Knock and MAF Frequency

[DKneble]

Did a few searches through the archive to see if anyone is having this exact problem but not seeing anything that matches up.

Never had a problem before with knock. See profile for mods but I am tuning with DSMLink and have the knock sensor turned off below 4500 rpm and 85% throttle. Just moved from florida to arizona and due to the elevation change I went out to retune. Also, they only have 91 oct at the pump here

I mixed some 110 oct and 91 oct (50/50) so that should be about 100 oct correct? Tossed in some new plugs and went out to tune. Was running way to rich at first but by the end of the night I had a bigger problem.

Knock starts at 5200 rpm on every pull with AFR between 10.9 and 11.5, only running 12 degrees of timing up top and only running 21 psi. I started out conservatively expecting to bump up the timing and boost throughout the night.

In florida a month ago I was running 93 oct, 24psi, 12 degrees of timing and no knock...whats going on here?

I changed plugs from bpr7es to bpr8es expecting to be running more boost on the 100 oct...could this have done it?

Thanks guys.

 

[donniekak]

Welcome to az. Try free reving the car to make sure it's phantom knock. Arizona is a whole different world to cars, 91 octane, and the ambient temps 2 feet off the road are in the 130's. I gave up on pump gas a long time ago out here. I went e85 and ditched the huge front mount that was making my car run warmer than i wanted to in traffic, with the ac on.

 

[Bluesideup]

If you've moved to Prescott for example you are now at 5280 ft +\- so the altitude could be causing a problem as well.

 

[slowgsr]

what do you consider 'knock' what voltage levels are you seeing on the knock sensor?

 

[DKneble]

fantastic!!!

Well thats why I was going to run 100 oct all the time or an equal mix of 91 and 110. Note that I was up in the lake pleasant area last night for tuning so air temps were actually pretty nice.

I don't have the raw voltage values but at either 5200 or 5500 RPM (every pull) I would get a jump from 0 degrees knock to 4 degrees and stay there the whole way through the pull or until I let off.

I thought going from bpr7es to bpr8es would help this problem but was that going in the wrong direction?

What afrs do you guys run at higher elevations?

 

[DKneble]

here is a pull from today...23psi 3rd gear. Please keep in mind I was running 24-26psi on just 93 oct previously with no knock (in Florida on 100+ degree days)

The afrs provided are off a little bit...the dynojet one is pretty close to the value my AEM uego reads off...about .2-.3 leaner then the display says. I really don't pay too much attention to the estimated AFR.

Also, no knock free revving


EDIT: Did a couple more pulls. Noticed that the knock was occurring right at 2000hz on my MAF...the switchover point. Changed this to 2400 hz and then the knock starts at 2400 hz. Therefore this is coming straight from the MAF frequency switchover point???? Huh?

 

[pneumo]

Is that the switchover point where the ecu clamps the maf signal?
Maybe your boost est is off?

 

[firebirdvert305]

here is a pull from today...23psi 3rd gear. Please keep in mind I was running 24-26psi on just 93 oct previously with no knock (in Florida on 100+ degree days)

The afrs provided are off a little bit...the dynojet one is pretty close to the value my AEM uego reads off...about .2-.3 leaner then the display says. I really don't pay too much attention to the estimated AFR.

Also, no knock free revving


EDIT: Did a couple more pulls. Noticed that the knock was occurring right at 2000hz on my MAF...the switchover point. Changed this to 2400 hz and then the knock starts at 2400 hz. Therefore this is coming straight from the MAF frequency switchover point???? Huh?

Theres tons of problems I see with this car the way its sitting. First off let me address the switchover point, the ECU isnt switching over because it simply isnt selected to do so in the options. You have to active fake maf in order for it too "switch over", and use map to have a more precise reading utlizing a map sensor which you dont have or dont have logged. None of which are selected so it wont be applied to the ECU. With that being said now lets take a look at your airflow...have you ever wondered why you have to take so much fuel away from the fuel sliders when trying to dial in the proper a/fs for the fuel your using? Its simply because your airflow sliders are over-estimating airflow by 14%. I dont know who played with this tune, but whomever did it, did a horrible job on the airflow tables and had no idea what they was doing. I highly suggest you learn to tune the gm-maf the way its suppose too be dialed in correctly, when its not, the car runs like ass (hiccups, hesistation, misfiring) is a common thing on an impropely tuned GM MAF car. First thing id do would be to dial in the fuel trims to make driveability 100% correct (I know there off, im 100% sure.. I can tell by looking at the airflow sliders). If it was a customers car I would ZERO-OUT the airflow sliders and start from scratch...if you want help im here to help, but theres alot of work to be had in this log. Im willing to help if your willing to "give" me what I need...I need the following.

For Good Fuel Trims and Driveability I need a regular 10-15 min drive around town with the car at operating temperature. No boost, just steady driving (stoplight to stoplight,stop and go, highway etc.) This will let me help tune the cars driveability. I want you to log LTFT lo, LTFT mid & STFT, Coolant temp (which you have logged) and also front02(which you have logged but appears to be dead. Try raising the global fuel table too higher to 25 and see if the 02 sensor starts to log, if it does, remove it and clean it with brake-clean and a paper towel and reinstall it should work from there on out. Also FUEL TRIMS WILL NOT CYCLE W/O A WORKING 02 SENSOR AND WORKING COOLANT TEMP SENSOR.Lastly Log openloop to verify the ECU is actually going into closed-loop mode, this will tell us alot.

For WOT tuning - The log you provided is good enough, but needs alot of work. As I stated before the ECU is over-estimating airflow by 14%, this happens soon as the car begings to go, right at peak volumetric efficency, this is exacly where it begings to knock as well. Its really "hard" to eyeball and give you a perfect answer without an map-sensor logged but on the airflow tables, the 1200-1600 and 2000 sliders I want you to put them too ZERO, the 2400hz put that too -2. This is going to make the car run rich on the top end b/c now the ECU is applying the "proper" airflow settings, so now what your going to have to do is re-adjust the fuel sliders back up some to achieve the proper a/f you want. Timing is very conservative but id leave timing alone till we can dial in the tune better and get rid of this knock. Then we can re-apply timing values to where you want them.

Last but not least, I want to note that the coolant temps are on the edge, you can see after the log they get as hot as 206degrees, thats way too high, I run all my cars and customer cars without thermostats w/the fans wired on 247, this helps keeps cars running cool, all the time and maintains temps in the mid 180s, even after a hard-track run. Last thing before I head to sleep, under the MISC tab de-select "fuel-pump", it isnt needed to be highlighted and will only strain your battery key on, engine off and might leave you stranded one day. Good luck, you can pm me or even call me if you have any questions, this goes for any DSMLINK'er im here to help

 

[DKneble]

Where to start

I do not have a MAP sensor, hence why it is not logged. On my list of parts to order but not for a few weeks. Also, there is no front O2 sensor, just the wideband and rear O2. I have been told this is all the car needs to operate by numerous people (doesn't neccessarily make it right). Reason it was yanked was because when running the race pipe out the side of the front bumper there was no need. Please correct me if this is wrong...I do know that it makes tuning much more of PITA but the information I have been given in the past is that the car only needs EITHER the FRONT O2 or REAR O2 to properly swtich between open and closed loop modes. But there is no way to get the fuel trims logged then.

I see what you are saying about the airflow settings, not sure why he pulled them down so far past the cruise hz sliders. Typically in the past I was at -4% at 2000 hz with the rest between 400-2000 "feathered." I realize this isn't the most accurate way of tuning of course but it had worked in the past and made 723awhp on the dyno. Please note in the past I was not using the "active fak maf" feature as well.

I agree about the operating temps. I came out here from florida last month and wasn't expecting the temp difference to be so much...even at night the ambient air temps off the pavement are insane. The fans are wired and run 24/7 but I need to bump the tstat down to a 170degree.

Fuel pump is set on because the bosch 044 inline doesn't like to try to "suck" fuel according to bosch so its used as a primer. I have been told this is the safer way to go but please correct me if I am wrong...I will note that with the walbro turned off and the bosch on at 'key on', it sounds horrible but with the walbro pumping it sounds normal.

I will make some airflow adjustments and post another log tonight.

 

[DKneble]

Finally got to trying some of the suggestions mentioned. Attached is the log file. I tried raising the airflow sliders but this seems to have made things worse as I thought it would. My boostest is already well over actual boost so wouldn't I want to yank out airflow?

Anyway, the knock is still there, just at a different maf frequency but roughly the same RPM. The amount of airflow correction at this high frequency indicates to me that something else is going on but my boost leak test was fine...could something else be causing airflow loss?

BTW I have never had knock this bad before on pump gas and this is on 110 octane now.

Pull at 25psi 3rd gear

Air temp, 77 degrees.

Please help before I start ordering all kinds of parts to rule things out...first on the list is knock sensor I guess.

 

[firebirdvert305]

Dkneble check your pms.

 

[DKneble]

Knock between 2.5 and 4.0 degrees on 25 psi and 100 octane gas, no knock at 21 psi...same 11.0 afr. This makes no sense....use to run 24psi on 93 octane.

 

[firebirdvert305]

re-pmed you DK.

 

[DKneble]

Replaced throttle body shaft seals...does not leak at all at 40psi. Cleaned MAF.

Yanked out the knock sensor and...surprise...the tell-tale gooey black stuff everyone talks about. Put a new OEM sensor in and car runs like a champ...no signs of knock so far.

 

[firebirdvert305]

Replaced throttle body shaft seals...does not leak at all at 40psi. Cleaned MAF.

Yanked out the knock sensor and...surprise...the tell-tale gooey black stuff everyone talks about. Put a new OEM sensor in and car runs like a champ...no signs of knock so far.

Glad to hear it.