Help with tuning out KNK

Quick · Post by **Quick** » Mon Sep 22, 2008 2:06 am

Ummm... (just in case) you probably don't want to tell them you're using a tuner. Just tell them you were using an OBDII scanner and noticed the unusual amount of knock retard.

DarthMuffin · Post by **DarthMuffin** » Mon Sep 22, 2008 11:03 am

Yes, of course. And fortunately I actually *do* have an OBDII scanner

DarthMuffin · Post by **DarthMuffin** » Tue Sep 23, 2008 4:29 pm

I hate dealing with dealers. They're saying they see this a lot and want to sell me a $169 fuel injector cleaning service rather than fix it under warranty. Anyone got any suggestions? This can't be normal on a 15k mile car...

Also, I added a bottle of octane boost just to see if it helped. It didn't -- no change in the knock even with the factory tune. So either it's false knock or something is really really broken.

Quick · Post by **Quick** » Wed Sep 24, 2008 1:31 am

DarthMuffin wrote:I hate dealing with dealers. They're saying they see this a lot and want to sell me a $169 fuel injector cleaning service rather than fix it under warranty. Anyone got any suggestions? This can't be normal on a 15k mile car...

Also, I added a bottle of octane boost just to see if it helped. It didn't -- no change in the knock even with the factory tune. So either it's false knock or something is really really broken.

That's crap. Call DCX customer service. Tell them you have an SRT8 and you want to talk to that department (forget what it is but you should get different customer service I think). Tell those guys you want to talk to SRT8 engineering because you're seeing an abnormal amount of knock. Hopefully you will get someone who will contact the dealer and tell them what to do. Is it a 5 star dealership? They get dinged bad for customer complaints so they're usually pretty responsive. Maybe try calling Chrysler and while that's going call the dealershop and see if you can talk to the service manager. Tell him you have a call into Chrysler but you'd like to see him address the problem.

At the very least the service department should have put a call into the regional office about it. Trust me, you have an unusual amount of knock/knock retard.

[edit] Jees, I thought about that more and had to add more. Tell them it's knocking and if they want to clean the injectors under warranty to fix the problem to to right ahead... You got any other dealers in the area? with a reputable service department? If they're 5 star dealers they're better. Doesn't mean they're good, but they're better.

DarthMuffin · Post by **DarthMuffin** » Sat Oct 25, 2008 2:29 am

Okay, I've been traveling for work and moving too so I haven't had any time to play with the car until now. I did get my oil catch can installed and it's not seeming to help a lot, if any.

I took it to the local dealer today and they could not find anything wrong. They could not find any data saying what an "acceptable" amount of KR is, there's nothing in the computer, the car runs fine and has the correct amount of power as far as they're concerned and they could not find any TSBs on the issue so they say they've done all they can. They are the only dealer within 150 miles of here and are a 5-star dealer.

So, I tried the Predator again and did some logging. Here's results:

Stock:
LT 2.0, ST 6.5. Some ST at mid RPM, most at high RPM. LT is only in a few brief spots.

91CAI:
LT 1.5, ST 7.0. ST at all RPM (when at WOT), slightly more pronounced at higher RPM. LT only at the 2-3 upshift.

91CAI +5/5/5% fuel:
ST 6.5, everything else identical to previous run

91CAI +10/10/10% fuel:
LT 1.5, ST 4.5. ST at all RPM, LT only at the 2-3 upshift.

91CAI +10/10/10% fuel, -1/1/1 degree spark:
LT 0, ST 3.5. ST at all RPM. Finally some improvement?

I can probably post pics of the data or the logs themselves if it'll help.
So what do you guys think? I was about ready to call it false knock until the spark retard helped. I might just call it good here and live with it. I suppose my options would be:
1) Call it good as-is and live with it.
2) Complain to DCX and see if they can get the dealer to fix the knock
3) Forget the Predator and just go back to stock
4) See if CMR or another custom tuner think they can help

Quick · Post by **Quick** » Sat Oct 25, 2008 4:19 am

Lets see the pics of stock and your last with -1/-1/-1

I'd add some more fuel too. Go +12 (and then maybe +14) with the spark retard and see if you can get it to mostly zero with some bumps. you should expect up to 3 at shift points and then maybe a bump or two up to 1.5 or even 2 in a couple of spots.

Have you thought about getting a wideband? I had gone to about 12.0 - 12.2 AFR and spark around 17 to 18 degrees.

DarthMuffin · Post by **DarthMuffin** » Sat Oct 25, 2008 1:06 pm

No, I haven't thought about getting a wideband. Sounds like a lot of expense and hassle just for a tune. If I were getting heads, headers, and a cam then I could see it -- but I'm not.

Stock:

+10 fuel, -1 spark

Thanks for the help so far!

Quick · Post by **Quick** » Sat Oct 25, 2008 2:04 pm

Well, not what I expected to see but the +10, -1 looks a whole lot better and like a normal graph. Odd on the first one where you have have 2 spikes of LT Knk while the ST stays flat through both of them (nice wheel spin though

, you have to work on that). Try adding a bit more fuel. Try adding a little more in the bottom range and a bit more than that in the middle range and see what happens. Try 12/15/10 maybe (leave the timing where you have it now).

hmmm, maybe try 10/13/10 first. then 12/15/10

DarthMuffin · Post by **DarthMuffin** » Sat Oct 25, 2008 3:50 pm

Thanks. I just got back from trying 12/12/12, results are pretty similar to 10/10/10 but this time I had a big LT knock spike right at launch. I'll post pics of that and 10/13/10 or 12/15/10 after I try them (maybe tomorrow night).

Yeah, the wheel spin caught me by surprise on that one -- tires were pretty cold. I'm not trying so much for the "best" launch here as I am trying for consistency to get valid comparisons between the tunes. I figure I can work on the launch once it's working correctly.

slip · Post by **slip** » Sat Oct 25, 2008 5:59 pm

I am no expert, but here is my interpretation, right or wrong.

The first log you had true ST KR that retarded your timing.
This is the first arrow showing the big drop in timing when ST KR was spiking.
But the rest of the log looks good, with your timing showing a strong upward trend through each shift. I am guessing your Actual Spk Cyl1 maxed out at near 23.5.

The second log is the one I don't like. The timing has a downward trend on the first shift and only a slightly upward slope on the second shift.
I doubt your timing reached it's max of 23.5.

I wish someone from Diablo would be more help how to read and interpret logs and KR. They obviously have the knowledge. Their absence of help is really noticeable.

Quick · Post by **Quick** » Sun Oct 26, 2008 12:42 am

He's peaking at about 18 in the first graph (actual spark cyl_1) and averaging around 14. About 17.5 in the second graph.
So the actual timing at cyl_1 is about the same in both graphs, just a lot smoother in the second. That's what we'd expect. In the first graph it's getting knocked down with Knr. In the second it's still getting knocked down to about the same point but with less degrees of Knr since the base timing is less. I'm pretty sure that the PCM can adjust for about 9* of Knr. After that you get actual pinging for a while.
I've rarely seen anyone get WOT timing above 21 or 22. Besides, you can't just go for timing, it's got to be in conjunction with fuel (how much and how good). Since there isn't any A/F gauge here DarthMuffin is sort of tuning by Knr... The PCM will adjust for Knr. The reason to get rid of most or all of the Knr is that the PCM handles it reactively. That means that actual knock occurs, is detected, and then the PCM adjusts for it. So you're actually experiencing a short period of knock before Knr is applied. The larger the intervention, the larger the event was. Secondly, once the PCM adjusts it gradually steps down the intervention. So if it was a momentary event your timing might be retarded for a short while when it's not needed.

I'm running a very good CMR tune and my timing is pretty flat through the whole range. Why do you think it's desireable to have it climb with the RPMs? It was pretty hot out for this one so I was seeing a bit more Knr than usual.

slip · Post by **slip** » Sun Oct 26, 2008 6:50 am

Quick wrote:He's peaking at about 18 in the first graph (actual spark cyl_1) and averaging around 14. About 17.5 in the second graph.
So the actual timing at cyl_1 is about the same in both graphs, just a lot smoother in the second. That's what we'd expect. In the first graph it's getting knocked down with Knr. In the second it's still getting knocked down to about the same point but with less degrees of Knr since the base timing is less. I'm pretty sure that the PCM can adjust for about 9* of Knr. After that you get actual pinging for a while.
I've rarely seen anyone get WOT timing above 21 or 22. Besides, you can't just go for timing, it's got to be in conjunction with fuel (how much and how good). Since there isn't any A/F gauge here DarthMuffin is sort of tuning by Knr... The PCM will adjust for Knr. The reason to get rid of most or all of the Knr is that the PCM handles it reactively. That means that actual knock occurs, is detected, and then the PCM adjusts for it. So you're actually experiencing a short period of knock before Knr is applied. The larger the intervention, the larger the event was. Secondly, once the PCM adjusts it gradually steps down the intervention. So if it was a momentary event your timing might be retarded for a short while when it's not needed.

I'm running a very good CMR tune and my timing is pretty flat through the whole range. Why do you think it's desireable to have it climb with the RPMs? It was pretty hot out for this one so I was seeing a bit more Knr than usual.

All my logs, stock or tuned, usually reach a max timing advance of 23.5 during WOT. And the logs usually show an upper slope during WOT, which usually peaks at 23.5. I assume this upper slope and peak of 23.5 is good and normal.
When I don't see this upper slope peaking at 23.5 under WOT I assume some timing has been pulled.

Here are 4 logs.
The first log is my stock 300c. Stock max timing advance is 23.5.
The second pic is a Predator 91 tune that I think looks normal and good, which also peaks at 23.5.
The third and 4th logs are Predator 91 tunes which timing is not consistently advancing during WOT and is stuck at 18. I think these 2 logs are not good.
I don't want to hijack this thread, so I'll try to minimize my input.

Quick · Post by **Quick** » Sun Oct 26, 2008 2:45 pm

I'm not real familiar with the 5.7s. DarthMuffin has a 6.1.

DarthMuffin · Post by **DarthMuffin** » Mon Oct 27, 2008 10:12 pm

Okay, here's the latest try. +13/14/10 fuel and still -1/-1/-1 spark--I didn't have my laptop with me so I could only log one run before coming home, so I split the difference on your recommendation.

That LT knock spike at initial WOT is 2.0 degrees in case you can't tell.

The +12/12/12 run graph looked exactly the same, with the LT knock spike at the start (of 3.0) and a max ST of 3.5.

Suggestions at this point? As soon as I have a shop again (December) I'm going to get this thing up on jacks and check the exhaust for leaks and rattling that might cause false knock, and check the engine mounts and knock sensors out too. I'm thinking that initial LT spike might be something banging around.

The good news is that I took a road trip this weekend. Gas mileage went from mid 19's to low 21's with the Diabo tune

Quick · Post by **Quick** » Mon Oct 27, 2008 11:14 pm

I don't know.. It put the long term on a different scale than the short term? Because that short term spike right after the long term is 4* I think.
That initial spike (4*) is happening right around 2200 rpm. You never see that rpm through the rest of the gears.
It looks like there is a bit at 4100 and then maybe sometihng at 5300-5500. It drops off going throught the upper range.
Adding a couple percent after adding 15 or so is a fairly small jump. How about trying to overshoot once and see what it look like instead of creeping up on it? Then maybe back off a bit.

Try 20/20/15. Leave the spark where it is. If you get rid of the Knr then spark will be about 18. Looks like you're breaking the tires loose a tiny bit on the 1-2 shift?

and knock is getting down to where it's not rediculous any more.