MAP sensor reading ~97 kPa all the time
 

I have a used enhanced MS2 for 1.6 and the MAP sensor always reads ~97 kPa. Even if I attach a short hose to it and blow/suck it will never change. Is there a chance it went bad? Do I need to calibrate it? I am able to start my car and it will idle pretty rough at ~10-11 AFR on the base map for this ECU. I am guessing because the MAP reading is so far off. Is that a correct assumption?

https://www.miataturbo.net/attachmen...ine=1460849825


This is where I attached the vacuum line originally on the car before I removed it and tried to blow/suck to test MAP reading.

If the sensor itself isnt blown, the circuit may be, but yes, its in the ECU. This isnt a calibration issue. No it will not run right with no MAP.

post the MSQ just in case, but yeah they can go bad. typically happens after a winter and when the MAP sensor is vertical. Condensation is able to drip into the MAP sensor and freeze.

Ok, Ill do that tonight when I get back to the car. Just to double check if I have to replace the sensor, the MPX4250 2.5 Bar sensor is most likely what Ill want to replace it with right?

I hope I added the attachment correctly.. This should be pretty close to the default map for this ECU. Thanks.

Ok, I replaced the map sensor and now it reads 94.5 kpa all the time and does not change when I blow/suck on the vacuum line. What do I do now?

is this a diypnp?

A used enhanced MS2 for 1.6.

https://www.miataturbo.net/attachmen...ine=1461292023

https://www.miataturbo.net/attachmen...ine=1461292023

https://www.miataturbo.net/attachmen...ine=1461292023

https://www.miataturbo.net/attachmen...ine=1461292023This is the connection after I replaced the original map sensor before I cleaned them up.
Bump with photos of board. I am really dead in the water until I can get this figured out.

Check to make sure the sensor is getting its 5v and has a good ground. Im assuming you can get MS2E schematics? start there

Update! I checked the voltage between pin 3 and ground. Pin 3 is labeled as MAP sensor input on the MicroSquirt module. As I suck and blow on the vacuum line I see this voltage raise and lower as expected. This tells me my sensor is good and is getting power and the issue lies either in the MicroSquirt itself or somewhere in TunerStudio. My guess is its not TunerStudio.
I also checked the resistance on C11 (From what I can tell is a 0.1 cap). I switched my multimeter to 20k ohms and the screen read 8.88. Is that what its suppose to be? Sorry, I dont know the proper terms to explain it better.
https://www.miataturbo.net/attachmen...ine=1461372911

Another thing to note, I am getting voltage going into C11, but I am not getting any voltage out of C11. Im assuming I should. Any idea?

I think I have found the correct schematics and will attempt to figure out what is wrong:
http://www.msextra.com/doc/pdf/Micro..._Guide-3.3.pdf

C designation is usually a capacitor.

Edit- if thats the correct schematic, it goes to gorund to filter noise or something. As long as it not shorted, its not an issue. Yours isnt shorted.

Do your voltage check whole blowing and sucking ( :giggle: ) on R5 on the C12 side.

edit edit- and from there you can check it on pin 27 of the processor chip

im going to be you have a software or settings issue

make sure the signal makes it to the CPU.

https://www.miataturbo.net/attachmen...ine=1461441917

I checked all points along the trace, the resistor, and pin 27. All getting the correct voltage, well at least the voltage changes when I blow/suck on vacuum line. I even reflashed the ECU just to make sure I didnt change a setting somewhere that would cause it to act like this. Just as a sanity check, I should be able to see the engine MAP gauge change even with the car off correct? Also, the only ini error that wasnt related to vvt, was this:

Warning: [GaugeConfigurations] gauge template assigned to undefined OutputChannel: maf, Problem at:
[mainController.ini]:[Line:6881]: mafGauge = maf, "Mass Air Flow", "g/sec", 0, 650, 0, 200, 480, 550, 0, 0

But Im guessing since it refers to gauge configs it probably doesn't effect anything.
Any other suggestions I can try?

I have a hard time believing that only 1 input pin on a chip like that could go bad, its just not how it happens in my experience. But anything is possible. It still sounds software to me, but youve already done everything i would have. Triple check that you dont have weird map/baro settings. And then I guess its time to swap that chip or board.

Lets see what brain has to say.

Ok, just to dump everything Im doing. I started a project using the 3.3.0f ini file I was told to use. Changed only these options from their defaults:
Megasquirt / Microsquirt selection: Microsquirt Module MSPNP/DIYPNP
EXPANDED CLT TEMP: Activated

Under sensor calibration I have the MPX4250 selected. I have tried what is the recommended values for this sensor and I have also tried modifying everyone of those fields and couldn't get the gauge to ever change.
Under general settings baro correction is set to initial MAP reading, MAP averaging lag factor is 98. MAP sample method is set to use average.
Non-Linear Baro correction is all 0s across 80-120 kPa.

I am unaware of any other options related to MAP/Baro. Is there a way to specify a MT.net user so they get a notification about a thread? I pretty sure this is built by Reverant and would like him to take a look at this and see what his opinion is.

Yes
@Reverant
@Braineack

Thank you Sean.

Bump, open to any and all suggestions.

So you've verified that the microcontroller is getting the correct voltage on the correct pin.
Do you see any other analog inputs (ADC) doing anything in tunerstudio? Or are all of them not doing anything? (coolant temp, air temp, 02, etc)
It is uncommon in modern microelectronics and even more uncommon in automotive spec stuff to fry a single pin; but it can happen.
If you did somehow achieve this, and other inputs are working, why not jump your map signal to the spareadc input? (pin29? (manual doesn't line up with silkscreen), edit: or pin16/17 based on silkscreen).

If none of the inputs are working then you have either a bad microsquirt module, corrupted firmware, or something more complicated at the microcontroller chip environment level.

I have the coolant temp, air temp, wideband, and vTPS all working. Lets say I did jump MAP to pin 29, how do I get TunerStudio to know this? Im guessing its a ini file change? Thats a pretty good idea but I would like to use that as a last resort.

Interesting. You'll need to take a good look at the silkscreen and make sure I got the pin right. The manual says pin29 but your silkscreen pic says pin16/17. I'd probably trust the silkscreen.
I actually have an ms3 so I'm not too sure about the differences in tunerstudio and the ms2.
For the ms3 there's a simple drop down menu under general settings that you can use to choose what ADC input the map signal is on.
I haven't messed with the ini files enough to know if you can tinker with them directly and have the java script still read it correctly.

That error about the MAF is interesting? You don't have the thing setup to use a mass air flow sensor do you?
Or have something odd in the ini file that the tunerstudio scripts cant read?

As others have said this sounds like setting issue.

Im pretty confident I got the right pin on the CPU. I mean I watched the voltage change when I sucked/blowed on the vacuum line. Ill feel really dump if its a settings issue. I swear I have went through every setting 10 times, but with that said I didnt see a drop down to choose the ADC input. Even if I was setup to use a mass air flow sensor, wouldn't the MAP sensor still register and change? Or just show 0 kPa?

Indeed, your map sensor is fine and its signal is getting to your microcontroller.
Now you need to tell the software that you're using the microcontroller pin that you've got physically connected to the map sensor.

I didn't write the software, but if I did, if you were to tell the system to use maf it wouldn't do anything with the map readout in the gui.
Maybe it shows the barometric value at startup or something. Maybe if you suck on the line and turn on the key it'll show something else haha.

Check your settings, make sure you're using speed density and your map adc input is correct.

From Googling, it looks like it is possible to move it over to a different pin but not as easy as MS3. This absolutely sucks. I will double check that I am using speed density. Thank you for your suggestions.

Also, in a perfect world this shouldn't matter, but I have been attempting all of this on a 64 bit computer running Ubuntu.

Tuner studio uses java and is pretty compliant across OS's.
Check out pages 67 and 71 of the manual. It has actual screenshots of the dropdown menus for the settings I've talked about.
Control algorithm should be set to speed density

If you're really confused post up pics of what you see.
http://www.msextra.com/doc/pdf/Megas...erence-3.4.pdf

This should be your bedtime story

also make sure the sensor calibration is correct.

Ok, I believe I have figured out the problem. I had to change the MAP sample method from the default of 'Use average' to 'Use lowest value'. Now Engine MAP changes when I suck/blow. Is there any reason why I shouldn't use lowest value? Otherwise Im gonna shout 3 "Hail Mary"s and start tuning this bitch. Thanks for all the help and input, I really appreciate it. Throughout this whole troubleshooting process I have learned so much about MegaSquirt and TunerStudio and how they work.

Never used that before. Something still sounds really wrong if you need that in order to get a reading.

Yeah, I agree. At the same time it has been running really really well so far so until I run into a problem I plan on just leaving it as is.