Reading voltage from OEM MAF?
 

I am working to get the MS2 going with MAF instead of speed density, but having problems getting a signal from the stock MAF.

Here is the relevant schematic from the NB 99/00 ecu.
http://1.bp.blogspot.com/-xNuje4v_zG...%2Bsensors.png

I started by verifying signal at the MAF connector. Three wires on maf plug, ground (black), +V (red/white), output(green/blk)

With black as ground, I am getting +14v on ref (good), and at idle the output of the MAF is 7volts.

According to the schematic, this 7v green/blk signal should be carried through to pin 2L on the ecu.

However, I am getting 0V when I read from 2L at the harness. If I unplug the Megasquirt the engine runs fine with the stock ecu, so I believe the MAF is good (recently passed emissions in this configuration.)

Am I misinterpreting the schematic? The green/blk carries through to pin 2L so there should be the same voltage as read at the MAF?

The thing that is strange is that at idle the MAF is showing 7v? The megasquirt ADCs expect logic level 0-5v and I would assume the stock ecu would expect the same.

Any suggestions on what is happening appreciated. I hesitate to connect the MAF output to the MS2 because 7v is not logic level. Is it possible that as flow increases, voltage drops? (output might be 0-7v) Or does voltage increase with flow?

I don't know how close the 1.6 AFM is to the NB MAF, but JoeP has some old threads about the AFM and working with an EmanageU. They might be somewhat helpful for you.

Thanks for tip. I found this thread which suggests that voltage to MAF should be +5, that output is expected 1.29v - 4/5v.

https://www.miataturbo.net/e-manage-...g-issue-63149/

I didn't change anything with wiring harness so perhaps stock ecu responsible for sending 5v signal to maf (as opposed to 14v)..? It could be that the maf is operating outside its normal range because input voltge is too high? But that wouldn't explain why I'm getting nothing on pin 2L.

This is weird... When I start the car with stock ecu and check voltage at MAF, +V is the same at 14v, but the output (green/blk) is 1.3v as expected at idle.

The MAF only has three wires: +V, gnd, and output. Given that the ground and +v are the same with both ecus, why would the output voltage be different with the Megasquirt given the same input voltage??

Are you using a DIYPNP or building your own MS2.

Its a DIYPNP I built last year, running great with MAP, need this for class rules.

There are two ADCs on the microsquirt board that I have verified and that are supported in Tuner Studio. One of them is listed specifically for MAF in the microsquirt manual. Miata MAF should be providing 0-5v logic level so it should be compatible.

Based on my readings, it looks like the MAF only works as expected when a stock ECU is hooked up. Givne same input voltage and ground, why is output different with different ecu? This doesn't make sense based on my (rudimentary) understanding of electronics, but would suggest I need to complete the parallel install before doing the maf.

That is correct.


You are not misinterpreting the schematic. You should see the same voltage at terminal B of the mass airflow sensor as at 2L of the ECU.


You are correct in that it is "impossible" for the MAF sensor to produce an output outside the range of 0-5v, and this will be the case regardless of what ECU is connected to it.

A few possibilities come to mind:

1: There is a fault internal to the Megasquirt ECU, which is causing it to apply a voltage to the pin to which you have the MAF sensor connected. This might be because an internal pullup resistor has been connected to that pin, as one would normally do when connecting to a passive (resistive) sensor or an open-collector sensor.

2: You may have had too much / too little to drink. There is a sweet spot about halfway between "stone cold sober" and "way too drunk to be working on a car" at which maximum productivity is reached.



Try disconnecting the MAF sensor entirely and see if you still have voltage on that line.

When I tested the voltage and read 7v on grn/blk, the megasquirt was running the car, but nothing was (intentionally) connected to 2L. I read the 7V at the sensor.

Probably too sober. I know what you mean.



Great idea and easy to test. That would tell me if the MS is putting voltage on that pin. It may be, as you say, that there is a pullup resistor inadvertantly connected to that pin and I'm possibly getting a voltage from somewhere else, but that doesn't explain why I was getting 0v directly from the harness 2L and 7v from the MAF itself.

I did trace continuity through the MS board from the header to the output to make sure it was clear, but I need a drink and start over.

My next step is to build the parallel harness and install the MS under the driver seat because reaching under the dash it a PITA for this kind of thing.

My biggest fear was that the stock ecu was doing something tricky like applying PWM to the input voltage or applying a filter to the output that was needed to make the maf correct, that my only option would be to read it through the harness with the stock ecu connected.

Are there any news?
I have the same problem.
Drove my MS for years with Speed density.
Now this MS came into a 1.8 NB which should drive with the MAF.
Now there is the problem that there are 7V at the signal while idle, which will not work.
Does anyone drive his NB with MS and MAF?

Looking at his public profile, you'll find that ChrisSTR hasn't been on the forum since 3/25/2015 so you're unlikely to get an reply back.

I am guessing that NB MAF sensor output is not voltage but current. In order to read voltage you need some resistor to GND (pull-down), that resistor is inside stock ECU?

I will post again if I confirm my guess.

And just to confirm - NB MAF with +12 power supply is very different from NA6 mechanical VAF.

so with 1K resistor to GND, not OEM ECU

1.4V engine not running
2.3V high idle
getting to 5V at 6-7K rpm