Lambda delay

Moderator: Tom Notch

User avatar
panel
Posts: 3628
Joined: Sun Sep 24, 2000 2:01 am
Location: Victoria B.C. Canada
Contact:

Re: Lambda delay

Post by panel » Sun Sep 30, 2012 6:06 pm

Another approach from Monster:
How Do We Do It: The delay chart is in the advance settings tab of Auto Tune. A cross
reference between idle, low mid, wot - rpms and at low, mid, high - map readings. Datalog at a
few different rpms. You want a big injector pulse change so you can spot it. Use acceleration
enrichment. So drive around and cruise at different rpm's and crack the throttle. Now go back
and look through the datalog. Bring up pw1 and o2 #1 in the top chart and then rpm and map in
the 2nd chart. Look for the injector spike from the acceleration enrichment. Click on the graph
exactly at the pw upward peak. Write down the time at the bottom of the screen. Note map and
rpm reading to know what cell in the lambda chart to fill in with the new delay setting. Now click
on the graph at the o2 sensor downward (rich) peak. Write down the time. Subtract the time's
and this is the sensor delay. (it takes that many milli seconds for the sensor to read that pw.) The
4 digit time is in seconds with the 3 digits after the decimal point being milli second (one
thousands of a second). So a number like 2456.166 compared to 2456.300 has a delay of 134
ms. It will always be the last 3 digits you will be looking at. Subtract the lower number (pw) from
the higher number (o2). This is your lambda/ o2 sensor delay. Enter that delay in the lambda
delay chart. As mentioned, it is different for every install and varies with rpm. Now go back and
change the log chart to pw2 and o2 sensor 2 and do the second cylinder. Fill in the chart for VE
table 2 lambda delay.

Hints:

Feel free to change the chart positions as well. (change 3000 rpm to 4500 or 6000rpm) Use the
same info your datalog shows. You do not have to try and make your datalog fit the chart. Make
the chart fit your datalog / rpms you used while riding. If map was at 60kpa and rpm at 4000-
then change chart to 60 & 4000 and enter your delay. Don't keep datalogging to try and get the
map at 49 and rpm at 3000. Just use what you get for a mid point.
Use low range to get the high rpm points. (its easier to drive 50mph than 80mph with a laptop)
Low rpm bins are easy to get. Datalog the following:.
· startup- this will give the high map.
· Idle- this will give the mid map.
· Decel- this will give the low map.
Map readings occur at the following:
· wot- gives high map readings
· cruising or steady state rpm- gives mid map readings
· close throttle/ deceleration- gives low map readings.

Example:
With Megalogviewer, If you click view at the top, you can add or remove the tuning console (fuel table
area) and dashboard (bottom gauges) to have a bigger chart if you want.
http://WWW.MONSTER-PRODUCTS.COM

http://www.monster-products.com/downloa ... _Guide.pdf

User avatar
Piledriver
Moderator
Posts: 21787
Joined: Sat Feb 16, 2002 3:01 am
Location: Van Alstyne, Texas

Re: Lambda delay

Post by Piledriver » Sun Sep 30, 2012 11:17 pm

Ran some tests today with my new SLC-OEM setup, with the AFR target set at 11:1 at 2200/3900/5400.
(I'm set for "include AFR target" so it just does what it's told once the ve table is cal'd)
I think i need to yank a filter cap or such out of the SLC carrier board as the analog out is far more heavily damped than I'm used to... I can handle the truth. :lol:

My normal AFR target >2600 and <83kpa is 18:1, gave a reasonably clear step change.

I decided I didn't really care what the delay was at idle, and to concentrate on the normal load/rpm area, since we only have 3 rows and columns.
(30 kpa is probably pushing that definition, but fuel cut gives a nice clear signal)

I run EMW HX pipes with a Bursch 914 exhaust on a 1.8L T4 & MS3, O2 sensor just past second collector.
The total pipe distance to sensor are ~close to a std 1.5" T1 exhaust, tri-Y in this case.
YMMV.

I recall my MS2 setup having notably faster response but i never ran this exact test with it.

After studying the logs in MLV, here's what I ended up with:
piles-SLC-lambda-delay.jpeg
You do not have the required permissions to view the files attached to this post.
I, for one, regularly embrace our new robot overlords, as I am the guy fixing the robots...

User avatar
panel
Posts: 3628
Joined: Sun Sep 24, 2000 2:01 am
Location: Victoria B.C. Canada
Contact:

Re: Lambda delay

Post by panel » Mon Oct 01, 2012 8:32 am

It sounds like most people are using deceleration as their area to view in a data log. Is it easier to see this in a data log or something? Rather than accelerating through the gears?

User avatar
Piledriver
Moderator
Posts: 21787
Joined: Sat Feb 16, 2002 3:01 am
Location: Van Alstyne, Texas

Re: Lambda delay

Post by Piledriver » Mon Oct 01, 2012 1:33 pm

Its hard to get decel fuel cut while remaining at 98kPa map and 6K.
It's trivial to set your targets rich/lean on a few columns and trivial to even manually revert.

Fuel cut works ~ok for the bottom row (low kPa) only.
accel shot also jacks up your MAP, so temporarily setting rich columns in the AFR target is probably the easiest/best way... but i'm always open to better ideas.
I, for one, regularly embrace our new robot overlords, as I am the guy fixing the robots...

User avatar
panel
Posts: 3628
Joined: Sun Sep 24, 2000 2:01 am
Location: Victoria B.C. Canada
Contact:

Re: Lambda delay

Post by panel » Tue May 16, 2017 10:55 pm

I was doing some more reading on the subject and found this great info from the 'Tuning' section on the MS forums.

http://www.msextra.com/forums/viewtopic ... 0&start=60
Here is how I personally find lambda delay. I find a place in the log where the driver suddenly lifted out of the throttle and use the time measuring function built into MLV to see how long it to for the wideband to register a change/stabilize. In this motor, the delay is about .2 sec as seen at the bottom of the screen in the white box.

http://www.nbs-stl.com/tuning/Turbo%20D ... 0Delay.png

Notice how quickly the motor the motor goes from full throttle and hard accel to backing down as seen in the RPM per Sec trace. This ability is rare in a turbo car.

Have fun tuning

Andy
Last edited by panel on Wed May 17, 2017 11:22 pm, edited 1 time in total.

User avatar
kangaboy
Posts: 745
Joined: Wed Jan 22, 2003 3:01 am
Location: St. Louis, MO

Re: Lambda delay

Post by kangaboy » Wed May 17, 2017 12:00 pm

So in five years, has anyone else come up with new procedures to do this? I have been using Pile's chart for the majority of my tuning, but really should get one figured out for myself.
I like the method mentioned by Panel above.
'74 Standard- 1776T, MS3X3
'76 Westy- Stock 2.0L L-Jet

User avatar
Piledriver
Moderator
Posts: 21787
Joined: Sat Feb 16, 2002 3:01 am
Location: Van Alstyne, Texas

Re: Lambda delay

Post by Piledriver » Wed May 17, 2017 12:55 pm

A tool to automatically get the data and setup the table is planned for Tuner Studio.

Note that different design WB controllers have different delays, all else being the same.

I just moved around a "holes/bump" in my existing ve table with very narrow rpm bins.
The delay was different for lean than rich in places.
I, for one, regularly embrace our new robot overlords, as I am the guy fixing the robots...

Post Reply