maf signal conditioning

[ol boy]

How much load? 1k? 220?

I can try some values tomorrow. I need to grab some stuff to test.

Sent from my SM-G920V using Tapatalk

[ashford]

How much load? 1k? 220?

I can try some values tomorrow. I need to grab some stuff to test.

Sent from my SM-G920V using Tapatalk

im not sure im thinking 5-20k

[subwoofer]

Just remember to place the load outside the series resistors.

[ol boy]

Does it matter how close or far from the sensor? I could tap into the signal wire near the sensor and grab the ground.

Sent from my SM-G920V using Tapatalk

[ashford]

well testing on the bench is successful using an actual maf. i used a 10k for a load on the maf. another thing i noticed was that voltage was about 50mv with no air flow, if memory servs me i saw .2v or so in the car with no flow, which brought me to look at the diagrams for the ms3x. there is a 2.49k 5v pullup on the secodary map port, this pullup may be a cause for some errounous readings i have seen.

[ol boy]

The element I have starts at 0.080 volts no-flow. I'm MS2 and using adc7. What results did you see with the load resistor and without the pullup?

Sent from my SM-G920V using Tapatalk

[ashford]

i didnt try it with a pullup on the bread board. looking at the schematics the adc's on the ms3x are a copy/paste of the clt and mat of the mainboard. the ad623 wont cope with that but the js 5/7 should be fine

[jsmcortina]

Remember that the MS3X ADC inputs have the pullup resistor non-populated. i.e. it is an empty space on the board for you to install a pullup only if you want one.

James

[ol boy]

Tossed a 10k across the signal out and it's ground return. I didn't see much difference in steadiness of the adc from idle to 1/2 throttle. I'd have to look at the log and compare it to and old one. I might try lowering the load resistor to 4k7. Let me know what test you want ran.

Sent from my SM-G920V using Tapatalk

[subwoofer]

Remember that that load has to sit on the MS side of the wires, adding it on the output of the MAF itself gains nothing in terms of noise immunity.

[ol boy]

So are we trying to reduce noise or get the proper analog out by loading the element properly and using the ad623? What's the end goal using which method?

Sent from my SM-G920V using Tapatalk

[subwoofer]

Ideally, you get both. By having a resistive line termination, you make sure the line (differential pair) is carrying current. This reduces the susceptibility to induced differential noise, with the side effect of loading the transmitter circuit correctly. Using the AD623 removes the ground loop.

[ol boy]

Are you working on a breakout board that has all this laid out? I really would like to go back to MAF for fuel control.

Sent from my SM-G920V using Tapatalk

[ashford]

this setup smoothed out the signal significantly, with a maf lag factor of 80 it is cleaner than before at 10

[ol boy]

Can you post a schematic of what you did and how it's all hooked up? Thanks Ryan

Sent from my SM-G920V using Tapatalk

[ashford]

Can you post a schematic of what you did and how it's all hooked up? Thanks Ryan

Sent from my SM-G920V using Tapatalk

here is exactly what i did. i put it on the proto area but ran out of room so i made another on a free proto board and screwed it to a transitor screw on the heatsink

maf conditioner.jpg

[ol boy]

Thank you so much.

Would a 9 volt reg work the same? What's the min/max input to the 623 we should use?

Sent from my SM-G920V using Tapatalk

[ashford]

a 9 volt would work as well. it can output vcc-.7, absolute min of 3.3v absolute max of 12v. liniarity of the output gets goofy at roughtly vcc-1.2v