Knock Sensor Test?
#1
Knock Sensor Test?
Hi guys, I was wondering if anyone had a way to test knock sensors? I've read through a number of the treads and they seem to suggest installing the sensor in the hook to see if it works before tearing into the top end.
Specifically, I was wondering if anyone with a known good knock sensor had hooked it up to an AC volt meter and played 7khz at it with a speaker to see if it would respond properly.
If I were to set this up myself, I'd just go to the youtube recording of 7khz
and set the sensor right on my computer's speaker with it hooked into the volt meter at the proper setting and see what happens.
Has anybody tried this? Is this old news that everybody knows?
Specifically, I was wondering if anyone with a known good knock sensor had hooked it up to an AC volt meter and played 7khz at it with a speaker to see if it would respond properly.
If I were to set this up myself, I'd just go to the youtube recording of 7khz
and set the sensor right on my computer's speaker with it hooked into the volt meter at the proper setting and see what happens.
Has anybody tried this? Is this old news that everybody knows?
#2
Registered User
Join Date: Sep 2007
Location: San Francisco East Bay
Posts: 8,254
Likes: 0
Received 822 Likes
on
649 Posts
The first problem is "proper setting" for the voltmeter. How high is high enough? This is compounded by the fact that inexpensive meters are NOT RMS ac meters, so they respond correctly (mostly) to 60hz, but who knows what they do to 7khz. (But if someone has successfully tried this, I'm all ears).
I would be surprised if sitting a knock sensor on a speaker would transfer enough power to be useful. (Unless you've got giant MF speakers on your computer!)
So if I were to try this, I'd use a frequency generator that I could sweep around 7khz, to see if the output (what ever it is) peaks around 7khz. I'd use an old speaker with a broken paper cone (ah! now I know why I kept that!) and just duct-tape the knock sensor to the driver. But I have no idea if this would work (I'm not willing to pull my own knock sensor just for this test). As you say, someone would have to set the baseline with a known-good sensor.
Last, I too have read threads suggesting you can attach the knock sensor to the lifting hook. I have never seen one where it was reported that it works. And I can't imagine that it could.
I would be surprised if sitting a knock sensor on a speaker would transfer enough power to be useful. (Unless you've got giant MF speakers on your computer!)
So if I were to try this, I'd use a frequency generator that I could sweep around 7khz, to see if the output (what ever it is) peaks around 7khz. I'd use an old speaker with a broken paper cone (ah! now I know why I kept that!) and just duct-tape the knock sensor to the driver. But I have no idea if this would work (I'm not willing to pull my own knock sensor just for this test). As you say, someone would have to set the baseline with a known-good sensor.
Last, I too have read threads suggesting you can attach the knock sensor to the lifting hook. I have never seen one where it was reported that it works. And I can't imagine that it could.
#3
#4
#5
Registered User
Join Date: Jul 2012
Location: Northern Colorado
Posts: 1,776
Likes: 0
Received 109 Likes
on
81 Posts
I would download one of the free sound card oscilloscopes for my laptop, connect the wiring from the knock sensor to the mic input, and tap the block with a hammer. You should see the ringing pretty easily. Set the horizontal sweep rate to about .5 msec/div or so. I like this one. http://www.zeitnitz.eu/scope_en
You may have a problem if the input impedance of the mic input is too low. Piezo transducers such as the knock sensor like to see a fairly high impedance, a megohm or more. However, they put out upwards of one volt at that impedance, so if your sound card is more like 100Kohms, you may only see a hundred millivolts or so. Just turn up the gain - your mic input is good down to a few millivolts of sensistivity.
All of this will be uncalibrated, but if you had a buddy with a working sensor in his/her engine you could make comparisons pretty easily.
You may have a problem if the input impedance of the mic input is too low. Piezo transducers such as the knock sensor like to see a fairly high impedance, a megohm or more. However, they put out upwards of one volt at that impedance, so if your sound card is more like 100Kohms, you may only see a hundred millivolts or so. Just turn up the gain - your mic input is good down to a few millivolts of sensistivity.
All of this will be uncalibrated, but if you had a buddy with a working sensor in his/her engine you could make comparisons pretty easily.
#6
When I'm done, I should be able to tell if the s10 in the hook is indeed a working solution and also if I can properly test a sensor out of a vehicle.
#7
Registered User
Post the fix please
I really hope you fix your knock code and that you share with us what the fix was. I am dealing with the same nightmare on a customers 90 4Runner from Hell! I have replaced the sensor because it was submerged in coolant, I replaced the wiring with shielded wire straight to the ECU and grounded the shield to the E1 ground, I even tried another ECU from my personal 4Runner and still have the code. I even put my customers ECU into my personal 4Runner and it ran perfectly with it in my rig. I am at my wits end with this truck. I don't know what else to do with it. If anyone has any ideas please post them. I did find older threads, ones about relocating the sensor, and the one guys post about Code 52 that was very helpful, but none of his ideas worked. So, please post your fix. Thanks
Thread
Thread Starter
Forum
Replies
Last Post