Notices
ECU Flash

94170008 ROM 2 byte load mod for dummies (like me :)

Thread Tools
 
Search this Thread
 
Old May 23, 2007, 07:01 AM
  #16  
Evolved Member
Thread Starter
iTrader: (5)
 
A418t81's Avatar
 
Join Date: Nov 2003
Location: Birmingham, Al
Posts: 849
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by C6C6CH3vo
JorgeT,

I just got the 2byte mod loaded on to my ROM from ECUFLASH and works with evoscan, show me what you did to get it logging on LW (without map,psibar jdm thing) and I'll see if mine does this as well.

???

What load is that occurring at?
I also use the latest version of Evoscan and have no issues with a spike. I remember reading somewhere about there not being an "answer" or something like that everytime and it could cause some weirdness. I believe that both MJ and evo4mad addressed this issue in their loggers, LW probably has not since we are having to add the 2byte math in ourselves. Try using mitsulogger.
Old May 23, 2007, 07:13 AM
  #17  
Evolved Member
iTrader: (18)
 
Jorge T's Avatar
 
Join Date: May 2004
Location: Houston, TX
Posts: 1,494
Likes: 0
Received 1 Like on 1 Post
I'm using the LW OpenPort 1.3 pluging from LW2 (it works fine with LW3) and added the xml lines to the protocols.xml [zip file] and use the formula included in zip file to caculate ECU_load while reviewing the data.

Run openport 1.3 / select MUT III, add channels loadmsb and loadlsb + timing,knock sum etc.(you'll have to add one at a time)/ hit start to log. Then run LW2 or LW3/new realtimelog.

[The protocols zip has the xmls lines to get EVOscan calc load method or MJ Mitsulogger calc load method]
Attached Files
File Type: zip
protocols.zip (1.9 KB, 38 views)
Old May 23, 2007, 12:11 PM
  #18  
Evolved Member
iTrader: (11)
 
mchuang's Avatar
 
Join Date: Sep 2005
Location: h town
Posts: 2,180
Likes: 0
Received 1 Like on 1 Post
Originally Posted by A418t81
I also use the latest version of Evoscan and have no issues with a spike. I remember reading somewhere about there not being an "answer" or something like that everytime and it could cause some weirdness. I believe that both MJ and evo4mad addressed this issue in their loggers, LW probably has not since we are having to add the 2byte math in ourselves. Try using mitsulogger.
The spike does not happen all the time. In fact out of the 20 or 30 times I have used .99 evoscan it has happened about 3 times.
Old May 23, 2007, 03:23 PM
  #19  
Evolved Member
iTrader: (18)
 
Jorge T's Avatar
 
Join Date: May 2004
Location: Houston, TX
Posts: 1,494
Likes: 0
Received 1 Like on 1 Post
Originally Posted by A418t81
I also use the latest version of Evoscan and have no issues with a spike. I remember reading somewhere about there not being an "answer" or something like that everytime and it could cause some weirdness. I believe that both MJ and evo4mad addressed this issue in their loggers, LW probably has not since we are having to add the 2byte math in ourselves. Try using mitsulogger.
I can't seem to get EVOscan log my LC-1, it may be the LMA-3 serial chain interfering or a laptop issue 'cause the xede LW pluging does not work nor commslog6. I can only log xede throut XMAP 3.2.17 . I'll give Mitsulogger a shot...but I hate to loose my LMA-3 channels and LW sample rate.
Old May 23, 2007, 06:04 PM
  #20  
Evolved Member
iTrader: (6)
 
nj1266's Avatar
 
Join Date: Nov 2004
Location: USA
Posts: 3,254
Likes: 0
Received 13 Likes on 3 Posts
Originally Posted by Jorge T
2 byte load vs. Calc load w/ MJ formula. In my case very close is it worth it for me, the spike really bugs me, what is it doing to the ECU?
What is the MJ formula for Calc load? Care to share it?
Old May 23, 2007, 07:39 PM
  #21  
Evolved Member
iTrader: (18)
 
Jorge T's Avatar
 
Join Date: May 2004
Location: Houston, TX
Posts: 1,494
Likes: 0
Received 1 Like on 1 Post
I got these off the innovate forum openport pluging thread

MJ mitsulogger method

?Injector Scaling = inj_scale[513]
?Injector Latency = inj_latency[0.432]
?i Name of Injector Pulse Width channel = inj_pulse[Fuel_Injector_Pulse_Width]
?i Name of AFR MAP channel = afrmap[AFR_MAP]

# engine load
calcload = 5 * inj_scale * (inj_pulse - inj_latency) / afrmap
MC(ECU_Load;%;0;350) = calcload

Evoscan method

?Injector Scaling = inj_scale[513]
?i Name of Battery Voltage channel = battery[Battery_Voltage]
?i Name of Injector Pulse Width channel = inj_pulse[Fuel_Injector_Pulse_Width]
?i Name of AFR MAP channel = afrmap[AFR_MAP]

# injector latency
inj_latency = (-0.1026*battery)+1.8741

# engine load
calcload = 5 * inj_scale * (inj_pulse - inj_latency) / afrmap
MC(LoadCalc;%;0;500) = calcload

Last edited by Jorge T; May 23, 2007 at 07:42 PM.
Old May 23, 2007, 07:56 PM
  #22  
Evolving Member
iTrader: (1)
 
FastEvoFL's Avatar
 
Join Date: Dec 2005
Posts: 115
Likes: 0
Received 0 Likes on 0 Posts
So what exactly are you storing in the array with the calcload formula?
Old May 24, 2007, 02:43 AM
  #23  
Evolving Member
 
JoeBee's Avatar
 
Join Date: Sep 2006
Location: Germany
Posts: 152
Likes: 0
Received 0 Likes on 0 Posts
My theory of the spikes:
https://www.evolutionm.net/forums/sh...=234060&page=8
Post 108 and 112

I think this is a kind of transition error bcause we log the bytes one after another. So this value can go wrong when the low byte is near 255.
Old May 24, 2007, 04:53 AM
  #24  
Evolved Member
iTrader: (18)
 
Jorge T's Avatar
 
Join Date: May 2004
Location: Houston, TX
Posts: 1,494
Likes: 0
Received 1 Like on 1 Post
So what exactly are you storing in the array with the calcload formula?
Not sure if I follow your question. LW uses log channels and/or entered values ie [513] to create a new calculated calcload channel in data reveiw mode.





Originally Posted by JoeBee
My theory of the spikes:
https://www.evolutionm.net/forums/sh...=234060&page=8
Post 108 and 112

I think this is a kind of transition error bcause we log the bytes one after another. So this value can go wrong when the low byte is near 255.
Thanks! In LW openport plugin I can specify the log order of each channel, will separating the order of the hi and low 2byte help prevent this?

Last edited by Jorge T; May 24, 2007 at 05:00 AM.
Old May 24, 2007, 05:22 AM
  #25  
Evolving Member
 
JoeBee's Avatar
 
Join Date: Sep 2006
Location: Germany
Posts: 152
Likes: 0
Received 0 Likes on 0 Posts
I think it is a good idea to log time critical values as near as possible.
For example: load lowByte, load HighByte, RPM, knock, AFRMAP and Ignition inline when you want to tune the maps. AFR should also be as near as possible, but I think in LW this is not possible.
After these values you can add additionally parameters such as octane nr., ...

But, not sure if this is such important.
Old May 24, 2007, 06:39 AM
  #26  
Evolved Member
iTrader: (11)
 
mchuang's Avatar
 
Join Date: Sep 2005
Location: h town
Posts: 2,180
Likes: 0
Received 1 Like on 1 Post
Originally Posted by JoeBee
I think it is a good idea to log time critical values as near as possible.
For example: load lowByte, load HighByte, RPM, knock, AFRMAP and Ignition inline when you want to tune the maps. AFR should also be as near as possible, but I think in LW this is not possible.
After these values you can add additionally parameters such as octane nr., ...

But, not sure if this is such important.
What can logworks not log? All that you have mentioned can be logged by logworks. It can log anything you can log thru mut protocol via the openport plugin and all add on sensors thru the mts chain using other innovate products. You can also add math calculations such as this

?Target AFR = tafr[11.3] <--11.3 can be changed to your target afr

MC(Tafr;AFR;7.4;22.4) = tafr

and it will put a straight line at 11.3 afr and when your actual afr traces across the log you will see how close are far you are away from target. If I were at my comp I could actually post a demonstration but I am not. And also logworks3 is great. It is currently in beta, but still so far working fine with VISTA woot!!
Old May 24, 2007, 07:32 AM
  #27  
Evolving Member
 
JoeBee's Avatar
 
Join Date: Sep 2006
Location: Germany
Posts: 152
Likes: 0
Received 0 Likes on 0 Posts
Sorry, misunderstandig.
(That happens when a german speaks english )
I mean you cannot arrange the AFR with the other openecu channels.
But I also think that dosent matter, because the AFR came from another source so that there is no time critical problems.
Old May 24, 2007, 07:34 AM
  #28  
Evolved Member
iTrader: (9)
 
C6C6CH3vo's Avatar
 
Join Date: Feb 2005
Location: sc
Posts: 4,223
Likes: 0
Received 4 Likes on 4 Posts
Originally Posted by JoeBee
But, not sure if this is such important.
You may be right, but more so I think as not being important.

What LW does do which I haven't seen with any other logger is log each parameter on an indivividual axis, this makes life so much easier for a human to interprete and analyze data.

All the other loggers I see will graphically display on same axis so you have a steep airflow line but the AFRs appear a solid straight line from 10.0 to 14.7.

Joe Bee, I guarantee if you try LW for a week you will continue to use it

As far as spikes are concerned I wish I could find the source of my dip in AFR at peak load/tourque - no matter lean spool enabled/disenabled, value in fuel map, maf smoothing, scaling, BE, and so forth - I allways get a dip in AFR at peak tourque/load.

All I can think of is it's mechanical, Perhaps this is the max pressure the FP regulator see's and therefore most fuel pressure

Any idears?

Last edited by C6C6CH3vo; May 24, 2007 at 07:44 AM.
Old May 24, 2007, 08:49 AM
  #29  
Evolved Member
iTrader: (11)
 
mchuang's Avatar
 
Join Date: Sep 2005
Location: h town
Posts: 2,180
Likes: 0
Received 1 Like on 1 Post
Originally Posted by JoeBee
Sorry, misunderstandig.
(That happens when a german speaks english )
I mean you cannot arrange the AFR with the other openecu channels.
But I also think that dosent matter, because the AFR came from another source so that there is no time critical problems.
Its ok hahaha, I guess I should have been more open minded when I saw your location lol. But also C6C6CH3vo is right, logworks is very user friendly and as of right now the gauges in logworks 3 are 2nd to none. They respond pretty fast and also show you peak spots and mark them automatically on the gauge.
Old May 24, 2007, 12:48 PM
  #30  
Newbie
 
klatinn's Avatar
 
Join Date: Jan 2007
Posts: 28
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by mchuang
What can logworks not log? All that you have mentioned can be logged by logworks. It can log anything you can log thru mut protocol via the openport plugin and all add on sensors thru the mts chain using other innovate products. You can also add math calculations such as this

?Target AFR = tafr[11.3] <--11.3 can be changed to your target afr

MC(Tafr;AFR;7.4;22.4) = tafr

and it will put a straight line at 11.3 afr and when your actual afr traces across the log you will see how close are far you are away from target. If I were at my comp I could actually post a demonstration but I am not. And also logworks3 is great. It is currently in beta, but still so far working fine with VISTA woot!!
Actually, LW3 has a better method than that.
Go to Tools->Marker Lines->Channel name in the log window. You will see 2 movable marker lines. Each channel has its own set but only the set of one channel is shown at a time. But their position for each channel is preserved. You can move them to whatever point in the scale by grabbing them with the triangular handle on the right end of the line and moving them up and down.
A shortcut to invoke the marker lines is to click on the channels current value in the measurement pane to the right of the graph pane. Clicking on the empty area below the measurements in the pane switches marker lines off.

As for scales. Using the magnifier tool on the scale of a trace itself allows you to "magnify" the scale of a trace to your liking to see details. You can then move the channel trace or scale up and down with the hand tool.

Another LW3 trick:
To get your log data into for example excel, select the area in the log you want to copy. Select Edit->Copy in the log window. Then open Excel (or whatever, including a text editor) and click on the upper left cell you want the log data to be in. Select Edit->Paste and you are done.


- Klaus

Last edited by klatinn; May 24, 2007 at 01:02 PM.


Quick Reply: 94170008 ROM 2 byte load mod for dummies (like me :)



All times are GMT -7. The time now is 05:31 PM.