High knock values

FlashPro questions & answers specific to the 2006-2011 Americas Civic Si
Post Reply
hapv
Posts: 31
Joined: Fri Feb 08, 2008 1:38 pm

High knock values

Post by hapv »

Hello Hondata,

I have a Central American SI with the next mods: Injen CAI, Toruku cams stage1, ingalls powerstick, vibrant race header and catback and NST pulleys.
Today I installed my flashpro, everything ok, but when I started the datalogs I found very high knock values.
I know I have to work with fuel values, but I am worried about knock levels.
I attach two dataloggins: stock ECU and Injen, header MAP.

Is this because some of my mods?
Attachments
stock.fpdl
(180.01 KiB) Downloaded 456 times
CAI sin tocar.fpdl
(111.01 KiB) Downloaded 354 times
User avatar
Hondata
Site Admin
Posts: 10434
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Post by Hondata »

Knock can be caused by mechanical noise (usually a header or valvetrain), but given that your car shows less knock with the stock calibration, this does not seem to be the case.

You are going to need to identify what areas and causing knock, and retard ignition in those areas. Select 'Knock Count' for the sensor overlay, and use that to show you where on the ignition tables the knock is. It should be no more then 4-5 degrees to remove all the knock.
Last edited by Hondata on Fri Feb 19, 2010 10:03 am, edited 1 time in total.
Hondata
hapv
Posts: 31
Joined: Fri Feb 08, 2008 1:38 pm

Post by hapv »

Hello Hondata,

today I take dataloggins with the normal ignition timming, -10 degreess in all tables, -5 degrees and +5 degrees.
I dont see any change in knock values and knock retard.
Please help.
Attachments
normal.fpdl
(13.57 KiB) Downloaded 254 times
+5 degrees.fpdl
(19.17 KiB) Downloaded 267 times
-10% degrees.fpdl
(31.53 KiB) Downloaded 242 times
-5% degrees.fpdl
(40.82 KiB) Downloaded 249 times
User avatar
Hondata
Site Admin
Posts: 10434
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Post by Hondata »

Knock retard is not related to the amount of noise the knock sensor gets - see the help file.

Your knock retard is fairly high, so I suspect that knock tables have been increased over stock.

Go back to the 'Civic-Si-RRB-1' Stock eqivilent and see what your knock values are. If they are normal, it is the calibration. If they are not, then it is engine modifications.
Hondata
hapv
Posts: 31
Joined: Fri Feb 08, 2008 1:38 pm

Post by hapv »

If you see my dataloggins stock (before any change with flashpro) you can see similar knock retards, so I think it is produced by my engine modifications.

What can I do?

Decrease the values in knock tables?

Sorry but I am very confused with this issue............. :oops: :oops:
VitViper
Posts: 273
Joined: Thu Aug 27, 2009 10:38 pm

Post by VitViper »

It's your cams generating a lot of noise.
hapv
Posts: 31
Joined: Fri Feb 08, 2008 1:38 pm

Post by hapv »

VitViper wrote:It's your cams generating a lot of noise.
Is there something I can do with the flashpro to reduce the knock retard?
VitViper
Posts: 273
Joined: Thu Aug 27, 2009 10:38 pm

Post by VitViper »

The 'knock count' from having a noisy valvetrain, no, not really.

The 'knock retard' you can by reducing the values in the knock low and knock high tables.
BV1
Posts: 28
Joined: Sun Aug 23, 2009 9:13 am

Post by BV1 »

I am having a similar issue on the same cams. I am starting to believe they are just noisy.

I want to clarify that the knock tables do NOT affect knock account. There is no knock retard in the datalogs (anymore) but 300-500 knock counts (still) over 40 minutes. The tune is basically completed but the counts are still on my mind.

Reducing the knock sensitivity appears to have no affect as these just affect the retard sensitivity. The knock counts appears false as AFR's and Ignition values are ideal when they occur.

Is there or will there be a way in the future to calibrate what counts as a knock count? I realize that the "knock count" may be a pre-defined noise limit or rather "rate of change in noise" in the Flash Pro Manager software but once the knock has been determined to be invalid/false/mechanical noise it would be nice to adjust the FPM count method, particular to that tune.

As in to make the .fpcal have a value in it that relates to FPM "count" sensitivity.

I did something similar for users that only have access to .csv files, by allowing them to change what determines a knock count from 3 variables. See attached.

Steve
Attachments
Knock on stock engine
Knock on stock engine
stockknock.PNG (94.27 KiB) Viewed 9651 times
VitViper
Posts: 273
Joined: Thu Aug 27, 2009 10:38 pm

Post by VitViper »

I've asked Hondata to allow us to adjust the ceiling for what counts as a "knock count". Hopefully we get that functionality as I'm sure people with cams are going a bit bonkers.
User avatar
Hondata
Site Admin
Posts: 10434
Joined: Thu Aug 15, 2002 12:13 pm
Location: Torrance, CA
Contact:

Post by Hondata »

The ECU determines a knock event based on the noise ratio at different parts of the engine cycle. It is not a simple adjustment.
Hondata
VitViper
Posts: 273
Joined: Thu Aug 27, 2009 10:38 pm

Post by VitViper »

Hondata wrote:The ECU determines a knock event based on the noise ratio at different parts of the engine cycle. It is not a simple adjustment.
Thanks for the info.
hapv
Posts: 31
Joined: Fri Feb 08, 2008 1:38 pm

Post by hapv »

Hondata,

do you think is safe to reduce the knock retard reducing the knock sensitivity if havent knock counts?
Post Reply