Project Heavy Metal - an all steel classic, remastered

[ian c2]

Originally Posted by Jonny042

g]

Gas gas ?

I used to ride sherco when I was back in England .
When I say “ride” , I mean mainly falling off .

[Jonny042]

Originally Posted by ian c2

Gas gas ?

I used to ride sherco when I was back in England .
When I say “ride” , I mean mainly falling off .

Yup!!!

That bike is pretty much the reason there aren't any updates on my project between May and October..... there is a small group of us riding up here and we try to get out 2 or 3 times a week. It's a blast.

I can't figure out why it's not a more popular sport in North America.

[Jonny042]

Originally Posted by Jonny042

My lack of updates doesn't necessarily mean a lack of progress, lots happening, but not much to share. Been fiddling with MFI stuff which I could write a book about. Luckily someone else has done that. I highly recommend:

PORSCHE 911
MECHANICAL FUEL INJECTION PUMP
1969-1973
A DIY STEP-BY-STEP GUIDE
Repair and calibration of the PED 6 KL... Mechanical Fuel Injection Pump At Home

By Michael J. Burgess

The book is every bit as detailed and descriptive as the title. I'm in the rebuilding phase, yesterday started re-assembling the pump. I mixed up the smallest batch of Gaurds Red Single stage ever and marked the middle of the 6 gear segments with my Loel-Corning Paint pen (a wonderful little device!)

Well........ todays update is: DON"T DO THIS TO YOUR MFI PUMP!!! Unless you want to take it apart and put it together again for a third time.

The clearances and tolerances of the MFI pump are incredibly precise. In particular, the rack and gear segments that rotate the control sleeves have very tight tolerances and ZERO backlash.

Even a single layer of paint sitting in the root of the rack is enough to jam up the rack and keep it from moving smoothly. I ended up removing the paint, cleaned up the rack, and used a red sharpie to mark the tips of the two teeth on either side of my original paint mark.

At least I am getting good at taking the pump apart and putting it back together! I did get the crank/cam shaft installed and was able to test the pistons by flooding the fuel supply port with WD40 and turning the crank/cam. SUccess!

Tomorrow I plan to pressure test the pump head before going too much farther with the governor and compensator sections.

All part of the fun!

[Jonny042]

Well the pump went together without too much fuss, even though I was freaking out a bit when I laid out all the parts!!



I installed the RS Space cam I've had sitting on my office desk for a few years now. Nice to be finally giving it a home!



Governor section almost complete.



Hopefully in the next few days I can get this thin running on my pump dyno and making some flow measurements! Lots of fiddling around to get there and likely a few trips to the hardware store since the dyno isn't built yet.

[jpnovak]

You may have mastered the inner workings.

Now... Solve the 3D mapping on the space cams while you are in there.

[Jonny042]

Originally Posted by jpnovak

You may have mastered the inner workings.

Now... Solve the 3D mapping on the space cams while you are in there.

I'm pretty confident I know and understand, 100%, how this thing works so I am sure I can get it as close as the space cam map will allow. It's like having an EFI system that's pre-loaded with a single map that you cant adjust except for fuel pressure. Well, maybe a little better than that, but not much!

I've been trying to balance the intake sizes to the intake port size (39mm) and displacement (3.2) of the engine and the camshaft choice to take best advantage of the 2.7RS space cam. I have anecdotal evidence from several sources that the 2.7RS space cam should serve well. I also chose (with the help of John Dougherty) cams to match the space cam, which is a bit of a backwards way of doing it but he has enough experience with it to have some confidence that it MAY work.

In the end, with his guidance, I chose DC43 cams on a 102 lobe center. Basically they are mod S cams with a bit more lift. Of course I asked him for the MFI drive on the left cam which comes at no extra charge with the billets.

Exciting times. Been cutting tubing to build my pump test rig all evening. Might have something for show and tell by this time tomorrow.

[Hcarraro]

Great project. I am late to the thread.

Henry

[lvporschepilot]

Great progress. Looks great. I tried a custom space cam from a German company that makes them, new. Very similar to RS but tweaked a bit. Run in a 2.8 twin plug setup. Wasn't cheap but it turned out excellent.

[Jonny042]

Originally Posted by lvporschepilot

Great progress. Looks great. I tried a custom space cam from a German company that makes them, new. Very similar to RS but tweaked a bit. Run in a 2.8 twin plug setup. Wasn't cheap but it turned out excellent.

Thanks! I am trying to keep it moving forward.....

That's great to hear about the custom space cam - I assume that you are referring to Eisenbrandt? They are my plan B. Well, maybe plan C or D if I can't tweak it myself. I wasn't sure if they would do a space cam to order without doing the mapping themselves. They actually have an electronically controlled master pump that they use to map an engine, much like you would an EFI system, then transfer that map to a space cam using CNC machining.

It strikes me as a really neat mix of old and new technology, to do that. I plan to have data-logging on my 45 year old MFI system for RPM, throttle position, and AFR to help me tune it, so I would probably be able to put in a pretty good order for a custom map.

[lvporschepilot]

Originally Posted by Jonny042

Thanks! I am trying to keep it moving forward.....

That's great to hear about the custom space cam - I assume that you are referring to Eisenbrandt? They are my plan B. Well, maybe plan C or D if I can't tweak it myself. I wasn't sure if they would do a space cam to order without doing the mapping themselves. They actually have an electronically controlled master pump that they use to map an engine, much like you would an EFI system, then transfer that map to a space cam using CNC machining.

It strikes me as a really neat mix of old and new technology, to do that. I plan to have data-logging on my 45 year old MFI system for RPM, throttle position, and AFR to help me tune it, so I would probably be able to put in a pretty good order for a custom map.

Correcto! Eisenbrandt worked out very well. I just bought the space cam from them, which they had immediately ready to go. I did not know about their analogue meets digital mapping system, very neat. Mine still required some data logging as you mentioned above, but I only did a bit on a dyno. Took a couple days but it eventually dialed in just fine.

[Jonny042]

Originally Posted by lvporschepilot

Correcto! Eisenbrandt worked out very well. I just bought the space cam from them, which they had immediately ready to go. I did not know about their analogue meets digital mapping system, very neat. Mine still required some data logging as you mentioned above, but I only did a bit on a dyno. Took a couple days but it eventually dialed in just fine.

They've posted once or twice in various MFI threads on the forum - here is a pic they posted of their "master pump".



One of my earlier ideas (before I went all-in on the MFI) was to replace the governor and compensator sections of the pump with a servo motor and control it with a microsquirt EFI - the programming of the efi would be the same except for injector pulse width 0-100 you'd have rack position 0-100% output. Throttle position would need a sensor but all the other governor and compensator functions such as speed, temp, barometric pressure, even cold start enrichment, decel cutoff, accel enrichment, and even closed loop lambda control would be a peice of cake.

Hmmmm........ I must not get distracted!!

[Jonny042]

I've been working on the MFI pump test rig (pics soon I promise) and also buying bits and pieces. New cog belt pulleys are self explanatory, but the red cap is a necessary change to the pump as I plan to isolate the oil supply. The cap is actually a breather/filter to allow for temp changes in the pump and keep it from pressurizing. I'm going to adapt it to the pump on the small triangular cover plate at the top of the cold start portion of the compensator casting.



I don't want to turn this into an engine build thread - I could go on forever! But a point of note I ordered some Glyco aftermarket GT3 bearings part #99610312194 with the intention of measuring them and returning them if necessary. I'd read so many bad things about the Glyco rod bearings that I didn't have high hopes but was pleasantly surprised:









I compared these to the nearly new looking, but used (approx 10,000Mi), #1 rod bearings and these are all within .0001" of the shell thickness of the used bearing which measured out on the low end of the clearance allowed (which is where I want to be) - There were two outliers in the measurements, two shells that were maybe .00007-.00008" thinner than the rest but were are talking parts of tenths of thousands here which is ridiculous. For practical purposes you can only make those sorts of measurements relative to each other. Point being these bearings are top notch and I'll be more than pleased to use them in the motor.

[Jonny042]

Well something that's long been on my mind is the electric fuel pump to feed the MFI system, and specifically the MFI fuel filter console. At first glance the fuel system feeding the MFI pump seems overly complex:



This post focuses on the thicker lines in the picture above. Even the arrangement at the fuel tank seems overly complex at first, but it's not quite as bad as it seems. The original style fuel pump had a third nipple (I'll come up with a limerick about this later) that was a return path for the internal pressure regulator. So that can be ignored, or, indeed, replaced with a "regular" pump.

Now a bit of a disclaimer, this is all very temporary, and will be improved upon later... I just wanted proof of concept before I made any more commitment to part locations and cutting actual fuel lines, etc:

My "normal" fuel pump feeding the MFI pump on my test rig (see above disclaimer!)



It's a run of the mill Bosch pump for a 84-89 3.2 Carrera. I chose it because it is readily available, inexpensive, and has a plain hose attachment on the suction end that matches the diameter of the tank filter screen outlet on the early style tank. Although it will produce much more pressure than called on in this application, it doesn't have to.



I was pretty elated when I hooked everything up in the factory configuration and the pressure at the MFI pump head was 12 psi or so. I tested the pressure at each point in the system and there were no surprises (see dry break test fitting on the banjo inlet of the filter console):



I initially used clear plastic tubing so I could see if I was getting any cavitation or air bubbles, etc. - I added some red colored fuel injection cleaner to my test fluid so that's why the hoses look pink:





The test fluid is more or less non-flammable, and somewhat low odor - it smells suspiciously like mineral spirits.

So that's that. But it still doesn't explain the fuel filter console and why it's so complicated. Why does the fuel go in, and out, to the MFI pump and back, then through it a second time? Turns out the passages in the console are connected somewhat. I always assumed this had to do with supplying high pressure, fresh, fuel to the cold start solenoid:

[Jonny042]

Something that's never sat right with me is that the early style tank has no provisions for capturing fuel, like a swirl pot, sump, etc.

I did spend a lot of time researching various options like adding an in-tank pump, etc. but decided to give it a go with the factory setup.

The problem I envisioned was that if the pump starved for fuel, it would be a problem. Carburetors are a bit tolerant of this since there's a small amount of fuel in the float bowl but I couldn't envision how the MFI system would deal with this.

I tilted my (not yet secured) test rig fuel cell so the pump would start to suck air and to my surprise the resulting bubbles went through the pump, to the filter console, and back to the fuel cell. But were NOT delivered to the MFI pump head. Even when startved to the point that the fuel lines were totally frothy, the fuel delivered to the pump head was clear and air free, as seen in the pic below:



So finally a light went off, that the filter console serves the secondary purpose of removing air from the fuel supply! As near as I can tell it works like this:

- Fuel (with air bubbles) is delivered to the filter console and first must go through the filter in a downwards direction.

- Due to the greater cross sectional area of the filter, the velocity going downwards is going to be very low, so much so that the bubbles will float to the top of the housing.

- There is a screen in the filter console that connects the cavities between the filter and the fuel return path - basically the fuel enters the housing, makes a 90 degree turn, then exits through the pressure regulating fitting and back to the tank.

- The differential pressure across that screen is equal to the pressure drop across the MFI fuel pump head, and the hoses and fittings it goes through to get there. This pressure will drive fluid flow through the screen at all times but it's a small amount compared to the total flow.

- One of the banjo fittings (above) has a built-in restrictor, presumably to "boost" the DP/flow through the screen. I always looked at it and wondered why. And indeed wondered if it was installed on the right port of the console.

- Should air become present at the screen, it will pass through much more easily than the fuel. and is swept away and back to the tank.

It's actually a pretty brilliant little bit of engineering! So many times I considered getting rid of it in order to simplify the system, but I now know that would be a bad move.

OK, I have an MFI pump to calibrate!!

[jpnovak]

Cool find/discovery on the filter console.

[Jonny042]

Porsche added a small restrictor in the return fuel line for 1973, apparently:



My fuel pressure is on the low side of the .8- 1.0 bar spec required, so I might add a bit of restriction at the tank return. Pinching the return line gives a pretty good idea of how that will work, and I should be able to dial it in to where I want it, about 15psi. That will have to wait until everything is assembled in the car.

[Jonny042]

Originally Posted by jpnovak

Cool find/discovery on the filter console.

Thanks Jamie,

It's a nice feeling to have all this stuff working out as planned - should be downhill from here with the MFI system EXCEPT for the tuning

[BURN-BROS]

Hi Johnny, system fuel pressure is set by the Filter console check valve. Part #901 110 915 01 I am not aware of any other restrictor in the return system. Can you let me know what it is and true location?

Also, FYI..the filter elements are not doing their job in regards to removing foreign debris! I have had to install multiple add on filters to solve the issue.

[Jonny042]

Hi Aaron,

Thanks for checking in, I appreciate it.

I priced one of those "overflow valves" as Porsche calls them. At $161.00, I'm glad mine is working well! My backup plan was to hide a modern regulator under the car.

This post by Early S Man describes the restrictor. Basically a 1mm orifice in the location I showed in my etch-a-sketch schematic markup.

Also, one of the banjo fittings (return from MFI pump head) on my fuel filter console has been brazed up and drilled a bit smaller (guess maybe 2mm) as described in my earlier post.

I was playing with the pump dyno tonight (more on that later) and I wasn't able to get the feed pressure to drop at all even at some ridiculous delivery volumes, so I am going to leave well enough alone.

Thanks for the tip on the filter, that never occurred to me! I'll be sure to add one at the suction of the pump up front.

Honestly, it's an honor to have you checking in on this thread! I know you put together that GORGEOUS MFI powerplant for STR 002 and it has been a big inspiration to me. I keep watching the Jay Leno video and hope I can get mine to run half as well! Any tips on tuning the MFI pump? I've got datalogging for throttle position, MFI rack position, AFR, and RPM planned. And I have a good idea what the WOT/full rpm fuel delivery should be.

I'm using the 2.7RS space cam, DC43-102 (Kinda like Mod S+) 39mm/36 ports and waiting on the new 41mm SSI heat exchangers. Might temporarily use ebay 1-5/8 headers if there's a wait on SSIs.

Last but not least I'll be installing a classic retrofit CDI+ so I can program the curve I want.

Thanks again!

Originally Posted by BURN-BROS

Hi Johnny, system fuel pressure is set by the Filter console check valve. Part #901 110 915 01 I am not aware of any other restrictor in the return system. Can you let me know what it is and true location?

Also, FYI..the filter elements are not doing their job in regards to removing foreign debris! I have had to install multiple add on filters to solve the issue.

[Jonny042]

Well, Rube Goldberg would be proud of this contraption!

I couldn't be happier to be finally testing my rebuilt pump! It's been a long time coming, researching and accumulating bits to make it work. The biggest thing was the 1-1/2hp, 600 Volt 3 phase (575 in US?) and the variable speed drive, but lots of other do-dads and fittings. Runs like a charm.

Spinning the pump up to 4000RPM and cranking the fuel flow up to max is pretty exciting.



Lots of fun ahead. Still need to make a tool for setting a lever angle inside the governor, and also a set screw thing for faking the temperature compensator into thinking the motor is fully warmed up.