MAP sensor problems - Mapdaddy4 and v3 board

[Chiburbian]

Last weekend I powered up my MS3 after doing the 5v pull up mod to enable alternator control per the most recent beta.

When I powered it up, I noticed my MAP was about 80kpa. This seemed strange considering that my local barometric readings usually run between 29.97-30.1 inches of mercury which is roughly 101kpa.

I pulled up the actual barometric pressure reading from the second port on the Mapdaddy and found that it was reading 98kpa (which I have since found to be below my minimum).

Anyhow the voltages as read from my map daddy at that point were:
1:  .932v (MAP out from MapDaddy4)
3:  4.80v
JS5: 1.104v (bark out from Mapdaddy4)

Using the y=mx+b formula to determine the standard calibration line it results in:
1: 80.5kpa
JS5: 94.682kpa

Both are low obviously but the MAP was the one that was totally off, so I ordered a new MapDaddy and installed it...

Now my values are:
1: 1.130v (96.805kpa according to formula)
JS5: 1.119v (95.9kpa according to formula)

Here is where it is strange...

Tunerstudio is giving me:
MAP: 135kpa
Baro: 97kpa

Just to confirm, this is using for both MAP and BARO the MPXH6400A settings listed on the DIYautotune website.

To be honest, the 1kpa or so difference between the two sensors isn't what is bothering me. What I want/need to know is wither the problem is in tuner studio and/or calibrations OR is the chip somehow reading 1.5v (135kpa'ish) and if so, where is that voltage coming from because it's not matching what is coming out of the MapDaddy4.

Where do I start? What should I meter?

[Chiburbian]

Another observation...

U2 pin1 reads 1526 ohms to the (in?) leg of R2.

From the tie point (viewing from top of board) about 8mm to the left and 2mm high of the u2 pin 1 to the "in" leg of r2 is 0 ohms.

IF I somehow burned out my trace from pin 1 to R2, why would I be reading higher kpa?

[Matt Cramer]

Can you check if anything is pulling down the VREF to below 4.9 volts?

If not, I'd suspect it being a settings / calibration issue. Try recalibrating the MAP and baro sensors.

[Chiburbian]

I will check tonight.

IF I see a VREF of less than 4.9v (say 4.8 or 4.85v for example) what could be causing this? Where do I start troubleshooting this? Just trying to think a step or two ahead because the hours that you are at work and thus able to help diagnose and my hours at work where I cannot work on this mean that it will take an entire day per item.

Just as an aside, this morning before work I removed the leg of R2 that goes to U2 Pin 1 and metered 600 mohm, which to me seems to be essentially an open correct? Would it be rational to jump directly from the pin 1 header to that side of R2 eliminating the influence of the trace completely?

At this point I am tempted to buy a whole new v3 board and start over LOL. I started with the standard map sensor, removed it for a MAPdaddy, removed that for troubleshooting earlier on in the project, reinstalled it, removed it when it failed, and installed new one. My board is pretty bad in that area.

[Matt Cramer]

I'd first look for stray flux on the MapDaddy itself, then check everything that is connected to 5V and ground.

But if your resistance reading on R2 is that high, it sounds like you may have a broken trace in need of bypassing with a jumper wire that is partially responsible for the problem.

[Chiburbian]

Why would a broken trace INCREASE the voltage seen by the MS chip though? That is what is puzzling to me. In order to see 135kpa with the mapdaddy calibrations the mapdaddy (or something in the circuit needs to be outputting 1.5v to result in that number.

In regards to VREF, the voltage seen at the 5v point in the prototype area and the voltage seen at pin 3 on U2 is the same, if that tells you anything (or excludes anything).

[Matt Cramer]

Why would a broken trace INCREASE the voltage seen by the MS chip though? That is what is puzzling to me. In order to see 135kpa with the mapdaddy calibrations the mapdaddy (or something in the circuit needs to be outputting 1.5v to result in that number.

Potentially, yes. The voltage on the ADC inputs tends to rise on its own if nothing is connected - I believe it's an internal capacitor there that just picks up charge and doesn't let it go.

[Chiburbian]

I removed the header from pin 1 of the U2 and soldered a wire directly from the pad that was pin 1 to r2.

That seems to have bypassed my problem and I was able to get a reading that was within 1 or two kpa of barometric reading. From there I modified the calibrations to make the barometric and MAP equal the barometric pressure from weather.com.

When I attempted to start the car my map was ALL off, which to me is a good thing. Obviously my map sensor had been failing for some time. It means a re-tune but otherwise we are golden.

(now on to alternator control. See my post in the alternator control thread on miataturbo.net if you would be so kind as to help)