Knock sensor question

Discussion in '2005+ Mustang GT 4.6L Tech' started by Laga, Aug 24, 2022.

  1. Laga

    Laga Senior Member

    903
    453
    I did download Forscan and waiting for cable to hook it up.

    In LiveLink, the label says “SPARK ADDER KNK SENSOR ARRAY[0]” That sounds like it is all the cylinders.
    The graph line goes up in a stair shape and goes flat for .5 to 1 seconds and falls erratically. The highest value is +7.5. Other datalogs the highest is +5.
    I’m tied up for next two days, so Wednesday is the earliest I can work on car.
    FDFE5AF6-A3D3-4E77-9E7E-633AE06601D8.jpeg
     
  2. Laga

    Laga Senior Member

    903
    453
    I sent Lito a datalog where the peak valves where only at +4. Waiting for his reply.

    GlassTop09, I installed forscan and this is what I got with driving around the neighborhood at 30 mph. DF1B1A2A-0FA0-4579-AAAA-052D130D391E.jpeg Later, I filled up with fuel and the ethanol content dropped from 80% to 76%, so I changed the tune. (Anything below 80% has a different tune), and the above values dropped by 40%. I think the new engine parts do not like the 80% tune. I then did a couple of short datalogs, and the knock went to around +4 except for a very sharp peak to +190 for a very short time. Do you know what would make such a high peak?

    Another test I did was to datalog just the knock sensors and drive over rough railroad tracks. If there was something loose causing the noises, this should register on the datalog. It did not.

    Redfirepearlgt, here are videos of it idling and a WOT run, and exhaust.
    https://youtube.com/shorts/JY_O06yn9DU?feature=share

    https://youtube.com/shorts/LFNsbTephT0?feature=share

    https://youtube.com/shorts/5QmULtL5UVI?feature=share


    I picked up a new knock sensor harness. The wires are not the same length for the two sensors, so the possibility of them being in the wrong position is slim.
     
    Last edited: Sep 1, 2022
  3. redfirepearlgt

    redfirepearlgt forum member

    2,496
    260
    Sounds real nice man.

    Love that pull clip.
     
    GlassTop09 likes this.
  4. GlassTop09

    GlassTop09 Senior Member

    880
    363
    Sorry for delay…….having major connection issues thru Centurylink so typing thru Verizon on IPhone…..

    That’s a lot of noise generation (I know you got a SC above these KS’s). What is the Forscan signal generated at idle?

    IMHO, this data pretty much rules out any vehicle side wiring/connector issue thus isolates the issue directly to the KS’s themselves. The only other item to check is the TQ to the mics……which you’d need to remove the SC.

    The sharp signal spike is 1 of the KS mics reacting to a perceived higher than normal frequency vibration from some noise/sound signal…..or a bad mic. This is where having these KS’s set up in Live Link to show individual cyl KS signals would help w/ tracking the issue down to which KS mic the issue is coming thru.

    Hate to say but you may have to address the KS’s directly. Good to know that B1 side is shorter than B2 side thus ID’s them……

    PS——What Lito is concerned about is any signal generation that is out of sequence w/ engine cyls firing…..the small spikes that occur at regular intervals….along w/ the height of signal spiking that doesn’t match the regular spikes from engine cyls firing…..the LiveLink graphs. That’s how he knows there is something else other than the engine cyls setting off the KS’s thus can’t isolate it from the tune side…….unless he had the car in his possession to do the physical follow up checks to find the culprit. This is what you’re having to do
     
    Last edited: Sep 2, 2022
  5. Laga

    Laga Senior Member

    903
    453
    Here are the knock sensors at a cold startup idle just now.
    A3401165-52BE-4EFE-8B06-62014929B33F.jpeg
    Yesterday, after getting fuel, and changing tune, they were in the 700-800 range.
    I’m hoping to hear back from Lito today. Last night, he asked if I was still on E85.
    I think you’re right about checking the TQ of the KSs.
     
  6. GlassTop09

    GlassTop09 Senior Member

    880
    363
    Finally, CenturyLink got off it & did their job.....................

    The reason why I asked for an idle run is to see if the KS signal changed from driving around at higher RPM's to idle RPM's or stayed around the same to indicate if KS's were stuck or not......

    Not apples to apples but here are some pictures of my engine using HPTuners VCM Scanner of KS's during a cold OL 1st start, then at OL\CL transition then at full hot idle to get an idea of what type of KS mV activity pattern you should also have seen: Note the KS's are being ignored here as engine load% is below the enable threshold setting.
    Knock Sensor 1st Start Open Loop 8-21-22.JPG Knock Sensor Open Closed Loop Transition 8-21-22.JPG Knock Sensor Hot Idle Closed Loop 8-21-22.JPG These are from my last tune revision. Here is a picture of a FORscan recording I made showing the KS activity during a full hot idle before I started doing my own tuning:
    FORscan Knock Sensor Closed Loop Hot Idle 1-3-22.JPG The KS's on my engine are the original sensors & have never been touched\disturbed to date.

    This is 1 of the main strength's of FORscan......the dashboard\graphing capability to be able to make baseline recordings of engine operations when all is good to then use to check engine operations later on to look for early signs of oncoming issues before they can become major issues\trip MIL's.
     
  7. Laga

    Laga Senior Member

    903
    453
    Took the car out to get a hot idle. 78E0C18A-1FB0-4EB3-9AC9-003B0746EA01.jpeg I have no idea why the values are that high.
    After that, I did a short run at slightly over half throttle.
    7C8A35C9-1C2B-45DE-A273-CC686FB729E1.jpeg
    I think the engine is not even close to being tuned, and maybe over torqued KSs. I wish Lito would get caught up.
     
  8. GlassTop09

    GlassTop09 Senior Member

    880
    363
    Yeah, since the KS mics are analog components that generate their own output voltages based on engine frequency\vibration level generation & have a specified TQ spec to prevent accidental mic overtightening thus distorting the KS voltage outputs, you have to go there if you cannot find the culprit of your KS issues anywhere else & you know these KS mics have been wrenched on.

    Sucks but is what it is...........

    After this is resolved, Lito then can work his magic.................
     
  9. Laga

    Laga Senior Member

    903
    453
    Update.
    It took 5 datalogs for Lito to adjust tune. Car is running great and it should be ok. He agreed it’s a good idea to pull SC to make sure KSs are in the right position, long cable on harness goes on drivers side, and torqued right. After that I’m going to add a couple of pounds of boost.
    Thanks again guys for your input. I was really worried there at the beginning. I’m going to skip the dyno for now, and just take it to the strip. When it gets cold, I’ll do the pulley swap and then dyno it. I also found a trans shop to beef up the trans to handle the new power.
    Glass top09, I am enjoying playing around with Forscan. Thanks for suggesting it.

    Redfirepearlgt, thanks for the help AGAIN. I’ll probably need your help in the future with a parasitic drain I’m chasing, but that’s for another day.
     
    Last edited: Sep 14, 2022
    GlassTop09 likes this.
  10. redfirepearlgt

    redfirepearlgt forum member

    2,496
    260
    NP man. Always fun and rewarding to help people that appreciate one's efforts. We can trace that down when you're ready. It will take some time, but I think we can hunt it down. In the meanwhile try to obtain an Amp clamp that measures DC amperage down to millivolts DC. Easier than trying to setup the MM to measure current and placing it in line with the Red terminal cable though we can do it that way. Most MM's are limited to 10Amps (fused inrternally), even the Flukes.

    In the meantime just make sure your battery isn't the source. I was just at the shop today and they had a 14 in there that demonstrated a current drain issue but ended up being a shorting battery cell on the battery. After being charged the residual was tricking their gear into indicating the battery was good though it was discharging internally.
     
  11. Laga

    Laga Senior Member

    903
    453
    I have an amp clamp. If it’s not the correct one, I’ll replace it. I’ll check battery this weekend.
    You might be right about it. I’ve pulled all the fuses and relays one at a time and didn’t find it. That was before I found the service manual procedure that states that they should be pulled one at a time, but left out until all are removed, then replaced.
     
  12. redfirepearlgt

    redfirepearlgt forum member

    2,496
    260
    LOL! Yip. You need to pull one at a time to find the circuit guilty of the problem. Send me a PM when you are ready. May be easier to work that way without distraction.
     
    Badd GT likes this.
  13. GlassTop09

    GlassTop09 Senior Member

    880
    363
    Good deal! Glad all got sorted out.

    If you already haven't figured this out concerning Forscan's dashboard use in Oscilloscope Mode, it uses the 1st 12 PIDs of the 28 total dashboard PIDs to draw the graph lines between the data points. So, whichever PIDs you want to have graphed, they need to be set up in the 1st 12 PIDs of the dashboard (starting from left side of dashboard counting in column from top to bottom).
    All recorded dashboard data can be saved & stored as a .txt file thus can be reviewed at any time in Forscan w\o needing to be attached to OBDII port.

    Hope this helps.
     
  14. Laga

    Laga Senior Member

    903
    453
    Yes, it does. Thanks
     
  15. Laga

    Laga Senior Member

    903
    453
    Update.
    I was finally able to remove the SC and get access to the KSs. Both were torqued to around 22-25 ft/lbs. Spec is 15 ft/lbs. Since the SC had been removed, I replaced the 18 year sensors and torqued them to the proper tightness. I changed the pulley on the SC to a 3.25” and sent a datalog to Lito. Naturally, he has to add more fuel, but he said the KS data looked more like what he would expect. This means the over torqued sensors were sending false knock signals.
     
  16. GlassTop09

    GlassTop09 Senior Member

    880
    363
    Good deal!
    Glad you got it all behind you. FYI, I've noticed on my engine that the KS mv signal voltage has increased approx 15-20 mv avg on both banks above where they ran before I started doing my own tuning, indicating that the tuning that I've done has increased noise resonance thus I'll assume for now that this is just that....noise from more powerful combustion events & not knock.

    I'll know soon enough though as we just got back home from seeing family over the holidays so I'm gonna get on installing my OBDII relocation harness to start datalogging.

    I think I'll also follow your lead & order myself a new set of KS's & replace them at 1st opportunity I get since they're 13+ yrs old.

    Question......did you apply some silicone dielectric grease between the KS's & block mounts or installed them dry? I was thinking to add a thin coat of dielectric grease to the block mounts before remounting them but I haven't found any info saying this is something that is a requirement thus am asking out of curiosity.

    That engine is gonna be singing off that 3.25" pulley!
    ;)
     
  17. Laga

    Laga Senior Member

    903
    453
    I installed them dry. I also did not apply any loctite to the bolts even though I was very tempted to.
    After searching on multiple forums, there were as many posts of guys using the loctite as those that did not. Since it’s not on the bolts from the factory, I decided against it. Plus, it could possibly affect the torque, acting like a lubricant. I did clean the mounting surfaces before installation. The KS wires are different lengths. The long side goes to the drivers side. Lito said it screws things up if they get swapped.


    I saw your plan to install a remote OBDII plug. Do you think there might be any problems splitting the signal while data logging? I know I cannot have my A/F WB O2 sensor plugged into the SCT4 FireWire and Interceptor gauge at the same time while using LiveLink, it gives false values that way. I have to physically unplug the wires from the back of the gauge. Even if the gauge is not plugged into the OBDII.
     
  18. GlassTop09

    GlassTop09 Senior Member

    880
    363
    The Dummy OBD Relocation harness I'm installing is not actually splitting the OBDII signal........the "fake" OBDII port is an exact replica of the car's original OBDII port (right down to the wiring schematics\coloring & conduiting) but it is not physically attached to the OBDII female connector (outside of the 12v+ & gnd) so any OBDII reader\tool attached to it will power up but cannot access the MS\HS CAN or any of the other wiring (is open terminated in harness) thus only the relocated OBDII port is fully active so should operate normally, just as if any OBDII tool\reader were plugged into the original OBDII male port.

    Now as for the HPTuners MPVI2+ interface usage concerning analog inputs, I'm assuming that this isn't an issue w\ HPTuners interfaces as they have set up the most used A\F WB O2 sensors\controllers natively within VCM Scanner software (my Zeitronics ZT-2, ZT-4 shows to be supported natively in VCM Scanner) & from all the videos I've seen\postings read on the HPTuners forums, should be a piece of cake & interface into VCM Scanner w\o issue. You can also set up other less known\used A\F WB O2 sensors & use them.....either on their own or thru a controller, some do need to have some twiddling applied to get them to work, but I'm counting on my Zeitronics ZT-4 to port both B1\B2 WB O2 sensors into VCM Scanner w\o a battle.......but this will be coming later on as I still haven't installed the ZT-4 yet so I'll be concentrating on all the other items for now, mostly thru CL-Normal Mode for starters since I can set up the tune to stay in CL to use the NB O2 sensors for fueling feedback here. I'll need the WB's for OL Power Enrichment\WOT & MAF\LWFM\SD tuning (PE\WOT runs in OL using the OL base fuel map w\o NB O2 sensor feedback thus a WB is a necessity & to accurately tune MAF\LWFM\SD tables a WB is also a necessity since NB O2 sensors can't meter A\F as opposed to a WB O2 sensor......all SO PCM airmass\load% calc's are products of some level of simulation......no actual MAP sensor installed to give accurate manifold absolute pressure downstream of the TB at all TPS angles thus accurate baro press upstream of TB.....even when using the MAF sensor to measure real air density\mass & temp for airmass correction of air flowing thru TB at all TPS angles).

    We'll see.
     
  19. GlassTop09

    GlassTop09 Senior Member

    880
    363
    Hey redfirepearlgt,

    While I was going thru some other post mortem diagnosis of parts I've changed out recently, I got around to running some checks on the bad set of KS's I pulled off my car's engine...................

    I ran tests using my Autel AL539B DMM using some back probes on the KS mics connector to test mic responses (found KS#1 mic was practically dead but will occasionally spike a very high voltage output on a light tap using a crescent wrench while KS#2 mic responded quickly & cleanly w\o spiking) & 1 of the items I found out is that using a DVOM to check KS mics olms readouts will be a useless endeavor as the KS mics will show up as an open circuit (OL or outer limit.....high resistance).......even though when lightly tapping the KS mics they'll generate a readable voltage signal on DVOM indicating that they're not blown thus still operate. I tried tapping on them w\ DVOM set up in olms to see if DVOM will show a value other than OL......no go, only shows OL or high resistance so they can't be tested thru the engine wiring harness using olms.......you'd have to physically remove them to test them for voltage generation output.

    Thought I'd pass this info along..........
     
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.