Testing for Ground Offset: Innovate LC-1, XD-16 and Trinity

[TIGwelder]

EDIT: There's a second post in this thread that should really be read before this one. I posted this and then did some more reading/testing.

I also renamed the thread

================

I just installed the Innovate LC-1 and the XD-16 gauge in my 08 Charger SRT Super Bee

LC-1 analog2 out (brown wire) to Trinity analog1
Used the default values on the LC-1 of 7.35 - 22.39
These were used on the Trinity.
As per instructions all grounds are soldered together including Trinity analog ground.

My observations and tests:

I noticed there seemed to be quite a difference between the XD-16 gauge and the Trinity. The Trinity showing a lower/richer AFR reading. Yes, I was taking into account the delay between the XD and Trinity. I recorded several videos of the 2 gauges. During playback I hit play/pause and watched the values first on the XD and then a few frames later on the Trinity.

I don't know if this is showing a possible voltage drop/difference between the LC-1 and Trinity. The LC-1 and XD share the same power and grounding points.

During logging I want the Trinity to be logging the actual AFR i'm seeing on the XD-16, not a richer mixture. I'd rather be equal to or error towards lean..by default it was showing an error towards rich. Less voltage means a richer AFR

My fix to get the Trinity to equal the XD reading was to modify the values in the Trinity analog input.

Calculating percentages of voltage for AFR display

So we have the output from the LC-1 and the default values for the Trinity
0v = 7.35
5v = 22.39

7.35 - 22.39 is a range of 15.04

5v divided by 15.04 is 0.33244 ( 0.33244 volts is 1.0 AFR )
5v divided by 150.4 is 0.03324 ( 0.03324 volts is 0.1 AFR )

So 0.1 AFR is 0.03324 volts

To shift the Trinity's display/logged AFR reading up by 0.1 AFR you need to add 0.03324 to BOTH the values
For a 0.1 shift up you need to change 7.35 to 7.383 and 22.39 to 22.423
For a 0.2 shift up you need to change 7.35 to 7.416 and 22.39 to 22.456
etc

In my case I ended up with a 0.5 shift ( 7.516 and 22.556 ) This seemed to have both gauges showing the same value.

Below is a slow motion video of the 2 gauges with the adjustment. Remember, the XD gauge will display the value first followed by the Trinity. There is a delay. Move to about the 1:20 mark. Towards the end it's easier to see when it drops into the 13's for a second



Note: You MUST change both values equally. Your just shifting the display reading range. Think of it as voltage correction. If you just change one value the AFR will not be displayed correctly across the entire range.

If your logging for tuning purposes, I think it would be a good idea to at least check this out.

I am not saying the Trinity is wrong or has a bug or anything. What I'm trying to do is correct the voltage being detected which may be incorrect.

============
I found a nice right angle plug at Radio Shack.
Part number 274-0298




Comments? Peer review? Anybody care to try this out and share your results?

[TIGwelder]

Ok. I've done some further reading and testing.

What I'm seeing is called the Ground Offset. Innovate describes how to test for it. My fist post describes how to adjust for it. I'll walk you through how to test for it on the Trinity.

The ground from the Trinity analog input is grounded and soldered to the LC-1 white and blue wire. Why is there still a voltage difference? Impedance? Got me. Guess it's a question for an engineer.

First, from the Innovate LC-1 User Manual

7.8 Analog Output tricks/hints

Sometimes there is a ground offset between the device that receives the analog output voltage and the LC-1. This can be especially true if the System ground is not connected to the same ground as the measurement device using the analog out. Both devices reference different grounds and therefore see different voltages. So to compensate for the ground offsets the analog output voltage points have to be shifted by the ground offsets. To measure what the real ground offset is, you can program the analog outputs temporarily to output a flat line voltage by entering the same voltage in both fields for the two analog out programming points. This way the analog output voltage will be fixed, independent of current AFR and can therefore be measured and compensated for easily.

And here is how you test for it on the Trinity.

First we'll set up the LC-1 to output a test voltage.

1) Connect your serial cable/usb adapter to the LC-1 serial out and power on the LC-1.
2) Start the LM Programmer software. You should see it connect and display the "Info/Set AFR" panel.
3) Choose the tab for the output your using. Example uses "Analog Out 2"
4) In the entry boxes for voltage set them both to 5 volts. This makes the LC-1 output a flat 5v

Example screenshot:


5) Click on the "Advanced" button and set the interval to anything but "instant". Choosing the 12, 6 ,3,etc will enable output smoothing/averaging
6) Return to the "Analog Out 2" screen and hit program.
7) Key off to reset LC-1
8) Disconnect your serial cable and plug back in your external gauge if you have one.

Now we'll set up the Trinity for testing.
In my example I'm using Analog1 of the Trinity.

We're going to make 3 "parameters" using Analog 1 as a input. You can use the same analog input and create multiple "parameters". Then you can use those multiple parameters to drive multiple gauges.

1) Go into options and choose "Configure Analog"

Standard settings
Choose "Add a new parameter"
"Enter the name" = UNCORRECTED
"Input channel" = Analog1
"Unit of measure" = AFR
"Low Value" = 7.35
"High Value" = 22.39

This one will be for your adjustments
Choose "Add a new parameter"
"Enter the name" = LC1
"Input channel" = Analog1
"Unit of measure" = AFR
"Low Value" = 7.35
"High Value" = 22.39

This one displays the voltage the Trinity is detecting
Choose "Add a new parameter"
"Enter the name" = GND_OFFSET
"Input channel" = Analog1
"Unit of measure" = Volts
"Low Value" = 0
"High Value" = 5

Now create a new layout.

1) From the main screen choose the "Logging" icon.
2) Choose "Create a new log"
3) Choose "An empty layout"
4) Add 3 small digital gauges to one page. For the parameters choose the following:

Top LEFT - gauge parameter is "UNCORRECTED"
Top RIGHT - gauge parameter is "LC1"
Bottom LEFT - gauge parameter is "GND_OFFSET"

Example gauge page layout:


5) Exit out and save your layout. Give it a name of AFR_TEST

Your now ready to test. The LC-1 should be sending out a flat 5v signal to the Trinity.

1) Key the ignition on without starting the vehicle
2) Choose "Monitoring" from the main screen
3) Choose your layout AFR_TEST
If it complains about starting the vehicle, just ignore it and load
your layout.
4) It may take a few seconds for the LC-1 to finish its sensor warm up before outputting a voltage. Be patient if the sensor is cold.

With the LC-1 outputting a flat 5v, the Trinity "should" be seeing an AFR of 22.39. The volt gauge should be seeing 5.00 volts. If this is not the case you are seeing the effects of "Ground Offset" which is why we're conducting the tests.

You may see values similar to the screenshot below



On the "UNCORRECTED" gauge, any difference between 22.39 and what it displays is the showing the effect of the ground offset.

Any difference between 5v and what the "GND_OFFSET" gauge displays is showing the effect of the ground offset and the voltage the Trinity is really seeing on the input.

Correcting ground offset

If you didn't read the first post in this thread..you might want to check out
"Calculating percentages of voltage for AFR display" and decide what values you want to test for the shift. If you would like to try some corrected values go in
Options->Configure Analog->Edit Parameter
and change the values for the "LC1" parameter.

That way you still have the "UNCORRECTED" parameter that will display the default AFR. Your "LC-1" will display the adjusted AFR. Log both if you want.

Remember when your done testing and adjusting to reconfigure your LC-1 back to 0-5v !!

In order to see real AFR readings again you need to get back to your 0 and 5v settings in the LM Programmer.
See above - "First we'll set up the LC-1 to output a test voltage". Just change the low value back to 0 volts

==========

Here's a video showing the test. I made the video first and then typed out this small book. Poor narration by me includes calling it "Innovative" instead of Innovate and other gems. I'm not a damn TV announcer ya know!! lol
And I'm wearing a red shirt while i'm filming..you'll see when I start the car. Even I was like "WTF was that?"




Den

[Animal451]

This is a big help! I have all grounds soldered to the same point but noticed the readings were off by approximately .4 (rich).

Thanks for the help . . .

[Knuckles]

You make good point on the difference in ground potential. I will probably check my out as well. By chance did you ever take a reading of what the voltage was at the 2.5 mm male connector that plugs into the Trinity with either a Analog or Digital Multi Meter, or even a Ocilloscope? I would also think that doing a free air calibration on the wideband sensor would be a good idea. That way we know that its calibrated and that its accurate.

[Animal451]

Personaly, I do a free air calibration before every tune adjustment. The problem I had is that without the engine running, the trinity kept giving mw a data stream errer and put me back to the main screen . . .

I was able to see the voltages for a few seconds though to allow me to get it close.

[toofart]

Glad you posted this. I resolved this several months ago by doing what you did in the 2nd post -- essentially, have the LC-1 output 3v during warm up, then use the Trinity to view the sensor voltage. Plotted my min/max values according to the error and off I went.

[TIGwelder]

By chance did you ever take a reading of what the voltage was at the 2.5 mm male connector that plugs into the Trinity with either a Analog or Digital Multi Meter, or even a Ocilloscope? I would also think that doing a free air calibration on the wideband sensor would be a good idea. That way we know that its calibrated and that its accurate.

I didn't check voltage at the plug. Also this was a new install so the open air calibration had just been done.

[TIGwelder]

Glad you posted this. I resolved this several months ago by doing what you did in the 2nd post -- essentially, have the LC-1 output 3v during warm up, then use the Trinity to view the sensor voltage. Plotted my min/max values according to the error and off I went.

I think doing it at the max voltage of 5v is a better idea. If you make an adjustment you want to it be correct over the entire voltage range

[cor430vette]

ok this is great but how do you adjust for the difference in the volt reading?
If I am getting 4.9 not 5.0?
I do not have another guage to compare it to so what would a difference of 4.9 to 5.0 be adjusted for?

[cor430vette]

Ok I think I figured this out...
If your reading less then 5v = offset
Set up the other two gauges:
1. Will show the difference of the offset problem
2. The corrected one will show your adjustments.

To figure the adjustment take the difference give or take between 22.39 and what the offset problem gauge is showing.
Example:
Your gauge shows 22.11 and should read 22.39.

Calculation:

22.39 -22.11= .28 difference

Take the calculation you provided of .1= .03324 volts
.28 is then divided by .1 .28/.1 = 2.8
You then take 2.8 x .03324 = .093072 added to both sides ....
7.35 + .093072= 7.443072
22.39 + .093072 = 22.483072

Correct?

This difference can have you tunning your car which for example mine should be 11.8 full throttle and if I am .28 off I am actually running = 11.8 + .28= 12.08 getting a little lean !! GREAT POST!!

[TIGwelder]

Looks correct.
Best way to picture the changes would be to imagine a radial dial indicator gauge where the indicator is left in position and then clocking the dialface to read correctly. We havent changed how the tuner detects the voltage, just how its displayed
Glad you found it helpful

[cor430vette]

ok I got in my car and did everything you said.
I even logged the output.
What I found was that my volts would drop for a split second to approximately 4.85 then immediately back up to 5.00.
My uncorrected gauge would drop temporarily to 21.95 approximately then immediately back to 22.39.
From reading your post it appears your voltage stayed below the 5.0 and the 22.39 number correct? Or did it fluctuate like mine?
Should I correct or is this split second drop not a problem?

[cor430vette]

Just called a guy at innovate and inquired about the split second fluctuation between the lower value and 5v higher value I was getting ...and he told me that this is ok ...it would be a steady level below the 5v that would be the problem not split second fluctuations. He claims this would smooth out when engine is on.

Update: I later tested again and I appear to be fluctuating around 4.8 to 4.9 ...this is after I redid my wiring because it was messy and I redid some crimps...( dont you hate when you think your making something better and make it worse?) ...I redid it over and over to try and get the 5v peak and nothing...

[cor430vette]

Here is another calculation I found on the innovate site....find your volt difference lets say your volt offset is 4.80 vs 5.
4.80 x 3.008= 14.43
Note: Why 3.008 ?
7.35-22.39=15.04/5 volts= 3.008

Then add in the bottom 0v preset 7.35
14.43 + 7.35 =21.78

This should give you a pretty close representation or confirmation of how much you are off on the gauge reading.

You then take the difference as expalined in thread above...
21.78- 22.39= .61

then take the calculation in this thread...
6.1X.03324 =.20 to should be added to each side...

Just another way to confirm ...