Wayne,
I was about to agree with you after saying "ah ha" while reading your email. And then I looked again at the charts. Remember that bottom scale on the MFI'd /21 is shifted left by 1000 RPM. It also doesn't help that the lines on the /20's graph touch (actually cross) at 5500 RPM. I suspect that this is why they changed the scale on the /21's graph

I'm eyeballing about 167 HP and just shy of 20 mkp of torque for the carb'd /20. AT the same engine speed, the MFI'd /21 is making 170 HP and ~20.8 mkp. Close, but the nod goes the the /21.

I think I'm going to try to map both curves on the same scale and convert the mkp to lb-ft. It might clarify things some.

Richard Parr makes a good case for carbs out-performing Mechanical injection. See:

http://www.pmocarb.com/sales_faq.htm

This is one of those debates without a right answer. I just like the looks of the carbs. The drive belt off the back of the cam always seemed a little awkward to me.

Here's the charts. The MFI'd 906/21 still wins out across the board.

http://forums.pelicanparts.com/uploa...1067000891.jpg

Methodology:
1) I took the torque number (in mkp) from each chart since torque is what dyno's measure. I converted back into HP at the end.
2) I converted from mkp to lb-ft by multiplying mkp * 7.2330139.
3) I converted from lb-ft of torque to HP by (torque * rpm)/5252

http://forums.pelicanparts.com/uploa...1067000808.jpg

http://forums.pelicanparts.com/uploa...1067000929.jpg

Observations:
- The HP curves are a little different then shown on the previous factory charts. But they both max out at the factory stated levels of 210 and 220 HP.
- Carb's (Weber 46's in this case) never come close to the MFI in terms of generating torque at any RPM.
- The gap between MFI can the carbs opens up even more at "low" and "high" rev's.

Conclusion: In absolute terms, when both options are set up correctly, MFI outperforms carbs in terms of the torque and hp output of an engine.

JGParker said
PMO has posted:
I'm not sure that I agree that he is making a good "case". I'd agree that he is making a good "sales pitch", but it doesn't supply any data at all, so it's really not a case. BTW - I don't behoove him for saying that, he's in the business of selling carbs. Certainly what he said is not 100% inaccurate either -- I would agree that a properly set up pair of carbs will outperform a poorly tuned MFI which is consistant with his statement. But to draw the conclusion that all MFI's are "too rich in the midrange" would be going a step too far.

Case in point,
check out Tyson's A/F mixture (the first chart) on his MFI'd T which is basically built as a Hi-CR 2.4E. Note that his engine is "lean and mean" at low rev's and peaks at a touch over 14:1 A/F at peak HP which looks pretty close to ideal (stoiciometric???) to me.

John,

I agree, a "case" is based on data, and all the data suggest that MFI is better.

I suppose it's fair to say that a lot of the people who have MFI don't have it adjusted properly, but you could say the same thing about any carburetor setup.

Do you think that the MFI's high injection pressure (220 psi) could result in better atomization of the fuel, better charge stratification, resulting in the power increase?

10 HP is a LOT, particularly when you net it against the parasitic losses caused by the MFI drive belt. Have you ever seen any data on how much horsepower the MFI pump consumes?

Why did Porsche use a Kugelfischer system instead of Bosch on the 935?

We need ROLAND for this thread!

I suspect that the MFI's higher injection pressure helps a lot with the low rev performance for the reasons you mention. HP has more to do with the delivery of air rather then fuel. Remember that for every pound of fuel, the induction system needs to ingest 14.1 pounds of air. That's A LOT of air. I think that the fewer restrictions of the MFI intake system are the big determining factor as far as the peak HP is concerned.

No, but I'd like to. Maybe Warren has seen the numbers somewhere.

According to Boschen's and Jurgen Barth's "The Porsche Book" (2nd ed) the 935 used a Bosche MFI system. BMW seems to have always used the Kugelfischer system. Per Boschen and Barth The 911 Turbo Carrera RSR (as driven by Derek Bell in the video) also used the same system. The 936 used the same system also.

John,

Check out the photos in Bruce Anderson's book of the various 935 engines, many of them use a Kugelfischer pump. I remember being surprised at that, as I always associated Kugelfischer with BMW (from the fabled 2002 tii, among others) but it's there, in the pictures. There are also a few references to "Kugelfischer" engines at the Gunnar racing web site, where Kevin describes the fuel pumps that feed the system, but unfortnately no good engine shots.

I'll check Bruce's book over the weekend, maybe he gives a reason.

update
 

After reading all your replies, I decided that it was time to swap engines. Out will come the 2.4 (w/o serial #); in will be a 3.2 from the PCA classified ad located a few towns away.

What do ya think?

refugee

If the 3.2 is in good shape, you'll have a real screamer for about the same cost as a rebuild on the stock 2.4. If the 3.2 is NOT in good shape, you'll have a real screamer for over twice the cost of a rebuild on the stock 2.4.

Life's a gamble!

Roger