MPS Web Page Update
 

OK, I've just posted the latest update to my MPS web page:

http://members.rennlist.com/pbanders/manifold_pressure_sensor.htm

It's taken a while, but I think I've finally bottomed-out on a consistent understanding of the adjustment and calibration of the MPS. Here's a short description...

There are three adjusters on the MPS: The full-load stop (the screw in the end of the unit, under the epoxy seal), the inner diaphragm screw (under the full-load screw), and the outer diaphragm screw (inner screw sits inside of it). The inner screw sets the overall mixture across full and part load. The other two adjusters are interdependent: the full-stop sets the maximum enrichment at full load, and the outer screw sets the vacuum level where the diaphragm touches the full-load stop (about 50 mbar below atmospheric pressure).

Previously, I'd thought that the outer screw set the vacuum level of the <onset> of movement of the full-load diaphragm. Not so. That onset value is set by the spring constant of the coil spring in the end of the main casting of the unit (behind the coil and armature assembly). There are two common versions of the MPS, the "short" nose and "long" nose, referring to the extension of the coil spring holder in the main casting (thanks to Gary Helbig for these descriptive names :-)). Units with the "short" nose have an onset vacuum level of about 100 mbar below atmospheric pressure, units with the "long" nose have an onset at about 150 mbar below atmospheric pressure.

I've verified all of the actions of these adjustments on a real full-functional 043 ('74-'76 2.0L) MPS that I recently picked up from Ralph Ricks. I was able to adjust this MPS to exactly duplicate the characteristics of either an 049 ('70-'73 1.7L) MPS or to my calibration curves I have for the 043 units using the theory of adjustments described above. Very cool.

I've modified the calibration procedure to reflect this new information, and have a graph that shows the effect of each adjustment on the final charracteristic. The only thing I have left to do on this page (hopefully) is to complete the electromechanical model.