A
Anonymous
Guest
The Following is on a modification to a BMW inline six with side mounted carbs removed in place of a 390 cfm 4 barrel carb.
Holley 4160/ model 8007, 390 CFM, p/n 8107. Dual 47mm mechanical primaries and 47mm vacuum actuated secondaries.
(Note: Tom Van Gunten has been running four barrels on 3.5 liter Big Sixes for several years. Read his FAQ and decide for yourself. The spreadbore carburetors are much easier to use on the stock manifold than the 4160/ model 8007)
The 4160/ model 8007 is designed for use with small V8 and V6 engines and the airflow capacity is 390 cfm (cubic feet per minute). Virtually all of the motor press that I have read to date specifies that the carburetors should flow between 1.8 and 2.0 cfm per cubic inch of engine displacement. Since a 3 liter BN|BMW engine has 182 cubic inches of displacement, the carburetor should be able to flow 364 cfm. The Holley carburetor is easily able to supply this amount of air to the motor, even if the BMW engine has a volumetric efficiency 7% higher than the robust 2 CFM per cubic inch calculation.
Airflow requirements for c.i.d and rpm’s. (per the Holley Carburetors Sales Brochure, ed. 11/96 )
Where: required CFM = (engine displacement (c.i.d.) * maximum rpm’s) / 3456
Note: 1 liter = 61.3 c.i.d
3 liter motor = (182 * 6500) / 3456 = 342 CFM
3.3 liter motor = (200 * 6500) / 3456 = 376 CFM
3.5 liter motor = (213 * 6500) / 3456 = 399 CFM
Some readers will note that I chose 6500 rpm as a maximum rpm versus the more typically quoted 7000 rpm. Given my normal driving pattern, even 6k is probably still too high: I rarely rev over 5500.
Using the above calculations, the Holley 4160/ model 8007 is acceptable for engines up to a street driven 3.3 liters. In my opinion, it would not be a good idea to put a larger capacity carburetor on a straight six BMW engine. Proper fuel atomization depends on a strong intake signal at the cylinder and the resulting high velocity of air through the carburetor venturis. Carburetors are designed to work best within certain airflow parameters. If a carburetor is too big for an engine, it won’t adequately meter and atomize fuel because the carburetor doesn’t receive a strong enough intake signal from the engine.
As rpm’s increase this will be less of a problem but at low engine speeds the low airflow will result in bogs and sags. Large diameter throttle plates and large diameter venturis look impressive when the air cleaner is removed but they are designed for engines more than twice as large as a Big Six. It shouldn’t be surprising if a big carburetor performs poorly until the Big Six was revved high enough to generate the same level of airflow as a larger displacement engine.
The aftermarket carburetor market developed to fill the needs of V8 engines that are 5 liters and larger and existing products reflect this premise. In the past ten years, fuel injection has replaced the carburetor on virtually all production cars. These factors combined severely limit the variety if carburetors that older BMW owners can use now, and will preclude any future development of aftermarket carburetors.
Necessary manifold Alterations
The 4160/ model 8007 has a squarebore base, meaning that the venturis are both the same size and equidistant from each other. The BMW manifold is a spreadbore pattern and the holes for the secondary venturis are larger and wider apart. The holes for the 38mm primary venturis are quite far part and the 58mm secondaries are separated by only 10mm of metal. To put it mildly - the Holley squarebore carburetor doesn’t fit on the stock manifold: mounting holes do not line up nor do the profiles of the carburetor base to the manifold deck.
I took the cheap and easy way out and purchased a squarebore to spreadbore adapter plate from Jeg’s automotive mail order house. It cost about $35. It was manufactured by Edelbrock though I am sure that other less expensive brands are available. Do not buy a cheap adapter that has one large opening rather than four separate openings (see below). Some matching was necessary and was done using Prussian blue and rat tail files. Since the adapter plate created a very sharp turn into the manifold underneath the primary venturis, I removed as many of the casting imperfections as possible. Reorienting the carburetor so that the primary venturis are over the part of the adapter plate with more friendly contours would involve changing the throttle linkage.
Single Plane versus dual plane manifolds
The stock BMW 4 barrel manifold is divided into two halves and each set of three cylinders is fed by one primary and one secondary throttle. It may be easier to think of the left and right side of the carburetors. This kind of divided manifold is usually referred to as a dual plane or divided plenum manifold.
The goal is to keep the cross sectional area in the manifold plenum as close to the original design as possible. If the area in the manifold under the carburetor is increased, there is a greater tendency for atomized fuel to linger since the air velocity would be decreased: less suction.
Conversely, a single plane manifold has an open area at the base of the carburetor where all of the cylinders can be fed by the whole throttle capacity of the carburetor. In the case of the BMW manifold, think of the manifold with the center rib removed. Dual plane manifolds usually outperform single plane manifolds at part throttle and offer better low rpm drivability and torque.
With an unmodified BMW dual plane manifold, using the 390CFM Holley as an example, each cylinder can be fed by a total of 195cfm of carburetor flow capacity. (One primary barrel at cruise plus one secondary barrel at full throttle). Referring to the CFM per c.i.d. requirements I included above: (182 c.i.d. / 2) * 2 = 182 CFM of carburetor capacity for 3 cylinders. If the manifold center were routed out and the manifold became a dual plane manifold, each cylinder could have fuel supplied by the full 390cfm of the carburetor. As a result of reading several books on performance engine modifications, I think that a dual plane manifold would be better than a single plane on an M30/ Big Six engine.
Most drivers are at full throttle only a few minutes per week of driving. The rest of the time, the combination of the carburetor and manifold will be too much for the engine and drivability will suffer. When you are starting from a stop, at low rpm’s, or cruising at part throttle, it reasonable to assume that using a dual plane manifold would cause: lower airspeed in the manifold and a weaker intake signal, resulting in poorer fuel metering and atomization, in layman’s terms: sags, bogs, and flat spots.
Other things to think about……
Getting fuel into the Big Six motor, ranked from best to worst:
timed / multi-point fuel injection with optimization capabilities – Motronic
multi-point fuel injection: L-Jet, D-jet
triple sidedraft carburetors (3 x 2 barrels)
twin carburetors (Weber 32/36’s)
single four barrel (Solex 4A1, Holley, Quadrajet)
Put simply, it’s always better to have a dedicated source of metered and timed fuel for each individual cylinder. Depending how you drive and your budget, you can be happy with one of the lesser alternatives. I chose a four barrel carburetor versus the others and I am satisfied.
Disadvantages of the Big Six four barrel manifold
Fuel distribution
None of the intake runners on the Big Six four barrel manifold are the same length, shape, and even their inside diameter varies slightly (different volume). Cylinders attached to a common plenum scavenge mixture from each other. What is unused fuel mixture doing floating around unused in the first place? This clearly affects fuel distribution, power, and drivability:
1). All six of the intake charges vary slightly from one another. Air/fuel mixture velocity is different in each intake runner because of the differing plenum volumes and the varying signals present at the jets. Cylinders closest to the carburetor will get richer mixtures than those which are further away.
2).
Air/fuel mixture from the carburetors gets bounced around and lost on its way to the intake ports because of the greater amount of common area in the manifold and the intake pulses from the cylinders.
In contrast, examine at a Big Six with fuel injection or triple carburetors with six independent runners. Those systems have six identical intake runners, both in length and diameter. In these intake systems, the air drawn into the engine may come from a common source but the fuel is introduced downstream, at the intake of each cylinder. The further our design moves away from six balanced intakes and individual fuel meters the worse off we are.
Dyno tests with V8 engines comparing 4 barrels against multi-point fuel injection and multiple sidedraft carburetor set-ups show that the 4 barrel lags behind the others in terms of torque and horsepower throughout the entire rpm range. Four barrels may produce the highest horsepower, but they lag behind the others at all other points. The same is (likely) true for dual downdraft Webers.
The four barrel manifold is clearly a compromise. Will 95% of Big Six owners notice the difference? I’m not sure. I simply can’t afford to run a conclusive set of dyno tests that would plot horsepower and torque of the same engine with the different intake systems.
Please contact me directly if you have questions. Martye9e12@hotmail.com or marty@cscoupe.org
Holley 4160/ model 8007, 390 CFM, p/n 8107. Dual 47mm mechanical primaries and 47mm vacuum actuated secondaries.
(Note: Tom Van Gunten has been running four barrels on 3.5 liter Big Sixes for several years. Read his FAQ and decide for yourself. The spreadbore carburetors are much easier to use on the stock manifold than the 4160/ model 8007)
The 4160/ model 8007 is designed for use with small V8 and V6 engines and the airflow capacity is 390 cfm (cubic feet per minute). Virtually all of the motor press that I have read to date specifies that the carburetors should flow between 1.8 and 2.0 cfm per cubic inch of engine displacement. Since a 3 liter BN|BMW engine has 182 cubic inches of displacement, the carburetor should be able to flow 364 cfm. The Holley carburetor is easily able to supply this amount of air to the motor, even if the BMW engine has a volumetric efficiency 7% higher than the robust 2 CFM per cubic inch calculation.
Airflow requirements for c.i.d and rpm’s. (per the Holley Carburetors Sales Brochure, ed. 11/96 )
Where: required CFM = (engine displacement (c.i.d.) * maximum rpm’s) / 3456
Note: 1 liter = 61.3 c.i.d
3 liter motor = (182 * 6500) / 3456 = 342 CFM
3.3 liter motor = (200 * 6500) / 3456 = 376 CFM
3.5 liter motor = (213 * 6500) / 3456 = 399 CFM
Some readers will note that I chose 6500 rpm as a maximum rpm versus the more typically quoted 7000 rpm. Given my normal driving pattern, even 6k is probably still too high: I rarely rev over 5500.
Using the above calculations, the Holley 4160/ model 8007 is acceptable for engines up to a street driven 3.3 liters. In my opinion, it would not be a good idea to put a larger capacity carburetor on a straight six BMW engine. Proper fuel atomization depends on a strong intake signal at the cylinder and the resulting high velocity of air through the carburetor venturis. Carburetors are designed to work best within certain airflow parameters. If a carburetor is too big for an engine, it won’t adequately meter and atomize fuel because the carburetor doesn’t receive a strong enough intake signal from the engine.
As rpm’s increase this will be less of a problem but at low engine speeds the low airflow will result in bogs and sags. Large diameter throttle plates and large diameter venturis look impressive when the air cleaner is removed but they are designed for engines more than twice as large as a Big Six. It shouldn’t be surprising if a big carburetor performs poorly until the Big Six was revved high enough to generate the same level of airflow as a larger displacement engine.
The aftermarket carburetor market developed to fill the needs of V8 engines that are 5 liters and larger and existing products reflect this premise. In the past ten years, fuel injection has replaced the carburetor on virtually all production cars. These factors combined severely limit the variety if carburetors that older BMW owners can use now, and will preclude any future development of aftermarket carburetors.
Necessary manifold Alterations
The 4160/ model 8007 has a squarebore base, meaning that the venturis are both the same size and equidistant from each other. The BMW manifold is a spreadbore pattern and the holes for the secondary venturis are larger and wider apart. The holes for the 38mm primary venturis are quite far part and the 58mm secondaries are separated by only 10mm of metal. To put it mildly - the Holley squarebore carburetor doesn’t fit on the stock manifold: mounting holes do not line up nor do the profiles of the carburetor base to the manifold deck.
I took the cheap and easy way out and purchased a squarebore to spreadbore adapter plate from Jeg’s automotive mail order house. It cost about $35. It was manufactured by Edelbrock though I am sure that other less expensive brands are available. Do not buy a cheap adapter that has one large opening rather than four separate openings (see below). Some matching was necessary and was done using Prussian blue and rat tail files. Since the adapter plate created a very sharp turn into the manifold underneath the primary venturis, I removed as many of the casting imperfections as possible. Reorienting the carburetor so that the primary venturis are over the part of the adapter plate with more friendly contours would involve changing the throttle linkage.
Single Plane versus dual plane manifolds
The stock BMW 4 barrel manifold is divided into two halves and each set of three cylinders is fed by one primary and one secondary throttle. It may be easier to think of the left and right side of the carburetors. This kind of divided manifold is usually referred to as a dual plane or divided plenum manifold.
The goal is to keep the cross sectional area in the manifold plenum as close to the original design as possible. If the area in the manifold under the carburetor is increased, there is a greater tendency for atomized fuel to linger since the air velocity would be decreased: less suction.
Conversely, a single plane manifold has an open area at the base of the carburetor where all of the cylinders can be fed by the whole throttle capacity of the carburetor. In the case of the BMW manifold, think of the manifold with the center rib removed. Dual plane manifolds usually outperform single plane manifolds at part throttle and offer better low rpm drivability and torque.
With an unmodified BMW dual plane manifold, using the 390CFM Holley as an example, each cylinder can be fed by a total of 195cfm of carburetor flow capacity. (One primary barrel at cruise plus one secondary barrel at full throttle). Referring to the CFM per c.i.d. requirements I included above: (182 c.i.d. / 2) * 2 = 182 CFM of carburetor capacity for 3 cylinders. If the manifold center were routed out and the manifold became a dual plane manifold, each cylinder could have fuel supplied by the full 390cfm of the carburetor. As a result of reading several books on performance engine modifications, I think that a dual plane manifold would be better than a single plane on an M30/ Big Six engine.
Most drivers are at full throttle only a few minutes per week of driving. The rest of the time, the combination of the carburetor and manifold will be too much for the engine and drivability will suffer. When you are starting from a stop, at low rpm’s, or cruising at part throttle, it reasonable to assume that using a dual plane manifold would cause: lower airspeed in the manifold and a weaker intake signal, resulting in poorer fuel metering and atomization, in layman’s terms: sags, bogs, and flat spots.
Other things to think about……
Getting fuel into the Big Six motor, ranked from best to worst:
timed / multi-point fuel injection with optimization capabilities – Motronic
multi-point fuel injection: L-Jet, D-jet
triple sidedraft carburetors (3 x 2 barrels)
twin carburetors (Weber 32/36’s)
single four barrel (Solex 4A1, Holley, Quadrajet)
Put simply, it’s always better to have a dedicated source of metered and timed fuel for each individual cylinder. Depending how you drive and your budget, you can be happy with one of the lesser alternatives. I chose a four barrel carburetor versus the others and I am satisfied.
Disadvantages of the Big Six four barrel manifold
Fuel distribution
None of the intake runners on the Big Six four barrel manifold are the same length, shape, and even their inside diameter varies slightly (different volume). Cylinders attached to a common plenum scavenge mixture from each other. What is unused fuel mixture doing floating around unused in the first place? This clearly affects fuel distribution, power, and drivability:
1). All six of the intake charges vary slightly from one another. Air/fuel mixture velocity is different in each intake runner because of the differing plenum volumes and the varying signals present at the jets. Cylinders closest to the carburetor will get richer mixtures than those which are further away.
2).
Air/fuel mixture from the carburetors gets bounced around and lost on its way to the intake ports because of the greater amount of common area in the manifold and the intake pulses from the cylinders.
In contrast, examine at a Big Six with fuel injection or triple carburetors with six independent runners. Those systems have six identical intake runners, both in length and diameter. In these intake systems, the air drawn into the engine may come from a common source but the fuel is introduced downstream, at the intake of each cylinder. The further our design moves away from six balanced intakes and individual fuel meters the worse off we are.
Dyno tests with V8 engines comparing 4 barrels against multi-point fuel injection and multiple sidedraft carburetor set-ups show that the 4 barrel lags behind the others in terms of torque and horsepower throughout the entire rpm range. Four barrels may produce the highest horsepower, but they lag behind the others at all other points. The same is (likely) true for dual downdraft Webers.
The four barrel manifold is clearly a compromise. Will 95% of Big Six owners notice the difference? I’m not sure. I simply can’t afford to run a conclusive set of dyno tests that would plot horsepower and torque of the same engine with the different intake systems.
Please contact me directly if you have questions. Martye9e12@hotmail.com or marty@cscoupe.org
