MFI Open Heart Surgery III - Reworking the Space Cam
 

Ah yes, my next big adventure in MFI tuning is now complete. I've successfully re-contoured my stock "T" space cam, the #55 from the 015 pump, to more closely match the fuel delivery requirements of my oddball 3.0. Don't get me wrong - it's ran great, and I've driven it with full satisfaction on the stock profile for over three years now, but I always felt it could be better - the A/F ratios had a wider spread than desired, like almost two points from richest to leanest. I finally decided to do something about that. And, the good news is, once I dove in, it really wasn't hard at all.

I began by carefully "mapping" the space cam. "A picture is worth a thousand words", so here is two thousand's worth:

http://forums.pelicanparts.com/uploa...1309562392.jpg

http://forums.pelicanparts.com/uploa...1309562417.jpg

The idea was to simply use a dial indicator to measure the peaks and valleys of the profile, using the idle position as "zero". The first photo show this "zero", and the second shows an adjustable parallel bar used to shim the cam up to the middle, and the cam rotated to full throttle. I hope they convey the idea - shim and rotate, them measure and record the deltas from idle, like this:

http://forums.pelicanparts.com/uploa...1309562515.jpg

Like the notes on the sheet say, negative numbers - where the profile is cut deeper than the idle position - denote richer, or more fuel delivery. Positive numbers, or the high peaks on the profile, denote lean, or reduced delivery. Note that fuel delivery increases with throttle, but decreases with rpm.

My goal was to simply reduce or eliminate the tendency to lean out at high rpm / full throttle. By doing this, I reasoned, I would be able to re-set the main rack to a leaner position, providing leaner running at lower rpm's and at cruise. So, in light of that, note how little I reworked the space cam - only in the area of interest, and very little at that. .10 mm - one tenth of a millimeter - makes a significant, measurable difference in delivery. I'm told the whole range of motion of the stylus is only 2 mm, so .10 mm is a 5% change.

Needless to say, this is pretty fine, tedious work. I trashed my first effort and ruined a space cam by going for the gusto and taking .2 mm off the high end right off the bat. It ran so bloody rich I couldn't use it. My second effort took a couple of days, removing .05 mm at a time (on a different space cam) and re-installing it so I could drive the car and get some A/F ratios from my trusty LM1.

I'm elated with the results. A/F at WOT now stays in the 12.2 to 12.8:1 range at all rpm's. Freeway cruise is about 13:1, and idle is about 14:1. This is an improvement from an 11:1 to 13:1 range at WOT, and a freeway cruise of about 11:1. Gas milage is up several mpg, and I don't burn my buddies' eyes out when I'm in front of them and getting on it.

Anyway, I hope this is of some help to you other guys that are back to messing around with MFI. I'm now satisfied that it can be pretty nearly optimized for any oddball combination of displacement, cams, compression, heads, and whatnot that a guy might like to try. Cheap, readily available wideband O2 sensors and A/F meters have taken the mystery out of a lot of this.

Wow that is cool. What did you use to recontour the space cam?

I really enjoy learning about MFI from your threads. Sorry to hear that you may have wasted one of your space cams. Why is it that no one seems to have cracked machining new ones?

Brian

The first cam (the one I ruined), I went after with a very fine stone in a die grinder. The second cam I just sanded with some diamond impregnated foam blocks, a 3M product they call "Diapads" (diamond pads). That made it much easier to slowly creep up on it, all the while checking progress on my little jig.

It shouldn't be all that hard for someone, in today's CNC world, to whittle out new space cams in all of the old profiles. It likewise shouldn't be that hard for some bright young man to model one, and recontour it in the digital world to match delivery requirements for custom applications. I just don't think there is the demand to warrant the effort.

Damn...just when I thought I really didn't need a CNC mill.
Now you have me thinking about spending money again....LOL.
Bob

Jeff,

Awesome stuff, I've been doing some work on an original RSR pump and loving every minute of it. Here is a picture of an RS cam on the left and the RSR on the right (note how much thinner it is the the T/E/S/RS cams). I can take measurements of the RSR cam if your interested. I know one of the guys I work with has modified cams by silver soldering and grinding by hand to achieve desired fuel delivery.

http://forums.pelicanparts.com/uploa...1309572535.jpg

Awesome thread! :)

Makes me want to get a wide band O2 sensor and dive right in.

Jeff:
Nice work ! You have way too much time on your hands :) I admire your skills.

The Germans really seemed to have a flair for MFI systems. And machining and mechanical stuff in general. Here is something a little off topic that I didnt know the Germans came up with; Rod end bearing - Wikipedia, the free encyclopedia

I just found this out a couple years ago.

This borders on the amazing.

Great Work Jeff!

I wish I was better with CAD or I would give it a shot. If you want to send the ruined space cam this way I am game to try and get it in a digital format.

http://forums.pelicanparts.com/uploa...1309736087.jpg

Old skool MBZ map...

Some complexity is in measuring the flyweight characteristics (fore/aft displacement vs speed with the control springs set at their deisgn baseline) for the standard and lightened flyweight sets

wow. nice work. makes me feel really dumb...

Surface hardness only?
 

This begs a question I've had on another subject... Since the space cam only needs to be hard enough to resist wear from the stylus, on should be able to 4 axis in aluminum then drop $20 to have it DLC coated, right? It doesn't have to be hard all the way through, just the surface...

t

start by having the space cam digitised - easiest way to 3D map it then off to CNC or then modify and CNC.

Am running carbs but friend is forever tinkeirng cussing over his MFI so am getting sucked ino to the worl and now beginning to wonder what if...

On a side issue does anyone have a set of MFI protractors for sale? Or dimensions so that a set could be made?

Investment casting? Make some latex molds of the modified one to make a wax pattern, then use lost wax casting.

Excellent work Jeff. Very important to know exactly where on the space cam you want to alter. If you cut too deep, that's it, no going back. You can only remove surface material for increasing the fuel delivery. Can't add it back to lean out things. I index the space cam at every 500 rpm and every 10 degrees throttle up to the 80 degree WOT.

http://forums.pelicanparts.com/uploa...1309810770.jpg

I also document the actual flow readings at every 500 rpm and 10 degrees.

http://forums.pelicanparts.com/uploa...1309810913.jpg

Then using a tool similar to yours, locate the exact spot or area on the space cam to be modified.

http://forums.pelicanparts.com/uploa...1309811068.jpg

Time consuming yes, but you end up with a space cam re-shaped to your engine specifications. It's not as easy as it looks.

Sheesh, doesn't even LOOK easy Mark.

I know where you can get one!!

Hello guys (Mark and Jeff ;) )

Since i'm on the way to tune a 3.0 MFI engine, i started to map some space cams to spot where i need to act ...
I've modelized the datas to see what the profile looks like (ok, my eyes are enough !)
Here are the pics :
sensor travel is rather small (depending on the cam type) (around 1.5 mm), we can clearly see where fuel is needed ... (low point areas) ...
I've collected up to 150 points, and hope to rework my space cams exactly where needed :cool:
I've a RSR cam too, which shows agressive fueling !

http://forums.pelicanparts.com/uploa...1312921584.jpg


http://forums.pelicanparts.com/uploa...1312921616.jpg

Philippe

Very cool to stumble into these threads that are way over my head =)

Jamie, I think you could manually digitize the space cam by taking your data points by just using the dial indicator and use a rotary table assuming you can get the axle position zero'd. I don't know how fine a mesh you would need though.