409whp Yay
- Thread starter CWiLL
- Start date
I thought I had made mention of this earlier but its all good. Dollar for dollar there isnt better protection for your motor. The sweet thing is you can bring it with you to your next car and it only cost a small fee to recalibrate it for most cars. Its an awsome device. And I sould like a late night infomercial.lol
TurfBurn said:
I dont think its restricktions are based on a static number like X amount of boost. I do know that they say if your goals exceed like 250 WHP then you should consider a better unit or at the very least get an AFC with timming control.
Maxx Mazda said:
)Thanks Steve. Very informative. Its weird going from forum to forum to see what trends are. Alot of different forums have different trends as far as the end all soltuion to their ems needs. Focus forums praise the sct tuner or tweecer. Probetalk is completely flooded with megasquirt users. I never seen anything on that site about j&s - especially from the 300-500whp guys so i figured they were just using ms w/ knock control. Thats why i first questioned it here. I was quite shocked to hear that alot of ems systems only retard the ignition 3-5 degrees. I would expect that out of a stock ecu with a knock sensor setup, but not a stand alone. Currently Im running a tfi ignition/ditsy setup because i have the fs02 motor. I was going to add 4 knock sensors and switch to a coil on plug setup, letting me control the ignition on each cylinder, but this j&s seems to already have some of the benifeits with the additional safety. I think im sold. thanks.
Last edited:
The J&S is a little weird in it's popularity because the problem is a lot of people don't understand it or why it's better. I haven't met anybody yet that once they fully understand it and/or use it that hasn't fallen completely in love with it. Some communities just have those people that swear by one thing or another and then everyone else follows, right or wrong... so that's what you see. It used to be that it was HUGE thing on Subie forums etc, but it faded off in popularity for some reason. Part of it is the units aren't that fast/easy to get a hold of, and the "sales force" for J&S isn't very wide.. I'm about the only reseller in the nation basically... and John doesn't get on a lot of forums a lot or promote a ton, so you find little pockets of J&S users, and that's about it. I've been working pretty hard to slowly correct that, but I'm still only one guy 
. The trick is simply building an understanding for people and then letting them decide if it is worth it or not for them... most believe it to be definitely worth it.
Later!
Steve
. The trick is simply building an understanding for people and then letting them decide if it is worth it or not for them... most believe it to be definitely worth it. Later!
Steve
John at J&S
Member
Thanks to all our supporters.
We have a following on the Focus (focaljet) forum, but the SCT tuner is big. Some have both. Besides the J&S Vampire, there is no easy way to add other piggybacks to the Ford ECU.
Competition (ECUTEK, UTEC, emanage) killed the Subie market for us, starting about '02, but these systems don't have individual cylinder control.
Igdrasil: Thanks for the link. You never know where stuff ends up.
Below are two screen shots, from two different engines. The first demonstrates the capability of the detector, the second shows the advantages of individual cylinder control.
Middle trace is the J&S retard signal, compared to AEM knock volts in the lower trace, on a 500hp Honda S2000:
Power curves from Ian Whiteside's NA F3 engine. The base curve (no J&S) provides a 2 safety margin from knock, as determined with knock sensing headphones. Remember, when tuning for "no knock", you are tuning for the cylinder that pings the easiest, robbing power from the others.
After installing the J&S and bumping the timing, the dyno operator could not detect any knock in his headphones.
We have a following on the Focus (focaljet) forum, but the SCT tuner is big. Some have both. Besides the J&S Vampire, there is no easy way to add other piggybacks to the Ford ECU.
Competition (ECUTEK, UTEC, emanage) killed the Subie market for us, starting about '02, but these systems don't have individual cylinder control.
Igdrasil: Thanks for the link. You never know where stuff ends up.
Below are two screen shots, from two different engines. The first demonstrates the capability of the detector, the second shows the advantages of individual cylinder control.
Middle trace is the J&S retard signal, compared to AEM knock volts in the lower trace, on a 500hp Honda S2000:
Power curves from Ian Whiteside's NA F3 engine. The base curve (no J&S) provides a 2 safety margin from knock, as determined with knock sensing headphones. Remember, when tuning for "no knock", you are tuning for the cylinder that pings the easiest, robbing power from the others.
After installing the J&S and bumping the timing, the dyno operator could not detect any knock in his headphones.
John at J&S
Member
Linux (first MP3/J&S): No issues that I know of. Picking up a used one? Make sure that it's two channel.
- :
- MSP 1145
with that, Steve, how hard/easy is it to tune using an MPI w/ turbo module? after the initial dyno tune and what not, I'm gonna have to be the one to take on the tuning of the car, and i dunno s*** aobut it. but maybe you can help me along..i know dana will too.
- :
- MSP 1145
ahh. very good.LinuxRacr said:
MazdaSpeed626
Member
- :
- Mazda 626 1998 5spd 2.0l FSDE
TurfBurn said:
I'm in no way knocking the J&S unit because i know its a great system.....BUT, I have a KnocksenseMS box, which was designed specifically for use with the megasquirt, and I can control timing on the fly just like the J&S unit does, and its only $65.
http://www.msextra.com/manuals/MS_Extra_Hardware_Manual.htm#knock
Explains the incorporation of knock detection, provides a diagram of how to build the circuit yourself and lastly gives a link to where you can buy the KnocksenseMS tuned circuit, which heres the link for that page: http://www.viatrack.ca/
The reason the J&S is so expensive is because its an independent system and its functions are controlled completely by the algorithms and user defined settings that are programmed into it. The KnocksenseMS unit i have is a tuned circuit from the distributor solely used to detect knock. When you buy one, they ask you for bore size and block material(cast iron, aluminum) to determine the approximate resonate frequencies of knocking based on your specific setup. So once it gets to you, its quite accurate at detecting knock. From there, it sends a logic signal from the KnocksenseMS to the megasquirt telling it whats going on(ie knock or no). From there, the megasquirt takes over and retards timing on the fly based on the settings you've given it. This to me is a more than capable system from my experience with it so far.
So basically, the cost of the knock ignition control is incorporated into the cost of the megasquirt because the megasquirt is still doing all the calculations and control. Not only is this option cheaper, but its also easier because everything is controlled by a central device....i run a very agressive igntion map on 15psi right now and my KnocksenseMS unit works beautifully, no knock here ladies and gentlemen. Oh and all the wire used with this unit is shielded, so theres no chance of getting interference from anything.
John at J&S
Member
In the 80's, GM used an external knock processor, called the ESC module. The detector was later incorporated into the MEMCal module, mounted inside the ECU. Similar concept. Different part numbers were used with different frequency sensors.
The Geraghty "Ping Alert" was one of the first knock detectors I took apart. Must have been around '85. It used a similar circuit as the one shown in your link, exept it had an audio alert instead of the LED. The instructions recommended that you pencil tick marks around the sensitivity dial for various driving conditions. It used a narrow band resonant sensor, tuned at 6.0kHz.
Carter Carburetor produced the EKE Engine Knock Eliminator about the same time, but it added circuts to increase the threshold with background engine noise. Didn't work that well on my noisy Fiat engine, though, so I got busy. Obsessed, actually.
One sensor per cylinder can't be used to detect which cylinder is knocking, since they all will hear it. The only reason I can think of would be to get the same signal level for each cylinder, but that could be fixed with a better detector. Has to be done before the signal gets to the MS, though.
The Geraghty "Ping Alert" was one of the first knock detectors I took apart. Must have been around '85. It used a similar circuit as the one shown in your link, exept it had an audio alert instead of the LED. The instructions recommended that you pencil tick marks around the sensitivity dial for various driving conditions. It used a narrow band resonant sensor, tuned at 6.0kHz.
Carter Carburetor produced the EKE Engine Knock Eliminator about the same time, but it added circuts to increase the threshold with background engine noise. Didn't work that well on my noisy Fiat engine, though, so I got busy. Obsessed, actually.
One sensor per cylinder can't be used to detect which cylinder is knocking, since they all will hear it. The only reason I can think of would be to get the same signal level for each cylinder, but that could be fixed with a better detector. Has to be done before the signal gets to the MS, though.
Speed626... you are now the owner of a sampling bandpass amplifier. It generates a fixed rate output pulse, and when the sensor gets a loud enough sound on the knock sensor it triggers a state switch to indicate knock. It has no logic, no automatic adjustment, and no correlation to cylinders. Yes, it's great that the MS will pull timing out when it gets that signal and up the amount of timing pulled to a limit etc, but you are still SOL on correcting for engine noise relative to RPM, still SOL on setting it "right" when you put it on the first time, and still out of luck for adjusting at anywhere near the rate and accuracy of the J&S. It's not a bad little device that you got, but it's no more intelligent or useful than any of the other knock systems like the AEM has integrated or so forth. Reference the graphs above that John posted and you'll see what we are talking about... if you can't see why there is a vast difference between the two situations I can explain it with examples if you want.
later,
Steve
later,
Steve
Last edited:
MazdaSpeed626
Member
- :
- Mazda 626 1998 5spd 2.0l FSDE
TurfBurn said:
I said logic with respect to the binary data sent to the megasquirt to tell it if it is knocking or not, a high state or a low state. That by definition is logic. True/False, 0/1, hi/lo whatever you want to call it, its telling the megasquirt when to retard the timing.....like i said at the very beginning, i'm not knocking the J&S unit because i know its capabilities. But, the KnocksenseMS is just one step closer to full integration of engine control into the megasquirt, and things like this circuit and its control can only be improved in the future....Hell, I might try improving on the concept further for my EE senior project.
Yeah, it's a step in the right direction. I agree, and see how it can be a useful low cost solution for some people. But the thing I was trying to emphasize is that it is a very "dumb" sensor as far as saying what knock is or isn't. It's just a threshold meter for noise in a frequency band is all it is. My point was that it's a far cry from the J&S as far as detecting/determining actual knock. People will come on here and see your comments and think that they'll get "close" to a J&S by using this meter and integrating it into an EMS (you could do something similar with it and the Microtech if you wanted mind you) but my point is that the accuracy will be wildly different... just because the device works well enough that YOU don't hear the knock doesn't mean your engine isn't knocking away at all sorts of conditions. There is plenty of knock that is basically inaudbile and that is why guys claim they are not knocking and don't get why they threw a rod through the side of the block or cracked off the top of their pistons.
So the point was that the device, while neat is still "dumb" as to what knock is and isn't giving you an accurate knock reading by any means... it may give you a relatively accurate gross knock indication, but is by no means finesse.
the statement of yours that I"m taking issue with is when you say specifically that the device is "data sent to the megasquirt to tell it if it is knocking or not" as it is NOT doing that, it is only telling the megasquirt when you have more noise at the rough frequency levels of the microphone/knock sensor than what you have set for a threshold.. that is NOT knock detection.
So the point was that the device, while neat is still "dumb" as to what knock is and isn't giving you an accurate knock reading by any means... it may give you a relatively accurate gross knock indication, but is by no means finesse.
the statement of yours that I"m taking issue with is when you say specifically that the device is "data sent to the megasquirt to tell it if it is knocking or not" as it is NOT doing that, it is only telling the megasquirt when you have more noise at the rough frequency levels of the microphone/knock sensor than what you have set for a threshold.. that is NOT knock detection.
Last edited:
That might be a dynmojet corrected number. The nearest dynodynamic to my place defaults to dynojet numbers for FWD cars unless you switch to dyno dynamic numbers before you run. And they correct 11.5%. The owners of the shop brought some cars to a dynojet and then back to their dynodynamic dyno to get a correction percentage.jrodhotrod said:
