Reading MLV Data Logs

Tuning concepts, methods, tips etc.

Moderators: jsmcortina, muythaibxr

Rick Finsta
Master MS/Extra'er
Posts: 548
Joined: Fri Dec 10, 2010 4:41 pm

Re: Reading MLV Data Logs

Post by Rick Finsta »

The histograms are also super useful for looking at datalogs where you don't have the tune file - by changing your z value you can easily see if things like ASE, AE, WUE, EGO Correction, etc. are active and what they are doing.
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

On logs that have knock retard, I like to created a filter that takes out all the records that have 0 knock retard and then plot in scatter plots or in histogram view the knock records that remain. It gets real apparent where the problem timing areas are.

I plot RPM, MAP and Knock Retard

http://www.nbs-stl.com/HarleyTuning/98f ... togram.png

Also see http://www.nbs-stl.com/HarleyTuning/Harley_Knock.png This happens to be off a Harley system but MS logs would look very similar. Only the field names change

Andy
dee808
Helpful MS/Extra'er
Posts: 37
Joined: Thu Feb 23, 2017 3:33 pm
Location: 808 HI \[]]]/ ¤shaka¤

Re: Reading MLV Data Logs

Post by dee808 »

Op, please keep posting! This is exactly what i wanted . A way to diag with logs and what to look for. Keep talking!
Efi Tuning Noob, so please go easy on me.
AE92 GZE'd 85 AE86 3dr(blue "cancer" pill)
ATS Racing 2.1 stroker GT3071R 92 MR2t t-top (red pill) DIYPNP w/5sgte in the works
"Nobody listens to Turtle"
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

What is it you want to see? Motors with issues? Well tuned motors? People posting logs?
Marek
Master MS/Extra'er
Posts: 573
Joined: Sun May 10, 2009 1:20 am
Location: Guildford UK

Re: Reading MLV Data Logs

Post by Marek »

What is most useful (and completely lacking in the manuals) would be a worked example of an engine with a particular issue with the resultant plot/graph and then then a back-to-back comparison with the issue solved (and what was changed) and the resultant plot/graph.

kind regards
Marek
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

I have some good logs of motors with issues. Funny thing is, I seldom hear the final solution or see the later logs once the owner fixes the issue. Huge amounts of AE aggravated by TPS noise is common. MAP noise causing wild swings in PW and DC with no matching changes in TPS and RPM come up quite a bit. Battery voltage coming from a sick voltage regulator is far from rare. Let me see what I can dig up...

Thanks for reading and have fun tuning.

Andy
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

Have you seen this thread?

http://www.msextra.com/forums/viewtopic ... 34&t=63074

Most of that is fair game here.

Andy
ChevelleFan
Master MS/Extra'er
Posts: 786
Joined: Sun May 02, 2004 5:55 pm
Location: Cleveland, Ohio
Contact:

Re: Reading MLV Data Logs

Post by ChevelleFan »

Andy,

There's another thread about measuring Injector Dead time from a log, http://www.msextra.com/forums/viewtopic ... 32&t=64972 . If you have any bad vs. good examples of how those scatter plots might look if the dead time was too short to too long.

This is a really good thread, btw. Thanks!

-Dave
'70 Chevelle ZZ383/700r4/3.73 MS3X, 36-1, LS Coils, Holley Stealthram -- success story
'92 Saturn 1.9 DOHC/5sp, MS3, Champcar racer success story
'79 Malibu DartSHP400/80mm/PG/3.55, MS3X/36-1/LS Coils -- success story
Porsche 944 S2 3.0 -- MS3X/36-1/LS Coils, 1:44s Mid-Ohio, Pro Course
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

Adding custom fields to the mix

Here is a fun experiment. If you have a registered version of MLV or MLVHD, create a new field.

Name the field for example Throttling

Then define it as

[TPS]<0?0:10*sqrt([TPS])

What this does is make the TPS a little easier to see what is happening to the MAP as the throttle is cracked open. Small angle changes at low throttle make beg changes in MAP

Here is a screen shot of what you get. This happens to be data coming of my ITB race motor.

http://www.nbs-stl.com/CRX/CRX%20Throttling%202.png

Here is how the math works out

TPS___Throttling
_0________0
_4_______20
_9_______30
16_______40
25_______50
36_______60
49_______70
64_______80
100_____100

Andy
dee808
Helpful MS/Extra'er
Posts: 37
Joined: Thu Feb 23, 2017 3:33 pm
Location: 808 HI \[]]]/ ¤shaka¤

Re: Reading MLV Data Logs

Post by dee808 »

This should be a sticky! title saying something like diagnosing/troubleshooting thru datalogs.
I was wondering if there were anywhere that can point me in the direction of using datalogs to diag certain issues, and what to look for and most importantly why, so everyone interested can actually understand why the poster is using stated data to do so ( if not obvious)
This thread has the potential to be chocked full of useful info. Thank you OP!
Efi Tuning Noob, so please go easy on me.
AE92 GZE'd 85 AE86 3dr(blue "cancer" pill)
ATS Racing 2.1 stroker GT3071R 92 MR2t t-top (red pill) DIYPNP w/5sgte in the works
"Nobody listens to Turtle"
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

Here is a motor that has what appears to be noise on the MAP sensor. If you look at the bottom of the screen, you can see at the screen shot zoom time goes from 545 sec to 550 or is only about 5 sec total at this zoom setting.

http://www.nbs-stl.com/tuning/MAP%20Issues.png

The first thing I notice is check out the static in the MAP signal. Notice that the other sensors look great so it is unlikely a serious grounding issue. Notice the MAPxRPM vs DC and the MAP vs PW. The tune is clearly being effected. In this case, I advised the owner is that the issues was either some sort of mechanical issue like an intake valve problem, cracked head... Something that was really making the MAP do that. Or was this some sort of electrical noise.... I simply don't know but what I do know is that there is something screwed up. We must find the source of the issue.

Andy
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

Here is an interesting case. This motor happens to be a high dollar ITB V8 Racecar. This is early in the tuning process and he is starting to roll a little into the power.

http://www.nbs-stl.com/tuning/ITB%20V8% ... 0Power.png

http://www.nbs-stl.com/tuning/ITB%20V8% ... catter.png

My first guess is that he is going so rich up top, the motor is going into a miss and causing the 02s to read lean. Look at the MAPxRPM vs DC.

Keep in mind that is is possible we are chasing an ignition problem that is giving the same look to the logs.

Andy
Rick Finsta
Master MS/Extra'er
Posts: 548
Joined: Fri Dec 10, 2010 4:41 pm

Re: Reading MLV Data Logs

Post by Rick Finsta »

Do you have fuel pressure data? That's what mine looked like when I needed to turn up the duty cycle on the fuel pump control...
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

Good catch. Fuel pressure (or lack of fuel pressure at WOT) is well worth looking into.
rukavina
Helpful MS/Extra'er
Posts: 58
Joined: Wed Aug 20, 2014 9:42 pm

Re: Reading MLV Data Logs

Post by rukavina »

Is there a way to take a long datalog and edit out all but a small piece so it is small enough to post?
4wheel drive 454 vortec on ms2 w/gpio for 4l80e Trans control
Marek
Master MS/Extra'er
Posts: 573
Joined: Sun May 10, 2009 1:20 am
Location: Guildford UK

Re: Reading MLV Data Logs

Post by Marek »

Use Excel.

Edit out the unwanted lines, keep the header lines and save down as a .csv format.

kind regards
Marek
Rick Finsta
Master MS/Extra'er
Posts: 548
Joined: Fri Dec 10, 2010 4:41 pm

Re: Reading MLV Data Logs

Post by Rick Finsta »

rukavina wrote:Is there a way to take a long datalog and edit out all but a small piece so it is small enough to post?
Yes, you can highlight areas and keep or delete them in MLV.
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

I prefer HUGE data logs that are hard to post here. The bigger the log, the better. 20 to 40 meg files are common if you really want to see what is going on. There are all sorts of file sharing sites, Google Drive, DropBox...

This will give you an idea what scatter plots look like with lots of data to pull together into one view.

http://www.nbs-stl.com/MLVDemo/020%20Sc ... ActAfr.png

Andy
NigeT
Master MS/Extra'er
Posts: 419
Joined: Wed Apr 08, 2015 1:14 pm

Re: Reading MLV Data Logs

Post by NigeT »

Is there a way to save the sets of graphed metrics. I've found the diagnostic set you suggested early on in this thread very useful but I appear to have lost the set of metrics I was using to try to work out what on earth the boost control is doing. Stupidly I assumed that exporting settings might save fundamentals like this. Thanks.
whittlebeast
Super MS/Extra'er
Posts: 2221
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis
Contact:

Re: Reading MLV Data Logs

Post by whittlebeast »

I was looking at ways to test a motors throttle response and relate that to AE settings. This log was taken with the new MS3 Ultimate logging at about 160 samples per sec.

I created two new fields

PW Base

and defined it as

100*[Field.pulseWidth]/[Fuel: Accel enrich]

and then

Duty Cycle Base

and defined it as

100*[Duty Cycle1]/[Fuel: Accel enrich]

Here is the plot that I got

http://www.nbs-stl.com/tuning/Throttle% ... esting.PNG

The gap between the red and white traces is the change in PW and DC due to the AE settings. Notice that the RPM per Sec peaks at 10322 RPM Per Sec and the RPM climbed 1573 RPM in .248 sec as seen in the white box at the bottom of the screen.

Have fun tuning

Andy
Post Reply