MFI Pump setting question

[DG58INNM]

Does anyone know a quick and dirty way to verify initial rack adjustment on the pump? Car is a '72 T. I have all the covers off for cleaning and gaskets. Although all looks nice inside, I have reason to believe someone in the past tinkered with this one, and I'd like to know its at a good starting point for "check, measure, adjust". I see there is a new book out about MFI that probably tells this, but I don't have a copy yet.
Thanks,
Dave

[edmayo]

I'd be amazed if there was a pump out there that wasn't 'tinkered' with! I know of no way to arrive at an 'initial' setting. I'm sure the pump rebuilders do, but they also have the advantage of their flow bench to see what the pump is putting out, which is how they arrive at their setting when they build a pump. Did the car run when you took the pump off? You'll need to 'set' the pump the old fashioned way, checking the mixture and adjusting to suit.

[304065]

See the pelican thread "MFI Open Heart Surgery"

http://forums.pelicanparts.com/porsc...t-surgery.html

You can see that the adjusting screws will bottom in one direction but not in the other. If there were a limit in BOTH directions you turn the screw all the way to the RIGHT until it bottoms (2,4 pumps have 12 clicks per revolution according to CMA) and then crank it all the way to the LEFT until it stops, then back half the number of clicks.

But you can't, if you crank it too far to the left you will pull the screw out and then be in a world of hurt that only Gus or Mark Jung or someone who is not afraid to pull the back cover off the pump can resolve.

But you don't need to do this. In order to set the mixture correctly you need a wideband 02 sensor, so just leave it right where it is and use clicks in either direction to iterate until the fuel curve looks right.

[DG58INNM]

Thank you Ed,
This car was running when I acquired it, but very poorly. The previous owner admitted turning both the part load and idle screws indiscriminately in an attempt to make it run better, with no success. The real culprit was the engine was extremely worn (rings, valves, and some really flat cam lobes) so it wasn't moving enough air to match the fuel being squirted. I've since rebuilt the motor and it is nearing start-up time. Putting the MFI back, the linkage was woefully out of adjustment (which I've now corrected). With all these changes I'm wondering if it will even start. I thought maybe one could set a dial indicator on the rack at the back of the pump while the rack was at resting position, and know that the rack was in approximately the right spot.

I expected to have to fine tune it as normal with the CO meter, as noted by 304065, but thought I might save some trouble by at least having it fire up sort of close.

It would be really fun to build a test rig to calibrate pump as shown by this new book (I did some more research last night), but odds are quite high, given the crazy prices of the cars now, that this will be the last MFI engine I ever have. So, probably not worth the effort.

DG

[edmayo]

If you have a situation where all other parameters are correct and the pump is so far out the engine won't start to even begin your mixture testing, then I have a way of dealing with that that is not in the books. With the car up on stands in the rear, I connect a remote starter to the starter motor, now I can crank the engine standing at rear of car. Then I disconnect the pump control rod from the cross bar (the one that is set to 114 mm) Now I can crank the engine and operate the pump independently of the cross bar and thus air throttles. This way I can easily alter the mixture to at least get the engine to start, and by manually working the pump rod and the air valves get a 'feel' for which way I need to adjust the pump to at least get to a starting running point. You mentioned the linkage settings were wrong,,,,that is possibly the most common error I find in these systems !!