When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
In the quest to make a very efficient CAI and datalog MAP values, I got the car on the dyno to do some controlled pulls rather than the street.
In doing so, I did 6 pulls between 3 different CAI setups (2 pulls per setup) only to find out later while looking at the data that the car was pulling upwards of 10.6* timing.
Key points:
-EVERY run the car pulled at least .3* right off the bat going WOT.
-Two of the 6 run retard climbs to 10* (on 2 different CAI setups)
-One run retard never went above 1*
-Car has been dyno tuned maybe 2 years ago by the same shop I just dyno'd the car and I trust them.
-Running 93octane with full exhaust
-Car did not spit/spudder through the pull and sounds normal.
-Daily driving car feels perfectly normal.
Possible causes I have come to find:
-Bad gas(currently 1/2 tank left and waiting to use up current tank)
-Exhaust hitting something (found tight clearance between header and frame, remedied but still have issues)
-Knock sensor going bad(confirmed voltage and KS resistance to be within spec)
These pulls lost me about 10hp/10tq with this timing issue. I was wondering if anyone with KS experience would have have an opinion on what to tackle first given my description.
Thanks a lot.
Last edited by smitty2919; 11-18-2015 at 08:09 AM.
it's definitely possible you're getting false knock.
in general, if you remove timing advance and the problem doesn't go away, it's false knock. you could try the eehack tool in my sig to reduce timing like 4 degrees. if it still knocks, you know its bunk.
is this a stock engine? roller rockers? headers? different springs?
on my manual transmission car, for example, hard shifting will hit the knock sensor and produce maximum retard. personally i always limit retard at wot to like 3 or 4 degrees, so at least if it happens, it's not going to kill my run. you wont notice a degree or two, but you'll notice 10 ....
EVERY run the car pulled at least .3* right off the bat going WOT.
that could be burst knock retard, which is normal. it's knock retard just from airflow calcs, designed to reduce the knock that happens from a momentary lean condition as you stab the throttle... .3 degrees usually just rounds off to nil.
Car is stock motor with addition of CAI, full exhaust and dyno tune. I need to start somewhere so I suppose I will do simple things like verify voltage to KS as well as test KS itself.
This is the first time having ability to see this since buying/downloading TTS Datamaster to my laptop. SO I appreciate the explanation of the retard being present at beginning of WOT condition.
I want to verify that I can see this issue on the street as well during a 1st gear pull so i can keep retesting. It's not feasible to getting back on the dyno every time. Since I found that I can have Datamaster give me a specific "Knock Warning" screen that glows red when retard increases I can have my wife watch as I drive.
Without the laptop in from of me, on one of the runs with upwards of 10* retard I remember the knock counts are always increasing. Again sorry for my poor explanation of what's going on. I admit that Datamaster is giving data that I have yet to wrap my head around its impact on drive-ability.
I will get more accurate rpm/knock count relationship values when I get home.
I kept getting false knock to where I just told Solomon to disable my damn KR. It's running on e85 so screw it. Never figured out why it had this low rpm false knock issue. No KR at WOT though.
Is there a way to find this through DataMaster? The car was tuned almost 2 years ago at the same dyno I just put the car on this weekend. Have not touched anything since.
4500rpm(where maximum power dip shown on dyno):
Advance 26
Retard 10.6
Knock Count 2293
From 4500rpm to redline the Advance will creep back up to 40 as Retard drops but Knock Count stays at 2293 to end of pull.
I have 4.88 Volts KOEO, and 3880 Ohms through KS. So per Shbox's website those check out OK.
I may have to put my foot in my mouth as I found a witness "dimple" in a header primary tube. This is 4mm clearance as shown. I have since took a little off the motor mount to gain more room.
Last edited by smitty2919; 11-16-2015 at 05:29 PM.
despite your exhaust, perhaps you're getting timing retard because your car was tuned by a retard?
i realize that three data points is nothing, and i can't see your MAP, but running 37 degrees timing WOT? that's way too steep for a stock lt1, even with 93 octane, it should make near peak power around 34-35 degrees.
again, try my eehack program to pull a few degrees of timing and see what happens.
The shop that did the tuning is well known and trusted. They still had the original dyno run of the car and comparing this weekend to the intial dyno tune, the power dips down right about 4100 rpm then comes back up to mimic the power curve. I have not changed anything in terms of tune or engine mods so IMO something funky has happened all of a sudden.
Honestly, I'm not comfortable with altering any tuning parameters (pull a few degrees per your suggestion) as I do not know enough about it. Don't take that as being difficult as I am the one asking for opinions/suggestions.
Car had been running perfectly fine, no CEL, no hesitation/sputter. Honestly, if I never got on the dyno with ability to datalog I may never have saw this issue.
Those data points were the "highlights" I guess of that run. I can pull/relay the MAP pressure at these same points if that will help. But what I think you are saying is that you need to see the overall trend not just "instantaneous snapshots" of data?
I know I have always had false knock since I did the headers and dumps (the cam doesn't help also), and tried everything checked for the exhaust hitting (fixed), fresh gas, 100 octane, nothing seem to fix the problem, so we just set the knock retard to 3 degrees and left it. Didn't feel like messing with it.
Come the new year a heads/cam is the plan, but this issue "should not be an issue".
I'll wait until I have 1/4 tank left then refill with gas to see if it was inferior gas that is the issue. If not, back to square one. Motor mounts MIGHT be on the list if the stock ones have deteriorated enough over the 2 autocross seasons that car has been through since dyno tuning it.
I keep seeing 3 main reasons for knock...bad sensor, low octane gas, or a physical vibration of exhaust components.
Something I see conflicting information on is if the car is supposed to go into Open Look during WOT pulls. Can anyone confirm or deny this? All pulls that I logged, the car stayed in Closed Loop.
Last edited by smitty2919; 11-19-2015 at 09:01 AM.
Something I see conflicting information on is if the car is supposed to go into Open Look during WOT pulls. Can anyone confirm or deny this? All pulls that I logged, the car stayed in Closed Loop.
WOT aka power enrichment is not a mode with its own fuel table, it's a fuel multiplier against closed loop.
in WOT, it stays in 'closed loop' but stops updating the BLM and locks the INT, which means the o2s aren't effectively doing anything. then it uses the already gathered closed loop trim as a base (even though there's no direct feedback, so it's technically in an open loop control state), then and adds your power enrichment fuel to it.
as this is usually not what people want for a hand-tuned car, there's a code patch called a 'blm locker' that always locks your BLM to 128 on entry to power enrichment, which means you're always adding PE fuel to your base calculations of 128 or whatever, providing more consistent results.
Looking more into the above, my STFT go to "0" upon WOT pulls, O2 voltage stays in the 900-1000mV (vs fluctuating during cruise).
I have put in fresh 93 Octane last night and did another log this morning, still no change and a lot of knock/retard around 4000rpm.
The last thing I have not checked is the knock module itself. I remembered that when i had my car tuned, the tuner did the last pull and was basically done. He then wanted to make one final tweak, flash the PCM and off I went. Upon making the last PCM flash, the PCM got fried. So they sourced another PCM from a junkyard. Flashed the new PCM on the table top, threw it in the car and I left. I was aware of the possibility of killing PCM's in these Fbody's when tuning. Just my luck that it happened.
SO, since a verifying re-dyno run was never done after putting the new PCM in the car (and not having ability to log any data until recently), I suppose it's possible that I have lived with this issue since that day??? Meaning it's possible the knock module has been bad?
Tonight I will look at the KM in my original PCM and the one in the new PCM. Possibly even swap them to see if that helps or not.
It appears to be intermittent...Within 1 logging session I did 1st-2nd WOT up to 5200rpm and a 3rd get pull. All these WOT pulls did not consistently pull the same amount timing.
I narrowed it down to 4000-redline RPM and 200+ g/s MAF airflow. But again it's not EVERY time that condition is met that I have a lot of retard.
The 1st-2nd pull I only saw .3* retard max. 3rd gear pull I saw upwards of 10* again.
While I would like it to be that easy, I can't do that. It's there for a reason and something is going on for it to be sensing knock. If there is a serious issue and it's not "false knock" and I disable the KS, game over.
I've replaced the spark plugs with some NGK TR6's. Originals were AC Delco brand. I was hoping that maybe the Delcos were getting tired from 2 autocross seasons...but new plugs did not change anything.
I think it's time to leave the car with the tuner and have him take a look. For all I know, the tuner disabled the KS during tuning then turned it back on when I left as a safety precaution.
engine harmonics are a bitch, that's why there are different knock filters for different engines.
some combinations of rockers and headers or whatever cause false knock, some dont.
i've tuned gigantic cammed cars that clatter like hell that didn't produce false knock, and mild cam-only builds that did.
i'll say again, drop like 4 degrees of timing and see if it goes away, if it doesn't, it's false. it's the only way to know for sure. i wrote software that allows you to do this without flashing.
11-16-2015, 11:50 AM
knock sensor killing timing 
