Sharing my trial to repair a 4agze afm knock sensor

turbo10

500 RPM
Senior Member
Joined
Dec 11, 2003
Messages
626
Points
3,028
Location
sg petani kedah
sorry for the crap photo quality, i just wanna share my trial repairing my damaged knock sensor that been bugging me for quite a long time..i cant find one for an AFM 4agze unit as other toyota types are not working for me so i decided to see what really happened with my hidden knock sensor under the intake manifold..first of all i got this code 52 from diagnostic test and that confirmed my car been feeling sluggish and no power and high fc due to the engine went into 'limp mode'..then a peek under the intake manifold confirmed my knock sensor damaged badly..

298129_10150345368265098_572550097_8425034_1905315152_n.jpg


so decided to remove the remaining part of the knock sensor (KS). needed to dismantle a few parts. first, the AFM unit, then the piping going to the intercooler and throttle body, ABV, inlet piping to the supercharger and finally the supercharger itself..i thought it was a difficult thing but it was straight forward. to remove the SC from the engine i need to remove the radiator too. to remove the SC need to release the tensioner with size 12mm socket until the belt can come off. later remove the long bolt and nut attaching the SC to the mounting bracket..wiggle wiggle a bit and it will come off..

310941_10150345368320098_572550097_8425036_357858007_n.jpg


305272_10150345368365098_572550097_8425038_59558879_n.jpg

as we can see the screw hole attaching the throttle body to the intake manifold didn't align due to the taller 7A block.

320075_10150345368430098_572550097_8425040_1938792856_n.jpg


300518_10150345368685098_572550097_8425046_460962179_n.jpg


294675_10150345368755098_572550097_8425048_329462630_n.jpg


312127_10150345368595098_572550097_8425044_732109410_n.jpg

finally i able to access the KS other half.

to remove it we need to use 27mm long socket.

302005_10150349857385098_572550097_8458853_181537879_n.jpg

sorry again for the crap phone camera quality but i prepared it for reattachment of the open piezoelement wires (2 of them) that will be connected onto the terminal at one point.

300846_10150349857435098_572550097_8458855_1817008004_n.jpg

here we can see temporarily i attached both decapitated KS together after re soldered both wires as mentioned above. after that the terminal were inserted into the cavity at the piezoelement (one hole in between the two wires). next we will test the KS using an Oscilloscope. connect the channel 1 probe to the signal out terminal and the ground to the KS body.

294011_10150349857530098_572550097_8458857_1475909178_n.jpg

set the scope accordingly..

379964_10150349857475098_572550097_8458856_1641497486_n.jpg


295863_10150349857555098_572550097_8458858_242149777_n.jpg

Test began by tapping the front face of the knock sensor (the surface attached/contacted to the engine block) using anything metal or hard so that the piezoelement/mic/crystal could catch the vibration. we can see the waveform generated by the KS shown in the scope. I did a comparison of the waveform to a good one which was taken from this site

knock_sensor.png

Knock Sensor automotive waveform captured using a PicoScope automotive oscilloscope

so seems like the waveform from my KS exactly as the good one from the picoauto site and all i can say i was so happy..partially..hehe..my KS is alive again..yay!

387053_10150349857635098_572550097_8458861_637296062_n.jpg

later after confirming the KS is alive, i used the epoxy gum to attached the broken KS permanently and let it dry and retest again and seems the results are good.



but i am still halfway..i need to test it on the engine and maybe it can work or maybe no..because KS is a highly sensitive component, a slight modification or alteration might alter the signal out of it..if the signals coming out are out of tolerance the ECU might reject it and still throw out code 52. so i hope the signals are ok and my ECU doesn't hate it...i hate to see that f'ing code again..It worth a try for me as i can know more about my engine..and even if it can't eliminate the problem, i am still partially happy coz i did successfully repaired my KS..to be continued...:driver:
 
Last edited:
update! update!..i got some free time slot..so i managed to install the knock sensor and all the parts back. tested the engine in idle...guess what..no more code, my engine as healthy as ever..idle was smooth at the time of test but high..checked and found some vacum leaks and sorted it out..readjust the afm and fuel regulator back to where it was..road test..Alhamdulillah problem solved and the car felt lively and strong, fuel consuption and response back to normal..so it means knock sensors are repairable but only for the one with broken or open wire inside the knock sensor..piezoelement/crystal damage? i highly doubt it can be repaired..
 
Wow! You're really good at repairing knock sensor. If it was me, I'd just get another replacement instead of repairing it. Congratz!
 
I used the digital multimeter in my engineering lab haha .. seems like you've practice what you've learnt
 
Wow! You're really good at repairing knock sensor. If it was me, I'd just get another replacement instead of repairing it. Congratz!
thanks bro..me too if i could find one :P ..i've been searching for months for afm 4agze knock sensor, tried 4agze map on my afm 4agze, the code still came out, an info from club4ag stated that 4agze afm only can use 4agze afm one only, unlike map 4agze..map one can easily be found, afm one very rare...i think desperation caused me to try to repair..hehe..can't stand the 'limp mode'..
i brought the knock sensor to my office and borrowed the portable oscilloscope to test..hehe luckily learnt during my study years..:biggrin:
 

A thread every 60 seconds


Search

Online now

Enjoying Zerotohundred?

Log-in for an ad-less experience
Top Bottom