard, so this may read like nonsense in the am. ::)

...

EDIT: I missed that the question for LDRXNZK being higher than LDRXN was already answered. It is indeed (at least for other EMS) to prevent that the driver feels an obvious loss of power.

O.k., that's good enough for me. I've revised the 1.8T guide to read:

"Remarks for LDRXNZK: It is noted in module LDRLMX 3.100 that the values should be set about 15% lower than LDRXN which would be reasonable from an intuitive understanding; i.e. overall ignition advance is reduced during knock so maximum allowable cylinder charge should also be reduced. However, in these examples, the values in LDRXNZK are actually higher from around 2,300 to between 5,000 and 5,500 rpm. The only plausible explanation is that a slightly larger cylinder charge is allowed to counteract the sub-optimal peak cylinder pressure and consequential loss of torque that would result from sub-optimal ignition timing due to knock. Otherwise, the driver would feel a noticeable step-change in power during the onset of knock. If in doubt, don’t try to second-guess the OEM calibrator; follow the advice in the funktionsrahmen and set the values in LDRXNZK a bit lower than LDRXN."

Thanks guys.

One issue I had which I failed to mention in the original post was that if there was to be a 15% or so difference between the maps with no other power-loss compensation, then the driver would feel a noticeable step-change in power every time continuous knock was encountered. That didn't seem correct and which, in part, caused my confusion.

Doug

Sorry for the off topic but can i ask something?when we say continious knock what does it mean? For how much time knocking must exist for to be considered by the ecu as continious?

I just made the same discovery on my AUQ engine. Thought I had the tables mixed up at first, but when I log the requested engine load it clearly follows the shape of the lower one (LDRXN), and this was during cold conditions with 98 octane fuel.

Link to my thread for reference:
http://nefariousmotorsports.com/forum/index.php?topic=1534.msg14573#msg14573