Tunerstudio reading very wrong AFR - Megasquirt pin 23 problem (RESOLVED)

[saturnv7890]

Issue resolved: Bad LC-1, analog output was connected to the wrong harness wire, probably killed the lc-1 analog output





So I recently picked up a car that was sold as not running, and it's got an MS1/extra in it.

The guy who sold it to me let the car sit with the battery dead for almost a year I think and when he got back he thought the Lc-1 went bad, which also made the MS go bad.

Anyways, I checked all the wiring for the lc-1 and it looked good, so I plugged MS into tunerstudio, made sure the wideband calibration was the LC-1 default, and turned it on. The innovate gauge read "22.4" which is right because the car is not on, tunerstudio read this:



Not really sure what that means, because the dial says one thing and the number says 7.35?


My questions:
-Does this indicate anything is broken?
-Is there a way I can test the lc-1?
-Is there a way I should be testing the megasquirt? I don't have a stim and I'm waiting on a paycheck to order one.

Let me know what I should be looking in to! Thanks!

[saturnv7890]

Now I'm realizing a few things:

1. The ECU must be seeing 0 volts from the wideband? But I'm not sure why the gauge needle would be at 18.6ish like that

2. The way the previous owner had this hooked up, the brown wire from the LC-1 went parallel to the gauge and the ecu, would this cause issues? I guess he did have it running like this before so it might not? Either way, I guess I will try to run the yellow one to the gauge and the brown one to the ecu.

[Jaxx_]

Now I'm realizing a few things:

1. The ECU must be seeing 0 volts from the wideband? But I'm not sure why the gauge needle would be at 18.6ish like that

2. The way the previous owner had this hooked up, the brown wire from the LC-1 went parallel to the gauge and the ecu, would this cause issues? I guess he did have it running like this before so it might not? Either way, I guess I will try to run the yellow one to the gauge and the brown one to the ecu.

The wideband sends a voltage to the megasquirt box, and the voltage should be calibrated to the proper table. You'll find the ability to burn this table in a menu, called "Calibrate AFR table." You should select the LC-1 option, or find the values online. It sounds like you have done this already.

If you read carefully the gauge is actually reading 7.35(the gauge is faux analog and it will wrap-around). This is probably no output from the LC-1, or just grounded. Confirm that the harness is okay to make sure this is not a physical issue. Usually widebands read incorrectly for a few seconds to a minute while they warm up, anyway. It's not really possible to tell if it is reading correctly without the engine on. Get the car running then worry about the wideband. You don't need it to get the car to idle or drive.

It is possible that the load from the gauge would lower the voltage or fuck up the signal to the ECU, or vice-versa.

[saturnv7890]

The wideband sends a voltage to the megasquirt box, and the voltage should be calibrated to the proper table. You'll find the ability to burn this table in a menu, called "Calibrate AFR table." You should select the LC-1 option, or find the values online. It sounds like you have done this already.

If you read carefully the gauge is actually reading 7.35(the gauge is faux analog and it will wrap-around). This is probably no output from the LC-1, or just grounded. Confirm that the harness is okay to make sure this is not a physical issue. Usually widebands read incorrectly for a few seconds to a minute while they warm up, anyway. It's not really possible to tell if it is reading correctly without the engine on. Get the car running then worry about the wideband. You don't need it to get the car to idle or drive.

It is possible that the load from the gauge would lower the voltage or fuck up the signal to the ECU, or vice-versa.

Thanks for the pointers!

I believe when I first started up TS I got the calibration right for the o2 reading, I'll double check with LM programmer to make sure my analog outputs on the lc-1 are the right calibration as well.

I guess I can try having the yellow wire go to the gauge and the brown only going to the ecu. Then I'll test the voltage across the pins on the harness and make sure they're correct and I guess I can run a test light across a few areas to make sure the harness is good.

The engine is a fresh rebuild and it is untuned, would it be able to run? And would it be safe to do so without knowing the AFR? I know being too rich will wash out the fresh cylinder walls and cause the rings to not seat right.


Any other things I should check? I'm at work now so I'm coming up with a list of things I'll need to check when I get home.

[FLG]

I remember I had to choose a different one as mine wasn't reading the lc1 correctly either. I'll take a look when I get home.

Either way it wouldn't cause a no start condition.

Sent from my Galaxy Nexus using Tapatalk 2

[saturnv7890]

I remember I had to choose a different one as mine wasn't reading the lc1 correctly either. I'll take a look when I get home.

Either way it wouldn't cause a no start condition.

Sent from my Galaxy Nexus using Tapatalk 2

Choose a different lc-1?

[FLG]

A diff selection in Tuner studio

Sent from my Galaxy Nexus using Tapatalk 2

[saturnv7890]

Alright I'll mess around with those settings a little too.

The only issue I have with trying to just start it is that it's a freshly built engine and I don't want to mess up the rings seating because this one thing is wrong, and I'd rather be more confident before trying to fire it up.

[FLG]

It's not like a stock ecu were it tunes based on afr. Ms couldn't care less what the afr is, it's up to the tuner to tell it what to do (unless you run auto tune obviously) I didn't have my lc1 connected to my ms for months.

Sent from my Galaxy Nexus using Tapatalk 2

[saturnv7890]

Oh well that is definitely reassuring then!

I guess I'll try and do some diagnosing, and if I can't find anytihng I'll start preparing to start it up then!




Where can I find a base map for MS1/extra?
It's an m20b27 with an h1c.

[FLG]

Double post

Sent from my Galaxy Nexus using Tapatalk 2

[FLG]

Try whowho or I'd be happy to send you my tune with 2.7i hx35 50lb injectors.

Sent from my Galaxy Nexus using Tapatalk 2

[saturnv7890]

Pm'd, thanks for the help!

[saturnv7890]

Went out and did some testing and found the following:

Observations:
Was able to hook up lc-1 to Logworks and get a good reading (20.8 afr)
->Gauge and MS did not see this (see "problems" below)

Was able to calibrate lc-1 by grounding calibration wire to chassis
->Calibration wire was observed to have 5v, probably normal

Double checked analog outputs with LM Programmer
->both were set to 0-5v wideband



Problems:
Brown and yellow wire (analog outputs from LC-1) were both giving 0v with respect to a very good ground
->This must indicate a bad lc-1? I can still get readings from logworks though.

The ecu was outputting 12v at the pinout (motronic pin 28, megasquirt pin 23)
->Is this bad?





Questions:
Is there a way to fix the lc-1 to make it give voltage at the brown and yellow wires?
->do I need to buy a new lc-1 for these to work?
->not a big concern for now, as I can just run logworks to view my afr


Should there be a +12v voltage at the o2 sensor pinout? (ms 23, motronic 28 )
->my ms isn't bad, is it?
->Matt Cramer from DIYat was saying "something is very wrong" - I might need a new ecu then.
->I'd love to see if it's just the main chip, so I can just replace it with the MS2 daughterboard ($100 vs $400)

[saturnv7890]

Took some pictures of the ecu for any MS gurus out there:

http://s000.tinyupload.com/?file_id=07851562756631110322

And, here's a picture from a few years ago when the board was running fine (for comparison):

http://smg.photobucket.com/user/JoeUser11/media/turbobuild/IMG_0476_zps711370c4.jpg.html?sort=3&o=4

Let me know if I should start shopping for an ecu!


UPDATE:
Some people over at MSextra are saying the board is probably fine and it's most likely an issue outside of the ms. After work I'll be chasing down the wiring issue in the harness.


UPDATE 2:
Issue resolved, car runs great now! See first post for solution.