Question: Coyote Knock Sensor Description Clarification

In the SCT Advantage suite, the description for knock sensor sensitivity says "A smaller number in these tables reduces the sensitivity of the knock sensor for that cylinder".

The description for that function in HPTuners indicates "This defines the detection threshold for knock. If the knock signal minus the noise floor is greater than this then knock is detected".

Does the description in the SCT Advantage software hold true for HPTuners as well? In other words, if I increase the values in the HPTuners software, does that also increase the sensitivity of the knock sensors?


Attachment 67267Attachment 67268

I've been curious about the same thing but haven't tested it out much. If I remember correctly HP Tuners description is correct and it defines the amount of noise before it is considered knock. I can't find the log but unless I'm backwards on my thinking I set the tables to like 0.1 on a good running car and started getting timing pulled from knock.

If the values are the amount of noise that must be filtered before the sensors actually 'see' knock, then if I were to raise these values, then it would tend to desensitize the knock sensors, which is opposite of the SCT description. If that is true, then lowering the values in HPT would increase their sensitivity. I can't understand why HPT would define the parameters in a way that is opposite of SCT Advantage. I started to question this when I pulled some SCT tunes from blower cars and noticed that many of the better tuners raised these values by 10-15% in order to provide some additional protection. I need to get confirmation from someone who knows definitively the answer to this question. I hope Eric or someone else can provide some clarity on this. riderunwv, I agree that this needs to be clarified and maybe a change in the description of this parameter is in order. I will send an email to support to see if they can provide an answer.

If I can borrow my friends Mustang later I'll check it again to confirm. Ford raises the numbers at high RPM on Boss tunes. I've noticed that and asked about it before but got no response.

If you get a chance before me you can test it by setting the values to an extreme and see which one pulls timing all the time and which one adds it. Obviously be careful of your knock advance settings and spark when doing this to make sure it's good and safe. I don't think they raised them to provide protection on blower cars but rather to get the sensors desensitized a bit because of the extra noise of the blower but I could be thinking about it backwards.

I sent an email to support for clarification. I also requested that they re-word the description to clearly indicate the impact of this parameter.

Quote:

---

Originally Posted by ridenrunwv

If I can borrow my friends Mustang later I'll check it again to confirm. Ford lowers the numbers at high RPM on Boss tunes. I've noticed that and asked about it before but got no response.

If you get a chance before me you can test it by setting the values to an extreme and see which one pulls timing all the time and which one adds it. Obviously be careful of your knock advance settings and spark when doing this to make sure it's good and safe. I don't think they raised them to provide protection on blower cars but rather to get the sensors desensitized a bit because of the extra noise of the blower but I could be thinking about it backwards.

---

I think their description is good. Maybe just add a simple terms of higher is less sensitive and lower is more sensitive or something like that.

I still believe that higher is more sensitive, but we will have to wait until Support gets back to me, or Eric or someone with a working knowledge of this parameter chimes in .

Quote:

---

Originally Posted by ridenrunwv

I think their description is good. Maybe just add a simple terms of higher is less sensitive and lower is more sensitive or something like that.

---

Yeah. I can't find the log from when I messed with it before so I probably didn't even save it. If nobody clarifies before I get access to something to test it on I'll let you know something for sure when I test it out.

Someone test it and report back. Raise the values way up, and way down. See what the results are.

I reused an already created parameter when I added this stuff, so the description could be incorrect.

I was wrong earlier about the Boss tune. They raise the number at high RPM. I edited my original post. They also raise the numbers at the lowest point of 1000 RPM.

Hard to tell about the factory calibration but GT files appear to have lower numbers around idle areas and higher numbers at higher load. I am curious to test again to make sure I'm remembering correctly now but I would think that would indicate the HP Tuners description being correct and SCT being wrong. Until one of us gets a chance to test it we're just guessing though.

Quote:

---

Originally Posted by ridenrunwv

I was wrong earlier about the Boss tune. They raise the number at high RPM. I edited my original post. They also raise the numbers at the lowest point of 1000 RPM.

Hard to tell about the factory calibration but GT files appear to have lower numbers around idle areas and higher numbers at higher load. I am curious to test again to make sure I'm remembering correctly now but I would think that would indicate the HP Tuners description being correct and SCT being wrong. Until one of us gets a chance to test it we're just guessing though.

---

Just a note, the 11-14 and 15+ Mustangs have completely different knock sensor parameters, and they could work differently. Let me know what you guys find.

I'm aware that the 2015+ seems to have more defined as far as knock detection goes. I've actually wondered if the 2015+ is better calibrated as far as this goes because it seems like I can get them to run more timing than an 11-14 with the same modifications.

When I read this calibration from an SCT coded tune, all of the knock sensor data were increased by 20% over the stock OEM amounts. Just as ridenrunwv and Eric suggested, testing this with significant increases should point us in the right direction. I will have the time tomorrow to do so. I believe that increasing it by a great deal should lower the amount of spark the sensors add, compared to the stock OEM parameters. I will also back down my spark first to give the sensors some head room. I will post my results. It is still hard for me to wrap my head around why the HPT code would work inversely from way SCT does it...

EDIT: I just checked my email and received a response from support at HPT. They indicated that the HPT parameters do mirror the SCT parameters and that a change in the description would be forthcoming. I will still test empirically tomorrow.

I guess I was remembering backwards then. You can just lower your spark advance table when testing so it can't add anything instead of having to mess with all the spark tables.

Looking at a stock 2006 Mustang GT read it looks like the opposite of how it appears on a Coyote for some reason. The numbers in the knock threshold table tend to go lower as load and RPM increases in the 2006 tune although the values are all over the place at times. Some cylinders have a blocked off section at higher RPM and load where it has a good bit higher value. If higher is more sensitive they must be really cautious with these cylinders or something. The Boss tune was doing this also.

Ok - Here are some quick observations: I tested this on my daily driver; 2012 Mustang GT, PD blower, 304 LTs (they ring) - a mechanically noisy car. First, I doubled the values in the knock threshold from OEM, made two hard pulls in 3rd and did not get any additional spark added (did see .5* briefly). I then wrote into the ECU a tune that used the OEM values for knock threshold. Even with a noisy engine environment, the ECU added 2*-3* to my WOT spark during a pull. Finally, I went back to my original tune (20% increase in threshold values over OEM), logged a similar hard pull, and the ECU added 1.5* - 2*.

When I raised the values in the KT, it increased the sensitivity to knock. I think some clarification to the definition of the parameter is required.

Yeah. Based on that it seems like higher is more sensitive.

Eric, any chance to get the parameters added to be able to log what value the ECU is reporting for this? I have thought of this on thea cars before when I wanted to see if there seemed to be a spike detected or if the general noise level just crossed what was in the table.

I was running behind my tail to find out the ways, and after checking it, it turned out that I had a false knock because of the flexible exhaust in my car, the amount is 3-5 KR, is this a real or a false knock, how can I reduce the sensitivity of the knock sensor and what is the proper setting for that in Attached tables below.
First, it is not stable in a certain place, it goes and comes, I have a strong sound, especially when it comes to direct pressure or the accelerator pedal suddenly and slowly, but when I press the accelerator pedal directly the sound disappears.

After examining the exhaust, I found that it was shaking with a strong vibration from the bottom.
Second, the injection system has been thoroughly cleaned with BG products.
Any advice would be appreciated.

Engine:stock
Octane: 95
Spark plugs: Ruthenium NGK; 31 Gap
I know that the solution is temporary until the exhaust is fixed. I have an idea to wrap it.
Thanks in advance...

Attachment 130751, Attachment 130752, Attachment 130753
Attachment 130754
Attachment 130755
Attachment 130756

You raise the "knock threshold" to desensitize