PerformanceDevelopments

We make Knock display called KNOCK LINK. If you are interested, contact Danno. He can source from normal Link Distributor and supply. You do need to have another sensor which we can supply to him. Knock signals cannot be shared.

Neil Harvey
Link USA

Bri Bro

You do need to have another sensor which we can supply to him. Knock signals cannot be shared.

The knock signal in this application is not from the sensor, it is a signal supplied from the DME computer to the KLR to change the timing. The 951 already has a sensor to detect knock.

If you want to add another sensor, I am sure PerformanceDevelopmentsLinkUSA has a great product to offer.

Danno

The really great thing about the Knock-Link product is that it doesn't just have a black&white, yes/no indicator of knock, but it also shows you the severity of the knock you're experiencing with a sliding LED display. It probably has a counter that records the number of knocks per time interval and shows higher-frequencies with greater urgency.

If you guys are interested, we'll order one and try it out so you'll know how it works (I suspect very well). I typically don't like to re-invent the wheel and usually we can find off-the-shelf products that performs the functions you desire.

If the trigger between the KLR & DME doesn't provide the proper signal (even with a high-impedance tap), we'll use a second knock-sensor on the head. There's actually a second threaded hole right next to the stock knock-sensor on the 951 head. And the S2 knock sensor would be a good upgrade over the stock 951 unit since it has a continous metal sleeve through the bolt-hole for better conductivity of sound to the sensor.

marksportcts

I've got the J&S safeguard engine knock eliminator on my car, turbo'ed n/a. I am using the older rectangular monitor, each led represents 2degrees of retard.<a href="http://www.jandssafeguard.com/" target="_blank">J&S Electronics</a>

Hans

So, What is the verdict?
Buy or built?

(Sometimes building these nice add-onn's yourselfs will give you that feeling that James Watt must have had when he did his first steam engine)
TakeCare

toddk911

Anyone build one of these yet? How accurate are they? How acurate is the Link sensor? I like that the Link model has the levels of knock. I would think this would be much more in line with tuning than just the a/f/ gauge. I also have GURU chips so do not need to be contstanly wondering what a/f is,they have set that perfectly, I just need to know how much I can crank up boost and be safe.

toddk911

Anyone???????????

dmoffitt

check out the hks knock amp (has a wideband too)

http://www.takakaira.com/asp/templat...KkGhHhceDfbiEd

Tom M'Guinn

I wanted to do this a long time ago, and recall coming to this conclusion: although it may be possible to build a circuit to monitor the knock sensor output, I could not find any digital logic level signal (yes/no signal) that could be used to trigger an LED. The knock sensor produces a signal like a microphone. The DME/KLR reads the signal and makes a judgment call about what is a knock and what is just noise (road vibration, rocks, etc.) Although it seemed possible to build a circuit to monitor the knock sensor output (or the DME/KLR connectors), the circuit would need to have its own rules about what is and what is not a "knock" since I could find no yes/no knock signal to tap. As a result, the circuit could tell you when it thinks there is a knock, but that will not necessarily be the same as when the DME thinks there is a knock and retards the timing. I suppose the same is true about aftermarket knock sensors, unless the signal characteristics of a true knock are obvious and/or industry standards. How about a real-time ignition timing meter and after-market knock sensor?

Keep in mind, of course, that I'm no engineer....so consider the source.

AlexE

So.......don't tap the knock signal........

Can you tap a signal from the dme when it is trying to retard timing????

Tom M'Guinn

It's been a while, but I could not find a signal anywhere that carried a yes/no type signal indicating knock. I think (could be wrong) that the input is an analog signal, which gets interpretted inside the DME/KLR (via some kind of filter circuit I'd guess) so that the DME knows when to pull timing. If there is such a signal, I'm all ears. I don't have a schematic for the DME/KLR and don't know where to find one. There may not even be a dedicated line for the 'knock happening' state inside the DME for all I know -- it could be a memory location, a register, or a multi-purpose input to the processor chip. Speaking of which, I was once told that the processor on the DME is an intel 8051, or some industrial version thereof. Anyone know for sure?

Again, consider the source....I still can't record on one channel while watching another.

toddk911

There is not some sort of voltage to tap into when knock is detected or timing is being pulled back to light just an LED??

Jon Schepps

Any progress on this thread??
I have been looking for similar info recently via a new thread (see "Knock sensor indicator"), but then came across this one.

According to the DME/KLR Test Plan manual, pin 15 of the KLR connector is labelled Knock Yes/No (output), as has been mentioned by others. However the circuit schematic shows no connection to pin 15 of the KLR. Has anyone determined if this pin really has a signal on it, and if so, what it looks like and where it comes from?

Jon

adrial

No progress by me, but I was just thinking about this thread the other day. I'll see what I can figure out...

Danno

I'll hook a 'scope up to that KLR line sometime this week to check it out. Would be interesting as I'll be dyno-tuning this weekend, I'm sure I'll be able to generate some knocking/pinging.

Also the way the timing-retard's done is all internal to the KLR. The Turbo DME passes a timing-signal from the reference-sensor to the KLR. Which then traps it and passes it back to the DME. The DME then fires the coil upon receipt of this trigger single. Optionally, if the KLR senses knock, it can hold onto this triggering signal for a certain amount of time before passing it back to the DME. So as far as the DME's concerned, it has no idea on knock or ignition retard.