SOHC V6 knock sensor replacement

Background
There is a knock sensor installed on the SOHC V6 engine to detect detonation. The PCM reads the output of the knock sensor and retards the ignition timing when detonation is detected to avoid damage to the engine. One of the parameters I data logged for James Henson when he generated my custom tune to accomodate my 75 mm racing throttle body and Lightning 90 mm MAF sensor was knock sensor retard. He noted that on some of my WOT runs there was as much as 8 degrees of ignition retard due to the knock sensor. This reduced the total ignition advance by an equivalent amount which significantly reduced power. I purchased a tank full of the highest grade premium fuel available in the local area but the knock sensor retard persisted. Since I had timing chain rattle at the time I assumed that the knock sensor was detecting the rattle as detonation. James desensitized the knock sensor retard in the tune so he could complete his task. When he was complete he provided me with two tunes - knock sensor retard on and off. Unfortunately, when I replaced my camshaft timing chain components I did not replace the knock sensor. Subsequent data logs revealed the persistant knock sensor retard.

Note: The knock sensor retard can be negative (retard) or positive (advance). I have seen as much as 2 degrees of advance due to the knock sensor.

I ordered and received from Rock Auto an Airtex knock sensor (PN 5S2221 shown below) and a Fel-Pro EGR tube O ring (PN 71202).

Anytime the upper intake manifold is removed it is prudent to have a replacement EGR tube O ring available. I did not order replacement gasket sets (a mistake) because my existing gaskets were only a couple years old from installing the 00M12 kit.

Today I began the knock sensor replacement. I thought I could save some time removing the upper intake manifold by not disconnecting the PCM connector. However, the PCM wire bundle between the firewall and the manifold prevented me from extracting the manifold from the EGR tube even though I tried multiple times. Eventually, I disconnected the PCM connector and moved the harness out of the way to successfully remove the manifold. I noticed that during the earlier exercise I damaged the gasket loop nearest the EGR tube. When I installed the 00M12 kit I thought the upper intake gaskets were a poor fit to the manifold so now I have a chance to try a different set. My local Advance Auto has a Fel-Pro lower gasket set for $17 and a Fel-Pro upper and lower set for $16 (go figure). I'll pick up the complete set in the morning.

In order to get good access to the lower intake manifold mounting bolts I disconnected the engine harness main connector, detached it from the valve cover bracket and moved the harness out of the way. The lower intake manifold gaskets are in good condition but I'll replace them since I'm getting a free set.

Once the lower intake manifold was removed I was able to see the knock sensor in the valley between the heads and under the fuel injector wiring harness.

The injector wiring harness is fairly rigid and there is a head intake port that blocks inline access to the knock sensor mounting bolt with a socket drive extension. I was able to get a socket to seat on the head of the bolt by using a 3/8 inch drive wobble extension. Once the bolt was completely loose I carefully raised it out with an extendable magnet wand. When I pulled the knock sensor out by the end of its wires the loom disintegrated. Fortunately, I had plugged the head intake ports with paper towels so none of the debris fell into the ports. I ran out of time (2 hours) before I could figure out how to disconnect the knock sensor connector from it mate.

Good thread. How many miles on your engine?

I now regret not replacing this sensor on mine when I had the engine out with the intake off and the wiring harness removed. I wonder if my sensor is bad causing odd timing issues resulting in lack of power as well.

I should log the knock sensor signal to see, but I don't know what signal would be normal.

Thanks for taking the time to do this thread. As many times as I have had my intakes off on both of my trucks I have never once noticed that was there(always wondered where the harness ran to). Looks like I will be doing this soon on both trucks (just to change them, only thing it hurts is the wallet right ) as I need to dig back in my 98 to get rid of this stubborn coolant leak at the t-stat housing. Wonder if my 98 might be detecting the timing chain rattle as detonation as well.... hmm..

Again thanks.

symptoms

99SportX said:

. . . How many miles on your engine?

Click to expand...

About 157,000 miles

99SportX said:

. . . I wonder if my sensor is bad causing odd timing issues resulting in lack of power as well.

Click to expand...

From my driving experience the retarded timing was not really noticeable on my Sport. It just didn't have as much power as it could have. When the knock sensor on my 1987 Volvo went bad it would lose all power at low engine speed (just as I pulled out into traffic). With no OBD available in those days and being intermittent it was very difficult to isolate. It acted like low fuel pressure or faulty ignition module.

99SportX said:

I should log the knock sensor signal to see, but I don't know what signal would be normal.

Click to expand...

My excessive retard showed up around 2,500 rpm at WOT. It would rapidly step down to 8 degrees and when I let off the gas gradually step back up to 0 degrees.

Knock sensor & lower intake installed

Today I was able to disconnect the knock sensor electrical connector. There is a press down tab to release the connector but there are two plastic positioning plugs protruding from the bottom of the housing which leaves no room for fingers to provide leverage. I grasped the side of the housing with my fingers and grasped the tab with needle nose pliers. Then I pressed the pliers downward and pulled back and the connector sections easily separated.

I positioned the new knock sensor using a long shaft flat blade screwdriver and then carefully lowered the retaining bolt into the bore of the sensor using my extendable magnetic wand. Once the bolt was in the bore I used 1/4 inch drive wobble extensions and socket to screw the bolt into the threads. I positioned the at the 9 o'clock position relative to the front of the block even though the stock one was at 11 o'clock. Unlike for my Volvo 850 turbowagon, Ford does not specify an orientation for the sensor. The 9 o'clock position allowed better positioning of the wire loom excess length. I torqued the bolt to 15 ft-lbs. I compressed the loom enough to force it between the head front intake port and the rear of the thermostat housing so it wouldn't interfere with the installation of the lower intake manifold.

Next I replaced the gaskets in the lower intake manifold and installed it. The bores of the manifold do not exactly match the bores of the heads so I loosely installed all of the bolts and then slid the manifold around for equal mismatch on both banks. I torqued the bolts from center to out side in four stages (60, 90, 110, 120 in-lbs) to prevent distorting the manifold. The specified torque is 107 to 123 in-lbs.

Upper intake installation

Note: My 2000 currently does not have the variable induction system (VIS). If you have a 1997 or 1998 with VIS then will be additional vacuum lines and an electrical connection. There will be a vacuum reservoir between the upper and lower intake manifolds.

I replaced the damaged upper intake manifold gaskets with the new ones from the Fel-Pro gasket set. The new blue gaskets when pressed into position stayed in position much better than the ones that came in my 00M12 kit. The old EGR tube green O ring was still in good condition so I only coated it with vinyl conditioner to allow the upper intake to slip over it easier. I remembered to reconnect the engine harness main connector. When moved out to the way to gain access to the lower intake manifold bolts there is not enough room to move the engine harness to the correct position after the upper intake manifold has been installed. I was very careful and patient when positioning the manifold onto the EGR tube and slipping it over the O ring. There are many things that get in the way at the rear of the manifold: EVR tubes and wires; ignition wires; PCV valve/hoses, PCM connector and harness, vacuum hoses. I made sure none of these were between the upper and lower intake manifolds. I also remembered to connect the underneath vacuum lines (right and left sides) when the upper manifold was in about the correct position. I also connected the EVR vacuum and electrical connections.

There are short alignment protrusions on the lower manifold. The upper manifold drops slightly when properly aligned. The EGR tube holds the front down a little to keep the manifold in position. I pressed the front of the manifold down with one hand while screwing in the front torx screw with the other hand until snug. I repeated the process at the middle and then the rear. Then I screwed in all of the remaining torx screws. I avoid using power tools on the plastic intake manifolds. If a screw seems to be binding I back it out and restart it. I torqued the screws starting at the middle and working outward in four stages: 20, 35, 50 and 60 in-lbs. The specified torque is 53 to 62 in-lbs.

I pressed the three ignition wires into the retainers at the rear of the manifold. I connected the PCV valve hoses to the intake manifold right and left side. I installed the PCM harness ground wire stud, terminal and nut and connected the PCM connector (snug the bolt but don't overtighten). I connected the power brake booster vacuum hose. I connected the vacuum splitter hose.

Installation complete

I finshed hooking everything up to the intake manifold: main intake tube, valve cover vent tube, cruise control and throttle cables, IAC valve and TPS electrical connectors. Then I connected the MAF sensor and IAT sensor electrical connectors and connected the battery ground cable. When I switched on the ignition the Accusump brought the oil pressure up so I cranked the starter. The engine immediately started and ran normally. I won't know if the new knock sensor fixed my ignition retarding issue until I perform a datalog. I can't do a datalog until I download the SCT datalog program to my wife's new laptop computer - the graphics processing unit on her old laptop overheated and I haven't been able to repair it so I bought her a new computer. While the Sport was in the garage I noticed oil drops on the concrete so I need to fix that before doing an engine performance test at WOT.

Excellent. Is there any way you could get a screenshot of the knock sensor log when you do your verification? Just to show what normal operation looks like?

Save screen utility

I'll search for a save screen utility to download. There's been times in the past I wanted to post something displayed on my computer but was too lazy or didn't have the time to find a utility to do it.

If you have windows 7 there is a 'snipping tool' that works real well. Otherwise in any windows you can use the 'print screen' key which copies the entire contents of the screen to clipboard. You can also use 'ALT + Print Screen' which copies just the contents of the active window to clipboard. I usually then paste it into microsoft paint to save as .jpg.

Alt + Prt Scr

Alt + Prt Scr works great! Thanks for the tip. Below is a section of a past WOT run pulling uphill in 3rd speed before my timing chain components were replaced. Just before accelerating the knock sensor retard was 0. At 2900 rpm the knock sensor retard started to kick in and by 3000 rpm had dropped to -8 degrees.

At 3800 rpm it started stepping back up and had increased to -6 degrees when I released the accelerator at 5800 rpm. Spark advance at 4000 rpm was only 9.75 degrees resulting in a significant loss of power.

Updated LiveLink

This morning I downloaded and installed the latest available version of LiveLink from the SCT website. Below is a display of the same WOT datalog. I think I like the new version better because of vertical scales for one reason but haven't figured out a way to rapidly move foward and aft to display PID values at a particular time. In the old version just clicking the mouse at any time would move a vertical bar to that location and the PID values were displayed. Now it appears you have to play the datalog and set bookmarks.

Knock sensor retard is the left vertical scale and RPM is the right vertical scale. Vehicle speed shows 0 to 88 mph in 32 seconds (uphill grade all in 3rd speed (2 on transmission selector). I've found for WOT testing an uphill continuous grade pulled in 3rd speed greatly simplifies determining what is going on with the engine. The uphill grade slows everything down as if expanding the time scale. I've done several 3rd speed WOT pulls on a dynomometer and things happen much faster (no air drag on the vehicle for one thing) but a dynomometer is much safer and legal.

Excellent. Glad you got it to work. Now we will be able to see the changes with the new sensor!

Communication error

When I tried to set up a configuration file on the new laptop LiveLink was unable to communicate with my Power Flash - it did not recognize the part number. I tried using the old version of LiveLink on my desktop computer and received the same error so I sent an email to SCT support. Their prompt response was that for some reason my Power Flash had lost its part number. They said to call them and they could remotely reload the part number via the internet. I'm not fond of the idea of giving someone internet access to my computer but since it is new there should not be much personal or financial account data on it - and the only other option would be to send the Power Flash back to SCT. I am disappointed that the Power Flash has lost its ability to communicate with a computer. The part number should be stored in nonvolatile memory just like the custom tunes and the stock tune. Fortunately, I have a copy of my stock tune and custom tunes stored on my desktop in case SCT overwrites that data. The Power Flash still communicates with my vehicle - just not with LiveLink.

My Sport seems to accelerate better at moderate throttle but that may just be my imagination. I haven't tried any WOT tests and won't until I can datalog them.

correcting communication error

I finally had time to call SCT technical support to get assistance correcting my Power Flash communication error. I restored my Sport PCM to the stock tune. Then I downloaded the SF3 Power Flash Programmer Device Update Software from the SCT website. Next I ran the SCT device updater.exe and installed the SCT device updater. Then I performed an upgrade of the SCT drivers. Even though the device drivers were updated my computer still could not communicate with my programmer. Eventually, an SCT expert remotely logged on to my computer, installed a missing file, and was able to start the upgrade of the programmer firmware which took 10 to 15 minutes.

I decided to attempt a data collection of an engine start with the stock tune. The engine started and died almost immediately because of my 90 mm lightning MAF sensor but I was able to confirm my data log capability. Then I reprogrammed my PCM to the full knock sensor retard capability. I was pleased to note that the stock tune and three other custom tunes previously on the programmer were still there after the firmware update. The engine started and ran normally with the custom tune and I have been driving the vehicle for several days with no problems.

Do you know if the system is still retarding the timing?
Did the new knock sensor help?

no time for data log

Jasonx521 said:

Do you know if the system is still retarding the timing?
Did the new knock sensor help?

Click to expand...

Due to a family member being hospitalized I have not had time to perform a data log. It may be a few weeks before I have an opportunity.

still a problem

This morning I had time to perform a short datalog.

You can see that the retard is still at -8 when the engine speed reaches 3671 rpm. The overall spark advance drops to 8.5 degrees resulting in a significant loss of potential power. I don't know what to try next to correct the problem.

Did you ever get this solved?

still not fixed

99SportX said:

Did you ever get this solved?

Click to expand...

Not to my knowledge but I haven't tested again since installing the copper Champion Truck Plugs. I recently noticed a noise from either the serpentine belt idler or the tensioner pulley. Maybe it is being detected by the knock sensor.