Alfa Romeo Forums banner

1 - 5 of 5 Posts

·
Premium Member
Joined
·
13,987 Posts
Discussion Starter #1 (Edited)
I have had Shankle headers on my Spider for over 20 years with no problems. My current motor makes 155 HP at the wheels at 6500 rpm and 135 ft-lbs at 5700 rpm 79 Spider on a dyno - The Sequel

The Shankle collector was opened up for a 2 1/4" tail pipe and it has the glass pack type mufflers - a big Borla in the center position and a small generic one in the tail section. The effect is to have what looks like a long straight pipe to the acoustic waves. The dimensions are: primaries 16", 18", 18" 20" for cylinders 1,2,3,4 respectively, 23" secondaries and effectively 108" tail pipe.

From reading and reviewing race car header information with Jim Steck and Steve Johnson (Lola Formula Ford) I decided that my headers were tuned for about 4700 rpm which makes sense for a stock motor but not for mine. So I cut 9" off the secondaries and added a 9" piece into the tail pipe to maintain the overall tuned length. I have data logged AFR and acceleration times for the two systems and the results are not what I expected. My original torque curve had a modest dip at around 4500 rpm due to a rich dip in the AFR. I tinkered with carb jetting and modified the emulsion tubes to minimize the AFR dip but I did not take the car back to the dyno. After the exhaust modification I now have a much bigger AFR dip that extends to higher rpm and appears to reduce power between about 4700 rpm and 5700 rpm but improves AFR below 3500 rpm and above 5700 rpm. For the full 3000 to 7000 rpm run the loss is greater than the gain and the total time is slower. It seems to me that acoustic wave effects are causing significant intake reversion and hence low AFR's in the region of the dip.

The orange curve is for 23" secondaries and the blue curve is for 14" secondaries.

It seems likely that I will return the headers to the original specification but I may try carburation changes first.

I would like to hear from anyone who has a good knowledge of 4-2-1 header tuning who can either confirm my thoughts or give an alternative explanation.
Thanks.
 

Attachments

·
Premium Member
Joined
·
13,987 Posts
Discussion Starter #2
I put the header configuration back to the original Shankle spec. I guess that John knew best!

Here is another copy of the comparative AFR curves. The orange highlited line is for the stock headers and the un-highlited line is the modified headers with shortened secondaries.
I drew in a horizontal "DMZ" between AFR 12.5 and 13.2 where I expect the two systems to be equal.
The blue shaded portion shows where the modified headers are "less too lean" than stock.
The red shaded portion is where the modified headers are "more too rich".
The small blue section from 5900 to 6400 can probably be corrected with a smaller air corrector and I will verify that soon.
It is easy to see that the large over rich area from 4600 to 5700 with the modified headers easily out weighs the gains at around 3200 and 6200 rpm
001.JPG
00 rpm
 

·
Administrator
Joined
·
12,540 Posts
Extremely interesting and I do not know the answer, but ...

25 years ago I built my first set of extractors to the sizes that my engine builder suggested. He pulled the sizes out of his head, and I didn't know any better so off we went. The engine made power but had a massive flat spot when it came on cam due to what I believe was fuel reversion. The next time we built this same engine, it dropped a valve due to cam follower housing wear and me being stupid in thinking "I'll just to one more race ... ", he had access to a computer program and used that to size the primaries, length of primaries, length of collector, length of secondary pipe until either the pipe ends or it is stepped up to a much larger size. The primaries were smaller and considerably shorter than the previous extractors, and the motor had a dead flat torque curve. So flat in fact that when it came on cam the only thing that really happened was the exhaust note hardened. You did not feel any great extra push or anything, but admittedly the car only weighed 512 kg's so 200 hp was not pushing much.

So if I was you, now that the tuning world has matured by 25 years, I would locate a shop with a program they trust and see what numbers you get. You might find that yeah if you change the length here and there you will make improvements, or you might find that the pipe sizing is all wrong and there are huge gains to be made by making a completely new system

Best
Pete
 

·
Premium Member
Joined
·
13,987 Posts
Discussion Starter #4
Hi Pete,
My understanding is that the programs do a decent job of designing a 4 into 1 system but not 4-2-1. The Nascar guys run essentially a pair of 4-2-1 systems on the two banks of the V8's and the word is that the primary lengths are not critical and the secondaries have the most effect and the teams do not share their knowledge. They are said to have different headers for different tracks.
My torque curve in post #1 of my dyno thread 79 Spider on a dyno - The Sequel is pretty flat. I wrongly thought that I could pick up a little power around 6000 rpm by messing with the headers. This is just a strong street motor. I have learned my lesson.
 

·
Administrator
Joined
·
12,540 Posts
Regarding programs doing a decent job of designing a 4 into 1 system, yes I forgot to say my system was a 4 into 1.

Wow, different headers for different tracks. Amazing. When you read about what the NASCAR guys do to their engines, it is quite something. Very clever

Pete
 
1 - 5 of 5 Posts
Top