EvolutionM - Mitsubishi Lancer and Lancer Evolution Community

EvolutionM - Mitsubishi Lancer and Lancer Evolution Community (https://www.evolutionm.net/forums/)
-   ECU Flash (https://www.evolutionm.net/forums/ecu-flash-179/)
-   -   Need advice on tuning style (https://www.evolutionm.net/forums/ecu-flash/399123-need-advice-tuning-style.html)

raasfaas Feb 9, 2009 08:02 PM

Need advice on tuning style
 
Ok i'm fairly new to tuning so I just wanted to get some feedback on how others tune their maps. I know there are many different styles to tuning and there is no 1 right way. So for instance, after you make a 3rd gear pull logging with evoscan, what review option do you use to decide where to make changes; the map tracer, graphing log data, or do you just pull it up in excel.
Also, how does the interpolation function of ecuflash work, when you make an adjustment do you interpolate from that point to max load for that given rpm range?
Sorry for the noob questions but I just want to be more efficient in tuning.

b16a95eg Feb 9, 2009 08:12 PM

dont worry, i've been reading up on this stuff for quite sometime, and i still dont know either :-P i keep learning every day, i finally got to start datalogging and what not yesterday so i still need to learn the stuff, so i need some answers too lol

Thoe99 Feb 9, 2009 08:16 PM

The best place to start is here:
https://www.evolutionm.net/forums/ec...-tune-evo.html

raasfaas Feb 9, 2009 08:25 PM

Yeah I fully went through that thread...I'm getting the results that I want, I just want to know if there is a more efficient way of doing it (smoothing out the tables instead of just changing the specific logged rpm/load values etc.)

dude Feb 9, 2009 08:33 PM

I actually use several tools posted on here. The first is the maptracer excel spreadsheet. I do a pull, import, clean and chart the pull. I tune for AFR first, then boost, then timing. Then I use the map smoothing tool to make sure I don't have any large transitions from one cell to another. So far it's worked out just peachy.

Adam@MMS Feb 9, 2009 08:38 PM

I've never used the interpolation function but I usually hit the cell(s) I want and then go the same number or leaner to the left and then the same number or richer to the right. The problem is you can make the table look nice and smooth except you don't know what the car is going to do in those cells unless you actually hit them. If the table was perfect you could just type in 11.5 for the wot cells and be done with it.... one day I'm sure. Don't make yourself too crazy about it. My 2c.

raasfaas Feb 9, 2009 09:04 PM


Originally Posted by dude (Post 6674042)
I actually use several tools posted on here. The first is the maptracer excel spreadsheet. I do a pull, import, clean and chart the pull. I tune for AFR first, then boost, then timing. Then I use the map smoothing tool to make sure I don't have any large transitions from one cell to another. So far it's worked out just peachy.

Hmmm, sounds interesting. Would you be able to point me to where I can get this spreadsheet. Also, which program do you use for that map smoothing function?
Currently I just use evoscan's maptracer and manually smooth my tables.

Mojo: I make my adjustments in about the same manner.
Also, I would love to see that day :lol:

The only thing that sometimes takes a little while for me to figure out is which cells to alter..for example, lets say at 5700rpm 289load I have 3 counts of knock...which cells would I need to alter 5500rpm or 6000rpm along with 280load or 300load?
I guess what I'm trying to find out is, does a specific rpm point (and load too) cover from that point up to 1- the next point? (does 4000rpm cover everything up to 4499rpm)

steadly2004 Feb 9, 2009 11:29 PM


Originally Posted by raasfaas (Post 6674123)
Mojo: I make my adjustments in about the same manner.
Also, I would love to see that day :lol:

There are aftermarket ECU's out there that use closed loop WOT wideband driven tuning. I think some turbo buick aftermarket stuff does it. Its not out of reach forever. Just needs enough processing power and fast enough wideband signal to make quick changes.


I actually had an e-manag ultimate that was supposed to use the "target" a/f map to adjust the "air-hz compensation table" to on the fly and make the the wideband input signal equal to the target a/f map, but it never worked right. It was a great theory, but didn't come together well.

raasfaas Feb 10, 2009 05:16 AM

I hope that one of the disassembly guru's will figure out something like that for the stock ecu...

steadly2004 Feb 10, 2009 09:22 AM


Originally Posted by raasfaas (Post 6674822)
I hope that one of the disassembly guru's will figure out something like that for the stock ecu...

I wish, but its not likely. The only thing already written is the idle trim adjustments and 02 trim, which adjusts very slowly compared to how fast we would need the adustments. The best bet would be to find an afftermarket ECU that can do it.

guess the haltech platinum 1000 and 2000 can do it
http://www.racetep.com/haltech.html

The Platinum Sport 1000
The Platinum Sport 1000 The Haltech Platinum Sport 1000 ECU is an ECU capable of controlling sequential fuel and ignition on 4 cylinder and 2 rotor applications (semi-sequential on 6 and 8 cylinders).With 8 Channels capable of controlling injection and ignition duties, the Platinum Sport 1000 supports modern engines with multi-coil ignition systems, as well as conventional distributor systems.

Features:

- Soft cut rev limiter
- Closed loop boost control
- Stepper motor and closed loop idle control
- Tuning via TPS with manifold pressure correction
- User-definable mapping points
- Anti-lag launch control with rotational idle
- Wideband closed loop 02 Control
- Sensor calibration capability
- 5 additional user-definable inputs
- 4 additional user-definable outputs
- Onboard Datalogging
- Numerous correction maps
- Windows Software
Although I don't think they are saying it would work WOT

Mr. Evo IX Feb 10, 2009 09:31 AM

delete

Mr. Evo IX Feb 10, 2009 09:38 AM


Originally Posted by raasfaas (Post 6674123)
The only thing that sometimes takes a little while for me to figure out is which cells to alter..for example, lets say at 5700rpm 289load I have 3 counts of knock...which cells would I need to alter 5500rpm or 6000rpm along with 280load or 300load?
I guess what I'm trying to find out is, does a specific rpm point (and load too) cover from that point up to 1- the next point? (does 4000rpm cover everything up to 4499rpm)

The rpm and load break points are centered at the value, and will affect regions before and after. If you have rpm points of 3500, 4000, and 4500 then changes at 4000 would affect 3750-4250. However the conditions your logging may be due to a condition that actually happened at 3500 and your seeing the effects at 3750. You'll pick this up with experience and trial and error.

The Evoscan map tracer works pretty good, I use that and I manually review the logs as well. I used map smoothing from 1 point to the next point (normally horizontal or vertical - not both). If you want to smooth a large area do it one line at a time and you'll probably get better results than grabbing several lines at a time.

raasfaas Feb 10, 2009 10:00 AM

Thanks Mr. Evo, that cleared up my misconception of values in between RPM points.
True, with time/trial/error I will learn, this forum has done a great deal so far...very helpful/informative.

Mr. Evo IX Feb 10, 2009 10:58 AM

The values between break points are interpolated by the ECU. This might be a better way of saying it. Example, if you have a 6 at 4000 and 8 at 4500, then the ECU will interpolate 4250 as 7.

raasfaas Feb 10, 2009 11:26 AM

ok got it...
also, is there a correct sequence for tuning. Based on the thread https://www.evolutionm.net/forums/ec...-tune-evo.html, it looks like the OP does timing first, then boost, then AFR...earlier in this thread "dude" says he does AFR, boost, then timing.
Personally I do boost, then timing, then AFR since ultimately my boost level will determine what load cells I hit so I know which timing cells need adjusting at that boost level. Also, altering the timing cells will also affect the AFR in that cell, so that why I do AFR last. Am I on the right path with this train of thought?


All times are GMT -7. The time now is 09:38 PM.


© 2024 MH Sub I, LLC dba Internet Brands