How to test a Knock Sensor (P0328) - Subaru

[@vessellsbuzz]

I had a subaru with this code the other day. Had 5 volts bias volts unplugged. Plugged in bias voltage fell to 2.5 volts. Ran vehicle until warm biased voltage plugged in would fall to 2 mV, unplugged knock senor bias voltage would go to 5 volts. put another knock sensor in did the same thing. Had to get a new knock sensor from a different supplier to fix the car. But the key to being 100% confident that I had a bad knock sensor was the bias voltage being pulled to low. Another great video. Thanks!

[@SteveRobReviews]

That was cool. Another vid that shows how good having a scope is.

[@ritchieschofield1185]

Makes learning fun. So good at teaching. Thank you

[@ORGANISM44]

I love the way he teaches, thank you for these videos

[@2slofouru]

The best way to activate the knock sensor is with something pointed and close to the sensor, to try and induce something in the frequency range of most knock....4khz-9khz. Nice vid

[@ljm200787]

Another cool vid I'm so pleased I come across your channel learning so much I used to ohm out circuits all the time as that's how the vw guided fault system tells you. Since adopting your method s been fixing so many more vehicles. Thanks man and nice to see vw vid the other day keep them coming!

[@nancylewis7142]

thank you for all this information.

[@henryrobinson9837]

good video,i my get some classes,i may be old but i cant be out done,for lack of knowledge. You damn know it alls need to watch this and pay attention and respect,he done got mine and i'm an a-hole

[@davidjacobson9463]

Thank you again, re: sensor vs. wiring, "good" vs. "bad."

[@zorant5741]

Hello,long time ago I had problem intermittent Knock Sensor code.It took me long time to figure out since 10 years ago I didn't know for Scanner Danner and didn't have Fluke 98 Scope.Ended up running new shielded wire from sensor to ecm along fire wall away from main engine harness.My conclusion was that signal wire shield was damage somewhere in main harness on engine which interfere with alternator DC voltage cable in bundle.Wire continuity was ok and was no short between wire and shield /ground. I have one suggestion for ScannerDanner to make one video for DIY people when buying used car ,how with use of scanner try to determent car condition etc. rollback milage.

[@mohammedcherifi9206]

Why do you read 0v on the scan tool while current flowing through the circuit?? While on a map or any three wire sensor when you jump the signal wire to the 5v reference the scan tool reads 5v

[@glen-ozzstarscars8474]

Like you always say, unplug the sensor check voltage readings. Thumbs up to the voltmeter once again

[@Macmillium]

That knock sensor with the pigtail reminds me of interchangeable parts the with the use of the pigtail can be adjusted to a specific application.

[@waqasazmat81]

I've learned so much from your videos and your book. Had I lived in Pen state I would have definitely went to the tech school where you teach. Definitely appreciate your hard work, you taking your time out and posting these informative videos. Thanks Paul!

[@ScannerDanner]

My younger brother went to Penn State. Been there many times my friend. Thank you!

[@felipeclayton6710]

i realize Im kinda randomly asking but does anyone know a good place to stream newly released series online?

[@gusmajor3706]

@Felipe Clayton Flixportal :D

[@felipeclayton6710]

@Gus Major thank you, signed up and it seems like they got a lot of movies there :) Appreciate it!

[@gusmajor3706]

@Felipe Clayton No problem :D

[@j.helvie6563]

Stop now & check the internal resistance of the sensor. Record that & reconnect. Start & run the engine up to full warm thru while monitoring the bias voltage only. If it rises, you have an internal resistance problem due to heat soak. Confirm by shutting the engine off & monitoring the resistance as the engine goes into a koeo heat soak condition. It will rise considerably..... Replace & Retest to verify before returning it to your customer. Oh, by the by... In Nissan, a bad knock sensor will not turn on the MIL, but it will adversely affect fuel mileage.

[@contagiousFX]

so if the knock sensor doesn't have a bias voltage, how would you condemn a bad sensor?

[@JoelAutomotiveInaction]

Paul does piezo electric sensor needs reference voltage to operate ?I hade a 97 toyota paseo could never fix. I cheked voltage at sensor no bias voltage? code p0325 any way I saw voltage when inducing knocks on the scope in the aceptable range , replace with factory sensor code came back, I check wire install 12 head lamp in one end and send power onn other end pass test wigle test no drops competer pin conector all passes so I replace pcm , code came back , money back to customer.

[@jesusmalagon4336]

thanks Paul Danner

[@larazaauto]

I ready change the nock sensor and still the check engine light come on and the scanner show nock sensor problem

[@gritpump]

I need a scope today.

[@stuzman52]

Nice video Paul demonstrating the knock sensor. Have you run across knock sensors where they are more tuned to a specific range of frequencies where tapping on the block doesn't show up anything? These sensors are supposed to be more accurate to only show the frequencies pertinent to an engine knock. Keep up the good work!

[@dreamer2355]

Another masterpiece! So why did you wait till now to post this video from 2012? And I need to talk to the wife about subscribing to your premium channel :)

[@michaelgalvin1024]

Hi Paul Mike from Down Under. Did you get my message re Pico basics No2?

[@kingstone2451]

got a similar case but my bias voltage is 0.5mv and t doesn't get pulled down when unplugged. is it a must for the bias voltage to be 5v? or my circuit could be shorted to ground? visual inspection on loom seems ok.

[@ScannerDanner]

+Kingstone Joseph It depends on what you are working on. Not all knock sensor circuit use a bias. But yes, if there is a bias of 5v and you are only reading .5, the circuit maybe shorted to ground.

[@kingstone2451]

How can i tell the expected bias for a 1998 Subaru Sf5?

[@ScannerDanner]

+Kingstone Joseph not sure and never heard of an Sf5, but every Subaru I've worked on here in the US has had a bias voltage on their knock sensors.

[@kingstone2451]

+ScannerDanner managed to sort the issue, circuit was fine, sensor was faulty after replacement sensor unplugged bias voltage remains 0.5mv and doesn't pull down when plugged (don't know what circuit design is this). Thanks a a lot for your input. We really learn a lot from you here Kenya Africa.

[@mathewpetrini9123]

What if your ground from your engine to ecu differ? as much as .04 is thought to be acceptable potential difference in ground measurements but if your knock sensor is grounded on the block, and your ecu is grounded on the firewall or chassis the ground of each to the negative battery cable should be inspected. If there is a difference, (which there is in a car I am working on right now) the grounds aren't equally pulling. It says high input on your po328, and you saw it function how do you go to needing a new sensor? Its sensing? Well I am getting a low reading on my clients car,Po233 but I got some funny symptom told me. She said reads knock, intermittent starting clicking . radio lclock time loss at start up and a few other strange things. The knock sensor is made from a piezoelectric plate and is super durable, as you saw it function fine , but your input is too high, Now how does it know the input high or low with just one wire? Unless the ground on the block is a different frequency than the one on the ecu by only a small amount .Less is able to transfer. I have been researching the internet on this 6 hours straight, It really bothered me that it didnt make logical sense to work but at to high or low when its just a vibration activated current. And everyone who changes that sensor always has problem with it working properly. Also a common beggining was water in the engine bay, but not even too much to flood it or anything. Also Pieso-electric plates are waterproof.

[@2slofouru]

The subaru ecus are usually grounded directly to the engine block, via a set of cables that goes straight to a ground ring at the passenger rear top side of the block.

[@contagiousFX]

brilliant

[@hddm3]

sweet thanks

[@j.helvie6563]

However, wiggle testing while monitoring resistance can isolate this issue. I like to Keep It Simple, Stupid until you know you need a closer look.

[@efrencerda5459]

Excelent,.,..

[@TheYTCable]

"knock the sensor"! so ... in this case, the fix for this problem was, wiggling...

[@Y10Q]

It's a $10 part. Just change it and see if it makes it any better. Takes 10 seconds to change it out.

[@gsgleason]

It's actually a $130 part, which can be found cheaper online for as low as $70. Certainly not $10.

[@Y10Q]

Gregory Gleason I replaced the part on mine. it was $12 with shipping, and it cleared the check engine code. $130????? lol all sensors on 2000 subaru are are about $10 now.

[@gsgleason]

+Y10Q On eBay, yes. genuine OEM parts are a lot more. not sure of the difference.

[@Y10Q]

Gregory Gleason no difference. It's a just a microphone. As soon as mine was installed, check engine cleared up.