MFI Pump - Open Heart Surgery

[356RS]

Ryan, I don't want to hijack Jeffs thread but, Yes, the black screws are the right ones. If its real rich, try backing them out 6 clicks each then see where you are with a A/F ratio meter.

[kenikh]

Pinging Jeff, I think you have some info to add to this thread!

[Jeff Higgins]

Originally Posted by kenikh

Pinging Jeff, I think you have some info to add to this thread!

Yup; I'm working on it. I need to take some photos to help explain my latest efforts. All I can say for now is that I have hooked the LM1 back up, and have been learning a bit more about adjustments within the idle/mid/high speed capsule. While the car has run like a champ for the last year or so, I've never been entirely happy with its range of variation in A/F ratio. I have managed to narrow that considerably. Stay tuned...

[Scott Clarke]

I'm on the edge of my seat!

[Jeff Higgins]

Well, I've finally had the time (and inclination) to revisit the adjustments of the three springs within the capsule in the governor assembly. As some of you may remember, I wound up last year running with the high speed (white) screws all the way in, the mid range (black) screws all the way out, and the idle screw all the way out as well. The whole idea of this had been bothering me, and I always knew I could do better. I believe I have.

I came to a very simple realization concerning the roles of the three separate screws: they are anything but independent of one another. In other words, it is altogether too easy to over-adjust one set of screws, and overwhelm the adjustment on the next set in the progression. So, in my case, I was effectively running on just the high speed adjustment. Here are some photos to help illustrate this concept:

This first photo shows a "normal" relationship between the inner high speed spring and the outer mid range (the idle spring is omitted for clarity). Notice how the lighter outer spring - the mid range spring - protrudes above the inner, heavy high speed spring. This is with the adjustment screws an equal number of turns out from bottomed:



This next photo shows the high speed screws all the way in, with the mid range screws all the way out:



Notice in the first photo that the the mid range spring protrudes above the high speed spring, where in the second the springs are level with one another. In the second photo, the high speed spring will overwhelm the mid range, rendering the mid range adjustments ineffective.

The bottom line is, these two adjustments work together in harmony with one another. To do so, they must be roughly equal. Even with about eight full turns of adjustment from bottomed out to falling out, we cannot take advantage of that unless we adjust them a more or less equal amount. I have found that about 2-3 turns difference is about the maximum allowable difference between these two springs.

The mid range spring does in fact affect high speed mixture. For that matter, the idle spring will to a degree. There is a cumulative affect on the high speed mixture, with both of the lower speed springs adding their spring pressure to arrest the outward movement of the flyweights. The idle spring has a greater affect on mid range than it does on high speed, but it will affect both.

This is kind of counter-intuitive, but the further apart the flyweights are allowed to spread, the leaner the mixture becomes. Remember, the idle mixture screw is "left is lean, right is rich", opposite of the main rack screw. The same holds true for the white and black, high and mid range screws in that little capsule - left (out) for lean, right (in) for rich. The more spring pressure keeping those flyweights from spreading, the richer (edit - thanks, Flieger!) it runs.

Anyway, my revised adjustments have the high speed still bottomed, but the mid range is now only one turn from bottomed, and the idle three turns from bottomed. These revised adjustments in the governor assembly have allowed me to lean the main rack out significantly, thereby leaning out the part throttle "cruising" mix and getting much better gas milage. Additionally, the full throttle mix now varies less than half a point across the entire rev range, from 2,000 to 7,000 rpm. That's an improvement over the former point and a half to two point variation I was seeing.

In closing, it seems to be very important to keep the three sets of adjustments within a couple or three turns of one another. Each individual adjustment is more effective this way. As a matter of fact, they become much more sensitive, with a click or two on the mid and high speed screws having a noticeable affect on the mixture. This also makes it a bit more difficult, and makes it all the more critical to keep good notes. It's a balancing act, a constant need to compromise, moving the governor screws more or less together, all the while having the main rack kind of ruling the roost. Remember, that main rack adjustment centers the range in which the governor adjustments have their affect.

[Flieger]

Originally Posted by Jeff Higgins

...the further apart the flyweights are allowed to spread, the leaner the mixture becomes...... - left (out) for lean, right (in) for rich. The more spring pressure keeping those flyweights from spreading, the leaner it runs...

So the mixture is lean when there is less spring pressure, allowing the flyweights to be further out.

There is a contradiction in your previous wording.

Thanks for documenting all of your efforts in discovering the workings of the pump.

[Jeff Higgins]

Originally Posted by Flieger

So the mixture is lean when there is less spring pressure, allowing the flyweights to be further out.

There is a contradiction in your previous wording.

Thanks for documenting all of your efforts in discovering the workings of the pump.

Oops... thanks for catching that. I edited that statement so I don't lead anyone astray.

[Jandrews]

This is really interesting, and I think particularly important because Jeff is running a bigger motor (3.0 IIRC). So many people say that these MFI pumps are "scalable", and will work with motors bigger than they were originally installed on. Well that is certainly true, and many will even run "good" in that configuration. However I believe what Jeff is uncovering here is that they will not be "optimal" until you get to this level of detail in your tuning.

Case in point: I have an original '73 2.4S motor (with the original 2.4S MFI pump). My motor has since been built to 2.7RS specs, with everything else remaining the same from the 2.4S. This motor absolutely howls, and has an extremely crisp throttle response and acceleration all the way to 7,300 rpms. However, I put an LM-1 on it recently, and tried multiple tuning attempts on the standard part-load and idle screws. No matter what I did, I could not get it to run leaner than about 11:1 AFR, even at WOT. My guess is that the relationship of the inner, mid and high speed springs that Jeff is adjusting, would change that, and allow me a much more "sensitive" adjustment range. This kind of "field research" and documentation is invaluable, and I appreciate the guys who dig deep enough to discover these nuances of the MFI system.

JA

[kenikh]

Thanks for writing this up! I think it is almost time to write a manual.

[jcge]

You've probably noted the "free play" in the capsule springs at rest...

The free play is central to the design of the "stepped curve" of the governor control springs - without free play to "take up", there can be no steps in the curve. (Exception is the IDLE spring which can be preloaded)

3 springs in parallel...IDLE, MID & HI (K1, K2, K3 respectively)

This drawing shows the interaction during adjustment of the screws.

The HI LIMIT is where the high speed screws have taken all the free play out of the high speed spring, thereby eliminating the MID to HI transition. (effectively goes straight from IDLE to HI) ("overwhelming the mid range" - as Jeff describes)

Similalry for the MID LIMIT - the MID screws have taken up all of its free play, and it effectivey eliminates the IDLE to MID transition point (kills the IDLE and goes straight to MID)

[jcge]

[356RS]

Jeff, you couldn't have picked a better space cam than the #55 "T" for your 3.0 MFI tuning. The mapping or contours that are formed in the T space cam are very minimal compared to the E & S space cams. In other words the formed ridges & valley's are almost unnoticeable compared to the E & S and even more, the RS. Pictures below show the comparison of the T, E & RS space cams. The blue "U" would be the approximate position of the "follower lever" on the space cam at 50% throttle and around 3000 rpm.



As you can see the T has lots of room for adjustment on the K2 mid range and K3 high end for keeping the flyweights from opening up and causing a lean condition. On the E space cam, you might run into a problem IMHO turning the K2 & K3 screws CW and cause the follower lever to run into a high spot, as you can see in the photo, and actually going lean. Even more so on an S space cam.
Your project worked out very well, Congratulations. Your patience and passion for tuning the MFI pump have provided all of us lots of information and data. Thanks

[kenikh]

I love this site.

[jcge]

Mark - great pic of the space cam contours....those poor engineers at Bosch and Porsche...it must have taken an eternity to map each engine, test grind, remap etc prior to the digital age.
Regards
John

[al lkosmal]

Mark,
the pix of the space cams are too cool! I have to see examples...or take things apart to eventually understand how they work. This helps.

thx,

al

[Jeff Higgins]

Great pics, Mark, and a great explanation of what happens as the follower travels fore and aft accross the space cam. I have a frozen up old 2.4 "E" pump that I have been using as my "sacrificial" learning pump, along with another operational "T" pump (that I'm not willing to tear quite as far into). I noticed the difference in the space cams right away - after all, isn't that one of the first things guys like us would look at? There is a lot of mystery and folklore surrounding that little guy...

I actually wanted to swap the "E" space cam into one of my good "T" pumps, assuming it would get me closer to being able to tune the bigger motor. Then I spoke with Henry at Supertec... He told me to keep the "T" space cam in it, as he routinely runs some very powerful 3.0 liter and larger race motors on the plain old "015" pump, "T" space cam and all. Your photo illustrates why this works - the "T" space cam does, in fact, make it easier to tune for oddball combinations, providing a much wider "sweet spot" in its profile.

The only downsides to not having a specific space cam tailored to a specific engine combination are some tradeoffs that must be made. My idle and part load cruise are a bit on the rich side, which I have found is necessary in order to keep full throttle, high rpm from leaning out. Granted, in a race motor like Supertec is doing this with, who cares... a street motor tuned like this probably isn't for everyone, though. As purely a hobby car, a weekend and DE fun toy, I'm willing to put up with it. The siren song of that old school MFI is just that alluring - it's such a large part of the personality of my car.

[dearangelo]

good day to everyone, I come from the direction of Mercedes, looking for knowledge about MFI pumps. I have an open pump on my bench and I need your assistance in locating replacement parts, mainly two things, plungers and the apparatus that goes with, and space cam, mine is worn a lot!
anywyas, it is 6 in the morning in Athens and I have spend at least three hours so far reading your incredible posts! I wish mercedes guys would care to mimik you in your effort.
regards!

[dearangelo]

Originally Posted by 356RS

Jeff, you couldn't have picked a better space cam than the #55 "T" for your 3.0 MFI tuning. The mapping or contours that are formed in the T space cam are very minimal compared to the E & S space cams. In other words the formed ridges & valley's are almost unnoticeable compared to the E & S and even more, the RS. Pictures below show the comparison of the T, E & RS space cams. The blue "U" would be the approximate position of the "follower lever" on the space cam at 50% throttle and around 3000 rpm.



As you can see the T has lots of room for adjustment on the K2 mid range and K3 high end for keeping the flyweights from opening up and causing a lean condition. On the E space cam, you might run into a problem IMHO turning the K2 & K3 screws CW and cause the follower lever to run into a high spot, as you can see in the photo, and actually going lean. Even more so on an S space cam.
Your project worked out very well, Congratulations. Your patience and passion for tuning the MFI pump have provided all of us lots of information and data. Thanks

hello, I simple and perhaps dumb question, are those valleys and ridges on the surface of the cams, wear? or that is the way they were suppose to be?

[Flieger]

The ridges and valleys are the coded information in the cam. It is a 4 dimensional "sapce cam", but the contours perform a similar function to the normal cam lobes on an engine's valve timing cams. This cam is "read" by a stylus, however.

[real550A]

This has been an enlightening thread, to say the least! Many thanks to
all who contributed. Bookmarked, for sure!