Notices
Evo X Engine Management / Tuning Forums Discuss the major engine management systems.

really bad knock all of a sudden

Thread Tools
 
Search this Thread
 
Old Feb 9, 2010 | 05:05 PM
  #196  
Golden's Avatar
Evolved Member
iTrader: (1)
 
Joined: Nov 2009
Posts: 1,456
Likes: 0
From: Omaha, NE
Tephra, how's that knock going? Any sounds of it since you swapped out the HFC?
Reply
Old Feb 9, 2010 | 05:23 PM
  #197  
tephra's Avatar
Thread Starter
EvoM Guru
15 Year Member
iTrader: (6)
 
Joined: Feb 2007
Posts: 9,486
Likes: 67
From: Melbourne, Australia
no knock.

BUT
1) I havn't done conclusive testing for fear of blowing something
2) I have changed the tune significantly
3) I have had a service ie oil change...
4) I am working on a MIVEC theory which may be the root cause, and coupled with #3 might be a significant discovery
Reply
Old Feb 9, 2010 | 05:26 PM
  #198  
Noize's Avatar
EvoM Administrator
20 Year Member
Liked
Loved
Community Favorite
iTrader: (24)
 
Joined: Mar 2003
Posts: 8,849
Likes: 138
From: Franklin, TN
Interesting, Tephra. Do share if it all adds up!
Reply
Old Feb 9, 2010 | 05:27 PM
  #199  
tephra's Avatar
Thread Starter
EvoM Guru
15 Year Member
iTrader: (6)
 
Joined: Feb 2007
Posts: 9,486
Likes: 67
From: Melbourne, Australia
ok so the theory is that under some conditions (ie old oil maybe) the mivec cams can't keep up.

particularly the exhaust cam since during spool it changes quite a lot.. and if its lagging behind means that the exhaust cam will be retarded too much on peak boost (for a short period of time)

P0011 and P0014 are meant to pop when the cams are out of range from their target, BUT it requires 5 seconds detection period... much longer than is useful.

its only a theory, but does make sense..
Reply
Old Feb 9, 2010 | 05:30 PM
  #200  
SiC's Avatar
SiC
Evolved Member
iTrader: (1)
 
Joined: Jun 2008
Posts: 1,018
Likes: 0
From: Kanagawa, Japan
Did a lot of testing today, the actual was around 1-3 degrees slower than target when going wot but while crusing it was pretty much spot on.

This is great, thanx mrfred & tephra
Reply
Old Feb 9, 2010 | 05:33 PM
  #201  
tephra's Avatar
Thread Starter
EvoM Guru
15 Year Member
iTrader: (6)
 
Joined: Feb 2007
Posts: 9,486
Likes: 67
From: Melbourne, Australia
I have a better formula for you:

(4096-x)*0.01953125

should be a bit more accurate...
Reply
Old Feb 9, 2010 | 05:38 PM
  #202  
tephra's Avatar
Thread Starter
EvoM Guru
15 Year Member
iTrader: (6)
 
Joined: Feb 2007
Posts: 9,486
Likes: 67
From: Melbourne, Australia
yes - my initial WOT log, looks like the exhaust cam never reaches the target.

now this maybe because the target map is based off a different load to what I am logging, but that would be a bit strange.. because its out by about 40 load points...
Reply
Old Feb 9, 2010 | 05:44 PM
  #203  
Clipse3GT's Avatar
Account Disabled
iTrader: (10)
 
Joined: Dec 2005
Posts: 2,185
Likes: 2
From: Cleveland, OH
Originally Posted by tephra
I have a better formula for you:

(4096-x)*0.01953125

should be a bit more accurate...
What MUTIII values for the GSR 52680020 do we log for MIVEC timing?

Use the same formula/?
Reply
Old Feb 9, 2010 | 05:45 PM
  #204  
tephra's Avatar
Thread Starter
EvoM Guru
15 Year Member
iTrader: (6)
 
Joined: Feb 2007
Posts: 9,486
Likes: 67
From: Melbourne, Australia
mrfred posted USDM memory addresses last page...

SiC posted the formula last page...

my formula above is a bit easier on the brain and a little bit more accurate
Reply
Old Feb 9, 2010 | 05:46 PM
  #205  
Clipse3GT's Avatar
Account Disabled
iTrader: (10)
 
Joined: Dec 2005
Posts: 2,185
Likes: 2
From: Cleveland, OH
Originally Posted by tephra
ok so the theory is that under some conditions (ie old oil maybe) the mivec cams can't keep up.

particularly the exhaust cam since during spool it changes quite a lot.. and if its lagging behind means that the exhaust cam will be retarded too much on peak boost (for a short period of time)

P0011 and P0014 are meant to pop when the cams are out of range from their target, BUT it requires 5 seconds detection period... much longer than is useful.

its only a theory, but does make sense..
So you think oil viscosity effects MIVEC timing?

I.e. a 5w-30 eventually breaks down to a 5w-20. So running something like 0w-40 might be better?
Reply
Old Feb 9, 2010 | 05:47 PM
  #206  
tephra's Avatar
Thread Starter
EvoM Guru
15 Year Member
iTrader: (6)
 
Joined: Feb 2007
Posts: 9,486
Likes: 67
From: Melbourne, Australia
don't know yet...

I probably should log target rather than using the map as a reference point
Reply
Old Feb 9, 2010 | 06:08 PM
  #207  
Clipse3GT's Avatar
Account Disabled
iTrader: (10)
 
Joined: Dec 2005
Posts: 2,185
Likes: 2
From: Cleveland, OH
Originally Posted by tephra
mrfred posted USDM memory addresses last page...

SiC posted the formula last page...

my formula above is a bit easier on the brain and a little bit more accurate
Thanks missed it earlier.
Reply
Old Feb 9, 2010 | 06:33 PM
  #208  
David Buschur's Avatar
Evolved Member
iTrader: (53)
 
Joined: Mar 2003
Posts: 14,622
Likes: 32
Sorry, just saw this for the first time. Only read up to page two.

The cylinders you listed in order from left to right I am assuming. So the low compression is in #3 actually. The last two cars we took apart had damage/low compression on #3 and seems to be a pattern to that cylinder.

I'd change the fuel pump relay for sure.

How many miles are on the car now?

If you need something let me know.
Reply
Old Feb 9, 2010 | 06:35 PM
  #209  
David Buschur's Avatar
Evolved Member
iTrader: (53)
 
Joined: Mar 2003
Posts: 14,622
Likes: 32
OK, skipped to this last page.

Oil pressure effects the MIVEC timing for sure, absolutely. Low oil pressure can also make a HELL OF A LOT OF NOISE in the MIVEC cam gears.
Reply
Old Feb 9, 2010 | 06:47 PM
  #210  
SiC's Avatar
SiC
Evolved Member
iTrader: (1)
 
Joined: Jun 2008
Posts: 1,018
Likes: 0
From: Kanagawa, Japan
Originally Posted by tephra
I have a better formula for you:

(4096-x)*0.01953125

should be a bit more accurate...
Thanx, I'll use that and also tell Hamish to add it to the next Evoscan release
Reply



All times are GMT -7. The time now is 08:22 AM.