PCM Diagnostics & Tuning HP Tuners | Holley | Diablo
Sponsored by:
Sponsored by:

No VE Tuning?

Thread Tools
 
Search this Thread
 
Old 06-24-2009, 11:52 AM
  #41  
TECH Enthusiast
iTrader: (1)
 
WeathermanShawn's Avatar
 
Join Date: Dec 2005
Location: Denver International Airport, Colorado USA
Posts: 549
Likes: 0
Received 0 Likes on 0 Posts
Default

Originally Posted by RedHardSupra
Virtual VE is a completely wrong direction. It eliminates all the pros that the parametric VE gives us, and allows us to proceed with all the wrong assumptions of VE tuning we've had so far.
Marcin, is that opinion, or do you have facts to back that assertion?

Asking with all due respect, not to start an argument.

Thanks.

..WeathermanShawn..
Old 06-24-2009, 12:10 PM
  #42  
TECH Fanatic
iTrader: (1)
 
RedHardSupra's Avatar
 
Join Date: Sep 2003
Location: Laurel, MD
Posts: 1,904
Likes: 0
Received 1 Like on 1 Post

Default

Shawn, I have a draft of a paper I wrote about it over a year ago. I never finished it, but I think it should cover enough details for you to form your own opinion. You want it? I do warn you though, it's like 13 pages of linear algebra...
Old 06-24-2009, 12:25 PM
  #43  
TECH Enthusiast
iTrader: (1)
 
WeathermanShawn's Avatar
 
Join Date: Dec 2005
Location: Denver International Airport, Colorado USA
Posts: 549
Likes: 0
Received 0 Likes on 0 Posts
Default

Well, I read your Speed density paper from start to finish.

Marcin no doubt your brilliant. I got through Calculus-3 and differential equations, but if I could make B's and C's I was lucky. So it is not easy reading.

No sarcasm here, but for 99% of us the 'cliff note' summaries of your work would really help to educate many of us as to your findings.

Either that or you need to market a new tuning software program. I like your ideas of VE..temperature bias and using STD in a program.

You just need to condense your knowledge for the rest of us mortals.

Can you briefly elaborate on your opposition to Virtual VE?

Thanks.

..WeathermanShawn..

.
Old 06-24-2009, 01:20 PM
  #44  
TECH Fanatic
iTrader: (1)
 
RedHardSupra's Avatar
 
Join Date: Sep 2003
Location: Laurel, MD
Posts: 1,904
Likes: 0
Received 1 Like on 1 Post

Default

I'll try to put up something tonight on my site, a discussion like that would be way off from the main topic of this thread.
Old 06-24-2009, 10:45 PM
  #45  
TECH Enthusiast
 
macca_779's Avatar
 
Join Date: Nov 2006
Location: Katherine N.T Australia
Posts: 620
Likes: 0
Received 0 Likes on 0 Posts
Default

Originally Posted by RedHardSupra
Virtual VE is a completely wrong direction. It eliminates all the pros that the parametric VE gives us, and allows us to proceed with all the wrong assumptions of VE tuning we've had so far.
The Maths of a Parametric equation utilising the co effecients is well above my head Marcian. Isn't VVE just a representation of those said coefficients anyway. If your just using it as a representation does it really matter. After all as long as the volume of air that the ecm thinks its getting is what its getting does it really matter how you got there.
The Raw data is still there if you want to use it after all.

One good thing is at least EFI LIVE has embraced the stock programming rather than hacking it and going old school like HPTuners has done.
Old 06-25-2009, 09:24 AM
  #46  
TECH Enthusiast
iTrader: (4)
 
BOTTLE ROCKET's Avatar
 
Join Date: Jul 2004
Location: GA
Posts: 720
Likes: 0
Received 0 Likes on 0 Posts

Default

"Boosted" differences were referenced a few times. Are you guys saying that if you are running boost (+ H/C) you definitely SHOULD tune both VE and MAF?

On a side note; Where is a "beginners" write up for calabrating the MAF?
Old 06-25-2009, 10:21 PM
  #47  
TECH Fanatic
iTrader: (1)
 
RedHardSupra's Avatar
 
Join Date: Sep 2003
Location: Laurel, MD
Posts: 1,904
Likes: 0
Received 1 Like on 1 Post

Default

http://redhardsupra.blogspot.com/200...explained.html

please comment, discuss, or just rip it a new one...
Old 06-26-2009, 05:05 AM
  #48  
Teching In
 
Pekka_Perkeles's Avatar
 
Join Date: May 2007
Posts: 43
Likes: 0
Received 0 Likes on 0 Posts

Default

Originally Posted by macca_779
You must of been doing it wrong.
Did you filter out large delta throttle inputs?
A bit off-topic, but how do you guys do that? Some special filter string perhaps?

Originally Posted by macca_779
Did you get over 50 hits in the main cells you wanted to tune?
Did you use a Wideband? And did you disable fuel trims when doing it?
Did you disable the MAF?

Each one of those is mandatory when VE tuning. Miss one and your wasting your time.

VE tuning is not hard. It amazes me most tuners outside of AUS still haven't seen the light of SD tuning. Hell we even had our high end performance models come out from factory MAFLESS SD tuned here.
Old 06-26-2009, 07:18 AM
  #49  
TECH Enthusiast
iTrader: (1)
 
WeathermanShawn's Avatar
 
Join Date: Dec 2005
Location: Denver International Airport, Colorado USA
Posts: 549
Likes: 0
Received 0 Likes on 0 Posts
Default

You are attempting to filter out throttle transients based on changes in TPS % per millisecond. Most of the tuning software gives you that ability. You can also filter out Fuel Cell >19 (decel, WOT, EVAP Purge, etc)..

Theory being abrupt throttle transients are not representative of a true VE RPM/MAP cell.

As far as accurate street SD/VE tuning, seems like the people who can do it successfully have little to no elevation changes on their course and can keep ECT/IAT constant during logging. Any elevation, ECT, or IAT change and you begin sampling multiple air density samples for the same RPM/MAP cell. BARO may only update at key on or open throttle on a sample run, so it can skew your results.

I.E...You can get big swings in your VE values (STD) for the same RPM/MAP cell.

By the way Marcin..great work on your recent work. Still thinking it over.

..WeathermanShawn..
Old 06-26-2009, 07:23 AM
  #50  
TECH Fanatic
iTrader: (1)
 
RedHardSupra's Avatar
 
Join Date: Sep 2003
Location: Laurel, MD
Posts: 1,904
Likes: 0
Received 1 Like on 1 Post

Default

eh, dyno tuning with a wrong methodology is still gonna return crappy results. they're gonna be less peppered with noise, but ultimately it's still crap. read into the paper, i think i've explained it at least partially in there, it's the whole 'proper attribution' bit.



Quick Reply: No VE Tuning?



All times are GMT -5. The time now is 09:49 PM.