COMEBACK at 551k miles?! "Bomb-Proof" Sienna (P0325) 

I have a really great local mechanic that works on only VW and Audi. I have seen him find work arounds like this that really work well. It is a college town and he has a tremendous amount of customer turn over as students come in and leave town. When he does a fix like this he prints out the alterations that he performed for the customer to keep and provide the next mechanic. His thought is it could help out the next mechanic being able to show the log of these changes. Could decrease troubleshooting times. Most might be just getting tossed but I thought is was a really try to help out the next guy.

Mine was the best car I ever owned. 2002 Went 303,000 before we donated it. I had a problem with the wiring harness to the knock sensor being eaten by a mouse and also around 250 k. had the same issue with my wiper motor, two of the male pins were totally corroded out. I ended up getting rid of the female connector, soldering all the wires to the buses inside the motor housing and epoxying the connector opening for water proofing, restored all the functionality of the wiper motor and delay function.

My 2000 Sienna has 572K and going strong! Thanks for the video!

I preformed a similar repair on my son's RX300 (same engine) with 171k miles. Because the knock (bank one) sensor failed overdrive would not engage ; same code set. I tried the aftermarket sensors mounted on the valve cove to no avail. Final fix was to splice the remaining functional sensor with the wiring for the dead one. I did the splice inside the cabin under the dash. Has been working like a charm for 2 years. Others on You Tube comment that non-OEM sensors don't work. I would have to agree.

To be fair Ivan. I did knock sensors on a 2004 Toyota 4Runner a while back. Both sensors and the harness, all new OEM. Those sensors aren't cheap either. And one sensor failed in under 4k miles. This appears to be a known issue posted all over Toyota enthusiast forums. So, I wound up replacing the one bad sensor with a new Dorman unit and the MIL hasn't come back on for over 10k miles. I know, cheap aftermarket unit crap shoot. But thankfully, I got lucky.

In the 80s and into the 90s I owned Mercedes automobiles and they were built like tanks. Then my wife and I had two 1998 models, an E-Class and a C-Class, and they were just junk compared to the ones we'd owned previously. So we got rid of them, and one of our purchases was a 2004 Toyota Highlander with the 3MZ-FE 3.3L V-6. It's now almost 19 years old with 400K on 'er and still going strong. Toyotas from that era were expensive, but worth every penny. I could walk away from the Highlander today and be satisfied, but it has years of life left.

Man I watched Ivan test driving with his seat belt off and just squirmed. Don't die mate, we need your expertise!

Just to throw my 2 cents in. This is more for those people still learning about engine performance systems. But I've seen cases of false codes for knock sensors from a couple of other causes. For example, I've seen a couple of Honda's in my time, throwing both p0420 and the knock sensors codes. I've seen Catalytic converters that have gone bad and had one of the ceramic pucks break loose inside, and start making a popping/banging sound. And those vibrations have carried back through the exhaust into the engine block, causing the knock sensor to start picking up those extra vibrations, and throwing performance codes. Also on some of the Honda's that I've seen this, you will notice the Ignition timing will start to back off and retard, reducing performance noticeably. I've also seen straight exhaust leaks also cause false knock sensor codes. I guess anything that can knock or bang near the engine block can confuse the ECM. Just something to keep in mind in the future, to avoid firing the parts cannon.

This was technically a come-back. If the part installed goes bad, then it gets replaced if you have a warranty. The mileage/condition of the vehicle shouldn't matter. Unless you charge less for that condition as well initially (doubtful). That's why most mechanics charge what they do for labor hours and installations, it usually covers the occasional come-back repair over time.

Great Vid! I love watching vehicles that have successfully lasted so long (High Mileage) and continue to run great! It's a testament to doing maintenance regularly! Love your work...always top notch! You continue to amaze! Even has the right bulb in his stock to fix the check engine light! You're the best!

I used to have a 2000 sienna and I had the same wiper problem. The middle pin corroded off, wonder if it is common. Sold the van to my neighbor with 495,000 miles and it now has 613,000 miles. Never have had any major issues. The 1mzfe is the best!

Great job, yeah it’s disappointing to see that check engine light out, but you did the right thing. Hope customer appreciates it.

I knew you were going to replace that bulb. Even though it hurt. Good mechanic. Bought a 2000 Sienna off of a church that had been sitting for ten years, 87k miles. Fuel pump was bad, needed tank work, replaced pump with one off eBay, looked good, wouldn’t start, pulled front bank injectors, cycled them with 12 volts till click was strong, put them back, engine started, ran rough, after 20 minutes all injectors started working. Runs great. Still have to address air conditioning (leak). Didn’t realize the longevity of these things

I was hoping that linking the 2 wouldn't work. For me, that's a botched repair. Be better to at least use the known good knock sensor from the last job. I know it takes time, but sometimes we need to suck it up and just do it.

Excellent poor man’s fix Ivan. That was my knock sensor fix for my 2000 RX 300, only I did it at the computer, it seems the harness would be easier. Engine never knew the difference. I felt the car was to old and not great shape to do a proper repair and I used it mainly for work.

Good going! I'll bet future ECM generations will reject signals if they are identical on both channels! I've seen some tricky stuff like that in industrial applications/ Made it impossible to fool the computer!

That's a bummer! Can't even trust a good brand name part! That extra little problem of the money light took more time and work than the reason for the comeback. Great solution and all is good now 🙂

Great job Ivan keep it up.

Good man on the wipers, and then... CEL failure... No good deed goes unpunished, Ivan! :D

Nice work around. Wow will have to be careful of NTK in future, not had any probs myself, however worth keeping in mind.

My fav Toyota was a 1977 Celica GT with a 20R engine. It had maybe 400k miles when I sold it. At 28 years old Back then they didn’t have 100k digit so it could of been more miles . I personally saw it roll over twice. While I owned it . Great vid. Electrical issues are always a painful. Right now I am battling a brake light switch issue. It keeps on burning out . After a week.

What ironic insanity! Tip of the hat to Mr Murphy and creativity.

That's a very good engine and transmission and car overall to make it that far with no huge problems these old Toyota's there just brilliant

Nice job Ivan 👍

awesome work around. there will always be people saying "i wouldn't of done that" just because they want to argue over it. However, I think 95% of them given the option of a $700 repair bill or whatever it would be parts + labor on a 500k+ van would take the work around no questions asked. Plus-- as many have stated-- knock sensors are rarely even coming into play the problems are usually pretty obvious when you are getting knocking and pinging and with one working sensor it would likely still pick up on it (unless the computer compares the two to see what is different on both banks). Regardless I think you're a legend for these videos and showing the work arounds.

You need to check out the check seatbelt switch it was flashing the whole time you were doing the test drive! I love your videos keep it up.

A great video as always .Thanks .Keep them coming..

Ivan, what to say. Great job man. Thanks

Great video as always!

The decline of after market parts really accelerated when Covid got here and all the problems that it caused to the supply chain. I been buying what I need at the local dealer when available if not the local junk yard a used OEM part is better that the alternative.

Ivan, you are the man. I wish you live near me.

When you decided to tie knock sensor channels together I cried foul. Then I remembered way back when you replaced rings on that tired Tercel without braking the glaze on the cylinders. Worked out just fine and the other day when I was doing the oil consumption repair on the aluminum LS engine I saw outlined on the TSB to replace rings but do not hone cylinders if the top end cleaning and PCV updates did not work. I remember when I was in a EFI class a number of years ago the Toyota instructor explained Toyotas operating strategy for knock sensors. I would not have thought you could cheat it.

That check engine light out reminds me of a customers car that had some cluster illumination bulbs not working as well as the check engine light bulb. The poor thing had been on for sooooo long it not only burnt out but the heat from the bulb distorted the socket and aperture it fits into. Your typical O2 heater circuit failure had set it and I was surprised they also allowed the change out of that sensor when told that was an additional problem.

Ivan, don't know what to say. In a quandary about your fix. I'm talking about the check engine light of course. There is probably a huge market out there for the burnt out one. I'd at least list it on Ebay. Ha! 🙃Thanks for Sharing! 🙂

Wow, ntk supposed to be good for toyota motor. 551k miles??? Damn. That's very reliable engine! Didn't know you can put both channels on 1 sensor!! I figured it might cause a problem when it can detect the knock other part of the engine. Great fix, Ivan! That's cool fix on the wiper motor. Mine crapped out. Had to tear t to the vent part to get to it. 3 bolts, is out. Not to bad for 20 min fix. 15$ from junk yard lol. Awesome video!

I usually do these for passing inspections to turn cel off. You can do the same with the knock sensors on the early Chevy v8s and the v6 Toyotas where the knock sensors are in the valley of the engine. Also for a 2 bank early Toyota with a 420 or 430 not both codes you can jump the signal on the rear o2 of the good bank to the bad bank to keep the cel from popping.

You can also do this hack under passenger dash. Yes, it's temporary but works. Did it in the Rx300. Will replace with new Denso. Great recommendation

Not just no parts required for the knock sensor, but eliminating parts required. The high miles Toyota isn’t yet knock-knock knocking on heavens door.

Hi Ivan, I tend to agree with stevestacey5793. The design engineers put two sensors on the engine for a reason. If they didn't think two were needed they could have saved the money buying sensors and installing them. I do understand why you did it that way but as steve says we don't know if it will pick up issues the same way as if there were two sensors.

@Pine Hollow Auto Diagnostics The way you work is "worth its weight in gold". I would say the same about the way you are honorable to your customers so that not everything has to cost your shirt when you want something done by a MAN who really knows his job :) Ps: And always with a smile -I love it. Edit = typo

Hope this works out OK, personally I'd be a little worried about splicing the sensors as you lose sensitivity to detect bank 1 knock events - the knock sensor positions are critical for getting an even representation of what is happening on each cylinder, across the full engine speed range and without interference from mechanical events like valve opening and closing etc - I can't help but wonder how different/attenuated the signal looks to the PCM when it is listening to bank 1 events 'remotely' via the bank 2 sensor (like listening to the TV from the room next door) I know it's not some super high performance high CR / turbo engine with a huge tendency to knock, but with high miles and probably some oil consumption I'd still be a bit concerned about potential knock damage at low speed / high load when driving up a hill in high ambient temps or something... Maybe I'm just too pessimistic 😁

Well it should be good for another 500k now just like new great work buddy....Thanks and take care....

way to go Ivan! used a lot of common sense and saved your customer a lot of $. love your videos

NTK and NGK (not spark plugs) recently have been letting me down. OEM or nothing at this point.....except a recent subaru MAP sensor that was over $300. AIP electronics on Amazon for $24 bucks is still going strong 5 months later

Coffee, pancakes ,eggs, bacon, half million mile 1MZFE getting a PHAD massage. What a great morning! 👍👍

Ivan, love your videos. I’d love to see you make many years worth more. A little concerned you sometimes take test drives without a seatbelt?

Yeah, I used to think NTK was a solid brand, but I got a bad NTK neutral safety switch for a '99 Saturn SL2 this summer! I was so mad! The one before that my buddy grabbed at Advance Auto since he was in a hurry, and that one lasted 30 days. This NTK one was DOA!! OE is NLA, so not many choices left out there. Grabbed one from NAPA and crossed my fingers, and that was the winner. Now it's on my lift with a bad timing chain tensioner or guides at 115k miles. Not one of GM's better brands, that's for sure, lol. Engine / trans is great, everything else on or around it is low bidder junk! LOVE the knock sensor fix on this one!! SWEEET!! 🤘

Ivan, those were some brilliant repairs. The Stealerships would have replaced the entire wiring harness and yanked the the Knock Sensors out.

I like BMWs, they always come with two spare bulbs for the cluster. You can even test them by pressing the "park anywhere" button.

Man ivean. These sensors have a 50%+ failure rate . I went through 4 in 2 months . They were good at time of install and bad within a week. Always the same thing low or high voltage on the wave form. We relocated the sensors to the heads .

I suspect the reason why you can't get a resistance from the knock sensor is it may be using a piezo electric ceramic/plated disk. The normal piezo will work both ways, speaker or mic. If you wanted to test pickup sensitivity, hack the volume pot on an old radio and connect the piezo knock sensor wires to it. (Or your scope) A small toffee hammer might be needed to make the sensor work lol :-D Yes i do think up some silly things :-D

way to go again Ivan.

That happened with my 2002 Holden VY SS Commodore sedan (similar to the Pontiac GTO or Holden Monaro 2 door coupe) when a cheap Chinese Knock sensor (number 2) was giving me a P0332 code in early 2019,I replaced it with the Wells brand of knock sensor until I got a P0327 code (number 1 knock sensor)6 months later on the freeway at 110 km/h (66 mph). I thought - oh bugger iit,so I bought a brand new aftermarket wiring loom for the 5.7 litre Chevrolet LS1 V8 engine from an eBay seller in Hong Kong for $20 AUD including postage to Australia & I fitted a brand new Wells Knock sensor which is made in the U.S. I have had no more knock sensor problems since then but I had to remove the Inlet manifold twice to do the job & the engine has a heap more get up & go afterwards. So 2 rules I follow with the engine :- 1. Only use a reputable brand of knock sensor such as the Wells brand 2. Do not pressure clean or hose down any Gen 3 Chevrolet V8 engine where the knock sensors are fitted into the lifter valley because they can get drowned in water then rust & corrode if anyone does that !

Over 500k the proof is in pudding of quality of a Toyota and Great maintenance.👍😎🤙

You have adopted a Sienna!

Unscrupulous shops are removing those bulbs Ivan is installing them . Job security. Job security.

Sir, here in the UK at least, you can buy a NOCK SIMULATOR, you place it next to the sensor or near it whatever and it pulses, it can be connected to a coil pack to ping a given cylinder. Got mine from dad many years ago, but a great bit of kit.

I have joined the two knock sensors together on a 2001 lexus rx300.

I tried this workaround on my ‘92 Acura Vigor once. Light came back on about 1 week later.

A work around? You fitted the faulty sensor! You should foot the bill for replacing it!

This was the pinnacle of quality from Toyota. They are suffering more and more as they caved to new car technology. There's no way my 2021 RAV4 will go 250k without major repairs, like the other 6 Toyotas I own from the mid 2000's

I'd like to add that doubling-up on the same sensor might result in the knock signal getting weaker because the signal is now going into two different circuits on the engine ECU and that might mean that there could be a higher rate of knock. It might not be a huge issue, just be aware that the engine could have a higher acceptance for knocking now. The NTK sensor could as well be made in the same production line as OEM sensors, just ending up in a different path of delivery and if the other sensor holds together then all is good.

A genius fix

Hey Ivan. I hope all is well with you and the family. Toyotas are amazing. Just ask Scotty! Happy Holidays!

You are the best

I had Volvo S60 with dead knock sensor. Engine started to knock when warm and loaded, but no codes ever. Knock sensors are under intake, so i did what you succesteg and it worked..

I wonder if this Toyota will keep running until the wheels fall off. Nice work.

Given the age and mileage of that van, I can see going that route

Probably also a warranty claim with the knock sensor part maker, only 4 months old, most parts are 1 year? That is another positive feature with Toyota...you can walk into their parts department and order anything you want on the car...I had ordered nuts and bolts years back as some parts were stolen off of my '87 4 Runner and I wanted the OEM appearance vs. standard hardware store bolts that were zinc coated vs. the OEM that were coated with that brownish finish plus the shoulder washers on the nuts and bolts all one piece. Then, I needed some plastic parts and a few eletrical and like in this video, Toyota offered them. They weren't in stock with the dealership but two days later, the Toyota warehouse delivered them ! Thats darn impressive. BTW, I had a similar issue with an 89 GMC V1500 on the wiper park feature. In that design, they used a nylon cam and a metal arm that would "park" the wipers when turned off. That assembly was a seperate "module" that was attached by plastic ears and clips which were no longer offering spring tension. Had to order a complete motor assembly and used just that one little part for the existing motor and ta da.

hopefully the customer gives you a nice tip.. where were the knock sensors actually bought?

But Ivan, isn't the amplitude of the signal affected by WHERE you are hitting the engine block?? I would expect the bank 1 signal to be must stronger than the bank 2 sensor if the knock is actually occurring on bank 1.

yes can't blame you for a workaround for now..NTK,NGK hmm don't scare me Ivan, i thought those would be ok as a replacement for OEM, i got 2 coils 6months back for our mitsubishi..guess time will tell.

Lol well pleased love to see him flustered lol. Woot!

Ivan you should have pulled seat belt warning light while you were fixing the cluster then you wouldn’t have me warning you about not wearing a seat belt while driving! Again🤣🤣

This van deserves 1 million miles.

Ivan, Great video and fixes - thank you! Just out of curiousity: in this case if the customer wanted the sensor _replaced_ who pays the bill? Does your services/parts come with any warranty? Do your suppliers warranty the parts to you? Thanks Paul (in MA)

Basically the knock sensor is likely either contains a capacitor and a magnet+coil or has a capacitor microphone config where it has a voltage on the incoming wire and the noise causes the voltage to vary due to varying capacitance, that's why it has infinite resistance. Basically the same way a capacitor microphone works in your video camera and phone. If you have a multimeter with a capacitor tester like the Fluke 325 then you can probably test the sensor with a multimeter. But you need to know what the good value is to know if it's good or bad.

that little spider running at 13:20 LMAO

13:00 I wonder if water is intruding from the wire side?

Ivan, with the thinktool Pro can I reprogram one key with out issues. I only have one key but the buttons to lock and unlock weren't programmed before I had the tool

But how would you know if there's knocking on bank 1 then but that is a great idea to join them together for a few months or weeks or whatever and then replace them with all OEM parts

You are great, you fix that rubbish knocksensor Hum must bee junk I drive my car 91652,25 miles no problem , still runing And the funny its a french car From Sweden 🇸🇪

Scotty Kilmer would be proud of that repair.

Ivan, you really come up with some great ideas, thank you!

What you could of done with the knock sensor is to put a capacitor say10 mfd in parallel with the sensor to reduce the peaks slightly .

Ivan sometimes things just turn to crap. Parts quality today does not exist and I feel it’s not going to get any better probably worse. Anyway nice problem solving. 😊

I see almost as many rockauto boxes and magnets as I have. Lol

I guess you could say, Ivan, got "Cluster Effed" by a damπ light bulb.

When I worked at Toyota when we got a knock sensor code on the 1mz or 3mz engine, we always replaced bolth sensors and the sub harness from main harness to knock sensors Edit: not saying it was totally "correct" or "the right way" but it seemed like if you touched it then it went bad later on

I'm just wondering in the future, if the knock sensor fails can it be replaced and relocated to an accessible area and abandoned the original location? Just thinking... Save a lot of labor.

I would wonder how it would go if customer refuses oem, then wants it fixed under warranty or for free, after replacement with oem is recommended.

Question I have 2012 Jeep Patriot with a crank no start. I repaired the relay box and had it running for about 10min then shut it off to get cleaned up. Now back to crank no start. I bought all new relays from mopar and then replaced the throttle body next I replaced both vvt solenoids and then both camshaft position sensors. It started then shut off again. The next sensor I am going to replace is the crank position sensor. If that doesn’t fix my issue what do you charge to find the real underlying wiring issue. As you say the part’s cannon was fired at it lol.

what's the suggestion if you only have one knock sensor so no on vehicle comparison ... how would you determine if the amplitude was too, especially since i've rarely seen a spec for this in FSMs????;-)

gotta love a Toyota, feed her oil, water and DENSO parts and shell go 500,00K more nice work!

I have given up on the replace both philosophy, if you cannot find a used original sensor and you cannot afford or get the proper oem one then just replace one...chances are the original will outlast it.

So did the new-remanufactured wiper motor go back as a core or does it go to the customer?

Great job and excellent work around.

Ivan, is it possible the Dorman wiring harness started to melt and either ground out or cause a diminished signal? I've replaced these knock sensor wiring harnesses on several Toyotas because of circuit codes, including my Dad's truck before it had 100K. The insulation would crumble away. Dealer always had these in stock, so you know it was an issue. If the OEM was having trouble with the wiring being able to take the heat, it's probably a safe bet that Dorman was far behind the OEM engineers.

What if the Dorman sub-harness is the problem. Could have a bad crimp or terminal. Just thinking.

If I was the owner I would be praying for a comeback 😅