Welcome to the North American Subaru Impreza Owners Club Sunday December 21, 2014
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Engine Management & Tuning

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 01-29-2004, 12:43 PM   #1
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default tapping into existing knock sensor--diy knocklink/detcans

as an alternative to a knocklink (which displays leds corresponding to various levels of noise from the engine, using it's own knock sensor), i have been tossing around tapping into the OEM sensor.

obviously you can't load down the circuit... the tapping must be as transparent as possible. however, this is not as hard as it sounds with modern OPAMPS, which have extremely high input impedances, to be used as buffers.

once the signal is tapped, it wouldn't be very hard to design a circuit to sequentially light up LEDs corresponding to RMS voltage input. hell, there are commercially available "power meter" kits designed to be hooked up into audio systems to monitor that very thing.

it would also be trivial to add a headphone output at that point.

has anyone ever attempted this? even considered it?

i always thought the knocklink's (and tuna's, for that matter) use of a completely separate knock sensor a bit silly. the only advantage i can see is redundancy.

thoughts?

ken
* Registered users of the site do not see these ads.
ride5000 is offline   Reply With Quote
Old 01-29-2004, 05:07 PM   #2
jblaine
Scooby Guru
 
Member#: 8512
Join Date: Jul 2001
Chapter/Region: South East
Location: St. Pete, FL
Vehicle:
2002 WRX chassis...
stage-infinity.com

Default

You know... well-engineered det cans for an EJ20 sure would be a good product for someone to make, market and sell... Hell, all I need is something that ends in a 1/8" mini stereo plug to hook to my aux-in and crank up my stereo.

<aside>It's a bummer the TXS Tuna doesn't speak the DeltaDash serial protocol. That seems like a glaring omission to me.</aside>
jblaine is offline   Reply With Quote
Old 01-29-2004, 06:53 PM   #3
Jon [in CT]
*** Banned ***
 
Member#: 2992
Join Date: Nov 2000
Chapter/Region: NESIC
Location: Connecticut, USA
Vehicle:
02 WRX Sedan
Silver

Default

Quote:
Originally posted by ride5000
obviously you can't load down the circuit... the tapping must be as transparent as possible. however, this is not as hard as it sounds with modern OPAMPS, which have extremely high input impedances, to be used as buffers.

once the signal is tapped, it wouldn't be very hard to design a circuit to sequentially light up LEDs corresponding to RMS voltage input. hell, there are commercially available "power meter" kits designed to be hooked up into audio systems to monitor that very thing.
Opamps are a good idea. Turning on LEDs is essentially what a Knocklink does and is pretty worthless for knock detection, although I have no doubt you could sell a boatload of these.

Near the end of thread http://forums.nasioc.com/forums/show...hreadid=492653 I offered some ideas and associated links for doing what you propose.

Last edited by Jon [in CT]; 01-29-2004 at 07:37 PM.
Jon [in CT] is offline   Reply With Quote
Old 01-29-2004, 07:43 PM   #4
Equilibrium Tuning
Former Vendor
 
Member#: 26933
Join Date: Oct 2002
Chapter/Region: BAIC
Location: Fairfield, CA
Vehicle:
2006 STI
CGM

Default

I've actually been planning on making a set of det-cans that tap into the stock knock sensor. All it takes is a small circuit and a tape recorder with a mic input

-- Ed
Equilibrium Tuning is offline   Reply With Quote
Old 01-29-2004, 09:46 PM   #5
dunk
Scooby Guru
 
Member#: 15359
Join Date: Feb 2002
Chapter/Region: MAIC
Location: Fairfax, VA
Default

I've actually made the unit Jon posted the link for. They work very well.

Duncan
dunk is offline   Reply With Quote
Old 01-30-2004, 08:02 AM   #6
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

dunc,

was there any evidence that the circuit posted did or did not affect the proper operation of the knock sensor in any way? i.e. did you ever get some kind of sign that the OEM knock detection was still working, and have the ECU pull timing advance? i'm very wary of loading down the knock sensor and not having it work properly, for obvious reasons.

i'd love to get a sample of our engine knocking. there may be stuff in the analog realm that we can do to filter out some of the noise, and leave the ping.

looks like another project on the books!

ken
ride5000 is offline   Reply With Quote
Old 01-30-2004, 11:17 AM   #7
john banks
Scooby Specialist
 
Member#: 11669
Join Date: Oct 2001
Location: UK
Vehicle:
Evo IX 345 WHP

Default

"Turning on LEDs is essentially what a Knocklink does and is pretty worthless for knock detection"

Have to disagree Jon. I can see every knock event of any significance with a knocklink and successfully doubled the output of my original EJ20 using a knocklink without any other form of detonation detection and it held together and was a healthy engine when it was removed to make way for a stroker. Correlate the activity with det cans and most agree it is an excellent in-cabin tool. It is a useful backup to see how my J&S is working - run too much base map advance and the knock is too sudden for the J&S to retard quickly enough. Knocklink tells you this. If it wasn't valuable I would have removed it when the J&S went on, but it tells me useful info so I keep it.
john banks is offline   Reply With Quote
Old 01-30-2004, 02:36 PM   #8
Equilibrium Tuning
Former Vendor
 
Member#: 26933
Join Date: Oct 2002
Chapter/Region: BAIC
Location: Fairfield, CA
Vehicle:
2006 STI
CGM

Default

The knock-link should have come with a headphone jack. That would have been a very nice feature.

-- Ed
Equilibrium Tuning is offline   Reply With Quote
Old 01-30-2004, 04:20 PM   #9
John at J&S
Scooby Newbie
 
Member#: 2036
Join Date: Aug 2000
Location: Garden Grove, CA USA
Vehicle:
2000 Intrepid
silver

Default

Mr. Banks:

In the 20 knock retard range, the J&S can retard as much as 6 on the first ping. Have you tried this mode?
John at J&S is offline   Reply With Quote
Old 01-30-2004, 04:30 PM   #10
john banks
Scooby Specialist
 
Member#: 11669
Join Date: Oct 2001
Location: UK
Vehicle:
Evo IX 345 WHP

Default

Yes John, but it was keen to excessively phantom retard at 3000 RPM throttle snaps even when it wasn't detting.

I could try 20 degrees with less sensitivity I suppose?
john banks is offline   Reply With Quote
Old 01-30-2004, 06:59 PM   #11
John at J&S
Scooby Newbie
 
Member#: 2036
Join Date: Aug 2000
Location: Garden Grove, CA USA
Vehicle:
2000 Intrepid
silver

Default

Sounds like a plan.

IIRC, you didn't hit on the ideal setting the first time out. Maybe it was during that time that you tried the 20 mode?

So the timing might be fairly advanced, and the A/F at stoich, when you stab the throttle? Maybe transitional knock?
John at J&S is offline   Reply With Quote
Old 01-30-2004, 07:04 PM   #12
john banks
Scooby Specialist
 
Member#: 11669
Join Date: Oct 2001
Location: UK
Vehicle:
Evo IX 345 WHP

Default

It don't think it is transitional knock, as I can hear "everything" through the det cans and it isn't even sparkly in this area, especially if it still does it at anything from 8 to 22 PSI with excellent fuel enrichment and timing adjusted across a wide range making no difference. I think it is hearing the fuel enrichment noise of the injectors. Sticking to a limit of 4 degrees excessive timing in the map seems to work well and allows it to be caught quick enough by the J&S. I will have another play with 20 degrees though.
john banks is offline   Reply With Quote
Old 01-30-2004, 08:51 PM   #13
turboICE
Scooby Specialist
 
Member#: 33979
Join Date: Mar 2003
Chapter/Region: Tri-State
Location: N.NJ
Vehicle:
2006 STi,'11 OBS
WRB & Black/Silver

Default

Good link that Jon sent me once when I asked his thoughts on it:

http://home.netcom.com/~bsundahl/kno...nockSounds.htm

Has a DIY link on it that works well.
turboICE is offline   Reply With Quote
Old 02-01-2004, 04:10 PM   #14
Turbo Tin Can
Scooby Newbie
 
Member#: 18383
Join Date: May 2002
Default

I also disagree with Jon,my knocklink has worked fine for over two years. I have made an adapter for my knocklink where my headphones and amp wire in. Works great, now if i had a better ear that would help. I just got my jdm WRX motor put in so i havent had it hooked up yet.
Turbo Tin Can is offline   Reply With Quote
Old 02-01-2004, 04:19 PM   #15
john banks
Scooby Specialist
 
Member#: 11669
Join Date: Oct 2001
Location: UK
Vehicle:
Evo IX 345 WHP

Default

Had another play with the 20 degree mode on the J&S and various sensitivities, it is far better on 10 degree mode on about 30-40% sensitivity. Too much sensitivity and it does pick up what is definitely phantom knock at 3000 RPM throttle snaps. Also at the top end you can feel as different cylinders have quite different (and excessive) retard that the smoothness of the engine has gone and vibration increases. 10 degrees is much more subtle, keeps things smooth and powerful, and works fine if I haven't gone mad with my base map. More than about 4 degrees too much and too much (IMHO) det gets through before it retards.
john banks is offline   Reply With Quote
Old 02-02-2004, 06:15 PM   #16
AZScoobie
Scooby Guru
 
Member#: 8785
Join Date: Jul 2001
Chapter/Region: SWIC
Vehicle:
02 c_turner@ix.
netcom.com

Default

My knock link is the single most important and usefull tool I have. It has saved my motor probably 100 times over. I have it mounted on the center flip down visor above the rear view mirror. Its always in my vision and it is very, very sensitive. I use a seperate bosch knock sensor. WOT runs at high power lights the second light solid. I have never had false indication of knock other then my AC compressor clutch clicking on. Any knock instantly shows up on the LED readout. Its very fast and after a few weeks of watching the lights you learn what the degree of the knock you just experienced was. Its a Must have for a Turbo charged Subaru that is being pushed to the edge. I would not even think of tuning up without it.

John: What kind of retard are you getting on the different cyls? Which cyl knocks the most ect..

Clark
AZScoobie is offline   Reply With Quote
Old 02-02-2004, 06:27 PM   #17
john banks
Scooby Specialist
 
Member#: 11669
Join Date: Oct 2001
Location: UK
Vehicle:
Evo IX 345 WHP

Default

Clark the usual spread of knock is 1 or 2 degrees, rarely is it just one cylinder knocking, usually two or more. If I go mad with the timing the J&S doesn't seem able to take out enough and quickly enough to stop the knocklink climbing the tree to at least the first orange and sometimes the red and then when it does sometimes the engine feels unbalanced and kills too much power as you might have 6 degrees out in one cylinder and 2 in another. Other times the J&S is incredibly effective at keeping things nice - at 7000 RPM only the bottom green light on the knocklink is flickering with maximum sensitivity. You can't tell which cylinder is which.

I back off when I see and feel this wide spread between cylinders as it doesn't feel at all nice. I hoped it would just get on with it and I could put my knocklink away but this is not the case. It is useful for up to about 4 degrees on top of the worst case timing I would usually map to. The knock behaviour can be quite random and erratic through the det cans, knocklink and J&S. Interesting that you can tune your timing just to run the quarter mile, only when you get into 5th gear out of 6 or 4th gear out of 5 do you really load the car up enough.... a nice compensation is to tune MORE boost in the lower gears because the engine can take it, as long as the gearbox and tyres can

I think to be truly effective the knock control must have learning behind it, but with a rapid learning cycle. The as it happens method just can't keep up - once the knock starts it needs a lot more retard to get rid of it than you could have retarded in the first place to make sure it never happened at all. I'm quite fussy about knock though - I don't tolerate any spike activity on the knocklink whatsoever as it is always associated with a thunk through the det cans. My first green LED illuminates sporadically from 5000 RPM, more solid from 5500-6000 RPM. If I see an illumination of first green at 3000-4000 RPM that as far as I am concerned is mild knock and tunes out. Paranoid yes, but I've not lost an engine because of knock so far.
john banks is offline   Reply With Quote
Old 02-02-2004, 06:34 PM   #18
AZScoobie
Scooby Guru
 
Member#: 8785
Join Date: Jul 2001
Chapter/Region: SWIC
Vehicle:
02 c_turner@ix.
netcom.com

Default

I kind sense that you like the J&S but its not a blanket solution that allowed you to relax and just enjoy the car. I think the bottom line is that we would both have to detune our cars WAY down to get to that point. Being on 91 octane fuel Its pretty rough to map my car. I map it out, Drive it for a week, Notice its not right and map it again. Its not childsplay to run 1.5 bar of boost on 91 octane fuel. Its fun, Its a challenge but its not something that everyone should try.

I have never lost a motor either and that is because I know when to say when. If I am knocking I let out and fix the problem or at least try too...

Cya

C
AZScoobie is offline   Reply With Quote
Old 02-02-2004, 06:42 PM   #19
john banks
Scooby Specialist
 
Member#: 11669
Join Date: Oct 2001
Location: UK
Vehicle:
Evo IX 345 WHP

Default

Spot on, totally agree.

Methanol is on its way after a lot of playing with different water injection jets and mixtures today. On Delta Dash at about 330 WHP at 19 PSI on 98 RON with the TD05/06-20G(7cm2). With ethanol injection it went up to 340 WHP with a lot of fuel removed and 2 degrees advance and 20 PSI. With a unknown composition methanol based screenwash used neat (freezing point -20C so probably only 25% methanol) it would also run the same but without removing the fuel or adding the timing (wouldn't take it without detting). Oddly with NF Race formula octane booster it will take 21 PSI (EJ257 remember) and makes 350 WHP, hope to get there with the methanol which I am inclined to use as 100% rather than 50/50 with water for my needs. Then onto a GT30R/P20 hybrid from iON next month for 400 WHP I am absolutely flogging this 20G to death, especially with its small housing and wheel on the hot side.

When I want an easy day I just turn it down to 18 PSI and run pump fuel and it is still enough to waste M3/M5, 911s and RS4/6
john banks is offline   Reply With Quote
Old 02-02-2004, 09:43 PM   #20
ride5000
Scooby Guru
 
Member#: 32792
Join Date: Feb 2003
Chapter/Region: NESIC
Location: lincoln, ri
Vehicle:
2003 GGA MBP
12.9 / 105+

Default

thanks for the comments, guys.

i will be sourcing parts and cobbling something together soon. if it works well i will offer up the design for others to do it themselves.

ken
ride5000 is offline   Reply With Quote
Old 02-03-2004, 01:11 PM   #21
AZScoobie
Scooby Guru
 
Member#: 8785
Join Date: Jul 2001
Chapter/Region: SWIC
Vehicle:
02 c_turner@ix.
netcom.com

Default

Quote:
Originally posted by john banks
Spot on, totally agree.

Methanol is on its way after a lot of playing with different water injection jets and mixtures today. On Delta Dash at about 330 WHP at 19 PSI on 98 RON with the TD05/06-20G(7cm2). With ethanol injection it went up to 340 WHP with a lot of fuel removed and 2 degrees advance and 20 PSI. With a unknown composition methanol based screenwash used neat (freezing point -20C so probably only 25% methanol) it would also run the same but without removing the fuel or adding the timing (wouldn't take it without detting). Oddly with NF Race formula octane booster it will take 21 PSI (EJ257 remember) and makes 350 WHP, hope to get there with the methanol which I am inclined to use as 100% rather than 50/50 with water for my needs. Then onto a GT30R/P20 hybrid from iON next month for 400 WHP I am absolutely flogging this 20G to death, especially with its small housing and wheel on the hot side.

When I want an easy day I just turn it down to 18 PSI and run pump fuel and it is still enough to waste M3/M5, 911s and RS4/6
John. Your car sounds strong. Those are some good numbers for DD. I ran a couple runs a few weeks back on pump fuel and it was in the 300 range. I had to do 2nd gear pulls though because 3rd takes me to 120mph+ and I had no room. I find the dyno feature "almost" usefull. Its so chopy that its hard to compare runs even when smoothed.

On the GT/P20 hybrid... What cartridge are you using and will it use the GT turbine wheel? You should be able to run ALOT more boost with the GT turbo. I ran my GT30R-15 cartridge (with a Turbonetics To4S compressor housing and .84 AR AVO turbine housing at 1.8 bar on C16 fuel without issue for a few days. I backed off it because I felt sorry for my motor.

Cya

Clark
AZScoobie is offline   Reply With Quote
Old 02-03-2004, 01:18 PM   #22
john banks
Scooby Specialist
 
Member#: 11669
Join Date: Oct 2001
Location: UK
Vehicle:
Evo IX 345 WHP

Default

Don't know the cartridge, but it is a Garrett GT CHRA, so this I think means the turbine wheel is GT.

I'm going more on UK results as there are a few in the UK around or over 500 BHP with this setup or very similar to it. There are so many "paper racer"/"bench racer" turbo resellers about, very few with results, many with a lot of promises that won't even get near my "small" 20G.
john banks is offline   Reply With Quote
Old 02-04-2004, 12:39 AM   #23
Equilibrium Tuning
Former Vendor
 
Member#: 26933
Join Date: Oct 2002
Chapter/Region: BAIC
Location: Fairfield, CA
Vehicle:
2006 STI
CGM

Default

Well I made a set of det cans as per the instructions on that website. I'm currently using my laptop to amplify the signal for my headphones. They seem to work, although I haven't actually heard any identifiable knock (good thing I suppose). The only issue I have is that there seems to be a lot of noise in the signal that sounds electrical. Anyone else hear this?

-- Ed
Equilibrium Tuning is offline   Reply With Quote
Old 02-04-2004, 12:47 AM   #24
turboICE
Scooby Specialist
 
Member#: 33979
Join Date: Mar 2003
Chapter/Region: Tri-State
Location: N.NJ
Vehicle:
2006 STi,'11 OBS
WRB & Black/Silver

Default

Check your ground reference. I used the same ground as the sensor. Also depending on the load of your laptop amplifier the low pass filter might need different spec pieces - I think - I am by no means an electrical expert.

I was pleased not to hear knock as well - which was confirmed further by my DD readings on timing settings. I didn't hear any whistle or electrical feedback though.

Ed.
turboICE is offline   Reply With Quote
Old 02-04-2004, 12:58 AM   #25
Equilibrium Tuning
Former Vendor
 
Member#: 26933
Join Date: Oct 2002
Chapter/Region: BAIC
Location: Fairfield, CA
Vehicle:
2006 STI
CGM

Default

I'm also using the same ground asn the sensor. I'm thinking a low pass and high pass filter would do the trick. I believe knock is in the 4Khz - 6.5khz range so filtering out anything bellow and above would probably help a bit. Maybe what I'm hearing is just the resonance associated with different RPM's. Most if not all of the noise seems to go away at higher RPMs.

-- Ed
Equilibrium Tuning is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Tapping the stock knock sensor, will this work? Ziggyrama Commercial ECU Reflashes 9 06-12-2007 08:02 PM
Accurate Knock Readings: KnockLink vs MY04 ECU Knock Sensor The Modaholic Engine Management & Tuning 3 12-05-2004 08:19 PM
FS: Knocklink with Bosch knock sensor modvp Private 'For Sale' Classifieds 3 05-14-2004 01:11 PM
WTB: Knock Sensor for KnockLink in 02 WRX parid Private 'Wanted' Classifieds 1 12-17-2003 01:06 PM
KnockLink and stock knock sensor Impreza Rider Normally Aspirated with bolt-on Forced Induction Powertrain 5 04-22-2002 01:29 PM


All times are GMT -4. The time now is 11:22 AM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2014 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.