Performance Carburetor Selection and Set Up
The British V8 Newsletter, Volume XV Issue 1by: Larry Shimp
There is no doubt that fuel injection will give the optimum results, but a properly set up carburetor actually works quite well. The key is "properly set up".
Float bowl and float setting
From my experience, the most important factor in carburetor selection is float bowl design. The fuel float is intended to keep the fuel at a constant level within the float bowl which permits the proper functioning of all of the fuel circuits. With an incorrect or fluctuating float level, a carburetor cannot perform properly. When the car is sitting and idling, any carburetor float/bowl works well. The key is what happens when the car is moving, especially at the extreme performance limits possible with a V8 powered MGB.
There are two basic float designs: center hung and side hung. Side hung means the pivot axis of the float is parallel to the direction of travel. Center hung means that the pivot axis is perpendicular to the direction of travel. Side hung floats tend to get upset on hard cornering. The float on the outside of the corner tends to stay closed, letting the fuel level drop, while the inner float tends to open, raising the fuel level. The result is an unequal, sub-optimum fuel mixture that differs on each side of the carburetor, leading to poor throttle response and a lower power output. Holley carburetors can be either side or center hung, but all Carter carburetors are center hung. However, because the Carter float bowls are on either side of the carburetor, they behave more like side hung floats. I first had a Carter carburetor on my car, and experienced poor throttle response on hard cornering. I could find no fix, so I went to a center hung Holley.
Please support the sponsoring companies who make BritishV8 possible, including:
The Holley center-hung design completely solved the cornering problem. However,
there were some straight line effects that did not show up with the Carter. The
most obvious was that the engine would flood and stall on hard braking. The cause
was that the fuel in the rear (secondary) fuel bowl would slosh up the vent tube
and into the carburetor throat. Holley sells a vent tube modification kit that
moves the vent intake from the area by the throat to the outer edge of the float
bowl (Summit part number HLY-26-89). This prevents any fuel slosh out of the vent
from either bowl on either acceleration or braking. I highly recommend this
modification, it solved the problem I had with the engine flooding out and stalling
upon hard braking. Installing the vent is easy, but it does require a #51 drill bit
to drill a small hole for a rivet/screw using the locating dimple in the metering
plate body. (If you have a holley carburetor model without a secondary metering plate,
this modification probably won't fit.) The #51 drill is available from Micro Mark
and similar hobby suppliers. An alternative to the vent extensions is made by Moroso
(Summit part number MOR-65221) which puts a foam baffle in the front of the bowl.
However, it requires that the bowls be moved out with a spacer and thus it changes
the spacing of the fuel inlets.
Metering Block (note the dimple that has to be drilled out for screw/rivet)
Vent Extension in Place
Vent Extension and Mr. Gasket Jetty Extensions
Another potential problem with the in-line holley fuel bowls is that the fuel in the secondary bowl can move to the rear under hard acceleration, partially starving the secondary jets. The fix for this is a kit that consists of a pair of jet extension tubes that move jets to the rear of the float bowl, and a notched float to clear the jet extensions (Summit part number HLY-116-19), or a kit by Mr Gasket (Summit part number MRG-6060) that consists of a pair of flattened sheet metal jet extensions that can be used with the standard plastic float. (Some carburetors come with a brass float, and this may be too large to work with the jet extensions.) Holley also makes plastic extensions (Summit part number HLY-55009HOL) that do not seem to need a notched float (but probably need a plastic float). Jet extensions are probably not necessary even on a very fast MGB because the drop in fuel level should be compensated for by new fuel entering the carburetor. It makes more sense for big block engines with 1000 cfm carburetors that use huge quantities of fuel during acceleration. However, it does not hurt to install the kit.
The next item to consider is the needle valve itself. The needle valve is held closed by float pressure once the bowl is full, but it must be able to flow very freely once the float level drops a small amount to facilitate keeping fuel level constant during periods of high fuel demand. A metal seat and needle is usually standard, but vibration, vehicle acceleration, and bumps can cause momentary unseating. When vibration causes the valve to bounce, running fuel level will be above the resting fuel level. A remedy for this is a needle with a rubber tip that is more tolerant of vibration and vehicle acceleration.
Naturally, all of the above steps are useless if the initial float level is incorrect. With a Holley, the float level is normally adjusted externally to the height of an inspection hole in the side of the float chamber. The best way to do this is with the engine off and the electric pump on. Holley (and others) make clear plastic plugs to replace the metal bowl plugs, but the plastic plugs cannot be left in permanently because gasoline will eventually attack them. Holley also offers replacement bowls with glass windows, but these are not really necessary. Some old Holley two barrel carburetors even had float chambers made entirely of glass. It was a neat idea, but that concept is long gone.
Remember that Holley has a long tradition in road racing competition, including Le Mans. Their success would not have been possible if the float levels were constantly being upset by acceleration, braking, and cornering!
Metering Block (note the dimple that has to be drilled out for screw/rivet)
Vent Extension in Place
Vent Extension and Mr. Gasket Jetty Extensions
Another potential problem with the in-line holley fuel bowls is that the fuel in the secondary bowl can move to the rear under hard acceleration, partially starving the secondary jets. The fix for this is a kit that consists of a pair of jet extension tubes that move jets to the rear of the float bowl, and a notched float to clear the jet extensions (Summit part number HLY-116-19), or a kit by Mr Gasket (Summit part number MRG-6060) that consists of a pair of flattened sheet metal jet extensions that can be used with the standard plastic float. (Some carburetors come with a brass float, and this may be too large to work with the jet extensions.) Holley also makes plastic extensions (Summit part number HLY-55009HOL) that do not seem to need a notched float (but probably need a plastic float). Jet extensions are probably not necessary even on a very fast MGB because the drop in fuel level should be compensated for by new fuel entering the carburetor. It makes more sense for big block engines with 1000 cfm carburetors that use huge quantities of fuel during acceleration. However, it does not hurt to install the kit.
The next item to consider is the needle valve itself. The needle valve is held closed by float pressure once the bowl is full, but it must be able to flow very freely once the float level drops a small amount to facilitate keeping fuel level constant during periods of high fuel demand. A metal seat and needle is usually standard, but vibration, vehicle acceleration, and bumps can cause momentary unseating. When vibration causes the valve to bounce, running fuel level will be above the resting fuel level. A remedy for this is a needle with a rubber tip that is more tolerant of vibration and vehicle acceleration.
Naturally, all of the above steps are useless if the initial float level is incorrect. With a Holley, the float level is normally adjusted externally to the height of an inspection hole in the side of the float chamber. The best way to do this is with the engine off and the electric pump on. Holley (and others) make clear plastic plugs to replace the metal bowl plugs, but the plastic plugs cannot be left in permanently because gasoline will eventually attack them. Holley also offers replacement bowls with glass windows, but these are not really necessary. Some old Holley two barrel carburetors even had float chambers made entirely of glass. It was a neat idea, but that concept is long gone.
Remember that Holley has a long tradition in road racing competition, including Le Mans. Their success would not have been possible if the float levels were constantly being upset by acceleration, braking, and cornering!
Fuel pump and lines
The final factor in maintaining proper float bowl levels is the fuel pump. Fuel pumps are rated on a gallon per hour basis, and even the smallest pump would seem adequate for an MGB V8 (after all, no-one uses 20 gallons of fuel per hour). However, in reality, a fuel capacity of at least 75 gallons per hour is necessary to be able to provide enough fuel during high demand acceleration conditions. In general, no fuel pump meant for a 4 cylinder MGB is satisfactory for a V8 MGB because of the much higher momentary fuel consumption possible with a V8.
The pump pressure is also important. Most carburetors are set up to work with 5 to 7 psi. Higher pressure forces open the float valves, and lower pressure cannot keep the level constant under high acceleration. (The pressure needed to force the fuel forward from the fuel tank under rapid acceleration also plays a role in this.) Pumps with a capacity of 150 to 200 gallons per hour or more often have pressures in excess of 7 psi, and therefore need a pressure regulator. However, a 5 to 7 psi rated pump can be found with a flow of around 100 gallons per minute that will be satisfactory.
Electric fuel pumps can be noisy. I have tried pumps manufactured by Carter, Holley, and Mallory. The first two were intolerably noisy, even with rubber isolation mounts. The Mallory is very quiet, and I strongly prefer it for that reason.
With an electric fuel pump, in the event of an accident, fuel can be pumped from a broken fuel line at a high rate and lead to a serious fire hazard. Do not install an electric pump without some type of automatic emergency cut off! I used a standard late model MGB inertia switch. Other inertia switches are available. There are also schemes that use a oil pressure switch that cut off the pump when the oil pressure drops to zero. Whatever system you use, be sure you understand how to install it or have someone who does install it.
The best pump location is back by the fuel tank, where MG originally put it. This keeps the fuel line pressurized and so prevents problems from vapor lock. Put a fuel filter before the pump. This will protect the high speed, precision, moving pump parts from damage from debris. And, of course, follow all of the pump manufacturer's instructions and recommendations.
The fuel line should be 3/8 inch in diameter. Smaller sizes may restrict fuel flow. Use steel line, not rubber line. Rubber hoses should be restricted only to the ends of the line where it joins the pump and carburetor. Be sure to route and clip rubber fuel lines to avoid chaffing or pinching damage. (Give a thought to how vehicle components are likely to shift in an accident.) The metal pipes where the rubber hoses connect should have proper beads to help retain the hoses.
Tuning principles
All of the above preparations (in addition to making sure the ignition system is properly set up) need to be carried out before tuning the carburetor. That way, tuning will only be affected by the carburetor settings themselves. With external influences, tuning will be impossible.
The following tuning recommendations are most applicable to Holley carburetors. Carter and other carburetors may work on different principles. Much has been written about carburetor tuning, so I will concentrate on issues that are especially relevant to a street tuned MGB V8. To begin, there are two basic carburetor types: those with vacuum secondaries and those with mechanical secondaries. For small block street engines, vacuum secondaries make the most sense and there is (in my opinion) no reason to consider mechanical secondaries. The vacuum secondary carburetors have the advantage that they are inherently very adaptable to a wide range of engine sizes and states of tune, while a mechanical secondary carburetor must be carefully matched to the engine as far as flow capacity is concerned.
Any carburetor has three types of fuel circuits: idle, transition, and power (main). Most cruising is on the transition circuit, which is the hardest to tune. Idle is adjusted by the idle jets, and power by the main jets. Transition behavior is mainly governed by fixed air and fuel bleeds that can only be modified by custom fabrication or expensive aftermarket metering blocks. My advice is to find a carburetor that naturally has the "right" transition settings for the engine. Holley has a huge number of carburetor models to choose from, and I have only tried a few. Basically, I have found that the Street Avenger 670 (Summit part number ???????) has excellent characteristics for my Ford 302 crate engine. I also tried the Street Avenger 570 which would seem to be better sized, but the transition circuits were too far off. I also tried various Demon carburetors, including one that was supposed to be perfectly set up for the Ford crate engine, but none were even close. (Luckily with eBay it's possible to buy and sell a bunch of carburetors without spending too much money.)
To assess the carburetors, I used a wide band air/fuel meter installed in the exhaust so I could monitor the mixture while driving. There are traditional ways of tuning such as timing acceleration runs and shutting off the engine at full throttle, then coasting to a stop and checking the plug deposits, but these are not very practical on public roads. Dyno tuning works, but an hour of dyno time can cost over $100, and many sessions may be needed. Therefore the approximate $250 cost of an air/fuel meter seems reasonable.
The goal with the transition circuits is to achieve an air/fuel ratio during cruise of 14 or 15 to 1. Then, upon the slightest throttle opening, the ratio should immediately increase to 11 to 12 to 1 to prevent hesitation. With some carburetors, the transition mixture was very rich (giving poor fuel economy), then went dead lean upon opening the throttle. The only cure for the leanness was very large main jets that resulted in much too rich of a full throttle mixture. I gave up on these carburetors. Note: these results were obtained with small, slow throttle movements that should not involve any accelerator pump action, so increasing the accelerator pump jet size did not help.
The final factor in maintaining proper float bowl levels is the fuel pump. Fuel pumps are rated on a gallon per hour basis, and even the smallest pump would seem adequate for an MGB V8 (after all, no-one uses 20 gallons of fuel per hour). However, in reality, a fuel capacity of at least 75 gallons per hour is necessary to be able to provide enough fuel during high demand acceleration conditions. In general, no fuel pump meant for a 4 cylinder MGB is satisfactory for a V8 MGB because of the much higher momentary fuel consumption possible with a V8.
The pump pressure is also important. Most carburetors are set up to work with 5 to 7 psi. Higher pressure forces open the float valves, and lower pressure cannot keep the level constant under high acceleration. (The pressure needed to force the fuel forward from the fuel tank under rapid acceleration also plays a role in this.) Pumps with a capacity of 150 to 200 gallons per hour or more often have pressures in excess of 7 psi, and therefore need a pressure regulator. However, a 5 to 7 psi rated pump can be found with a flow of around 100 gallons per minute that will be satisfactory.
Electric fuel pumps can be noisy. I have tried pumps manufactured by Carter, Holley, and Mallory. The first two were intolerably noisy, even with rubber isolation mounts. The Mallory is very quiet, and I strongly prefer it for that reason.
With an electric fuel pump, in the event of an accident, fuel can be pumped from a broken fuel line at a high rate and lead to a serious fire hazard. Do not install an electric pump without some type of automatic emergency cut off! I used a standard late model MGB inertia switch. Other inertia switches are available. There are also schemes that use a oil pressure switch that cut off the pump when the oil pressure drops to zero. Whatever system you use, be sure you understand how to install it or have someone who does install it.
The best pump location is back by the fuel tank, where MG originally put it. This keeps the fuel line pressurized and so prevents problems from vapor lock. Put a fuel filter before the pump. This will protect the high speed, precision, moving pump parts from damage from debris. And, of course, follow all of the pump manufacturer's instructions and recommendations.
The fuel line should be 3/8 inch in diameter. Smaller sizes may restrict fuel flow. Use steel line, not rubber line. Rubber hoses should be restricted only to the ends of the line where it joins the pump and carburetor. Be sure to route and clip rubber fuel lines to avoid chaffing or pinching damage. (Give a thought to how vehicle components are likely to shift in an accident.) The metal pipes where the rubber hoses connect should have proper beads to help retain the hoses.
Tuning principles
All of the above preparations (in addition to making sure the ignition system is properly set up) need to be carried out before tuning the carburetor. That way, tuning will only be affected by the carburetor settings themselves. With external influences, tuning will be impossible.
The following tuning recommendations are most applicable to Holley carburetors. Carter and other carburetors may work on different principles. Much has been written about carburetor tuning, so I will concentrate on issues that are especially relevant to a street tuned MGB V8. To begin, there are two basic carburetor types: those with vacuum secondaries and those with mechanical secondaries. For small block street engines, vacuum secondaries make the most sense and there is (in my opinion) no reason to consider mechanical secondaries. The vacuum secondary carburetors have the advantage that they are inherently very adaptable to a wide range of engine sizes and states of tune, while a mechanical secondary carburetor must be carefully matched to the engine as far as flow capacity is concerned.
Any carburetor has three types of fuel circuits: idle, transition, and power (main). Most cruising is on the transition circuit, which is the hardest to tune. Idle is adjusted by the idle jets, and power by the main jets. Transition behavior is mainly governed by fixed air and fuel bleeds that can only be modified by custom fabrication or expensive aftermarket metering blocks. My advice is to find a carburetor that naturally has the "right" transition settings for the engine. Holley has a huge number of carburetor models to choose from, and I have only tried a few. Basically, I have found that the Street Avenger 670 (Summit part number ???????) has excellent characteristics for my Ford 302 crate engine. I also tried the Street Avenger 570 which would seem to be better sized, but the transition circuits were too far off. I also tried various Demon carburetors, including one that was supposed to be perfectly set up for the Ford crate engine, but none were even close. (Luckily with eBay it's possible to buy and sell a bunch of carburetors without spending too much money.)
|
Enjoying this article? Our magazine is funded through the generous support of readers like you! To contribute to our operating budget, please click here and follow the instructions. (Suggested contribution is twenty bucks per year. Feel free to give more!) |
To assess the carburetors, I used a wide band air/fuel meter installed in the exhaust so I could monitor the mixture while driving. There are traditional ways of tuning such as timing acceleration runs and shutting off the engine at full throttle, then coasting to a stop and checking the plug deposits, but these are not very practical on public roads. Dyno tuning works, but an hour of dyno time can cost over $100, and many sessions may be needed. Therefore the approximate $250 cost of an air/fuel meter seems reasonable.
The goal with the transition circuits is to achieve an air/fuel ratio during cruise of 14 or 15 to 1. Then, upon the slightest throttle opening, the ratio should immediately increase to 11 to 12 to 1 to prevent hesitation. With some carburetors, the transition mixture was very rich (giving poor fuel economy), then went dead lean upon opening the throttle. The only cure for the leanness was very large main jets that resulted in much too rich of a full throttle mixture. I gave up on these carburetors. Note: these results were obtained with small, slow throttle movements that should not involve any accelerator pump action, so increasing the accelerator pump jet size did not help.
The way to influence the richening upon small throttle movements is with the power
valve, not the accelerator pump. The power valve is a vacuum operated valve that
opens at high vacuum conditions to lean the mixture. It then closes as vacuum drops
upon acceleration to richen the mixture. I have a relatively mild cam in my engine
so I am using one of the highest rated power valves (closes at 10.5 inches of mercury).
With a vacuum gauge I found my cruise vacuum is about 17 inches of mercury, so the
power valve is open. Slight throttle opening drops the vacuum to 10 or less inches
of mercury, and the power valve instantly closes and richens the mixture. Be aware
that the standard power valve in the Holley Street Avengers is rated at about 6.5
inches of mercury, and so will be closed until relatively large throttle openings,
unless you have a very radical cam that gives a low cruise vacuum. (Note: The power
valve model number is equal to the value of the vacuum at which the valve closes.)
This leads to a flat spot at small throttle openings and in most cases the standard
power valve needs to be replaced.
Holley secondary metering systems don't normally have a power valve because the secondaries only come into play at relatively low vacuum conditions. But without the richening effect of the power valve, the secondary jet sizes need to be bigger than the primary jet sizes. That's why Holley carburetors normally come with secondary jets about 8 to 10 sizes bigger than the primary jets. If you buy a used carburetor, always replace the power valve just to be sure, and check to see what jets are installed.
Ignition timing
Ignition timing is also important to success with a lean cruise mixture, and a vacuum advance distributor must be used. This will keep the timing well advanced so the slow burning lean cruise mixture will fire properly. Then, as soon as the throttle is opened slightly, the vacuum advance drops out, and the timing reverts to a power setting. Without a vacuum advance, the engine will run poorly on a lean mixture and will almost certainly hesitate with small throttle openings, even with a properly chosen power valve. (My engine runs fine on a 15 to 1 air/fuel ratio and gets about 26 mpg on the highway.)
Power valve protection
From my experience, one problem with power valves is that they can rupture from a carburetor backfire. The Holley Street Avenger (and apparently all Holleys made after about 1992) has a check valve that protects the power valve in the event of a backfire so it's not a problem with this model series. There are also kits made by several manufacturers to retrofit older Holley carburetors with the check valve. I highly recommend this modification.
Jetting and secondary vacuum canister springs
With the carburetor properly set up and the right power valve installed, final tuning is relatively very simple. The goal is smooth throttle response with no surging or flat spots associated with rapid throttle openings. If throttle response is good, chances are that the full throttle mixture will be reasonably correct as well. First of all, if there is some hesitation upon throttle opening at low engine speeds, it is probably the accelerator pump circuit. Holley a has a number of accelerator pump cams and even a larger capacity accelerator pump itself (50 cc versus 30 cc). I have found that simply putting in the next size larger accelerator pump jets (0.033 inches versus 0.030 inches) worked fine for my car.
Throttle response
At higher speeds the secondary throttles open. When this occurs there should be no surging, just a smooth increase in power. Some people think there should be a surge as the secondaries open, but this only signifies a lack of power. (A surge means either the engine was down on power until the secondaries opened or there is a flat spot coming and going.) There is no accelerator pump associated with the secondary throttles, the only tuning variables are primary and secondary main jets, and the secondary vacuum capsule spring. If the primary jets are too lean, there will be a surge as the richer secondary jets cut in. On the other hand, if the secondary jets are too lean, there may be a flat spot as the secondaries open. However, too soon of a secondary throttle opening will also cause a flat spot even if the mixture is correct. This can be addressed by putting a stiffer spring in the secondary vacuum capsule. For some reason, with the Holley 570 carburetor, I had a flat spot when the secondaries opened even though the eventual full throttle mixture was fine. Installing much richer secondary and primary jets solved the flat spot, but the full throttle mixture was then too rich. Fitting a stiffer spring in the vacuum capsule worked well, but then the secondaries did not open all the way. For some reason the bigger (670 cfm) carburetor works well with a softer vacuum capsule spring. (Note: the degree of opening of the secondaries can be ascertained by putting a clip on the secondary throttle rod. The clip will be pushed down the rod during driving and give a record of how far the rod has moved. Be very careful that the clip does not jam the throttle linkage and remove it immediately after the test is finished! Note that the secondaries will never open when the engine is revved up in neutral.)
There is a major difference between Holley and Edelbrock (Carter) carburetors in when the vacuum secondaries come into action. With Edelbrock carburetors, the secondaries are interlocked so that they cannot open until the primaries are fully open. With Holley carburetors, there's no interlock and the secondaries can open whenever primary vacuum gets sufficiently high. The Edelbrock interlock is good for fuel economy because the secondaries never open unless the throttle is floored. However, this can lead to sluggish part throttle response, especially with the small primaries that Edelbrocks tend to have. The Holley allows the secondaries to open under part throttle, high RPM conditions. This contributes to a more responsive feel.
Optimal Return Spring Orientation
Incorrect Spring Position
Throttle linkage
Finally, make sure the throttle is opening fully. This is an area that is often overlooked in carburetor set up, and it can cost a ton of power, I had to drill an extra cable mounting hole in my carburetor throttle arm to get the right match between the MGB accelerator and the carburetor. To check, with the engine off, have someone push the throttle to the floor, and then see if you can open the throttle at the carburetor any further. First be sure that the carburetor is fully warmed up and the choke is off because some choke mechanisms limit the throttle opening when cold.
For connecting a throttle cable, I recommend cable ball joints from pegasusautoracing.com (either fixed, # 1276 or quick-disconnect, # 1274).
For safety, use dual throttle-return springs and make sure that at full-throttle gas pedal movement is either limited by the floorboard or "pedal stop". The carburetor-mounted linkage is NOT designed to withstand the load you can exert with your leg muscles! Read the carburetor's manual for additional installation recommendations.
The throttle return spring should be installed so that it directly opposes the throttle cable, and not on the opposite side of the throttle shaft. A spring connected to the opposite side of the throttle puts a load on the throttle shaft as the cable pulls against the spring. This creates much more rapid wear in the shaft and bushings which leads to air leaks that will cause an erratic idle and a poor transition response.
Choke
A choke is not really needed with a manual transmission car, but it is a convenience. An air cleaner is, however, essential. With limited hood clearance, often all that will fit is a triangular foam air cleaner by Edelbrock. It is, however, rather poor because it does not filter well, and can upset the flow into the carburetor. An alternative that is almost as compact is the Mr. Gasket dropped-base air cleaner, which really works much better except for one problem. The top of this air cleaner is so low that it can interfere with the choke plate. This is a consequence of the recess in the lid where the attaching nut goes. No other drop case air cleaner has this recess, and it is the main reason why this air cleaner fits under low hoods. I solved the choke interference problem by simply removing the choke plate and shaft. I left the electric choke actuating mechanism in place, as well as the rod that went to the choke shaft. The rod and its plate seal what would otherwise be a hole that would let unfiltered air into the carburetor. The choke mechanism still actuates the fast idle control, and this is enough to make the engine livable, even down to zero degrees. The air cleaner lid just touches the float bowl vent tubes. In the primary carburetor side this looked like it might provide a restriction to air flow because of the choke tower. I eventually cut off the choke tower, and it seemed to make a difference in that the engine seems freer revving before the secondaries open (but it could also be my imagination).
Choke Tower Cuts
Finished View of Modified Choke Tower
One other tip to gain space for the air cleaner is to have the carburetor mounting flange angle milled to tilt the carburetor forward. Just a few degrees will allow the air cleaner to follow the slope of the hood, and this does not necessitate any change to the carburetor mounting studs. This is a tip I picked up from Glen Towery.
Example settings
The optimum settings I found for my car are:
Holley 670 street avenger
Standard secondary spring
Standard accelerator pump and cam
0.033" Accelerator pump jets
10.5 Power valve
58 primary jets
68 secondary jets
Your results may vary because different engines, cams, etc. will influence what the optimum settings will be.
With the above settings my carburetor is "on the edge". This shows up as a slight temperature sensitivity, where below about 40 degrees there is a slight flat spot with small throttle openings (mainly in 5th gear below about 1800 rpm). This gets slightly worse as the temperature goes lower, but even at zero degrees it is still not bothersome. However, this is why car manufacturers adopted heated air supplies to the carburetor with the advent of emissions controls.
A word about manifolds
There are several types of manifolds available including dual plane, single plane and X type (or 360 degree type). One important property of these manifolds is the plenum volume. A large plenum volume is needed for optimal high rpm power. Mixture is drawn alternately from one side of the plenum and the other. The gas develops an inertia as it flows, and this inertia makes flow reversals more difficult. At high engine speeds the inertia problem is worse, and is one of the reasons why high speed cams open earlier on the intake cycle than low speed cams. A large plenum volume reduces the velocity of the gases in the plenum and so minimizes the inertial problem. Performance type manifolds (single plane and X type) already have large plenum volumes, but dual plane manifolds, because they are not intended for all out performance engines, may or may not have decent plenum volumes.
With an engine that has a mild cam and restrictive heads, such as a typical 215 V8, plenum volume of any manifold is sufficient. However, for engines with power peaks in the 6000 rpm range, plenum volume becomes important. For hood clearance reasons, I am using an Edelbrock Performer manifold on my Ford V8. This has a much smaller plenum volume than the Performer RPM manifold, but is otherwise similar. Unfortunately, the small plenum volume of the Performer manifold is probably costing me 30 to 50 hp. This loss can be somewhat mitigated by using a larger carburetor. If there is hood clearance, then a carburetor spacer can be used to increase plenum volume, although it is best to go for a better manifold. In any case, at least with a Ford engine, the availability of relatively cheap stroker kits can make up for the power loss (and more).
Manifolds can be ported. The Edelbrock performer is designed to match the standard iron Ford head port size, but the ports on the GT40X aluminum heads are considerably larger. I ported my manifold to better match the intake ports. The key is to especially enlarge the top portion of the manifold openings to reach the top of the taller ports. Most air flow down the port is on the outside of the port radius, and this starts at the top of the port, so the manifold should match here. The bottom should also be enlarged, but a perfect fit is not as essential. I also enlarged the passages back into the manifold to the point where they opened up as they approached the passage junctions. To make sure the passages were reasonably consistent, I bent coat hanger wires to use as gauges, one for the width, and one for the height. Ideally, the manifold should be flow bench tested to ensure all passages flow equally, but I did not do this.
I really do not know if the power has increased because any increase would be at higher RPMs, and the high RPM power was already really good, with the power peak arriving at 5900 RPM. In any case, there was no noticeable loss of low RPM torque or response. Eventually, I plan to get the car dyno-tested to see what difference this made. (I already have a baseline from before the modifications.)
Disclaimer: This page was researched and written by Larry Shimp. Views expressed are those of the author, and are provided without warrantee or guarantee. Apply at your own risk.
Photos by Larry Shimp. All rights reserved.
Holley secondary metering systems don't normally have a power valve because the secondaries only come into play at relatively low vacuum conditions. But without the richening effect of the power valve, the secondary jet sizes need to be bigger than the primary jet sizes. That's why Holley carburetors normally come with secondary jets about 8 to 10 sizes bigger than the primary jets. If you buy a used carburetor, always replace the power valve just to be sure, and check to see what jets are installed.
Ignition timing
Ignition timing is also important to success with a lean cruise mixture, and a vacuum advance distributor must be used. This will keep the timing well advanced so the slow burning lean cruise mixture will fire properly. Then, as soon as the throttle is opened slightly, the vacuum advance drops out, and the timing reverts to a power setting. Without a vacuum advance, the engine will run poorly on a lean mixture and will almost certainly hesitate with small throttle openings, even with a properly chosen power valve. (My engine runs fine on a 15 to 1 air/fuel ratio and gets about 26 mpg on the highway.)
Power valve protection
From my experience, one problem with power valves is that they can rupture from a carburetor backfire. The Holley Street Avenger (and apparently all Holleys made after about 1992) has a check valve that protects the power valve in the event of a backfire so it's not a problem with this model series. There are also kits made by several manufacturers to retrofit older Holley carburetors with the check valve. I highly recommend this modification.
Jetting and secondary vacuum canister springs
With the carburetor properly set up and the right power valve installed, final tuning is relatively very simple. The goal is smooth throttle response with no surging or flat spots associated with rapid throttle openings. If throttle response is good, chances are that the full throttle mixture will be reasonably correct as well. First of all, if there is some hesitation upon throttle opening at low engine speeds, it is probably the accelerator pump circuit. Holley a has a number of accelerator pump cams and even a larger capacity accelerator pump itself (50 cc versus 30 cc). I have found that simply putting in the next size larger accelerator pump jets (0.033 inches versus 0.030 inches) worked fine for my car.
Throttle response
At higher speeds the secondary throttles open. When this occurs there should be no surging, just a smooth increase in power. Some people think there should be a surge as the secondaries open, but this only signifies a lack of power. (A surge means either the engine was down on power until the secondaries opened or there is a flat spot coming and going.) There is no accelerator pump associated with the secondary throttles, the only tuning variables are primary and secondary main jets, and the secondary vacuum capsule spring. If the primary jets are too lean, there will be a surge as the richer secondary jets cut in. On the other hand, if the secondary jets are too lean, there may be a flat spot as the secondaries open. However, too soon of a secondary throttle opening will also cause a flat spot even if the mixture is correct. This can be addressed by putting a stiffer spring in the secondary vacuum capsule. For some reason, with the Holley 570 carburetor, I had a flat spot when the secondaries opened even though the eventual full throttle mixture was fine. Installing much richer secondary and primary jets solved the flat spot, but the full throttle mixture was then too rich. Fitting a stiffer spring in the vacuum capsule worked well, but then the secondaries did not open all the way. For some reason the bigger (670 cfm) carburetor works well with a softer vacuum capsule spring. (Note: the degree of opening of the secondaries can be ascertained by putting a clip on the secondary throttle rod. The clip will be pushed down the rod during driving and give a record of how far the rod has moved. Be very careful that the clip does not jam the throttle linkage and remove it immediately after the test is finished! Note that the secondaries will never open when the engine is revved up in neutral.)
There is a major difference between Holley and Edelbrock (Carter) carburetors in when the vacuum secondaries come into action. With Edelbrock carburetors, the secondaries are interlocked so that they cannot open until the primaries are fully open. With Holley carburetors, there's no interlock and the secondaries can open whenever primary vacuum gets sufficiently high. The Edelbrock interlock is good for fuel economy because the secondaries never open unless the throttle is floored. However, this can lead to sluggish part throttle response, especially with the small primaries that Edelbrocks tend to have. The Holley allows the secondaries to open under part throttle, high RPM conditions. This contributes to a more responsive feel.
Optimal Return Spring Orientation
Incorrect Spring Position
Throttle linkage
Finally, make sure the throttle is opening fully. This is an area that is often overlooked in carburetor set up, and it can cost a ton of power, I had to drill an extra cable mounting hole in my carburetor throttle arm to get the right match between the MGB accelerator and the carburetor. To check, with the engine off, have someone push the throttle to the floor, and then see if you can open the throttle at the carburetor any further. First be sure that the carburetor is fully warmed up and the choke is off because some choke mechanisms limit the throttle opening when cold.
For connecting a throttle cable, I recommend cable ball joints from pegasusautoracing.com (either fixed, # 1276 or quick-disconnect, # 1274).
For safety, use dual throttle-return springs and make sure that at full-throttle gas pedal movement is either limited by the floorboard or "pedal stop". The carburetor-mounted linkage is NOT designed to withstand the load you can exert with your leg muscles! Read the carburetor's manual for additional installation recommendations.
The throttle return spring should be installed so that it directly opposes the throttle cable, and not on the opposite side of the throttle shaft. A spring connected to the opposite side of the throttle puts a load on the throttle shaft as the cable pulls against the spring. This creates much more rapid wear in the shaft and bushings which leads to air leaks that will cause an erratic idle and a poor transition response.
Choke
A choke is not really needed with a manual transmission car, but it is a convenience. An air cleaner is, however, essential. With limited hood clearance, often all that will fit is a triangular foam air cleaner by Edelbrock. It is, however, rather poor because it does not filter well, and can upset the flow into the carburetor. An alternative that is almost as compact is the Mr. Gasket dropped-base air cleaner, which really works much better except for one problem. The top of this air cleaner is so low that it can interfere with the choke plate. This is a consequence of the recess in the lid where the attaching nut goes. No other drop case air cleaner has this recess, and it is the main reason why this air cleaner fits under low hoods. I solved the choke interference problem by simply removing the choke plate and shaft. I left the electric choke actuating mechanism in place, as well as the rod that went to the choke shaft. The rod and its plate seal what would otherwise be a hole that would let unfiltered air into the carburetor. The choke mechanism still actuates the fast idle control, and this is enough to make the engine livable, even down to zero degrees. The air cleaner lid just touches the float bowl vent tubes. In the primary carburetor side this looked like it might provide a restriction to air flow because of the choke tower. I eventually cut off the choke tower, and it seemed to make a difference in that the engine seems freer revving before the secondaries open (but it could also be my imagination).
Choke Tower Cuts
Finished View of Modified Choke Tower
One other tip to gain space for the air cleaner is to have the carburetor mounting flange angle milled to tilt the carburetor forward. Just a few degrees will allow the air cleaner to follow the slope of the hood, and this does not necessitate any change to the carburetor mounting studs. This is a tip I picked up from Glen Towery.
Example settings
The optimum settings I found for my car are:
Holley 670 street avenger
Standard secondary spring
Standard accelerator pump and cam
0.033" Accelerator pump jets
10.5 Power valve
58 primary jets
68 secondary jets
Your results may vary because different engines, cams, etc. will influence what the optimum settings will be.
With the above settings my carburetor is "on the edge". This shows up as a slight temperature sensitivity, where below about 40 degrees there is a slight flat spot with small throttle openings (mainly in 5th gear below about 1800 rpm). This gets slightly worse as the temperature goes lower, but even at zero degrees it is still not bothersome. However, this is why car manufacturers adopted heated air supplies to the carburetor with the advent of emissions controls.
A word about manifolds
There are several types of manifolds available including dual plane, single plane and X type (or 360 degree type). One important property of these manifolds is the plenum volume. A large plenum volume is needed for optimal high rpm power. Mixture is drawn alternately from one side of the plenum and the other. The gas develops an inertia as it flows, and this inertia makes flow reversals more difficult. At high engine speeds the inertia problem is worse, and is one of the reasons why high speed cams open earlier on the intake cycle than low speed cams. A large plenum volume reduces the velocity of the gases in the plenum and so minimizes the inertial problem. Performance type manifolds (single plane and X type) already have large plenum volumes, but dual plane manifolds, because they are not intended for all out performance engines, may or may not have decent plenum volumes.
With an engine that has a mild cam and restrictive heads, such as a typical 215 V8, plenum volume of any manifold is sufficient. However, for engines with power peaks in the 6000 rpm range, plenum volume becomes important. For hood clearance reasons, I am using an Edelbrock Performer manifold on my Ford V8. This has a much smaller plenum volume than the Performer RPM manifold, but is otherwise similar. Unfortunately, the small plenum volume of the Performer manifold is probably costing me 30 to 50 hp. This loss can be somewhat mitigated by using a larger carburetor. If there is hood clearance, then a carburetor spacer can be used to increase plenum volume, although it is best to go for a better manifold. In any case, at least with a Ford engine, the availability of relatively cheap stroker kits can make up for the power loss (and more).
Manifolds can be ported. The Edelbrock performer is designed to match the standard iron Ford head port size, but the ports on the GT40X aluminum heads are considerably larger. I ported my manifold to better match the intake ports. The key is to especially enlarge the top portion of the manifold openings to reach the top of the taller ports. Most air flow down the port is on the outside of the port radius, and this starts at the top of the port, so the manifold should match here. The bottom should also be enlarged, but a perfect fit is not as essential. I also enlarged the passages back into the manifold to the point where they opened up as they approached the passage junctions. To make sure the passages were reasonably consistent, I bent coat hanger wires to use as gauges, one for the width, and one for the height. Ideally, the manifold should be flow bench tested to ensure all passages flow equally, but I did not do this.
I really do not know if the power has increased because any increase would be at higher RPMs, and the high RPM power was already really good, with the power peak arriving at 5900 RPM. In any case, there was no noticeable loss of low RPM torque or response. Eventually, I plan to get the car dyno-tested to see what difference this made. (I already have a baseline from before the modifications.)
Disclaimer: This page was researched and written by Larry Shimp. Views expressed are those of the author, and are provided without warrantee or guarantee. Apply at your own risk.
Photos by Larry Shimp. All rights reserved.
