cbarno042

This is a writeup on a very affordable standalone knock monitoring setup.

Here is the setup: https://gumroad.com/d/032c2221026264fc11cf0ac21f068427

It requires a wideband Bosch knock sensor, a headphone jack, 3.5mm to USB adapter, and the software. FYI you get 5 licenses per purchase. It's essentially electronic DETCANS that can be recorded, manipulated, and fine tuned for your specific setup. As you may know, I've been fighting some what-seems-to-be phantom knock since the beginning of July. I'm using the software to verify no knock is present and to just learn more about how resonant frequencies behave in an engine.

So the setup is super simple. Connect one wire to each pin coming out of the knock sensor and then solder a headphone jack to the other end of those wires. I installed this sensor on an empty port that's just above the factory knock sensor. An M8x1.25x 45mm bolt did the trick. (Should it be an empty port? Idk. Works though.)




Here is the dashboard:




The software will recognize the USB adapter as a microphone and you can select it from the dropdown above the analyzer plot thing. At first everything worked great. Then after a while I would select the USB microphone and it would still listen through the built in laptop mic. It was driving me nuts as I wasted a ton of pulls trying to figure this out. I eventually decided to reinstall the software and everything is fine now.

You'll see on the dash that you can select bore size. This changes the resonant frequency that is monitored while you're listening to the recordings. You can also adjust them however you want with the hi/low pass knobs. There is also a hi frequency selection that filters everything out except sounds that are at 2x your bore's resonant frequency. The software comes with some example recordings of knock and this hi frequency setting is crazy. Ain't no knock getting by you on this setting.

Every recording you do is saved as a full-range, 1kHz- 20kHz, .wav file. The hi/lo pass filtering is done after the fact.

Gain and sensitivity are adjustable and that essentially dials in your particular setup's happy-spot. There is a false knock filter as well. I don't really screw with this as I don't even pay attention to the knock count. I just listen and it's pretty obvious when there's knocking. For your convenience, you can have the software say the word "knock" whenever it receives a count. I don't like Stephen Hawking yelling at me so I disable that.





First step of this process is to get a baseline run with no counts of knock. The first run of the day is always knock free for me and today was no different. I got a good clean pull, then followed it up with 3 more pulls. 1 with 6 counts throughout, 1 with 9 counts, and 1 with a whopping 19 counts. I was excited to look at that train wreck.

I then set all of my parameters based on my clean pull and went from there. Between the pulls I couldn't hear ANY discernible knock. Funny enough, the 19 count pull was the quietest of all and was well below the knock threshold on the software. This was probably because throttle was cut. I then ran the recordings of known 300zx knock given to me with the software. I'm confident now that this is phantom knock.


I made some desktop recordings of the software. The audio is kinda messy and very low because the microphone isn't so good on this laptop but this'll give you the idea. *****Be careful headphone users when the full range 300zx plays***** I show 4 of my pulls in each video and in the high frequency/ full range videos, I show what real knock looks like with a provided sample from the software creator. The high frequency mode REALLY picks up on knock well from messing with the knock samples. All pulls are using the same map. Low boost and timing.

The 4 pulls from me are:

Lower Base (a 1-2 count pull according to the log, clean pull)
Lower 1 - 6 counts (6 count pull according to the log)
Lower 2 - 19 counts (Yea, a 19 count pull according to the log, throttle and timing were reduced by Tephra)
Lower 3 - 9 counts (9 count pull according to the log)

Then I throw in the 300zx sample recording to compare. You can see that even though the 300zx has an 87mm bore size, the knock is still easily detected in the high-frequency setting at 89mm.

First video is full-range, next is high frequency, last is standard frequency


My thoughts in the end are that this software is really valuable and it fits right in to where a hobby mechanic can justify standalone knock monitoring software. It's really user friendly, gives good results, and doesn't cost much. It can also be set up for analyzing 2 banks for V or flat engine configurations.

They have a free 5 day trial if you're interested. They supply you with a few different .wav files of true knock so you can mess around. https://gumroad.com/d/032c2221026264fc11cf0ac21f068427

If you DO have problems, tech support is non-existent from what I can tell. So good luck.

jmlegacy

The correct link for the Tuner Knock Monitor is
https://gumroad.com/l/tunerknockmonitor

Updated to version 1.12, so far much more sensitive to knock and more precise in isolating knock from noise.

They are also working on an RPM module, so you know what RPM they got the knock!

LetsGetThisDone

iTrader: (1)

Seems like a lot of work vs taking the car to a real tuner and getting it tuned correctly.

donperry

And what if he's his own tuner? Sometimes false knock does exist though and I believe forged pistons may be behind his peril.

cbarno042

If it's been determined that it's not real knock, then the tune is fine, right? It's a phantom source at that point. Unless you're talking about my tuner filtering the sensor or other techniques.

q_dubz

so two wires to a sensor and a software download is a lot of work?

jmlegacy

Hello,

A new release is available. There is a demo download.
http://tunernerd.com/download/knock-monitor-pro-demo/

jmlegacy