Knock sensor/ octane/ timing--how does the SC engine work?

1FSTBRD

Registered User
I was searching around on the forum for a definitive answer, but couldn't find one via the search function (though I did find this thread http://www.sccoa.com/forums/showthread.php?74279-knock-sensor-question&highlight=knock+sensor , which was an interesting read on how the knock sensor works). Anyways, obviously the knock sensor pulls timing with lower octane gas or diluted gas, but how much timing is increased if you're running 94 octane? Is there a point in the octane level where it has a max timing increase, or does it keep going with octane levels and intelligently increases it?

Obviously i'm worried about potential detonation, but was also wondering if there's a legitimate reason to run 94 instead of 91, in terms of actual performance gains.
 
The knock sensor doesn't know anything about octane - just knock. A rod knock can set it off, same as if you slapped the side of the block with a hammer. When it sends a signal to the EEC, it decides what to do based on it and/or other inputs at the time. Ignoring those several other inputs, the knock sensor is a fail safe that tries to mitigate disaster in the form of damage to one or more cylinders.

Or are you simply asking under what conditions can knock occur?
 
Last edited:
Ah, okay! I know the conditions in which spark knock occurs, I was just wondering if the knock sensor actually had automatically adjusted/ advanced the timing as it saw better gas, but it makes sense for it just to pull timing if the engine is going to be in trouble.
 
Ah, okay! I know the conditions in which spark knock occurs, I was just wondering if the knock sensor actually had automatically adjusted/ advanced the timing as it saw better gas, but it makes sense for it just to pull timing if the engine is going to be in trouble.

The knock sensor senses knock. It doesn't do anything with timing. The EEC-IV or Engine computer takes the input from the Knock sensor and makes adjustment to timing based on what it sees. the knock sensor is one of the many things that the EEC-iv takes into account for engine timing.

The input of the knock sensor will only result in reduced engine timing. Lack of knock simply means the timing commanded is not reduced, it does not increase timing. A certain amount of knock noticed at certain rpm ranges for a certain length of time will result in timing being reduced for a certain amount of time.

the most common thing to cause engine timing to be reduced on the SC is excessive intake air temperature. The engine computer monitors that and will reduce ignition timing if it is excessive. (starting at about -2 degrees to a max of around -8 degrees)
 
The knock sensor senses knock. It doesn't do anything with timing. The EEC-IV or Engine computer takes the input from the Knock sensor and makes adjustment to timing based on what it sees. the knock sensor is one of the many things that the EEC-iv takes into account for engine timing.

The input of the knock sensor will only result in reduced engine timing. Lack of knock simply means the timing commanded is not reduced, it does not increase timing. A certain amount of knock noticed at certain rpm ranges for a certain length of time will result in timing being reduced for a certain amount of time.

the most common thing to cause engine timing to be reduced on the SC is excessive intake air temperature. The engine computer monitors that and will reduce ignition timing if it is excessive. (starting at about -2 degrees to a max of around -8 degrees)

That's good to know that the entire computer will monitor various factors to reduce timing, especially IAT's.
 
As far as reducing heat soak, in another thread, someone had suggested a double IC and/ or an IC fan, but is there any gaskets that can isolate heat exchange away from each component, like the supercharger away from the manifold and the throttle body/ plenum/ plenum/ sc? I'm also of the frame of mind that having the air filter cone right out in the engine bay is going to do more harm than good in terms of sucking in hot air.

Also, how hot does the IC tube get in the summer for you guys? I'd got mine a week ago and haven't ran the car in summer time temps, but I'm thinking that a big, thick metal IC tube might act heat up a fair bit (like the Turbo T-Bird's IC being right next to the turbo creates a lot of heat). I've also ran ceramic headers on another car of mine, and it had reduced heat soak temps down--all those types of things added up to the car's engine bay never exceeding ambient temps on hot summer days. Is there anyone that makes ceramic shortys for the SC's? I was looking for some but didn't see any online.
 
The stock IC is behind a full width A/C condenser. Two things that will help are to convert to a narrow condenser (search here on details) so the IC gets a clean air flow and to install an IC fan [ http://www.siliconeintakes.com/prod...=1024&osCsid=a9aca53d320f08726918ca745125332b ].

Then you can add air intake temp monitoring to help judge how much more work may be needed to control temps on your car and which ones will have the best ROI, etc. Just remember that once you start making mods, it can be a balancing act to determine how to proceed...what works on someone else's car may work out differently on yours, etc. so be flexible as you proceed.

As for isolating the supercharge from the manifold - it already is. Three bolts clamp it down on three standoffs. There is an open valley between the supercharger body and the upper manifold surface.

About the CAI - there are kits that place the filter down below the stock opening, inside the fender and there are kits that will create a box for a cone filter to sit above the fender, like this:
SC_MAF_CAI.jpg
 
Last edited:
The knock sensor is a piezoelectric device and only fires at its design resonance range.

http://www.oldfuelinjection.com/?p=34

94/95 SC use the 4.6 sensor.

The sensor is providing a signal all the time though it is optimized for a specific frequency, so it isn't so much of a "fire". The EEC-IV samples the signal at specific load and RPM ranges looking for specific behaviors that are indicators of knock and if it sees these indicators for a number of samples, it will adjust timing.

Intercooler tubes get very hot. Though not as hot as a aluminum radiator will get.
 
Building a stockish gauge cluster. Looking at the standoff for Boost/ Fuel thinking room for a Knock sensor light that would light up top portion of the fuel gauge. Just need to sort volts coming off the feed to get a proper sized bulb.
Creighton
 
Building a stockish gauge cluster. Looking at the standoff for Boost/ Fuel thinking room for a Knock sensor light that would light up top portion of the fuel gauge. Just need to sort volts coming off the feed to get a proper sized bulb.
Creighton

You'll need to build a somewhat sophisticated circuit if you want to illuminate a warning light. Think of the knock sensor as more of a limited-band microphone than a resistive element. The signal itself cannot drive a light that would be useful for visuals.

I'd bet if you do some internet searching you can find some ideas or maybe even a schematic where someone has built one before, if you really want one.

On datalogs - such as generated by the Quarterhorse tuning "chip"- you can watch commanded ignition timing and deduce (based on other inputs being above/below their "pull timing" threshold) when the computer thinks the knock sensor is picking up knock. It would cost more to get setup with the tuning hardware / software, but you'd get so much more out of too!
 
Back
Top