2nd Generation Specific (1986-1992) 1986-1992 Discussion including performance modifications and technical support sections.

Help: SAFC showing Engine knock even at idle

Thread Tools
 
Search this Thread
 
Old May 13, 2007 | 12:38 PM
  #1  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
Help: SAFC showing Engine knock even at idle

Hey guys, ever since i installed my SAFC this winter i have been able to monitre my engine for knocks. Since this summer has started, my engine has been shoiwng some knock (7-18) even today, at idle, after 5 mins of running on a cold start, its been showing knock. No audible knock. 54K on origional motor. Ive been running 93. On cold nights it doenst knock AT ALL but any time its a warm day it knocks. DOes anyone know the cause or cure for this? Even while i water-treated my motor today it STILL knocked during treatment. I dont understand. I thought knocks was caused by pre-ignition, which i thought was caused by the rotor surface or housing surface getting too hot, high enough to support pre-ignition. Sucking cold water directly into the upper intake manifold shouldve cured this. Any suggestions is appreciated.

87TII
Reply
Old May 13, 2007 | 12:41 PM
  #2  
SirCygnus's Avatar
whats going on?
Tenured Member 20 Years
iTrader: (1)
 
Joined: Mar 2004
Posts: 4,930
Likes: 8
From: atlanta ga
some knock is ok as much as i understand. when it gets very high i think triple digits is when it gets bad.

i dont know. this was on my bros wrx. i know a rotary might be a bit more sensetive. but in any case, it would still show knock on that car. around 1 to 10 sometimes. just interpherence i think.

i would worry about it unless it gets higher than normal.
Reply
Old May 13, 2007 | 01:01 PM
  #3  
AllMotorRotor's Avatar
Banned. I got OWNED!!!
 
Joined: May 2007
Posts: 170
Likes: 0
From: Home
What you are possibly seeing is just electrical interfierence or noise, which ever is the correct term. Its says it right in the manual. I have the same on mine and every SAFC I see.
Reply
Old May 13, 2007 | 01:09 PM
  #4  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
im having doubts that its electrical interference...Even with a 1% correction increase in fuel, the knock goes away. I thought our motors already ran rich as hell lol. Replacing my o2 sensor might help i guess. But yeah, it couldve been doing this it's whole life, just now i can monitor it. hmmmmmmmm
Reply
Old May 13, 2007 | 01:15 PM
  #5  
j9fd3s's Avatar
Moderator
Community Builder
Liked
Loved
Community Favorite
iTrader: (3)
 
Joined: Mar 2001
Posts: 31,835
Likes: 3,232
From: https://www2.mazda.com/en/100th/
the knock sensor is a microphone, it picks up noise. it is tuned to pick up knock, but it will also pick up other drivetrain noises as well.

but if you richen it up and the knock goes away, you have your answer. "rich as hell" is relative.
Reply
Old May 15, 2007 | 03:28 PM
  #6  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
So far, what I have seen is that the knocking is DIRECTLY related to ambient air temperature. Cold start or hot, coolant at ANY temp, at idle or in vacuum or in boost (when not correcting), it knocks. Today its 82F and i started it cold and it was already knocking at 7. Im wondering if my IAT sensor is bad. Or, it could just be bad OEM fuel maps for high intake air temps... But it is true knock, adding fuel completely eliminates it...This takes some more thinking. Any ideas guys?
Reply
Old May 15, 2007 | 03:58 PM
  #7  
SirCygnus's Avatar
whats going on?
Tenured Member 20 Years
iTrader: (1)
 
Joined: Mar 2004
Posts: 4,930
Likes: 8
From: atlanta ga
through deductive reasoning i have concluded that....


your car is broken.
Reply
Old May 15, 2007 | 04:22 PM
  #8  
cptpain's Avatar
Torqueless Wonder
Tenured Member 15 Years
iTrader: (1)
 
Joined: Oct 2004
Posts: 2,020
Likes: 3
From: Texas
well the S-AFC not known to be an accurate knock sensor, its just a damn piggy back
Reply
Old May 15, 2007 | 04:51 PM
  #9  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
cptpain - My SAFCII is not a knock sensor at all, all its doing is showing what my OEM sensor is reading, which i believe is accurate. And i realize all it is is a "damn piggy back", but that still doesnt explain why my motor is knocking. I see alot of threads on here with people having this same condition, yet noone seems to have an answer.

Haha, and yes, my car could be broken. I'm just trying to play it safe.

Anyone have luck with alcohol injection?




P.S. Since our IAT is incorporated in our AFM's, im gonna recheck the AFM recall we have to make sure mine isnt one of the defective ones, i gotta wait till tomorrow to check the database again (ALLDATA)

Last edited by BoostinFC; May 15, 2007 at 05:02 PM.
Reply
Old May 15, 2007 | 05:53 PM
  #10  
Sideways7's Avatar
Lives on the Forum
Tenured Member: 20 Years
Liked
iTrader: (7)
 
Joined: Dec 2003
Posts: 6,598
Likes: 10
From: Temple, Texas (Central)
Unless turbos are different, the IAT sensor is completely separate from the AFM.

Also, they generally only run super rich under WOT.
Reply
Old May 15, 2007 | 09:21 PM
  #11  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
hmm, ill have to check my sources again. (I see what you mean Sideways7, the sensor below the throttle body i believe? More research i guess. If this proves true, ill test it at the shop probably next week or the week after) Either way, ill double check the serial # for the AFM recall.

The problem is, ive been tuning based on my car's twin's AFRs which I know is a bad idea. Ill have a wideband in a couple months, I just gotta get outa college first (End of June Thank GOD)
Reply
Old May 15, 2007 | 09:24 PM
  #12  
RotaMan99's Avatar
Rotary Freak
Tenured Member 10 Years
 
Joined: Dec 2006
Posts: 1,791
Likes: 0
From: New Hampshire
cptpain - My SAFCII is not a knock sensor at all, all its doing is showing what my OEM sensor is reading,
The reading you see will also be higher then what is actually happening because of electrical interfierence.

Unless turbos are different, the IAT sensor is completely separate from the AFM.
There is a IAT sensor in the AFM. Between that flap and the IAT in the AFM, thats how the ECU determins the amount of air flowing into the the engine.

Not to be confused with the IAT on the manifold.

Last edited by RotaMan99; May 15, 2007 at 09:36 PM.
Reply
Old May 15, 2007 | 09:34 PM
  #13  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
Originally Posted by RotaMan99
The reading you see will also be higher then what is actually happening because of electrical interfierence.
This i do understand, but it would make way more sense if it showed knock ALL the time, as in, cold days (all through winter no knock at all, or any day lately where its colder than 65-ish) but it only shows it on warm days (i.e. today, 82F outside, knocked the entire time 8-14)), which is why I'm a little worried. It steadily climbs the warmer the engine gets (peaks around 17 sometimes). The only time it'll do it on a cold day is when i take her home (300 mile trip) after about an hour itll show slight knock (7-12).
I'd almost think I'd see LESS knock now, if this was caused by electrical interference (mainly because heat increases resistance, and on a warmer day, the wires between the sensor itself and the SAFCII would have increased resistance, therefor showing less voltage(BTW I know what the SAFC shows isnt voltage from the sensor, but thats how piezoelectric knock sensors work))
Reply
Old May 15, 2007 | 09:45 PM
  #14  
RotaMan99's Avatar
Rotary Freak
Tenured Member 10 Years
 
Joined: Dec 2006
Posts: 1,791
Likes: 0
From: New Hampshire
Well I havn't delt with knock before but could it also be caused by a little to much advanced timing?
Reply
Old May 15, 2007 | 10:18 PM
  #15  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
i've been thinking about that. I'm going to unplug my FCD and get back to ya
Reply
Old May 15, 2007 | 10:20 PM
  #16  
SirCygnus's Avatar
whats going on?
Tenured Member 20 Years
iTrader: (1)
 
Joined: Mar 2004
Posts: 4,930
Likes: 8
From: atlanta ga
i dont trust an safc. period. could be ANTYHING!!

try modifieing and redoing some grounds on the main harness and see if that helps. its probobly just noise.
Reply
Old May 16, 2007 | 08:15 AM
  #17  
RotaMan99's Avatar
Rotary Freak
Tenured Member 10 Years
 
Joined: Dec 2006
Posts: 1,791
Likes: 0
From: New Hampshire
I'm going to unplug my FCD and get back to ya
the FCD doesn't advance timing right?
Reply
Old May 16, 2007 | 08:34 AM
  #18  
jackhild59's Avatar
Rotary $ > AMG $
Tenured Member: 20 Years
iTrader: (7)
 
Joined: Apr 2005
Posts: 3,785
Likes: 30
From: And the horse he rode in on...
So, how much fuel (%) have you pulled out of the fuel maps? You know that as you fool the ecu, the timing is not retarded correctly, because the ecu 'thinks' the airflow and engine load is lower?

You may just be exposing and alalyzing the basic flaw in piggyback a/f controllers.

Can you reset your fuel adjustments to zero and still read the knock numbers? This might be with trying, just for a referrence.

Good luck
Reply
Old May 16, 2007 | 03:30 PM
  #19  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
well, S4 ecu's use the boost sensor for two things only: Fuel Cut, and timing retard at certain boost pressures. The FCD is a resistor to fool the ECU into seeing less boost, therefore no fuel cut. This will leave timing advanced more than intended from the factory, which could be causing my issues. But that still doesnt explain why it ONLY knocks at higher ambient temperatures. As stated before, on S4 Turbo II's the IAT is infact part of the AFM, but my dumb self forgot to get the serial number for the recall again, so ill have to check it tommorow.

As far as pulling fuel, im not pulling any fuel at all, im adding fuel. When Zero'd out it knocks more, with added fuel the knock goes away completely(0). Today is 60F. I started it, let it warm up, boosted all the way to 14psi and back, took it home, let it cool, and shut it off. No knock at all. More and more it seems to be directly related to ambient air temperature and nothing else. Im testing my IAT asap. I'll post what the results are
Reply
Old May 16, 2007 | 03:40 PM
  #20  
Richter12x2's Avatar
Taste great, more filling
Tenured Member 10 Years
iTrader: (1)
 
Joined: Feb 2005
Posts: 2,012
Likes: 0
From: Dallas, TX
Maybe I misunderstand the whole basis of tuning fuel, but what does your air/fuel ratio look like when you add enough fuel to get the knock to go away? Is it showing mega rich, or normal? Do you smell raw gas from behind the car?
Reply
Old May 16, 2007 | 03:48 PM
  #21  
RotaMan99's Avatar
Rotary Freak
Tenured Member 10 Years
 
Joined: Dec 2006
Posts: 1,791
Likes: 0
From: New Hampshire
well, S4 ecu's use the boost sensor for two things only: Fuel Cut, and timing retard at certain boost pressures. The FCD is a resistor to fool the ECU into seeing less boost, therefore no fuel cut. This will leave timing advanced more than intended from the factory, which could be causing my issues.
Thanks for clearing that up. So without the FCD, as boost creeps up, the ECU will retard the timing more?
Reply
Old May 16, 2007 | 05:13 PM
  #22  
RotaryEvolution's Avatar
Sharp Claws
Tenured Member 15 Years
iTrader: (30)
 
Joined: Oct 2003
Posts: 5,107
Likes: 50
From: Central Florida
the knock sensors equipped on these cars are pure crap, disregard any info it may be feeding you.


the main reason you are seeing more knock in higher ambient temperatures is because the sensor is located in a gel like substance, as it warms the gel becomes more soft and pliable and also picks up more vibrations, mostly of which are normal vibrations from accessories and engine operation. over years the gel liquifies and melts out of the sensor and the readings become very high.

as i said though, disregard what the knock sensor is telling you because it likely is nothing to worry about. they may have worked "okay" about 20 years ago but not anymore..
Reply
Old May 16, 2007 | 06:12 PM
  #23  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
Originally Posted by Karack
the knock sensors equipped on these cars are pure crap, disregard any info it may be feeding you.


the main reason you are seeing more knock in higher ambient temperatures is because the sensor is located in a gel like substance, as it warms the gel becomes more soft and pliable and also picks up more vibrations, mostly of which are normal vibrations from accessories and engine operation. over years the gel liquifies and melts out of the sensor and the readings become very high.

as i said though, disregard what the knock sensor is telling you because it likely is nothing to worry about. they may have worked "okay" about 20 years ago but not anymore..
that is something i completely over-looked. I guess replacement is the only fix for that?
Reply
Old May 16, 2007 | 06:13 PM
  #24  
BoostinFC's Avatar
Thread Starter
Piston Hater
Tenured Member 05 Years
 
Joined: Aug 2006
Posts: 41
Likes: 0
From: Rochester, NY
Originally Posted by RotaMan99
Thanks for clearing that up. So without the FCD, as boost creeps up, the ECU will retard the timing more?
Yes, without the FCD the ecu WILL retard timing more (to the best of my knowledge) even though I believe it is only a couple degrees.
Reply
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
trickster
2nd Generation Specific (1986-1992)
25
Jul 1, 2023 04:40 PM
Kyo
2nd Generation Specific (1986-1992)
6
Apr 13, 2019 09:24 AM
ncds_fc
New Member RX-7 Technical
1
Aug 15, 2015 10:06 AM




All times are GMT -5. The time now is 10:28 AM.