As promised, here’s more info on my project. But first, to give credit where due, I finally located the thread that first suggested this approach and want to give hats off to DIY4Penny who explains it very well in this thread (see posting #4): https://www.clublexus.com/forums/rx-...placement.html This is required reading.

Next, a few disclaimers:

DIY4Penny said it well: “I am not a car mechanic. Do it at your own risk. I just share my DIY experience.” Same goes for me.

Also, as mentioned before, this is a temporary work-around until I feel like tackling the serious project of repairing the leak(s) under the intake manifold. Which unfortunately is what eventually will need to be done on all of our cars that have this issue.

OK, so I read DIY4Penny’s write up and with that in mind here’s what I did:

For me it was necessary to remove the lower section of the air cleaner, along with the inlet snorkel. Making sure first to take photos of, and mark, the vacuum lines and connectors.

The first photo shows the overall orientation of both the connector that connects the engine harness to the chassis harness and the new “spare” sensor. Photo #2 is a close up of the connector (attached to the right side of the engine) that I had access to once the air cleaner was fully removed. This is my “after” photo with zip ties holding it back in place. Originally it was snapped into a plastic holder attached to the engine. It was fairly easy to disconnect the two parts of the socket, but really difficult to unsnap the holder from the engine – which was necessary in order to pull the engine harness far enough up to do the wiring work. I eventually had to pretty much destroy the plastic holder. (Thus the zip ties.)

The engine harness has four wires: red and green go to the front sensor (Bank 2), and black and white to the rear (Bank 1). I wanted to wire my new sensor to the engine side of that connector so I peeled back as much of the cable sheath as I could and cut the wires (black and white in my case) as far away from the connector as possible. I think I was able to get 3 or 4 inches to work with.

Then, using solder and heat shrink tubing, I connected a foot or two of similar sized wire to the existing wires going to the connector. After that, I was able to reconnect the two parts of the socket and zip tie it back in place - since now I had the “extension” wires within easy reach.

The next step was to find a place to mount the new sensor and I noted a bracket that attaches a metal pipe to the top of the intake manifold. This is the metal plate that’s underneath my new sensor in photo #3. The round sensor did not fit flush on top of the plate because of the metal pipe so I needed a spacer. I found that larger diameter nut (9/16) worked well as seen in photo #4. This is not threaded onto the bolt - it slips over it to act purely as a spacer. Then it was off to the store to buy a longer 8 mm x 1.25 bolt to hold everything in place.

Finally, I needed to connect my extension wires to the new sensor. Luckily finding the correct sensor connector was not as hard or expensive as I had assumed – see photo #5. I was originally looking at buying a whole new engine sensor harness for around $50 and cutting off the bank 1 connector, but this is the same thing and was only around $10.

Then it was just a matter of connecting the wires from the socket to the extension wires. The challenge here was that the connector wires are not color coded – both are black, so it was not clear which way to wire it. DIY4Penny said to just try one way and if it doesn’t work, swap them. That’s what I ended up doing but I’m not sure it really matters because I may not have waited long enough the first time. (See note below) Also, there are some posts out there that say knock sensors do not have a polarity.

So that was it. It’s been a couple of weeks and so far all is well. No lights, no limp mode.

Note: If I didn’t have my own cheap ODB code scanner this would not have worked well at all. It seems that every time a sensor is disconnected and reconnected, the ECU needs to go through its self test procedure that can take hours or days to complete. Until it finished and set all of the monitor flags, I was unable to tell if the problem really was gone or if the error code would reappear. Being able to watch the self test progress on the scanner was invaluable.

EDIT/UPDATE: Regarding which way to wire the sensor. As noted, the aftermarket connector I used has two black wires with no indication of which one goes to which of the colored (in my case, black and white) wires headed to the ECU. It occurred to me that if I'd sprung for buying the whole OEM sensor harness and cutting off the needed connector, the color coding would be obvious. So I searched for a good photo of the OEM part and it's included below. Note that each connector has the retaining clip area off to one side of the socket. So it appears that for bank 1, the white wire goes closest to the clip and for bank 2, the green wire is closest. This info would have made it easier for me to figure out which way to connect everything. And FWIW, I just checked and after my first trial, I did indeed end up with the correct arrangement.


Orientation

"After" photo of the engine to chassis connector

 

After reading this Thread & doing a bit more research, I purchased 2 new knock sensors ($60 each vs $300 each from Lexus), wiring harness, set of gaskets, & bypass coolant hose for about $225 from RockAuto. Trusted mechanic did labor for $300. Coolant leak was the problem. But engine performance is back to its beautiful pre P0328 self.

 

I just hope i dont get any codes popping up any time soon. so far i havent had any in the almost 4 years ive had my 07 rx350. thankfully. I hope i dont anytime soon as i dont have the money to fix it right now.

 

Sounds like a good outcome. Did you save and/or inspect the old harness and sensors? Curious to know if the one that was throwing your code had corrosion from the leak in the connector, or if the sensor actually failed internally.

Thanks,

 

Can I please get a break! lol

Got my emissions code fixed last month and it hasn't been back- BUT- for the last couple of weeks I have been getting a P0328 Bank 1 knock sensor code and for an extra bonus I sometimes also get a random P0301 cylinder #1 misfire code with it too for extra added fun...

The other day I swapped out the cylinder#1 coil and the misfire code hasn't been back but its's only been a couple days so too early to tell if that did anything.
Luckily I swapped the ignition coil this time without pulling the manifold- YES this is possible because I just did it and I didn't think it was possible- took 10 mins, will post pics in another thread.

Really not looking forward to pulling the manifold to access the knock sensors but at least the sensors are above the coolant valley plate so I won't have to mess with that again.
I appreciate this thread and the info, it looks like rock auto is recommended for parts and if anyone else has tips please let me know! Thank You.

Just looked at Rock Auto and they have sereveral brands of knock sensors, which ones do ya'll recommend???

 

Of the brands rock auto has, I have only heard of BECK/ARNLEY, DELPHI and STANDARD MOTOR PRODUCTS. Of those 3, I used a part made by Delphi once and it worked ok. Don't remember what the part was, but it was not a knock sensor. I would be real afraid of the one they have for $17.

Don't know what the difference between the ones they call ECONOMY and DAILY DRIVER might be.

 

I have bypassed my knock sensor, I have an 04 and a couple years ago my Bank 2 code kept coming up. The car ran fine if the code didnt come on. I checked with toyota professionals and they thought it was a common faulty sensor or bad wiring harness. 1500 to replace, I decided to bypass it. You can find the information here in the forum and by searching on the net. Basically you find the two bank sensor wires under the dash and cut the wire that goes to the failing sensor and piggy back it onto the good sensor so the computer thinks both are working. Its been two years and i havent had any issues at all. Car runs and drives great.

 

Good info thank you.
I'm glad I'm not the only one afraid to buy the economy parts.

 

Great idea, I can confirm a bad knock sensor by doing this hack first if it works. Thank You, Brilliant!
I appreciate you letting me know this because I didn't think it would work but it doesnt look like it would hurt anything to try it, looking at the drawing(5V pullup resistor).



The KNK wires below look like they are the ones carrying voltage and the EKNK wires are the gound.





Knock sensors are using Plug E9 on the ECM in case anyone needs it, RX330.






Plug E9 pins pins 20 and plug E9 pins 28 are the knock sensor ground wires from the knock sensor drawing above and I can confirm they are internally jumped in the ECM, I just tested it on my spare.




So it looks like only the single KNK wire needs to be moved from the bad sensor to the good sensor as mntntrls71 is saying above.

 

Yep thats all the info i found a couple years ago. It worked great and we have put about 14k miles on the car after the change. I had just done a complete water pump and timing belt change due to a idler pully freezing up and shredding everything. It happened the week before the appointment to change the timing belt. My Toyota guys I use are great and figured it was related to what had happened. Anyway I figure my car is worth about 6500 dollars with 114k miles on it. I didnt want to spend 1500 for a bad sensor if I didnt half to.
just dont cut the wrong one and leave the good sensor wire in tact and splice the bad sensor onto it.
good luck.

 

The local Lexus dealer near me wants over $300ea for the knock sensors and I'm thinking they should be replaced in a pair since the other one can fail next.
So that's almost $700 for sensors plus whatever the knock sensor replacement harness costs plus labor so I can see a $1,500 bill easy.

Your hack only took me 20-30 mins to pull the glove box and make the splice taking my sweet time and running tests.
I haven't put the glove box back on, I'll do it another day.


Spliced the KNK1 & KNK2 wires together at ECM and made s short jumper to the KNK2 wire that goes to my knock sensor 2.
KNK1 is my kock sensor 1 that is bad and throwing codes so that wire was left unhooked going to the sensor.




The code is uaually thrown on the first start of the day as I'm 1/2 block down the road so I have to pull over and clear it to get out of limp mode.
Took it to the store and it's good so far and no codes or pending either so far.
Hopefully this will buy me some time to get the sensors and new harness from Rock Auto.

Update 09-02-2021: No check engine lights so all is well so far. I will be ordering the knock sensors and the knock sensor harness from Rock Auto.

 

Thank you Dennismik,

Please help, getting ready to order knock sensors from Rock Auto and I really don't want to pull the lower intake manifold twice for bogus parts!!!

>> Anyone used any of these with success as an alternative to OEM from Lexus???

Thank You in advance!

 

I see this thread is a couple of years old, so i hope that it is still alive. I have the P9328 code and it shows to be "sensor 1 Circuit high. Bank 1 or single Sensor." My vehicle is a 2009 RX350. Looking at the info here, I see the images of the ECM and connectors, refer to an RX330. I wonder if the connections are the same for the 350.

Anyhow, I am finding that sometimes, the CEL and other indicators goes away. That is, without my clearing the codes. i would think that even if the problem is due to a connection which is intermittent, that the CEL would not clear by itself. But it does make me suspect a connection problem rather than a bad sensor.

Another point which makes me curious about this alarm is that when it is present, I can detect no change in engine performance while the code is present. No sluggishness, which I have read is a symptom. No change in RPM or reduction in MPG per the instant or average MPG indicators.

Anyhow, the vehicle is a 2009 with 266 K miles on it. Runs great and we hope to keep it for a while but I'm not sure I want to fork out the $1500 or so to get to the actual sensor. I'm thinking that the fix described here should do fine.

Dan

 

The answer is NO. The electrical systems of of the RX330 and the RX350 have many differences. Many electrical parts/modules may look the same, but inside they are different. The theory of this knock sensor hack will work on the 350 as well as any make of car that has 2 knock sensors. The ECM for the RX330 is slightly different from the RX350.

RX330



RX350 knock sensor wiring







The only information for the RX350 ECM that I do not have is the connector numbers. Both ECMs have the same number of connectors and the connectors are keyed the same from left to right. Both ECMs have the knock sensors going into connector E9 of the ECM, but connector E9 may not be the 2nd from the left on the RX350 like it is on the 330.

The thing to understand is that the ECM will detect a problem and turn the check engine light on. If the problem goes away, the ECM does not turn the check engine light off. Some of the codes will be reset after some amount of time, but not all of the codes. Maybe what is going on with your knock sensor is intermittent. Go to an auto parts store and ask them to hook up their OBD code reader, check and see if the P0328 is still there and reset the code if it is. Then drive the car and see if the check engine light comes back on right away.

 

thanks for the clarification on the differences between the wiring of the 330 to the 350. I sort of suspected that would be the case.

As to the code issue, I do have a OBD reader and that's how I knew it was the 328 code. Most of the time the code and the CEL have remained until I reset it. A few times, the, upon starting, or while driving, the light went off. And it might stay off for a time and then return. Sometimes while driving and other times upon restarting the engine. I have not had the chance yet to plug the OBD reader in during the time the CEL light was off by itself to see if the code is stored. I suspect that it is, but I can't be sure.

This extinguishing of the CEL without me doing a reset through the reader has happened only a few times. Most recently a few days ago,. But now the light is back on. Most of the time only after clearing the code does the light go off.