To avoid these ads, REGISTER NOW!

Header design - lean on one cylinder, how about some back pressure?

spv

Well-known member
Joined
Aug 6, 2010
Messages
301
I am currently building up a set of headers for my single carby (40DCOE) car. The engine I am building is notorious for running lean on #4 (it is a 4 cylinder) due to the design of the intake manifold. I am wondering from a theoretical stand point whether there is merit in adding a sharper bend close to the head on #4 to create some back pressure to try to stop it from leaning out with my free flowing headers?

My basic header design is equal length primaries that are free flowing. There are no bends greater than 45 degrees by design. This engine is known to perform well with a free flowing exhaust system. That said, if it runs lean on #4 I wonder if a bit of back pressure might help things?
 
To avoid these ads, REGISTER NOW!
OP
S

spv

Well-known member
Joined
Aug 6, 2010
Messages
301
Keep your tube length the same,if its lean so be it!!!!

Primaries length will be 36 inches before it hits the 4-1 collector. My question is really if I should add a couple of tighter radius bends to restrict the flow somewhat on #4 whilst keeping the length to 36 inches. I have quite a few options here. My rule of thumb is no bends tighter than 45 degrees. That said, I could add tighter bends on #4 to create back pressure on that cylinder alone in an effort to stop it running lean when compared to #1 through to #3.

- as a side note changing the intake manifold is not an option and this is a single carb setup.
 

fredybender

Well-known member
Joined
Apr 16, 2012
Messages
141
Just make the tubes equal length;
If #4 is known to be lean, just jet it one size larger on the main: DCOE40's are one of the easiest carb to work with, IMO ;
Why would you compensate mixture with back pressure? Take the easy route, and do it right, with the mixture itself... Webers will let you do that!
What engine are you talking about?
PS: I got two cars with DCOE's...
 
Last edited:
OP
S

spv

Well-known member
Joined
Aug 6, 2010
Messages
301
I think he's thinking less scavenging during valve overlap of the intake air/fuel hopefully keeping more of the mixture in the cylinder... :confused:

That is the logic behind the idea. Re-jetting the DCOE is not going to be the answer. When the jetting is perfect for #1 to #3 it is lean on #4. It is a result of the intake manifold design. The easy answer would be "run injection" but it is not an option either. The carb and manifold setup is here to stay. Playing with a single exhaust primary seems to be a logical thing to do to try to balance out #4.

Edit: The engine is a Fiat DOHC 4 cylinder.
 
OP
S

spv

Well-known member
Joined
Aug 6, 2010
Messages
301
Never seen it done successfully. Add all the bends you want,if there all the same length, it wont change a thing.

I was hoping to hear from you Nastyzen given all the fabrication work you have done on race cars.

I find it interesting your comment that if they are equal length it wont change a thing. If I for instance have say a tube that has a few 45 degree bends a decent distance from the head, and another tube that has two 180 degree tight radius bends (like an "S" bend) close to the head and then a straight section, they will both flow the same and back pressure will be identical between the two pipes because they are the same length?
 

mayday0017

Well-known member
Joined
Oct 20, 2010
Messages
1,715
Location
Houston Texas
SR20DET has the same problems with 1 lean cylinder, I believe this is common among 4 cylinder cars. It is related to the intake manifold and the only way to solve the problem is to change the intake manifold, anything else is working around the problem...
 
To avoid these ads, REGISTER NOW!

NASTYZEN

Well-known member
Joined
Jun 11, 2010
Messages
2,823
Location
St-Colomban,Que. Canada
I was hoping to hear from you Nastyzen given all the fabrication work you have done on race cars.

If I for instance have say a tube that has a few 45 degree bends a decent distance from the head, and another tube that has two 180 degree tight radius bends (like an "S" bend) close to the head and then a straight section, they will both flow the same and back pressure will be identical between the two pipes because they are the same length?

Same volume. Maybe a lab could measure a diff. though.
 

meatness

Well-known member
Joined
Nov 4, 2011
Messages
54
Location
Sultan. WA
Another thing to think about is how the engine is cooled. Because of how coolant runs through a 4 cylinder the 4th cylinder will usually run hottest. Depending on head and port design of the engine there may be things you can do to help. Is there plumbing on the back of the head for a heater core? Just an idea.
 

phill u7c

Active member
Joined
Jan 9, 2009
Messages
38
Location
Cambridge, UK
How lean are we talking here, like detonation lean?? I know that the Fiat engine does have many issues, is the vacume takeoff for the brake servo coming off that inlet track?
Also no 4 is furthest from the water pump so maybe its getting too hot, on race engines they enlarge the waterways at that end of the head.

Hope that gives you some ideas, good luck
 

BWS

Well-known member
Joined
Sep 3, 2006
Messages
923
Location
Mnts of Va
Probably shouldn't reply....duh.

meatness,more than likely has your problem?How you go about "fixing" it should see you addressing the root cause.....and JMO,dealing with spent gasses ain't it.

If you're rolling your own header,why not weld in some fittings a few inches downstream from head so you can at least moniter any "improvements" upstream?

Other than inherent heat issues....is there an intake problem?Thats better fixed at the point of the problem.Dickin with header design is simply,"after the fact".
 

larry_g

Well-known member
Joined
Apr 28, 2007
Messages
16,858
Location
oregon
I thought about this for a day and it seems to me that if you restrict the flow out of the cylinder you will be diluting the intake charge more which would lean things even further. Being this is a known problem per your first post then haven't some tuners come up with a mod for the intake?

lg
no neat sig line
 
OP
S

spv

Well-known member
Joined
Aug 6, 2010
Messages
301
This thread has turned into a hive of activity. I figured it would get some people thinking. I should really give a bit more background on the build. This engine I am building is to full race specification. This is a very detailed build right down to every nut and bolt being replaced. It has all forged internals (crank/CP pistons/Pauter rods), custom Kent intake camshaft, big valves (inc sodium cooled exhaust valves), full port job (including oil and water passages), modified crankshaft (eyelets), custom designed pistons (in progress), big custom radiator, Mocal oil cooler setup, custom fabricated baffled sump with trap-door, custom ignition system and the list goes on (and on..). Everything is customized right down to the custom built pulley driving the supercharger. In addition I have a full data-logger setup incorporating wideband, charge temp pre and post supercharger, ex temp etc (all the usual suspects are also being logged).

There are two things however that must remain. The carb and intake manifold. I am not beyond making another manifold with internal velocity stacks and revised layout but it is not something that I am willing to do at this stage. The manifold redesign will come with my "next years" 2.0 build which will have a 16valve head, redesigned manifold, and possibly injection.

The headers are what I am working on at the moment. I figured rather than going to the usual crowd, I would see if anyone on GJ had encountered this issue in other engines and if they had played with the exhaust to improve the situation.

I am completely aware that changing the fueling (dual carb is not possible due to the historic supercharger) and manifold will solve the issue. #4 only runs lean on this engine with the supercharger setup, which is an increasingly rare combination. These superchargers are becoming scarce. People do not go down the "authentic" path, they generally bolt a small snail to the engine and get great performance. I chose to go down what I call the "historic" with modern materials path.

As for these engines being unreliable or a bad design, there is nothing further from the truth. People regularly get 100hp per liter, with twin 45DCOEs out of these engines in N/A setups. These are extremely good flowing heads and are very robust engines in stock form. The Fiat name may not be highly regarded, but the Lampredi designed engines are. These were copied by many manufactures in the 70s. They are the basis of truly legendary cars including the Lancia 037, Lancia Delta S4, and the Lancia Delta Integrale.

Chris Harris has recently done a small segment on his ride in some of the Lancia Rally Legends at a Pirelli tyre launch. The engine I am building will eventually be much like the 037 which you can witness here:

If you would like to see what the engine is capable of, our crazy friends in Italy have a YouTube video of a running 1970's 1.8 liter Fiat engine (the same engine I am working with), (running injection) at 10,000 rpm putting out 250hp in N/A form:

Thanks to "meatness" and "phil u7c" for the ideas. It has given me some food for thought.
 
Last edited:

Dan Babb

Well-known member
Joined
Dec 20, 2010
Messages
85
Instead of trying to put bends in thr pipe, how about making a smaller versipn of a Car Chemistry insert. You could add or remove discs to change the pressure and see if it helps. You may have to do that before you weld on a collector, but if you have a straight run on the pipe, you should be able to fit something.

I have no experiece with this thing...so i dont know if it will work, but thought it could be an easier way to tune just that one pipe.

Heres a link to the insert
http://www.carchemistry.com/shop/index.php?l=product_detail&p=17

They dont make one small enough...but maybe you could make a mini version
 
Last edited:
To avoid these ads, REGISTER NOW!
Top Bottom