Pelican Parts Forums

Pelican Parts Forums (http://forums.pelicanparts.com/index.php)
-   911 / 930 Turbo & Super Charging Forum (http://forums.pelicanparts.com/forumdisplay.php?f=222)
-   -   knock sensor (http://forums.pelicanparts.com/showthread.php?t=405586)

John at J&S 06-19-2008 05:31 PM

raceman:

Thanks for your interest.

Not to be confrontational, but it's already been tested and there is nothing to dial in.

Who has to say it works that you would believe?

Is your guru good enough for the next guy, or is there a head guru?


Todd from Protomotive used to use it to develop his chips.

http://retro.co.za/efi/DIY_EFI/1996/efi1-259.txt

JohnJL 06-19-2008 06:28 PM

I have an acquaintance here who has installed, tuned and runs a knock sensor on his supercharged 911 race car. I have seen his dyno logs and the overlayed knock sensor readings. It definately worked for him. That car is actively campaigned and races regularly with no engine problems and more than acceptable power. Check www.supercharged911.com

His knock sensor is tapped into the case underneath the crankcase breather cover. I count at least 3 mounting surfaces between the combustion chamber and the sensor yet, sure enough it works. He is using a standard bosch sensor and a GM ECU he's hacked and remapped. I've seen where knockcounts creeped up under boost on the dyno and when timing was marginally pulled back the counts dropped off. He has reams of dyno logs and track lap logs with all the key variables correlated.

I don't know yet if my approach will work but I have tapped the heads to accept 993-style knock sensor bridges. My ECU (Adaptronic) has some highly configurable filters/amplifiers which I hope to be able to isolate a true knock count and minimize false-positives.

John at J&S 01-15-2009 06:37 AM

Les: No hard data for you, but another anecdote to consider, this one from Ernie J. at Mantis Racing.

I thought with his racing and tuning experience he might be able to tell the difference whether or not the unit on his 911 properly detects and controls knock.


I gave him the spiel on the new version and he replied that he has been using one for years on his 911 (invoice says '99), reminding me "you changed the driver current so I could drive two coils. The engine is twin plugged. it makes great power, fuel mileage."

Quotes were posted with his permission. I asked if the 911 with the J&S was on his site, but he didn't answer that one.

Pretty good video of him battling it out in a 944 is on his site:
http://www.mantissport.ca/

nize 07-02-2009 02:21 PM

i guess i'm a latecomer to this thread, and i have a 951 so you could argue it's not really comparable to the 930 motor, but thought i'd chime in because i've been running the j&s for over two years now with great success and i thought i'd share my story.

i first installed the j&s with the wolf stand alone EMS system on the stock motor, details in this thread;
http://forums.pelicanparts.com/showthread.php?t=336610

http://gallery.silentmedia.com/main....serialNumber=2

since then, i've gone to a completely built darton big bore 2.7 with custom pistons;
http://forums.pelicanparts.com/showthread.php?t=405959

http://gallery.silentmedia.com/d/121...IMG_4004-2.JPG

and then to a custom big-valve head;
http://forums.pelicanparts.com/showthread.php?t=482147

http://gallery.silentmedia.com/main....serialNumber=2

the car will soon make over 420whp/wtq on regular 92octane everyday pump gas, that's over 100whp per cylinder. i have plans to make over 500whp using e85 fuel.

recently i've helped develop the next-generation wolf v500 system with j&s knock control;
http://forums.pelicanparts.com/showthread.php?t=479564

http://gallery.silentmedia.com/main....serialNumber=2

the j&s has a gauge with 10-LED knock indicator lights, that show you when knock is detected and how much timing is being pulled. this makes it very easy to tune the car, and a side benefit is that it also makes it easy to tell when you've got a tank of bad gas, which happened to me on more than two occasions and which would have otherwise caused engine damage.

i can tell you that without the j&s installed, these performance tunes would not be as easily obtained since it would not be possible to safely run the engine on the ragged edge of knock, extracting maximum performance. i'm sure you are all aware of how critical knock control is on a turbo motor.

les_garten;
you claim that the first time you pulled your motor apart, 'every compression ring was broken into uncountable pieces, all twelve.'

considering how much time and money you must have invested in your motor, why not at least try the j&s for yourself and see first-hand if it makes a difference? i'm thinking for less than $500 it's cheap insurance. i think it's rather amusing that 930 guys seem to be slow to adapt to new technology, y'all sure are a bunch of skeptics. :) whatever happened to the 'try it and see for yourself' method?

personally, i think the j&s is one of the best investments i've made and i think we're all very lucky to have someone like john stick around and take punches and criticism and feedback and continuously and actively improve the product. i only wish all other manufacturers stood behind their product as well as john at j&s does.

nize 07-03-2009 11:46 AM

here's a photo of the 10-LED knock gauge for the j&s with automatic dim based on ambient light;

http://gallery.silentmedia.com/main....serialNumber=2

JohnJL 07-03-2009 02:46 PM

I got mine working as described in my post above. Bridge across the 456 cylinders with a bosch sensor. It it monitored by my Adaptronic ECU. Additionally I use one of ther ECU outputs to trigger a superbright flashing ECU on the dash when knock is sensed. Works awesome.

kenikh 07-06-2009 04:02 PM

Quote:

Originally Posted by JohnJL (Post 4013309)
I have an acquaintance here who has installed, tuned and runs a knock sensor on his supercharged 911 race car. I have seen his dyno logs and the overlayed knock sensor readings. It definately worked for him. That car is actively campaigned and races regularly with no engine problems and more than acceptable power. Check www.supercharged911.com

His knock sensor is tapped into the case underneath the crankcase breather cover. I count at least 3 mounting surfaces between the combustion chamber and the sensor yet, sure enough it works. He is using a standard bosch sensor and a GM ECU he's hacked and remapped. I've seen where knockcounts creeped up under boost on the dyno and when timing was marginally pulled back the counts dropped off. He has reams of dyno logs and track lap logs with all the key variables correlated.

I don't know yet if my approach will work but I have tapped the heads to accept 993-style knock sensor bridges. My ECU (Adaptronic) has some highly configurable filters/amplifiers which I hope to be able to isolate a true knock count and minimize false-positives.

This post is worthless without pics! ;) Sow us how you did it, because I'd sure love to.

JohnJL 07-06-2009 05:07 PM

Ask and ye shall receive...


Here's a pic of the sensor and bridge.

http://forums.pelicanparts.com/uploa...1246928784.jpg


Here's a link to the build...http://forums.pelicanparts.com/showthread.php?t=444521

Ray installed his by drilling and tapping a thread into the case in the baffle underneath the breather cover.

kenikh 07-06-2009 05:10 PM

Thanks - good food for thought!

nize 07-06-2009 05:10 PM

kenikh; i see you're in seattle. have you talked to bob yet at random ems? i know he has several 911's running with the wolf system using j&s safeguard successfully. his contact info is here;
http://randomems.com/

kenikh 07-06-2009 05:13 PM

Quote:

Originally Posted by nize (Post 4762823)
kenikh; i see you're in seattle. have you talked to bob yet at random ems? i know he has several 911's running with the wolf system using j&s safeguard successfully. his contact info is here;
http://randomems.com/

Not yet, but sounds like I will be. :) I was planning on going MegaSquirt. I just looked it up - pretty fancy stuff. Not cheap, but may be worth a look!

John at J&S 07-06-2009 06:17 PM

That's a GM knock sensor.

mark houghton 07-06-2009 06:20 PM

Quote:

Originally Posted by krasuskyp (Post 3926905)
Kinda related but figured some yous might get a goof out of this one...

A bud of mine does alot of chip tuning in the Audi circles, been working closely with VEMS lately, etc.

He just rig'd up this neat little concoction as he too is skeptical of knock sensors and 'normal' engine 'noise'. He wired up an adapter harness to a factory audi knock sensor to a standard microphone jack. Plugged that into a laptop PC with a set of race style headphones plugged into the headphone jack.

Which allowed him to produce this on a stroker '95 //S6 pushing around 400allwhp:

http://www.theswansonfamily.us/modules/gallery2/d/57834-1/brian_s6_onramp.mp3

Pretty neat. He's a madman, gotta love it.

Just watched the video....felt like I was on some sort of illicit drug. Must have been a flashback.
Knock knock, who's there....I didn't hear any.

nize 07-06-2009 11:50 PM

Quote:

Originally Posted by mark houghton (Post 4762960)
Knock knock, who's there....I didn't hear any.

i always get a kick out of people who claim they can 'hear' knock while driving and will let off on the gas to protect the motor. by the time a human can actually 'hear' real knock, it's too late and the motor is probably trashed. the early onset of knock is not audible to the human ear while driving.

that's like saying someone can manually trigger each spark plug by pushing a button with every other crank revolution at 6000rpm. crazy talk.

John at J&S 07-07-2009 12:57 AM

Here is a video of one installed on an F150 Lightning. Forget that it's not your favorite engine, just watch what individual cylinder knock control can do to the timing. In the last frame of the clip, cylinder #5 is retarded 8°. No others are retarded. No tuner can do that.

A pro racer told me that he installs a plug three heat ranges colder in #5. This tends to confim that it's not engine noise that causes the activity seen.

http://video.google.com/videoplay?docid=-3115045722815533155

If for some reason the link doesn't work, google "knock controller demo".

kenikh 08-07-2009 11:35 PM

Quote:

Originally Posted by John at J&S (Post 4763328)
Here is a video of one installed on an F150 Lightning. Forget that it's not your favorite engine, just watch what individual cylinder knock control can do to the timing. In the last frame of the clip, cylinder #5 is retarded 8°. No others are retarded. No tuner can do that.

A pro racer told me that he installs a plug three heat ranges colder in #5. This tends to confim that it's not engine noise that causes the activity seen.

http://video.google.com/videoplay?docid=-3115045722815533155

If for some reason the link doesn't work, google "knock controller demo".

How does it detect each cylinder? Does it use one sensor or one per cylinder?

kenikh 08-07-2009 11:38 PM

Quote:

Originally Posted by JohnJL (Post 4762816)
Ask and ye shall receive...


Here's a pic of the sensor and bridge.

http://forums.pelicanparts.com/uploa...1246928784.jpg

I just bought a pair of bridges. Much cheaper than expected and available on Pelican! $45 shipped for two!!!

John at J&S 08-08-2009 03:03 AM

One sensor per cylinder is no help. If one cylinder knocks, they all hear it.

It can all be done with one sensor, and no cam or crank sensors are needed.

The knocking cylinder is the one that just fired. Software keeps track of it, and retards THAT cylinder next time it comes up on its compression stroke.

Coil on plug is not a requirement. It can all be done through the distributor.

Couple of posts from a 928 guy with one:
http://forums.rennlist.com/rennforums/5446950-post1755.html

http://forums.rennlist.com/rennforums/5934265-post82.html

hobieboy 08-08-2009 03:52 AM

I understand that knock sensing system needs to be tuned to recognize knocks?

So a question for those who installed a system - is my understanding correct? If so, how did you tune it the 1st time? Make the engine knock?

kenikh 08-08-2009 06:55 AM

Quote:

Originally Posted by John at J&S (Post 4824018)
One sensor per cylinder is no help. If one cylinder knocks, they all hear it.

It can all be done with one sensor, and no cam or crank sensors are needed.

The knocking cylinder is the one that just fired. Software keeps track of it, and retards THAT cylinder next time it comes up on its compression stroke.

Coil on plug is not a requirement. It can all be done through the distributor.

Couple of posts from a 928 guy with one:
http://forums.rennlist.com/rennforums/5446950-post1755.html

http://forums.rennlist.com/rennforums/5934265-post82.html

That was both shockingly intuitive and in retrospect, obvious. Thanks!


All times are GMT -8. The time now is 03:13 AM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2025, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2025 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website


DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.