Oh great gods of MFI...

tadd · 05:46 PM

please take pity on a humble acolyte!

I got the pumps out of the housing and cleaned up tonight and have a couple of questions. There are two different part numbers for the rods 004 (4/404) and 030 (1/303) with each bank (row) of three being the same (i.e. 3, 2, 1 or 6, 5, 4 as called out in the MFI pdf). All of the pump 'bodies' are listed as 90. Should this be? Do the part numbers have meaning (i.e. are they bore sizes or something)?

I know someone has been in the pump, and was rather ham-fisted as the two of the 'guide collers' (what the rack gear clamps too - the pdf calls them 'control sleeves') are pretty beat up (looks like a screwdriver as a prybar), two have light gouging, and two are untouched. Can these pieces be bought or do I need to just 'dress them' and get on with life.

I bought this on ebay as an educational experience, but I am sure the deal itself was shady anyway as three of the 'caps' on the pumps were loosened so the drive could turn over, even if it was 'lumpy'. When I snugged them back down, the pump wouldn't 'turn over' at all.

So I have no idea what 'should be' inside.

Any help would be really appreciated!

tadd

Scott Clarke · Scott Clarke's Garage

Tadd-
I can't answer your question, but it would be great to see some pictures of these parts. If you have time, I think that many of us would like to see them. I commend you on your educational endeavor.
-Scott

tadd · 05-16-2006, 07:35 AM

I have been taking lots of pictures and notes.

I figured if this works out I would do a pdf. Something like: 'DIY MFI or What You Are Paying Big Bucks For'.

It may be cause I am a macintosh user, but I have never been able to get pics on pelican up. They upload fine, but don't show up in the post. I've tried all sorts of browsers. If I knew the pelican URL they upload to, then I could call that manually. My home box finally died and I have not replaced it yet.

I'll See if I can pester Kenikh to put some up for me if you are really interested.

tadd

Scott Clarke · Scott Clarke's Garage

Tadd-
If you email them to me I'd be happy to post them.
-Scott

kenikh · 05-16-2006, 08:26 AM

Here are the pics.

The Good:

The Bad and Ugly:

kenikh · 05-16-2006, 08:27 AM

BTW, Tadd, have you tried using photobucket, then linking to the photos from there? www.photobucket.com.

kenikh · 05-16-2006, 01:01 PM

More from Tadd:

Injector numbers: do they mean anything??

Crankshaft (camshaft?)

nabilious · 05-16-2006, 02:02 PM

I'm not certain but I bet those numbers must correspond to the sizes, Like a carb jet. Just a hunch.
Nabil

nabilious · 05-16-2006, 02:03 PM

....and also very thankful that you opened yours up. I've been curious for some time and have debated opening up my spare one. I'll be following closely!!
Nabil

tadd · 05-16-2006, 02:10 PM

and tossed a caliper on the holes drilled in the piston and they were both within 0.01mm. Doesn't seem to mesh with 004 and 030 which should be an awful big jump. It may have to do with the angle of the slot at the top that controls the fueling as the plunger is rotated...

Tadd

Jandrews · 05-19-2006, 06:13 AM

Correct if I'm wrong, but is this the FIRST photo-documentary of a rebuild of the mysterious MFI pump??!! Tadd, thank you very much for jumping into this and posting your exploits here! We have learned a lot here over the past few years on MFI pumps, but I don't believe anyone has actually tried the open-heart MFI surgery DIY until now!! Please keep this going, and we will follow closely!

Thanks,

JA

tadd · 05-24-2006, 06:55 PM

this post has simmered. No one can help me on what these numbers mean and why each 'bank' of three should be different?

tadd

nabilious · 05-27-2006, 11:18 AM

Tadd, your right, you got all us wanna-be MFI guys here but the real experts are guys like Grady Clay or Henry Shmidt. You should get henry's attention as if I;m not mistaken, he has rebuilt these.
Lemme know what happens.
Nabil

tadd · 05-27-2006, 11:24 AM

I was kind of hoping for one of these folks to chime in.

I specificly tried to avoid any kind of question that would dent anyones bussiness repairing these things cause I respect a way for someone to put food on the table.

If worst comes to worst, and no one wants to answer, it just means buying a few more pumps and stripping them to see whats inside. Just money and time.

What I do need to find is parts... like a gasket kit...

tadd

NevenM · 05-27-2006, 03:11 PM

Hi

Since the control rack goes down the centre the left and right banks would be left and right handed (you can check this for us) The 'twist' on the plunger for example, The pumps are a precision instrument and the 3 critical things are the plunger to body fit, (this run in gas which is 'dry' ie non lubricating unlike diesel) The non return values on the top and the injectors. Gus i'm pretty sure can rebuild any pump, I just depends how
many parts need replacing, The chances of you doing it nil..The setup rig is quite extensive

HTH
Neven

tadd · 05-27-2006, 06:08 PM

then I'll never know, will I? I don't see some ginormous test rig in the factory MFI manual. it calls for a test under pressure and under vacuum. Now I can't belive that it is anymore difficult to test 'running' than an electric motor to spin it and a precison mesuring device to get a volume/time-rpm figure. LOOK at the pictures above. That is a COMPLETE pump subassembly. There ain't that much there to fiddle with - just the control sleeve. The regulating parts all have specific measurements called out in the MFI manual. That's it.

Now as for fuel lubrisity... Are you lumping #1 sweet in with #2? Cause just on the thin film strength of the given hydrocarbon length #1 diesel has almost the same value as gasoline - and a big portion of that difference is in the sulfur content, so by the time the phase 3 sulfur regs are fully in force, I can see #1 not any different.

They did build these things (MFI) in WWII so it can't be that complicated. Some pitfalls, perhaps, but nothing a bit of thought can't sort. I build multi-million dollar next generation lasers for a living - I got a good price on the pump and figured 'why not try it'. Becides, wouldn't it be nice to have a 'new' manual to do this incase you wanted to? Does everything have to handed over to someone? IS porsche ownership just about spending money? Is the trip no longer a worthy adventure. Should I just stand around a car someone else built, drink beer, and talk about how cool porsches are?

I guess you pushed my buttons... sorry.

tadd

NevenM · 05-27-2006, 06:34 PM

Tadd, Hey no problem

I'm also a guy that doesn't think Porsche ownership is bout money (see attached pics of my 73 resto). You may well synch the plungers but recalibrating the control circuit another thing (historically it appears even adjusting the pump seems to beyond most people which is why so many are for sale!)

My pet project would be to hack a megasquirt to control a servo to control the rack, 70's look but 21st century control

Neven

davidppp · 05-28-2006, 01:41 AM

Tadd, Of course you are right.

Its ineveitable that the knowledge about these pumps will eventually reach the public domain..even then, I'll bet that 99% of us will still send them out to Henry etc to be done..

Its just not possible to be an expert in everything.

I still understand why guys who have a big investment of time and money in finding it out for them selves are not keen to give it away for free..and I know you do too..

Neven, yes, The MS-controlled servomotor idea had occurred to me too..
Any firmer thoughts on that?

Kind regards
David

tadd · 05-28-2006, 11:31 AM

NevenM, let me just say that when I figured out that that was the floor seperated from the car like a giant can of tuna... well... may I prostrate before you in worship? Talk about cajones!

As for the MFI pump... I assumed that there were a bunch fixed in the mud of the eastern front. I mean, they couldn't just send back to the factory to get that ME 109 up in the air... Maybe I am fooling myself, but it will make for a nice puzzle.

Davidppp: I usually try anything once for grins. After just finishing up a second set of heads for a second plug and boring of the ports, I see why people charge 500 bucks for this. If I get this MFI pump to work right then I will pdf the info and those that don't will have a clue as to what they are spending on. Running the end mill thru those fins at that angle is nerve racking (even with the wax).

Speaking of heads... I pm'ed you (from a thread on the early board I think) about some heads you had 'extra'. Might you still have them?

Kenikh and I have also tossed this MS-MFI hybrid thing around a bit - might I respectfully suggest instead of controlling the main rack, just replacing the barometeric contol? That way you don't need to have speed and precision. There are very nice, near nm precision servos for optical mount control for about $300. At first blush, these would seem to work with the direct MS output of a fixed voltage, variable width pulse. Just set the pulse width for the ammount of rotation needed out of the servo. What I don't know is if MS can reverse polarity so the motor can reverse. This way you could have the benifits of closed loop with the robustness of mechanical injection AND not need a very unique actuator.

Thoughts on this?

Tadd

NevenM · 05-28-2006, 12:36 PM

Guys

I've got a pump and an ms but thats as far as I have got as I have a number of other projects on the go. I am pretty sure that all mfi pumps use pretty much the same metering plungers, but I don't know if the rack pos to fuel delivery relationship is linear (not that its important, as you can correct for this), It would be nice to know though.

David, Hacking the ms code to control a servo is relatively simple, changing from an event triggered pwm (the crank/ignition sensor) to a fixed timebase, There are numerous servos that wuold do the job (I'd go for quite a large futaba with metal gears), I ewas going to gut the control end of my pump and put the servo in a sealed box (this is a lot of oil flung around there)

Tadd, The problem with not directly controlling the rack is that because there is not a direct correlation between servo position and delivery you could not do open loop control, efi uses open loop for everything but economy (where the wbo come into play), ie at part throttle and high revs.

If someone want to collatorate on this further I'm happy to look at the ms code.

Neven