Status
Not open for further replies.
That is a very nice unit but costs more than his ecu. With an ms2 For 85cents he can build a knock sensor conditioning circuit..... it of course is a crude method comparably and the circuit must be tuned to the sensor but it's easy when you have the manufacturers data.... also it's fully integrated and can pull spark, add fuel and cut boost when using a boost controller.


A very simple method to test for knock is to make a set of headphones. Using copper tubing smash one end flat so it can be bolted to block, on the end of the copper tubing add a length of nylon tubing and run that to a set of earmuffs. Drill a hole in the muffs to push the tubing into and tape it so it stays in place, remove the wading from that side of the muffs. Voila knock sensing muffs to tune with.
 
  • Sponsors (?)


Rehash of a Jan 17 2017 post

with a huge turbo, 9:1 compression, huge boost 250 I6

Aussie EECIV's used them too. Pin outs on the US Big Six, different. Carb, EECIV and TFI had them on the 4.9.

EFI unleaded 4.1 had TFI and the same US derived EECIV stratergy.

The common E6RF 12A69BA sensor is the orange inserted item in one of a guy on another forum.

All later Aussie EECIV EFI's had Pin 23 hooked up to the Knock Sensor (Unleaded Only, we went unleaded only in 1986, not 1975 as in the US)

xfeec4knocksesnorboltsensor.jpg


xflowintake.jpg


You just need to come down on total advance.

All X-flow Canted valve heads have 302 Boss/ 351 Boss closed chamber properties, so there is poor gas flow incomming, and excessive quench.

Ford Australia started with Leaded 97 octane the heighest lead value in the world untill 1986.

Then they dropped the compression ratio from 9.35:1 to 8.6 or 8.8:1, and added 1986 E6 Mustang style high swirl ramps and EECIV TFI spark reduction stratergy.

That allowed 91 octance to give the same power as 97 leaded.


So you have to do the same, and quell detonation by tipping in advance retard under boost. High static advance, then no more than 23 total, and maybee even less than 16 under boost.


The front head stud takes a knock sensor on X flow Fords. ARP stud kits aren't drilled for the battery earth which also fits the KS.

Early log head Falcons had a front US passenger side thread drilled into the head bolt.

The bolt is basically the Cleveland type, 7/8" taller than the old 200/250 non cross flow logger. It has a screw in thread, and a Piezo electric knock sensor was used to trim the spark advance.

You can use it to peal back advance under load.


Don't be affraid of knocking back advance...low advance tollerance and a lack of incipient knock is proof of efficient combustion. Thats what all box code WA1 EECIV ecu's have, essentially 1987-1997 4.9 F150 and 1980-1991 5.8 G code VV7200 P71 knock sensor logic.


Its the whole reason why Ford Special Vehicles killed most of the 1970 to 1978 Cleveland race engines, and why dumb a$$e$ who put earlier Durasparks on TFI EECIV 4.1s ended up taking them out....


Oh yeah, now I remember..

When I did mine, I used the coomon 60 pin plug EEC 4 Ford Falcon XF 1986 to 1991 E6RF-12A69BA item, which looks similar to the old EEC3 and EEC4 Panther platform items for 302 and 351 from 1981 to 1991. It mounts on the common earth strap for the log head bolt number one, US drivers side front bolt. The Alloy head bolt, also used on Classic Inlines engines, is about 0.875" taller, but uses the same sensor bolt as my 1972 XA Falcon engine.

The sensor is Fig 13 on the Page 221 Tomco website listing http://www.tomco-inc.com/Catalog/knock sensors.pdf, looks the same as the F150 4.9 six as well. Tomco 22 or 42 (29022 and 29042)

Tomco listes them differently to Standard Parts #KSxx listings, but the Ford knock sensor was the first to the US market in 1981 as a type 21 item, Tomco part number 290021.

The Aussie ford item just polls an on off 3 volt signal back to the EEC4, and that cuts the TFI timing back.

Since its a Honda head, it needs Larry Widmer Honda logic.

He uses the GM's 1986-1996 1997562,10456017,10456287 (Tomco 29014) knock sensor on his 480 hp VTECH engines


p74174_large-1989_honda_civic_si_coupe-knock_sensor.jpg


He also uses a GM knock sensor, which he learned about while he was in the stock car biz. “I got busted for running them in NASCAR…that and traction control,” he recalls with a grin.


Those Metal Lines Running Underneath The Headers Are Strain Gauges That Measure Cylinder Distortion. If The Cylinders Start To Flex Too Much, The Engine Management System Automatically Reduces Timing And Adds Fuel.

Good fun with the Knock Sensor tune, its easy when you can map it by Tuner Studio
 
  • Like
Reactions: 1 user
I grabbed a knock sensor for a f150 with a 5.0. Haven't wired it into MS yet but plan to get it done in the next ten years....
 
  • Like
Reactions: 1 users
You use that your muffs I describe building while you turn the circuit for the Knock sensor. You wear The ear muffs While you are data logging. When you hear knock you look to see if the ECU pulled timing if it did not then you turn the voltage up on the potentiometer. Like I said it's sort of a crude method but once it's done it's done for good The newer software with MS3 is much more OEM like MS2 you have to do-it-yourself
 
How does getting a knock sensor for a factory ecu going to work with a non-factory aftermarket unit?
Where is the reference standard going to be for what is regular engine harmonics, and what is detonation?

The knock sensor input is already built in my MSPNP2. It has an earphone jack on the ecu and has settings in TS for knock sensor frequency and adjusts timing and... I believe fuel tables.

The threaded hike at the back of a 302 block is for a factory knock sensor but only the trucks came with them from the factory.

View: https://youtu.be/hEWB0oqbdfo
 
Much easier knock sensor solution and it's a jy friendly one at that. Only requires small mod to the input circuit and is OEM SOLUTION..... you just setup how you want the signal used (I can help here I'm doing this to mine as well)

http://www.megamanual.com/ms2/knock.htm

Knock sense is a product that diy sells to interface with the ms units.... it's all of 55$ built and ready to go you just hook it up.
 
All MS's are really 60 pin EECIV's with more open source electronics.


Down here in the Antipods, we are stuck between a mixture of the old European ignition retard systems, and the later Link ECU system controls, which are for Japanese turbos. Ford Turbo Falcons and Supercharged or turbo Holdens with GM engines, there is normally a reflash, so they use the stock stuff.

Most guys are using a Jaycar KC5444 knock sensor by John Clarke.

It has a PSD= Power Spectural Density determination

ks_calibration_006.jpg


ks_calibration_007.jpg


Early knock sensors...well, all the same really, a knock sensor is a knock sensor, although the GM ones are more sensitive.

Later, the normal Ford 3 volt signal polling rate with a change to less than 3 volts being the descriminator has changed to the AFM type amp draw algorithim.

Just an Anolgue Digital "Do If True" ignition peal back.

Like the very bad 4.9 1-bbl feedback F150 truck system that screwed over so many shade trees who couldn't be frigged pullin the codes.

For the earlier resonant type, they are tuned to a particular engine’s knock frequency of interest.

Later KS's have a Power Spectural density determination, with a sampling cut off to sense the deap seated knock. Jaycar's does it that way, and so does GM now.

The combustion shock wave excites a characteristic frequency in the engine, which is typically in the 5000Hz–7000Hz range. Cylinder bore diameter and combustion chamber temperature are the main variables that affect this fundamental frequency. Variations in the fundamental frequency for a given engine configuration can be as much as ± 400 Hz in terms of combustion chamber design.

In terms of bore, the knock frequency can be greatly influenced.
Approximated using :

900/(Pi * r) where r = the radius of the bore in Sillymeters, um, millimeters.

Given a bore of 76mm the knock frequency will be 7540 Hz.
A bore of 3.7" or 93.98 mm, the knock frequency will be 6100 Hz
A fat bore Chevy or Ford with 4" bore, its down to 5600 Hz, with the very smooth combustion of a wedge head dragging its audiability down to a 5200 Hz threashold

Larger diameters and/or lower temperatures result in a lower fundamental frequency. Signals received by a remote sensor contain additional vibrational modes, which are structural resonances in the engine excited by the shock wave as it hits the cylinder wall. Typically, two to four additional frequency peaks are evident between the fundamental frequency and 20000 Hz. (PSD= Power Spectural Density determination)

Engine structure can have different higher vibrational modes. Sensor mounting location can affect which modes are detectableand the amplitude of each with respect to the background mechanical noise.

The later sensor’s all follow the Siemens or Bosch resonant frequency is above 20kHz with a wide pass-band to give a relatively flat response over a frequency range between 3kHz and 15kHz. This allows this sensor to monitor either the fundamental knock frequency or 1st harmonic for most engines.

The pinouts and use of MegaJolts EDIS6 means its a straight down the line FoMoCo.

On the stock Aussie EECIV for 1986-1992 Port EFi X flows, Ford Australia just copied the USA protocols. If knock occurs, the base 3 volt signal is pulled low.
xf_250_efi_pinouts_001.jpg


"Do if true" ignition timing reduction if knock is present. How it does it is just switch related.

"Cut me a switch, bouy"
 
Last edited:
All MS's are really 60 pin EECIV's with more open source electronics.


Down here in the Antipods, we are stuck between a mixture of the old European ignition retard systems, and the later Link ECU system controls, which are for Japanese turbos. Ford Turbo Falcons and Supercharged or turbo Holdens with GM engines, there is normally a reflash, so they use the stock stuff.

Most guys are using a Jaycar KC5444 knock sensor by John Clarke.

Early knock sensors...well, a knock sensor is a knock sensor, although the GM ones are more sensitive.

Later, the normal Ford 3 volt signal polling rate with a change to less than 3 volts being the descriminator has changed to the AFM type amp draw algorithim.

Just an Anolgue Digital "Do If True" ignition peal back.

Like the very bad 4.9 1-bbl feedback F150 truck system that screwed over so many shade trees who couldn't be frigged pullin the codes.

For the earlier resonant type, they are tuned to a particular engine’s knock frequency of interest.

Later KS's have a Power Spectural density determination, with a sampling cut off to sense the deap seated knock. Jaycar's does it that way, and so does GM now.

The combustion shock wave excites a characteristic frequency in the engine, which is typically in the 5000Hz–7000Hz range. Cylinder bore diameter and combustion chamber temperature are the main variables that affect this fundamental frequency. Variations in the fundamental frequency for a given engine configuration can be as much as ± 400 Hz in terms of combustion chamber design.

In terms of bore, the knock frequency can be greatly influenced.
Approximated using :

900/(Pi * r) where r = the radius of the bore in Sillymeters, um, millimeters.

Given a bore of 76mm the knock frequency will be 7540 Hz.
A bore of 3.7" or 93.98 mm, the knock frequency will be 6100 Hz
A fat bore Chevy or Ford with 4" bore, its down to 5600 Hz, with the very smooth combustion of a wedge head dragging its audiability down to a 5200 Hz threashold

Larger diameters and/or lower temperatures result in a lower fundamental frequency. Signals received by a remote sensor contain additional vibrational modes, which are structural resonances in the engine excited by the shock wave as it hits the cylinder wall. Typically, two to four additional frequency peaks are evident between the fundamental frequency and 20000 Hz. (Power Spectural Density determination)

Engine structure can have different higher vibrational modes. Sensor mounting location can affect which modes are detectableand the amplitude of each with respect to the background mechanical noise.

The later sensor’s all follow the Siemens or Bosch resonant frequency is above 20kHz with a wide pass-band to give a relatively flat response over a frequency range between 3kHz and 15kHz. This allows this sensor to monitor either the fundamental knock frequency or 1st harmonic for most engines.

The pinouts and use of MegaJolts EDIS6 means its a straight down the line FoMoCo.

On the stock Aussie EECIV for 1986-1992 Port EFi X flows, Ford Australia just copied the USA protocols. If knock occurs, the base 3 volt signal is pulled low.
xf_250_efi_pinouts_001.jpg


"Do if true" ignition timing reduction if knock is present. How it does it is just switch related.

"Cut me a switch, bouy"
Great info.


Have anything application specific?
 
Um, yeah, most of us are using Propane and have to face long drives before the gas station, so lots of dual fuel gasoline and LP Gas V8 applications in pommy cars like MG Magnttes and Rover P5 3500's.

Same with the Ford I6 stuff...it was actually propane that caused such a smack in the head for Ford Australia when aftermarket and dealer fitted LPG systems wouldn't cow tow to the new EEC5, so Ford Australia suspended the six cylinder EDIS for two long years as a running change, went back to TFI and EECIV, and then upgraded to EECV without OBDII. All the changes ment a lot of guys were experiementing with extreme spark advance like the GM 3800 engine, which often had its D4 Delco hacked in open wheeler race cars.

All our early EDIS systems were in 6 and 8 cylinder Falcons, and skipped OBDII, so a stock EECIV or EECV down here is missing the extra protocols, but has the extra ECM fire power. Not everything Knock Sensor related needs an aftermarket MJ.

Knock Sensors were all two wire, and a lot of guys were able to gene splice the Aussie EECIV to the Vane meter and 75 mm throttle body to other in line engines. Its similar to the big 4.9, but has the vane meter, turbo 2.3 VAM style.

Texas Instruments provides the details, http://www.ti.com/lit/an/spra039/spra039.pdf

http://www.rhinopower.org/knock/knock.html the PSD info

Jaycar the kit http://archive.siliconchip.com.au/cms/A_108910/article.html

The knock sensor allow 11:1 compression ratios with aluminum blocks. Although the guy hand tuned it with audiable knock, this this version doesn't have the K5444 kit, the Jaycar kit does the high and low pass correction on the square wave form

1_SC_Module_s.jpg


The 268 cubic inch version of the Rover Buick 215
http://roverp5.proboards.com/thread/9873/silicon-chip-programmable-ignition-module?page=1

Link in NZ make non OBDII internal computer refits which then have to interface with stock Bosch KS sensors via the M8 or M10 block bolts.

http://www.nzefi.com/product/knock-sensor-bosch-donut-type/
 
Last edited:
  • Like
Reactions: 1 user
I think he explained it quite well. I now have a better idea that I don't know cht about spark knock sensors. I just thought it was bad, now I know it's really bad.
Backup the timing in boost
Add a little more fuel
Use a colder plug
Hit the go peddle and have fun!
 
Damn I drove down to visit my 85 yr old Dad in Corpus Christi and missed a ton of crap! Keep driving it as long as you can because it looks like this fall er summer is going to last a while.
 
  • Like
Reactions: 1 user
The question woz

How does getting a knock sensor for a factory ecu going to work with a non-factory aftermarket unit?
Where is the reference standard going to be for what is regular engine harmonics, and what is detonation?

#6445 madmike1157, Monday at 9:45 AM

The answer is, its conditioned data from a switch. For 0.85c and the time taken to fit it, you can use the anaoluge filter to count triggered knock if you wanna go back and add the stuff into your system....its just time.

I don't know what all the hub bub about knock sensor is about round here,.. .im not gonna use one.

It was just an idle thought.

I guess your not the one who knocks

spbcs1_003.jpg


Your the one who.....
spbcs1_002.jpg


And just to confirm

I'm too busy look'n up all those big words xecute was using in his posts to be bored. Definitely not a product of the public school system, bet he has an accent.

Yeah, a Kuwi Uccent...like this....
View: https://www.youtube.com/watch?v=u7t2f-L6Z04


I like the system.

Tick the trick Snow maker. A really mean meth does not a Bake Make....heh...
 
  • Like
Reactions: 3 users
The question woz

How does getting a knock sensor for a factory ecu going to work with a non-factory aftermarket unit?
Where is the reference standard going to be for what is regular engine harmonics, and what is detonation?

#6445 madmike1157, Monday at 9:45 AM

The answer is, its conditioned data from a switch. For 0.85c and the time taken to fit it, you can use the anaoluge filter to count triggered knock if you wanna go back and add the stuff into your system....its just time.



I guess your not the one who knocks

spbcs1_003.jpg


Your the one who.....
spbcs1_002.jpg


And just to confirm



Yeah, a Kuwi Uccent...like this....
View: https://www.youtube.com/watch?v=u7t2f-L6Z04


I like the system.

Tick the trick Snow maker. A really mean meth does not a Bake Make....heh...

Now your fitting in.:D

However you seem very versed in ecu hardware I would love your technical jargon over in the megasquirt help thread. We will just need to water it down a bit so it's easy for the regulars to understand.
 
  • Like
Reactions: 1 user
Status
Not open for further replies.