JNorris

iTrader: (4)

I recently bought an Innovate Motorsports LM-1 Wide band O2 sensor. The only problem is that I had to log data in the LM-1 as well as in EFILive. So I decided to use an existing sensor input into the PCM to output the LM-1 date into.
Below is a link to a Word doc that explains what I did to achieve this.

Link to the document


Below is the text from that document minus the pictures.

Below is how I integrated the Innovate Motorsports LM-1 wide band O2 meter or any wide band that has an output range of 0V-5V into the PCM of a 98 Z28 using EFILive.
I am sure that this method will work for any 98 - 2002 F or Y body car. You will need to consult your service manual to confirm this.
If you have any questions or corrections email me at jnorris404@aol.com

Thanks
John Norris

Preface

The goal is to pipe the analog output (voltage) from the LM-1 into the stock PCM and log it with EFILive as you are logging other data as well. I used the 0V-5V input into the PCM that is used to monitor the A/C Pressure.


LM-1

I used the analog output port number 2 as configured from Innovate Motorsports.
1V=10:1AFR 2V=20:1AFR.
The analog interface cable I used is part number 42-2561 from Radio Shack. You will need to cut the ¼ inch female end off and strip back insulation to expose the wires and copper shield. The shield is ground, the red wire will be output 2 and the white wire is output 1.
Ultimately you will be using the ground and the red wire to interface the LM-1 with the stock PCM.

PCM

Under the hood on the passenger’s side towards the back you will the pressure sensor for the A/C system. See picture below.
There is 3-wire plug attached to the sensor. On a 98 F body the red wire is the input into the PCM, the gray wire is the 5V reference voltage supply, and the black wire is the ground.
I ran the analog output from the LM-1 into this plug using wiretaps to connect to the red wire from the LM-1 to the red wire going the PCM and the black wires from the LM-1 to the black wire going to the PCM. I passed the analog output cable from the LM-1 through the rubber grommet located below and the left of the PCM

The A/C pressure plug must be unplugged from the sensor while you are using it to receive data from the LM-1. If it is not damage to the LM-1 or your PCM could result. Additionally the A/C will not function when the plug is not connected to the sensor.

EFILive

With some help from Paul (EFILive) I created the formula below which is based on the GM.ACP PID.

(RAW({GM.ACP})/5.1)-correction factor

The RAW function tells EFILive not to perform any calculations or conversions to the data for the specified PID.

GM.ACP – A/C Pressure has a range of 0-255 in its RAW state.

Dividing the GM.ACP PID by 5.1 is for the following reason.
The RAW voltage range from the PCM for the GM.ACP PID is 0-255 so 255/51 = 5. This converts the raw data to a 0V-5V range
The voltage range the LM-1 is going to output is 1V-2V so 1.47 will really be an AFR of 14.7:1. Remember on the LM-1 that 1V=10:1AFR 2V=20:1AFR
You would need to multiply the data by 10 in order to see the 14.7 value.
So 51 divided by 10 equals 5.1 the completed formula would be written as
GM.ACP / 51 * 10 which can be simplified and rewritten as GM.ACP / 5.1.

In order to log the wide band O2 data in EFILive you must select the GM.ACP PID as well as the new wide band PID you are creating.

Correction factor

For whatever reason the analog output from the LM-1 is not the same as the A/F value as logged by the LM-1. In order to make the values appear correctly in EFILive You will need to apply a correction factor.
What I did was to let the car idle for about 15 minutes and recording the A/F with the LM-1 as well as EFILive with no correction factor applied to the formula. Afterwards I averaged the A/F numbers from the LM-1and EFILive logs. The difference is the correction factor. I also did some WOT logging and the correction factor was inline with the correction factor as determined from the idle test.

When I measured the voltage on the red wire going to the PCM from the A/C Pressure with nothing else attached or plugged in there was a voltage floor of .02V. My correction factor was -.425. The formula I put into CALC.PIDS.TXT file EFILive was (RAW({GM.ACP})/5.1)-0.425. This file is located in the following directory “C:\Program Files\EFILive\ScanToolV6\Configuration”
Also when you re-calibrate the wide band sensor you will also have to confirm your correction factor is still correct. The first correction factor I used was 0.522. With a good correction factor in place when I compare the logged values from the LM-1 to the values logged in EFILive there is less that .1 difference.

The LM-1 will output the AFR even if you are not recording the data. So after you have determined the correction factor you will no longer have turn on/off the record button to get the data as long as you are recording with EFILive.


Below are the formulas in my CALC_PIDS.TXT file.
# Calculated
# ----------------
#Units Low High Fmt Expression
#------------ ------------- ------------- ---- --------------------------------------------------------------
*CLC-00-1000
V 10.0 20.0 .3 (RAW({GM.ACP})/5.1)-0.425
# ================================================== ============================
*PRN - Parameter Reference Numbers
# --------------------------------
#
# Calculated
# ----------------
#Code PRN SLOT Units System Description
#------------------------- ---- ------------ ---------------- ---------------- ------------------------------------------
CALC.WBO2 F600 CLC-00-1000 V Fuel "Wide Band O2 Sensor"

Nickn20

very cool stuff

S_J_H

This is a very good idea! It has been done before I think.
I am using v5.26EFI. To convert the PID to voltage all you have to do is select edit PID and use .01976 for a factor and -.02 for an offset.
I used a couple 1.5 cell batteries to input to the pcm and verfied what EFI showed with a DMM. It was dead on with the DMM at 1.62 and 3.22 volts.
So I'll go ahead and hook it all up.
I am using a Techedge wideband with the NTK L1H1 sensor.
Stoich is 2.5 volts. The curve is not linear with voltage so I can't use a simple formula like used. But no matter all I need is the voltage logged as I have the table.
Thanks for the idea!
Steve

deezel

Nice write-up! Thanks for sharing!

Mallett396

Now Deezel, you need to try this first, I'm busy at work

JNorris

iTrader: (4)

Thanks

John

XLR8NSS

iTrader: (4)

Steve - I tried fitting the Vout and Petrol stoich numbers to an equation and couldn't get an equation accurate enough. Looks like logging voltages for the techedge setup is the only way to go.

One other thing, which wire are you using to get the Vout signal from? It looks like pin 7 is the one I want to tap off of. I wonder if it'll affect the reading on my 5301 display to tap the signal. I'm going to hook this up tomorrow and find out.

PS - I have the same techedge/sensor setup as you.

deezel

Can you post a copy of the curve/table?
Did you try an exponential fit? I don't know if EFILive can do logarithms, but I think it can do exponents (x^y).

S_J_H

XLR8NSS, I have it all hooked up. Works great! It does not effect the displays output. Just use pin 7 which is the white wire for Vout and the copper strand for the ground for the pcm hookup.
At a logged 2.5 volts I see exactly 14.7-1 on the display.
I just keep a printout of the table handy to check A-F ratio to logged voltage.

deezel,
here is the table- http://techedge.com.au/vehicle/wbo2/wblambda.htm

Steve

XLR8NSS

iTrader: (4)

Deezel - Yep, tried a exponential fit and even with a correlation coeficient of 99+% the data just didn't match up very well. I emailed Peter @ Techedge and he said there is no equation and that the conversion is done through a lookup table.

Steve - If you want to try and log stoich values "directly" take a look at what pin 4 on the Techedge box outputs(Vlin). Peter pointed that out also in his reply. Here is part of his reply.

That looks like an easy enough equation for EFILive to handle, just have to make sure the linear output on pin 4 matches up to the real Vout data. Maybe pick a certain routine to log and log both ways enough times to get a meaningful average run for both. Then check them both against each other to see if the linear output matches up.

Sunset01

I did the same with my LM-1. Thanks for the info. I was heading toward the rear O2 harness instead of the AC high pressure switch, but 66impala has had problems with that, Chris Benight has supposedly made that work, but I haven't gotten his attention. Two points:

1) A filter is recommended before sending to ECM by Innovate: 10K Ohm resistor on the signal line and after than a 0.022 uF capacitor across the
far side of the resistor on the signal line and to the analog out ground. Radio shack 271-1335 and 272-1066 respectively. You only need 1 pc each, though those part #s are multi-packs.

2) The voltage differences that caused the original poster to use of -.425 is due to ground differences between ecm and ground used by the LM-1 unit. Programming the lm-1 to output flatline volatage for all af is a great way of testing the offset. I got:

lm-1 output at ecm
1.1 1.0
2.0 1.72
3.0 2.56
4.0 3.4
5.0 4.25


Dean

Sunset01

A couple of clarifications:

I used Innovate's analog out cable
I am monitoring the volatage directly in Ease, rather than raw
My voltage offset is possibly opposite direction ? But this is complicated by the formula used..

kp

iTrader: (34)

Couple comments on this since I seen it close to the top..

I did this a while back with a version 1.0 tech-edge usng the egr sensor and it worked fine. Luckily the new 2.0 TE has a linear 0-5v output and you can use the simple formula V*2+9 to read out directly in EFIlive. I used the standard EFIlive conversion factor in EFIlive but Steve's (S_J_H) seemed more accurate when testing at various voltages so I ended up using his.

I found with the EGR codes and EGR set to 'false' in LS1edit I cant use the egr sensor anymore (plus it would trigger a SES) the AC sensor works OK except it turns the cooling fans on a couple seconds after starting. I assume that doesnt happen to JNorris since his max votage is 2v and that isnt enough for the PCM to think the a/c high side pressure is way up and kick on the fans.. Not a big deal since I only log occasionally.

Either way it works well and keeps me from staring at the a/f display at 100mph+ which is a bit safer. Here is a sample export of an EFIlive file - it works great and compared to a dynojet wideband its pretty close..

JS

iTrader: (4)

Pretty kewl setup Jay...
I talked to EFIlive,they sell a kit to upgrade my AT so I can run WB w/there EFILIVE scan tool.It cost 340 though and I dont kow if I'm gonna spend the $$$.