Marauder Knock Sensors on a Teksid

The teksid is 12x1.5 and the ones in our cars are 8x1.25, so does that mean I would be looking for a 12x1.5 to-> 8x1.25 helicoil?
 
F1erceK said:
The teksid is 12x1.5 and the ones in our cars are 8x1.25, so does that mean I would be looking for a 12x1.5 to-> 8x1.25 helicoil?
Click to expand...

It does not work that way. You may need threaded inserts which are larger than helicoil. It do what Zack says abd eliminate
 
Ya, just delete them. Especially if you're going FI. Your ears are far better knock sensors anyways.
 
Do your self a gigantic favor turn them off any buy yourself a J&S Vampire


Sent from my iPhone using Tapatalk
 
I got rid of my knock sensors and had them shut off in the tune. Same block and intake you are using.
Like was already said, the stock knock sensors don't do much, no reason to keep them.
 
I kept my knock sensors. With the blower and forged piston clearances, the sensitivity was turned down but they're still active in my tune.
 
FordNut said:
I kept my knock sensors. With the blower and forged piston clearances, the sensitivity was turned down but they're still active in my tune.
Click to expand...

Were you able to determine the piston rattle false triggered them, or was this just a general precaution? Knock sensor signal processing has improved over the years and is more resistant to false triggers, but it appears the consensus is the Marauder PCM is not very smart.
 
Dynamically adjusted. On the dyno. Reduced sensitivity until it quit pulling timing for no apparent reason. Then added spark until it started pulling timing for a known reason.

Maybe the reason for false triggering was piston clearance, maybe blower noise?
 
Yes, the Vampire works great. Here's a vid I made in 2008, on bgstew6's Lightning. At the time, he was a moderator on lightningrodder:

https://www.youtube.com/watch?v=CeuD2-GIlb0

This was his daily driver, with pulleys for 19psi, though he held it to 8psi during this test.

The unit had been out a year already, but this was the beta install for a Lightning.

The Lightnings were equipped with a centrally located knock sensor, but Ford did not enable it in the tune. We were running out of daylight when it came time to tap into the factory knock sensor.

Rather than trace the wire in the harness, we decided to add a separate knock sensor. Problem is, the only spot that looked easy was a stud on the very front corner of the driver's side cylinder head.

His plan was to test on the street, on 89 octane, but tuned for 91. He knew it would audibly rattle at 0 psi, without hurting the engine, so we decided to install the sensor on the end of the head. We figured if it didn't work, we could spend the time to connect to the factory sensor.

Well, it did work. Fifteen minutes later he was through testing on the street, and headed for the 5 freeway.

I bought along a headphone amplifier with two outputs. Knock sounded like birds chirping. The gauge would light up at the same time as you heard it in the headphones.

This run was near Camp Pendleton, where he worked at the time. We were running with traffic about 65-70 mph, then accelerating to about 85 mph to pass, at 8psi.

Two weeks later he connected to the factory sensor, and performance was identical.

Photo of the sensor on the end of the cylinder head:
https://picasaweb.google.com/100126742631003862483/JSVampireInstallation#5200857299633473250
 
HUH???

But Zack said:

Quote:

"You don't need the knock sensors. They don't do squat"

Quote:

"Knock sensors don't do a god dang thing"

(he said it twice)

So, how can it be that " ... the Vampire works great"?


==
 
The Vampire is a stand alone unit. It senses knock and retards timing on individual cylinders that are knocking and keeps pulling timing until the knock goes away. It's good for safely tuning to the limit, then backing the timing off to where there is no knock. This will save a motor if you get bad gas.

I think the stock knock sensors are what don't do much.
 
It is all in the signal processing and implementation of knock retard algorithms. Knock exhibits a hysteresis effect, which needs to be accounted for also.

Knock sensors can and do work. Unfortunately, this does not mean the implementation in a 2003-4 Ford PCM is anywhere near correct.
 
GreekGod:

Do you understand what was happening in the video I posted?

The truck was tuned for 91 octane and 19 psi. It had been running that tune for three years before I showed up, so it was proven to be safe, reliable, and had the power he wanted. Stock block, and the heads had never been off the engine.

I show up on the forum he moderated (lightningrodder), and offer him a unit, in return for being the beta install on a Lightning. This is his daily driver, so to lessen the chance of engine damage, he says we will test on the street at 0 psi with 89 octane. He knows it will knock but not bad enough to hurt the engine.

We do the install and take it out for a spin around town. Pulling away from a light, it would knock very audibly, for about two seconds, until it had enough knock retard built up in it to stop the knock. We played with the sensitivity for a bit, then gave up on it and went back to his place.

I told him the unit is not designed to work under conditions like that. Remember, we were lugging the engine at 1500 RPM on 89 octane, while tuned for 91.

His wife served cake and coffee, then my wife and I said our goodbyes to them, and we (dejectedly) head home.

Back in my lab I thought about what happened, and decided rather than give up on Lightnings running 89 octane, I would try to MAKE it work under his conditions.

The DETECTION part of the system worked perfectly. I had a portable scope on my lap and knock headphones. We could see the unit respond to the knock pulse immediately, but it didn't dial in the retard quickly enough.

The existing scheme at that time was it would retard one, two, or three counts per knock event, depending on knock intensity. Each count was worth one degree or two degrees, depending on one of the mode switches.

We had the switch set for two degrees per count, so that means six degrees retard on the first knock event was not enough to kill the knock the next time that cylinder fired.

In addition, the original code had a "feature" that lessened the retard increment, the farther it retarded. Kind of like a bicycle pump. The tire fills slower and slower as you pump it up. This was to prevent knock retard from going crazy if there was engine noise, which did not turn out to be a problem.

I reworked the CONTROL part of the software to allow UP TO seven counts, or fourteen degrees retard, with one knock event. In proportion to knock intensity. And I removed the accumulator code, described above.

We returned the next week with the software revision, plugged the unit in and went out again, on the street with 89 octane. Pulling away from a light, lots of knock retard, zero audible knock. Only the knocking cylinders get retarded, so you can't even feel it. It's just enjoyable to mash the throttle and watch the thing take over.

After about twenty minutes playing around on the street, Stew says screw this, and we head for the onramp. We drive North, from Oceanside to San Onofre, about fifteen miles round trip, goosing the the throttle when we can, but the unit kicks in and prevents audible knock. Stew is a happy camper.

He continued testing for a month, running on a mix of 89 and 91, and performed weekly plug checks, looking for signs of detonation on the porcelain.

We return for a final test at the end of the month. Everything was still working great, so we gave him title to the unit.

Very long thread about it here:
http://www.lightningrodder.com/foru...individual-cylinder-knock-control-system.html

More information in this old group buy thread:
http://www.lightningrodder.com/foru...tronics-engine-anti-blowup-safety-device.html
 
You must log in or register to reply here.
Back
Top