bank 2 knock sensor repair bill
#1
bank 2 knock sensor repair bill
well I ve been putting it of far awhile and now car is at dealership getting it replaced, at id get restricted performance light and cel and it was nusense anyway... im waiting for the car to be done and was quoted 11hrs labor at $112.50 an hour..... I just wanted it done correctly and if there was an issue afterwards I would be able to take back and have it covered...
#3
They could be right on at 11 hours, they have to R&R the supercharger to get at the knock sensors.
There are probably some hoses and such under there that you should replace while they have the supercharger removed.
I need to replace my left bank knock sensor and I was thinking about trying to find another place to bolt a new one down and just leave the bad one in place under the SC and attach the wiring to the new one.
I'm going to post a thread with that question.
Vector
There are probably some hoses and such under there that you should replace while they have the supercharger removed.
I need to replace my left bank knock sensor and I was thinking about trying to find another place to bolt a new one down and just leave the bad one in place under the SC and attach the wiring to the new one.
I'm going to post a thread with that question.
Vector
#4
#5
I can't see any reason to put it in such a stupid place - you should 'hear' the knock anywhere on the head - at the worst you'll have to smooth the area down or use something like heat sink compound to get good acoustic coupling.
This is applicable to lots of 'R' engines - if it works please PM me or post a link in the other relevant sections.
#6
#7
Join Date: Sep 2008
Location: Damon /Houston, Texas
Posts: 7,254
Received 2,187 Likes
on
1,356 Posts
they could be right on at 11 hours, they have to r&r the supercharger to get at the knock sensors.
There are probably some hoses and such under there that you should replace while they have the supercharger removed.
I need to replace my left bank knock sensor and i was thinking about trying to find another place to bolt a new one down and just leave the bad one in place under the sc and attach the wiring to the new one.
I'm going to post a thread with that question.
Vector
There are probably some hoses and such under there that you should replace while they have the supercharger removed.
I need to replace my left bank knock sensor and i was thinking about trying to find another place to bolt a new one down and just leave the bad one in place under the sc and attach the wiring to the new one.
I'm going to post a thread with that question.
Vector
The following users liked this post:
Robert Page (03-02-2013)
Trending Topics
#8
It is rare (unknown) for me to disagree with you, Brutal.
But I do.
It is impossible to program an ECM to know the torque the knock sensor was fixed to the head.
'Pinking' has very specific frequency components.
The ECM is incapable of performing fast Fourier transforms on the knock sensor data so it can only be looking for frequency dependant peaks in regions not normally seen.
This will be a crude algorithm insensitive to sensor placement or acoustic coupling.
There may be some background adaptive function (listening to the general noise and adjusting gain to set that to an established level) but the output from the sensor on true knock will be at least an order of magnitude over background.
But I do.
It is impossible to program an ECM to know the torque the knock sensor was fixed to the head.
'Pinking' has very specific frequency components.
The ECM is incapable of performing fast Fourier transforms on the knock sensor data so it can only be looking for frequency dependant peaks in regions not normally seen.
This will be a crude algorithm insensitive to sensor placement or acoustic coupling.
There may be some background adaptive function (listening to the general noise and adjusting gain to set that to an established level) but the output from the sensor on true knock will be at least an order of magnitude over background.
#9
It is rare (unknown) for me to disagree with you, Brutal.
But I do.
It is impossible to program an ECM to know the torque the knock sensor was fixed to the head.
'Pinking' has very specific frequency components.
The ECM is incapable of performing fast Fourier transforms on the knock sensor data so it can only be looking for frequency dependant peaks in regions not normally seen.
This will be a crude algorithm insensitive to sensor placement or acoustic coupling.
There may be some background adaptive function (listening to the general noise and adjusting gain to set that to an established level) but the output from the sensor on true knock will be at least an order of magnitude over background.
But I do.
It is impossible to program an ECM to know the torque the knock sensor was fixed to the head.
'Pinking' has very specific frequency components.
The ECM is incapable of performing fast Fourier transforms on the knock sensor data so it can only be looking for frequency dependant peaks in regions not normally seen.
This will be a crude algorithm insensitive to sensor placement or acoustic coupling.
There may be some background adaptive function (listening to the general noise and adjusting gain to set that to an established level) but the output from the sensor on true knock will be at least an order of magnitude over background.
Vector
#10
well I ve been putting it of far awhile and now car is at dealership getting it replaced, at id get restricted performance light and cel and it was nusense anyway... im waiting for the car to be done and was quoted 11hrs labor at $112.50 an hour..... I just wanted it done correctly and if there was an issue afterwards I would be able to take back and have it covered...
#11
It is rare (unknown) for me to disagree with you, Brutal.
But I do.
It is impossible to program an ECM to know the torque the knock sensor was fixed to the head.
'Pinking' has very specific frequency components.
The ECM is incapable of performing fast Fourier transforms on the knock sensor data so it can only be looking for frequency dependant peaks in regions not normally seen.
This will be a crude algorithm insensitive to sensor placement or acoustic coupling.
There may be some background adaptive function (listening to the general noise and adjusting gain to set that to an established level) but the output from the sensor on true knock will be at least an order of magnitude over background.
But I do.
It is impossible to program an ECM to know the torque the knock sensor was fixed to the head.
'Pinking' has very specific frequency components.
The ECM is incapable of performing fast Fourier transforms on the knock sensor data so it can only be looking for frequency dependant peaks in regions not normally seen.
This will be a crude algorithm insensitive to sensor placement or acoustic coupling.
There may be some background adaptive function (listening to the general noise and adjusting gain to set that to an established level) but the output from the sensor on true knock will be at least an order of magnitude over background.
Without detailed knowledge of the software, we cannot judge the likelihood of software independence from the characteristics of the hardware interface. Educated guess ... the programmer had access to data regarding the response characteristics in the as designed configuration and exploited it fully. In other words, the software baselines against an expected mechanical environment.
You could get around it if the parameters could be tuned ... but then we wouldn't be moaning about the inability to tune.
My .02 euros.
#12
One easy solution would be to put a twin channel scope on the RH KS and a 'spare' fixed somewhere accessible on the L head.
The KS body needs to make good electrical contact to the head.
It's hard to believe the whole head doesn't transmit the 'ping' - KS location on different engines can be on the head, inlet manifold or block.
It looks as if there are two sorts of KS, resonant and 'flat' - ours looks like the Toyota type
http://highlanderclub.ru/files/manua...4h/cip0325.pdf
If that's correct the ECM listens to the KS background noise from the engine. IT has to see a certain amplitude or it sets a fault code so that would give an indication of good acoustic coupling but not, of course, the ability to hear a knock.
Does anyone know if the timing is trimmed individually per cylinder or globally?
The only video I could find showed a global timing shift when knock was detected.
The KS body needs to make good electrical contact to the head.
It's hard to believe the whole head doesn't transmit the 'ping' - KS location on different engines can be on the head, inlet manifold or block.
It looks as if there are two sorts of KS, resonant and 'flat' - ours looks like the Toyota type
http://highlanderclub.ru/files/manua...4h/cip0325.pdf
If that's correct the ECM listens to the KS background noise from the engine. IT has to see a certain amplitude or it sets a fault code so that would give an indication of good acoustic coupling but not, of course, the ability to hear a knock.
Does anyone know if the timing is trimmed individually per cylinder or globally?
The only video I could find showed a global timing shift when knock was detected.
#13
The following users liked this post:
steveinfrance (03-02-2013)
#14
Talking of replacing the hoses under the SC why you have it off, are there any better hoses to fit than OE ones or any way to protect them better so they will last longer. Apart from replacing both a knock sensors is there anything else under there you should just replace. In for a penny in for a pound!
#15
Join Date: Sep 2008
Location: Damon /Houston, Texas
Posts: 7,254
Received 2,187 Likes
on
1,356 Posts
Yes a knock sensor does go on frequency, and maybe i wasnt clear in relating my thoughts. A knock sensor can be made more or less sensitive based on contact with the surface which is why torque is important. Maybe moving it to a differant location wouldnt matter, this i cant say for sure. But i do feel that its a very important sensor which is why when they go out you get limp home/restricted performace because it leave the engine unprotected against knocking. And this is my point, i just wouldnt feel comfortable trying to reengineer an important sensor and its location based soley on lasyness/cheapness. Its penny wise dollar(pound) foulish imo. Use it as the oportunity to change the often leaking coolant hoses under the supercharger and maybe a pulley upgrade at the same time too
The following users liked this post:
steveinfrance (03-02-2013)
#16
I think you're probably right but the S/C, like the convertible top hoses, seems to have been a 'chuck it on' item - easy to do when building the car but a right pain to fix later on.
Are the two KS's in exactly the same place on an R as an 8?
I don't want to experiment now but if my LH KS goes I'll probably order one and try it 'off piste'.
Are the two KS's in exactly the same place on an R as an 8?
I don't want to experiment now but if my LH KS goes I'll probably order one and try it 'off piste'.
Last edited by steveinfrance; 03-02-2013 at 10:57 AM.
#17
Here's an idea: someone - whose (bank 2 I suppose, given the thread title) KS is working - bolt on a second KS on the same bank then look on a 'scope at the signals from both. If you can't tell them apart then moving it should be OK.
Not one for me.
(Don't connect the extra one to the PCM.)
Not one for me.
(Don't connect the extra one to the PCM.)
#18
Join Date: Sep 2008
Location: Damon /Houston, Texas
Posts: 7,254
Received 2,187 Likes
on
1,356 Posts
The following users liked this post:
steveinfrance (03-02-2013)
#20
Wouldn't play with it either as Brutal is suggesting, another position (and possibly also differently angled) will give a different reading, and not knowing how the software is set, you do risk more knock, as now it’s pretty fine-tuned as it detects it per cylinder.
But I find the 11 hours pretty high for the replacement, I would be thinking more like 6 or 8 (as long as the EGR will budge…).
But I find the 11 hours pretty high for the replacement, I would be thinking more like 6 or 8 (as long as the EGR will budge…).