SVX Network Forums Live Chat! SVX or Subaru Links Old Lockers Photo Post How-To Documents Message Archive SVX Shop Search |
IRC users: |
#1
|
||||
|
||||
Function of the Knock sensors
What is the function of the knock sensors. I know what they do but what are the voltages? I am data logging the air to fuel ratio, throttle position sensor, mass air sensor, and I would love to data log the voltages of the knock sensors. This being said, I know the pin voltages at idle (should be zero retard) read 2-4volts as per the voltage pin charts. What happens to the voltages when they pick up noise and the ECU sends the signal to reduce timing? Not sure whether we will see a change in voltage in either direction or whether a waveform signal is sent. If anyone knows please fill me in
Tom |
#2
|
||||
|
||||
Tom when you say you are data logging I assume that you are doing it with the Laptop? I might be telling you nothing but the value you will see will be simlar to what is in the raw data table Phil post not the adjusted value. Remember the ECU just retains a number that is liner. Here are the number from the table for maxium load
RPM Load 64 600 0 800 0 1200 0 1600 5 2000 6 2400 7 2800 11 3200 18 3600 15 4000 15 4400 10 4800 5 5200 5 5600 4 6000 5 6400 2 Tom in doing this I just realized something our SVX can put out heaps more power keeping the Knock sensor at 20 from 3200 to 6400, the buggers are pulling the power by pulling the knock sensor. Sorry back to the point and please jump in any one if i get it wrong, there are tables for the timing and the ECU will run thoses untill it gets a reading on the Knock sensor that gives a number that exceeds the value in the table at which time it will reduce the advance (could have this the wrong way). Keep in mind that a number of things are causing the Knock, Fuel & Timing the knock reading generated when a given amount of fuel, the current timing and the actual load or physical factors effecting the engine. An example of physical factor could be when try the engine on the 1/4 mile the engine will have less knock then when I am trying to speed up in sand, my engine is under greater physical load and can't rev out. I may be wrong here but I am sure that you can allow a higher knock at a higher rev with in reason.
__________________
1995 - SVX 700,000 K Mine, DMS Struts to lift car 2in. Tyres Wrangler Silent Armor 235/70R16, PBR Radiator. 6 speed with DCCD and R180 rer diff, Heavy duty top strut mounts front and rear. Speedo correction box fitted. New stero (gave up on the old one). Back seat removed and 2 spare tyres fitted for desert driving. ECUTune SC sitting in the box for the next SVX. 1992 - SVX 255 K Wife (Want to stay Married so not allowed to fit SC) 1992 - SVX Pearl with black roof race car roll cauge etc ready to race. Ex Tasman Targa car. 1995 - SVX Green low k mint condiation. 1995 - SVX Rally car, ex Matts car. Now to be used on track. 1992 - SVX red & Black being converted to Mid Engine. 1995 - SVX Red 143,000 bit rough. Owned 5 others Subaru back to a 1974 1400 GSR. |
#3
|
||||
|
||||
I am using an outside data logger that is meant to datalog the wideband with 4 other ports for 0-5v sensors
Tom |
#5
|
||||
|
||||
Tom,
Since the knock sensor is a piezoelectric sensor (accelerometer), with the piezoelectric element of the sensor designed to generate voltage whenever a certain level of vibration is applied to it (okay, so this is "knock sensor 101"), you should be able to hook a power supply to a spare sensor and rap on the sensor to see what its output is. -Bill p.s.I seem to remember the shop manual saying something about using a 'scope to see a waveform, but you should still be able to break the wave into discreet voltage "steps". You might want to give OT a call to see how the SSM outputs the state of the knock sensors.
__________________
Retired NASA Rocket Scientist Most famous NASA "Child" - OSIRIS-REx delivered samples from asteroid BENNU to Earth in Sept. 2023 Center Network Member #989 '92 Fully caged, 5 speed, waiting for its fully built EG33 '92 "Test Mule", 4:44 Auto, JDM 4:44 Rear Diff with Mech LSD, Tuned headers, Full one-off suspension '92(?) Laguna, 6 spd and other stuff (still at OT's place) My Locker Last edited by SVXRide; 11-09-2007 at 01:53 PM. |
#6
|
||||
|
||||
Quote:
The relatively high impedance of 560,000 ohms, would indicate that the ECU must incorporate amplification in order to interpret the raw signal, and the final output could consist of very clean pulse.
__________________
Trevor, New Zealand. As a child, on cold mornings I gladly stood in cowpats to warm my bare feet, but I detest bull$hit! |
#7
|
||||
|
||||
TOm when you say Log will you be driving the car or statanery?
__________________
1995 - SVX 700,000 K Mine, DMS Struts to lift car 2in. Tyres Wrangler Silent Armor 235/70R16, PBR Radiator. 6 speed with DCCD and R180 rer diff, Heavy duty top strut mounts front and rear. Speedo correction box fitted. New stero (gave up on the old one). Back seat removed and 2 spare tyres fitted for desert driving. ECUTune SC sitting in the box for the next SVX. 1992 - SVX 255 K Wife (Want to stay Married so not allowed to fit SC) 1992 - SVX Pearl with black roof race car roll cauge etc ready to race. Ex Tasman Targa car. 1995 - SVX Green low k mint condiation. 1995 - SVX Rally car, ex Matts car. Now to be used on track. 1992 - SVX red & Black being converted to Mid Engine. 1995 - SVX Red 143,000 bit rough. Owned 5 others Subaru back to a 1974 1400 GSR. |
#8
|
||||
|
||||
I will be driving the car. I know if I hook up my select monitor interface it will show me a full pull of 10* of timing under boost. I haven't even looked yet but I can almost guarentee it. The purpose of trying to datalog the knock sensors it to see the surrounding conditions using the TPS, MAF, and wideband to see what is causing the timing to be pulled... Am I making sense or does it just sound right to me?? Usually the manual states if the output is waveform or not and there is no mention of it for the knock sensors. I will tap the ECU inputs and just try it to see what I get
Tom |
#9
|
||||
|
||||
Quote:
I do not think that a vibration or a wave form as such. is involved The knock sensor incorporates a damping weight, so that a single output pulse is likely to follow each shock wave. The relatively high impedance of 560,000 ohms, would indicate that the ECU must incorporate amplification in order to interpret the raw signal, and the final output could consist of a very clean pulse.
__________________
Trevor, New Zealand. As a child, on cold mornings I gladly stood in cowpats to warm my bare feet, but I detest bull$hit! |
#10
|
||||
|
||||
queston is, how can I tap this output of the ECU??
Tom |
#11
|
|||
|
|||
Knock sensors are usually tested and observed using a scope, The diagnostic tool will display it graphically.
__________________
2007 GS 450h-Active Stabilizer/Radar cruise 1994 L Blue 3.70 VTD ASE Master Automobile ASE Advanced Level Specialist Toyota Master Diagnostic Technician (former life) |
|
|