Page 1 of 1

odd knock

Posted: Thu Sep 07, 2006 3:13 pm
by EPA2
I want to start by saying I plan to get the car tuned ASAP, but work has kept me from making the 4 hour drive to the shop to get it tuned. Anyway..
I have a A2 powered EP3 and everything is running fine with the cal that I have except when I have the cruise set or going a constant speed up a mild-severe incline I get some knocks. I have datalogs, but can't post because I am at work. Does anyone have any suggestions. I will try to post datalogs later.
What do I need to change to get rid of the knocks?
Thanks

Re: odd knock

Posted: Thu Sep 07, 2006 5:16 pm
by Razathorn
EPA2 wrote:I want to start by saying I plan to get the car tuned ASAP, but work has kept me from making the 4 hour drive to the shop to get it tuned. Anyway..
I have a A2 powered EP3 and everything is running fine with the cal that I have except when I have the cruise set or going a constant speed up a mild-severe incline I get some knocks. I have datalogs, but can't post because I am at work. Does anyone have any suggestions. I will try to post datalogs later.
What do I need to change to get rid of the knocks?
Thanks
Retard ignition in the cruise area / light load area that it knocks in the map. That's really all there is to it. Open your datalog and open the tables, sensors, display, and graph window. Add knock count to one of your graph templates and notice where knock count increases. Make sure you have the options "tables follow cam angle" and "tables follow vtec" enabled. Select the ignition tables in the tables window, then click on the places in the graph where knock count increased, go back to the tables window and remove 2 degrees ignition from the hilighted red 4 cells that by hilighting the 4 red cells and hitting ctl-j for adjust and typing in -2 in the center box that says "add to selected values". That will remove 2 degrees from the ignition in that part of the map. If the CAM in sensors is between two cam angles, say it's at 25, you may want to remove ignition from the 20 and 30 degree cam ignition tables in the same place (ensure edit all tables is not enabled before doing this). Be sure to not remove 2 degrees from the same spot in the map twice unless it continues to knock. If you find yourself removing more than 4 degrees from the cruise / light load areas, you might want to check for mechanical noise causing false knock pickup or ensure your wideband is accurate as it is likely using closed loop for fuel adjustment at that point and could very well be running leaner than expected if the o2 sensor is damaged or the map is a little lean and your'e not even in closed loop.

It's always a good idea to verify your wideband o2 sensor readings with another (in the same spot in the exhaust system).

There is an example video here: http://hondata.net/media/kprotraining/7 ... nknock.mov

More videos can be had here: http://hondata.com/techk-protraining.html

Wayne

Posted: Sun Sep 10, 2006 9:43 pm
by EPA2
-2 ignition cleared up some of my problem, but I still have some problems. I took another 2 out of the area but didn't know if I should take more after reading everything. The problem is just when I am going uphill in 6th going from a cruise to increased load to make it up the hill. If I give it more gas I don't have any problem.

Cals posted.
Mods:A2,jrrh,injen cai,neuspeed exhaust
Thanks

Posted: Mon Sep 11, 2006 11:08 am
by Razathorn
EPA2 wrote:-2 ignition cleared up some of my problem, but I still have some problems. I took another 2 out of the area but didn't know if I should take more after reading everything. The problem is just when I am going uphill in 6th going from a cruise to increased load to make it up the hill. If I give it more gas I don't have any problem.

Cals posted.
Mods:A2,jrrh,injen cai,neuspeed exhaust
Thanks
I won't be able to look at the datalog for a while, but it sounds like it's the transition from closed loop to open loop and you got some fuel thrown in and lowered the afr when you gave it gas and quelled the knocking.

Personally, I wouldn't worry about it if it's just a few knocks every now and then. If you have to pull out more than 4 degrees ignition, I would consider trying another brand of fuel or running higher octane.

Wayne

Posted: Mon Sep 11, 2006 11:31 am
by EPA2
I use 93 octane. I get about 10 knock counts but they are on my way to and from work. What are the chances of causing damage. It looks today like maybe I need to do something with the AF.
Thanks for your help.

Posted: Mon Sep 11, 2006 11:42 am
by Razathorn
EPA2 wrote:I use 93 octane. I get about 10 knock counts but they are on my way to and from work. What are the chances of causing damage. It looks today like maybe I need to do something with the AF.
Thanks for your help.
10 knocks total during a drive to work is nothing to worry about in my opinion. The K series motor has a bad habbit of knocking at light to medium load in the 2k-3500 range. You can tune most of it out, the rest you just have to live with. That level of detonation shouldn't hurt anything in my opinon, but you should at least try to tune it out. If you cannot, I wouldn't lose any sleep over it. Concentrate on high load knocking and get the light load when you can isolate it. Watch your MIL for flashing when driving it daily to spot trends in when it knocks.

Wayne

Posted: Tue Sep 12, 2006 4:34 pm
by kornsined
I get some knock under boost. not much maybe 6 at the most but i'm still gonna try and iron it out.

Posted: Tue Sep 12, 2006 7:04 pm
by Razathorn
kornsined wrote:I get some knock under boost. not much maybe 6 at the most but i'm still gonna try and iron it out.
You should have zero knock under boost. That's full load detonation and should be avoided at all cost. Are you running colder plugs? What kind of exhaust setup do you have and what boost level are you trying to push? Do the knocks lessen as you remove ignition?

Posted: Tue Sep 12, 2006 9:50 pm
by kornsined
colder plugs (BKR8EIX NGK Iridum plugs), 3" ... exhaust, tops out at 9psi on a3 rsx motor. I retarded ignition -2 from 4500rpm-5000rpms where knock was occuring in the datalog. I was good for a few days until today and got a mil flash again. it's very inconsistent on when it occurs. I'll have to datalog again tomorrow and see.