Modified Cadillac
Welcome, Guest. Please login or register.
Did you miss your activation email?
July 22, 2014, 06:52:34 PM

Login with username, password and session length
Search:     Advanced search
Visit our Documents section to reference 1000s of pages of Cadillac & LaSalle factory & dealer literature items including:
Master Parts Lists, Illustrated Parts Catalogs, factory sales brochures, technical bulletins, advertising, data books & more.
89,095 Posts in 7,850 Topics by 747 Members
Latest Member: Brassnuts
* Home Help Search Calendar Login Register
Modified Cadillac  |  Home  |  Discussion  |  Topic: 331/365/390 header design 0 Members and 1 Guest are viewing this topic. « previous next »
Pages: [1] 2 3 Go Down Print
Author Topic: 331/365/390 header design  (Read 4184 times)
c322348
Registered Guest

Offline Offline

Posts: 125



« on: December 21, 2010, 12:15:48 AM »

OK, my next dumb question...

How do you select the header pipe diameter and length for the siamesed center exhaust port? I've seen info on designing headers based upon displacement and RPM assuming 8 pipes, but what do I do when I only have 6 pipes relieving 8 cylinders?Huh??

I know I could just make something work well enough (got to be better than the cast iron manifolds), but it couldn't be much harder to size it properly for best performance either. Who knows how the math works here? Sorry, I'm an engineer by trade so I like to do things by the numbers...
Logged
The Tassie Devil (Bruce Reynolds)
MCLC Member
****
Online Online

Posts: 8,834


Now home, and loving it.


« Reply #1 on: December 21, 2010, 04:45:19 AM »

Oh, the posing questions.

But, when working out what dimensions the primaries should be, you have to look at the firing order, and see what sequence the gases leave the cylinders, and if there is any close-firing of the ports that are together.

I am not sure of the 429, but in most cases, the diameter of the centre pipe could be the same as the front and rear.   Making the centre pipe too large could cause a problem with the flow characteristics further on down the pipe, before the Collector, or Joint is reached.

For street use, a Tuned Length is best, but not individual pipes into one Collector, but maybe the centre one shorter than the others, so that the gas in this pipe, each time the two cylinders actually get rid of their gas, the first "exhausted" gas gets into the collector, before the next lot of "exhausted" gas.   This way, the first lot, plus the other pipe-lots can assist in "extracting" the second lot, without pressurising the gas from the front and rear pipes.

Make sense?

Bruce. Evil 
Logged

CLC # 18992
'72 Eldorado Convertible (LHD)
'70 Ranchero Squire (RHD)
'55 Buick Special Post Coupe (LHD)
'72 Chris Craft Gull Wing (RHD)
Past President of the Modified Chapter
jw
Registered Guest

Offline Offline

Posts: 204



« Reply #2 on: December 21, 2010, 07:18:53 AM »

maybe look at some old flathead v8 ford racers. they had the same deal.
Logged
classic cruiser
Registered Guest

Offline Offline

Posts: 92



« Reply #3 on: December 21, 2010, 07:24:33 AM »

They used to make a lot of 3 tube headers for pontiac v8s with D port heads back in the day. they seemed to work fine.
Logged

Coffee fueled-Beer cooled
STDog
Webmaster
Administrator
*****
Offline Offline

Posts: 5,471


CLC# 24051


« Reply #4 on: December 21, 2010, 10:24:35 AM »

But, when working out what dimensions the primaries should be, you have to look at the firing order, and see what sequence the gases leave the cylinders, and if there is any close-firing of the ports that are together.

I am not sure of the 429, but in most cases, the diameter of the centre pipe could be the same as the front and rear.   Making the centre  

He's looking for the 331/365/early 390.
18463572 (front to rear: Right bank 2-4-6-8, left bank 1-3-5-7)

So the center cylinders are 90o apart on both sides. Not good from a performance standpoint.
So, I think the center tube should be a bit bigger, but not sure how much.


Logged

Carnut
Registered Guest

Offline Offline

Posts: 794


« Reply #5 on: December 21, 2010, 12:48:13 PM »

look at this.

http://www.sandersonheaders.com/Sanderson-CAD390-Header-Set-for-Cadillac-390/429-Engines-1966-Coupe-DeVille-and-Street-Rods.html
Logged

1984 Coupe Deville
1954 Pontiac Chieftain Catalina
1950 Chevy Convertible
2008 Infiniti G37S
1989 F250
c322348
Registered Guest

Offline Offline

Posts: 125



« Reply #6 on: December 21, 2010, 01:56:44 PM »

My definition of exhaust pulse is from when the exhaust valve opens until it closes. Not sure if that is the proper terminology.

The question seems to be, how long does the exhaust pulse remain in the center pipe (if you look at the pulse as a string in the pipe) and does it go fast enough to get into the collector before the adjacent cylinder's pulse? If it goes fast enough, the exhaust pulse should be in the collector and another can come behind it with no interference (although that tube might get hotter because it has twice the heat energy going through it). That must not be the case however, as the center port exit area is larger than the single pipe ports and the Sanderson header seems to have a larger center tube as well. The Sanderson header seems to also shorten the larger center pipe compared to the other pipes.

The larger diameter center pipe with its larger cross sectional area slows down the gas flow, so maybe it needs to be shorter to get the gases into the collector in the same time-frame as the other pipes do. So I'm not sure if he larger pipe is merely a convenience factor for packaging or whether it is necessary because of the volume of exhaust going through the center pipe.

The above discussion relates to the gas flow only and ignores the resonance of the exhaust pulses (tuning).

I would love to just get this over with and buy the Sanderson and install it, but the dimensions are too vague for me to know whether it will fit or not, not to mention it doesn't list the 331/365/early 390 as an application. Anyone have a set I can borrow?    Grin

If I fab a custom header, I might as well engineer a design rather than just throw something together.

Thanks all! Keep the ideas coming.

Logged
STDog
Webmaster
Administrator
*****
Offline Offline

Posts: 5,471


CLC# 24051


« Reply #7 on: December 21, 2010, 03:52:22 PM »

The late 390/429 firing order different though.
18726543

So 6 and 4 are now 180o apart (as are 5 and 3)
So the optimum for that would be different. Even if the Sanderson headers are well tuned, they would not be best for the the earlier engines, where the center cylinders are only 90o apart.


I suspect the early engines need an even larger and shorter tube, to get the pulse out of the way faster and to the collector in time.

As for calculation the proper size/length, I suggest calling header design company.
Either a manufacturer like Hooker or Sanderson, or a smaller place.
I think a call to Headers By Ed http://www.headersbyed.com/
might be worthwhile, and more productive than a big commercial firm.

Then again, Ed might just push you to buy his "INFOPAK" or design services.
Logged

Carnut
Registered Guest

Offline Offline

Posts: 794


« Reply #8 on: December 21, 2010, 06:35:35 PM »

I wonder if you could change the firing order on the earlier engines with a different cam?
Logged

1984 Coupe Deville
1954 Pontiac Chieftain Catalina
1950 Chevy Convertible
2008 Infiniti G37S
1989 F250
STDog
Webmaster
Administrator
*****
Offline Offline

Posts: 5,471


CLC# 24051


« Reply #9 on: December 21, 2010, 06:57:34 PM »

I wonder if you could change the firing order on the earlier engines with a different cam?

I think you'd need a different crankshaft too.
The cylinders just don't appear to line up to me, but maybe I missed something.

As far as I can tell the cylinders are numbered the same way
18726543, 1 & 6 both at TDC (2&3, 8&5, 7&4)
18463572, 1 & 3 both at TDC (6&2, 8&5, 7&4)

Granted it's odd. Maybe I have the firing order for the older engines wrong?  Huh
(I looked up up earlier, but maybe I copied it wrong)


Logged

Carnut
Registered Guest

Offline Offline

Posts: 794


« Reply #10 on: December 21, 2010, 08:18:18 PM »

I am getting 18436572 for 1951 Cadillac (two different websites) That would still leave a minor issue with the 6 cylinder firing after the 4 cyllinder and the 5 cylinder firing after the 3 culinder. There are no pulses from the front or rear cylinders to scavage gasses out of the center tube (between the 6 and 4 or 5 and 3) . The same situation exists with the later engines. In effect you have the 4 pulse in the pipe when the 6 valve opens.
Logged

1984 Coupe Deville
1954 Pontiac Chieftain Catalina
1950 Chevy Convertible
2008 Infiniti G37S
1989 F250
Misfit
Guest
« Reply #11 on: December 21, 2010, 08:44:20 PM »

You have to remember, these engines weren't originally designed to put speed products on. They were made to power smooth luxobarges. The designers weren't interested with putting headers on a Cadillac.

Fins
Logged
The Tassie Devil (Bruce Reynolds)
MCLC Member
****
Online Online

Posts: 8,834


Now home, and loving it.


« Reply #12 on: December 21, 2010, 08:49:48 PM »

When working out the tubing diameter, a lot depends on the diameter of the Exhaust Valve, and in street applications, it is a case of the closer the diameter of the Primaries to that of the Valve, the faster the transition of the gas through the pipe, and then this in turn creates a vacuum, or extracting effect on the adjacent pipes as the gas passes the junctions.

If going to Supercharging, and the like, then the size of the pipes isn't that important.

Bruce. Evil
Logged

CLC # 18992
'72 Eldorado Convertible (LHD)
'70 Ranchero Squire (RHD)
'55 Buick Special Post Coupe (LHD)
'72 Chris Craft Gull Wing (RHD)
Past President of the Modified Chapter
c322348
Registered Guest

Offline Offline

Posts: 125



« Reply #13 on: December 21, 2010, 10:32:35 PM »

Headers by Ed has the following to say on their website:

Quote
      Dissimilar Tube Collectors are used to make headers for engines that have siamesed exhaust ports (where two cylinders share one exhaust port).  Flathead Ford V8s, Studebaker V8s, earlier Cadillac V8s, earlier Oldsmobile V8s, earlier AMC V8s, earlier Chev and GMC inline 6s, etc. are just a few of the many earlier engines have had siamesed exhaust ports.  There are also a number of earlier foreign built engines that also have shared exhaust port cylinder heads as well.
      The reason why two different tube sizes are used is to keep the exhaust gas velocities in all of the header tubes about the same.  The larger tube has approximately twice the cross-sectional area of the smaller tube so that it can "carry" double the exhaust gas volume but at the same velocity so that the exhaust gas scavenging ability of the header is kept near equal for ALL of the cylinders.  This not only maximizes performance gain but also reduces and/or eliminates the tuning problems and air/fuel mixture distribution problems in the intake manifold that are often created by headers where the tubes are NOT sized to equalize exhaust gas velocities.

http://www.headersbyed.com/_collector.htm

They also offer flanges, some with stubs welded on here: http://www.headersbyed.com/__cadillac.htm#20202D

« Last Edit: December 21, 2010, 10:34:58 PM by c322348 » Logged
STDog
Webmaster
Administrator
*****
Offline Offline

Posts: 5,471


CLC# 24051


« Reply #14 on: December 21, 2010, 11:46:16 PM »

Headers by Ed has the following to say on their website:

I glanced at the site earlier, but didn't look at the right page.

So, he says double the CSA on the center tube. Which sounds reasonable.
I'll mention that he is a firm believer in equal length headers, and long tubes.

Logged

Pages: [1] 2 3 Go Up Print 
Modified Cadillac  |  Home  |  Discussion  |  Topic: 331/365/390 header design « previous next »
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.18 | SMF © 2006-2011, Simple Machines Valid XHTML 1.0! Valid CSS!
Page created in 0.223 seconds with 21 queries.