I am reprinting this article that CD (Cyber Dave) of Direct Parts and Bike Talk wrote in 2004. It is interesting to note how the results clearly showed how well the Harley Davidson Keihin CV carburetor maintains intake velocity better than the S&S or Mikuni. When I received permission to reprint this article CD wanted to make it clear that this test was done in order for him to have a better understanding of the CV slide function. It is not a scientific test by any means and is not meant to be taken as such.
Reprint of the original article follows:
I get asked the same question many times a week. It usually goes like this. “I have a Harley Davidson 80” EVO with a cam, exhaust and adjustable ignition. I can’t seem to get the Keihin CV carburetor tuned properly. What carburetor would you suggest? A Mikuni HSR? An S&S Super E carburetor? Edelbrock slide carburetor? Please help!
I reply and explain that I need some more information in order to help. The first thing I would ask is what specifically is wrong with the way the CV works? Does it bog down, hesitate, do you have off idle pop? Do you have poor mileage? What jets are installed? What mods have you already done to the carburetor?
I also ask for specifics on what they have done to the bike. What type of air cleaner? What cam is installed? What brand and model of ignition and coil is installed? I also ask what may be the most important question of all. What brand and model of exhaust is installed?
Knowing what exhaust is installed is important because the style of the pipe often can have a dramatic effect on carburetor tuning and performance. Drag pipes tend to be problematic in carburetor tuning and a nightmare on the EFI bikes due to reversion. Reversion refers to the reversing of the exhaust pulse back up the inside wall of the pipe eventually pushing part of the intake charge back out the intake port creating an intake stand off wave. The result is a bog right in the mid range and part throttle response suffers.
Knowing what camshaft is installed is important because the lift, duration, overlap, lobe separation angle and centerline dictate what RPM the engine performs best at. Some cams can cause idle and low RPM throttle response problems and others can make top end tuning a chore.
The ignition is important because being able to adjust the base timing and the timing curve (advance curve) to match your engines needs makes tuning a whole bunch easier.
When I was drag racing and running a small garage I often had customers bring their cars into the shop for some basic performance increases. I would add a high flow air cleaner element, open up the air cleaner or use an open element unit, add some free flowing exhaust, bump up the timing and some basic carburetor tuning. This always produced a nice gain in performance. Once in a while a kid would come in and complain that his car would bog and pop and stall when he tried to accelerate quickly from a street light. More than once I opened the hood only to find a Holley 650 CFM double pumper with mechanical secondary's. The car would have stock exhaust, no headers, stock cam and stock torque converter. The CFM demand for this combination is usually less than 400 CFM and definitely no mechanical secondary's. I would explain this to the owner and help him choose the right parts for his car. I would talk to him about air flow and how much air his engine could use. I would also explain the importance of air flow velocity.
In its simplest form, an engine is an air pump. You could design your air pump to be most efficient at a certain RPM where you get the best cylinder fill and pump the most air through. The intake and exhaust ports and valves would be designed and sized for the specific RPM range as would the cam shaft specs. The intake would be sized and shaped to help maintain the most flow at the optimum velocity. It doesn’t matter if it is a single cylinder, V2, inline 4, V8, V10, V12 or even an 18 cylinder radial aircraft engine the same principle applies. You can hang the biggest carburetor you want on it but if you cannot maintain velocity, you will never get optimum performance and throttle response will suffer.
Recently I was visiting with Rod Tiner, my old racing partner and a true wizard with small block MoPar's we started discussing HD carburetion and the strange way the HD intake pulses due to the uneven firing order. I was joking about wanting to see how the CV slide cycled without burning my face if the engine backfired when Rod reminded me that he still had our simple flow bench we built to help match the flow in our race engine heads. It is nothing fancy and uses a set of four 10” high speed fans to create a draw (vacuum) in a box below a metering plate. The fans are wired to a control unit that allows switching on and off each fan and varying the fan RPM. In the base cabinet we mounted a sensitive pressure / vacuum gauge so we could simulate intake vacuum levels. There is a 2” x 4’ long clear plastic tube sloped upward meeting a larger 6” x 4’vertical tube and is mounted so that air is pulled directly down into the metering plate. The 2” tube is mounted into the 6” tube at an angle and is back cut so the air flowing past it creates a siphon effect. At the base of the 2” tube we mounted a small 2 gallon aquarium that acts as the reservoir tank for the indicator fluid. The 2” tube ends slightly above the bottom of the tank. The tank is filled to an established mark with transmission fluid which is easy to see and mark. As air flow increases down the 6” tube it creates a consistent siphon effect on the fluid causing it to rise up the 2” tube making observation and marking different air flow amounts easy. We could generate an intake vacuum of over 15” while testing a 650 CFM Holley. So, getting enough air flow for these carburetors wasn’t going to be a problem. It is nothing fancy especially compared to what is used now but it did what we needed allowing us to match port flows as closely as possible. We decided that we could use the bench to help me figure out how the CV carb maintained its flow velocity since we could observe the slide’s movement easily. Then we decided it would be interesting to test each type of carburetor commonly used on Harley's. Butterfly, CV or Constant Velocity and a flat slide carburetor like the Mikuni HSR 42 and Edelbrock carburetors.
I removed the S&S Super E from our shop mule. It is a high performance butterfly type carburetor. It is a simple, straight forward carb and pretty easy to tune for racing.
We twisted Sam’s arm and “borrowed” his Mikuni HSR 42 which uses a slide type throttle. This is an excellent performance carburetor and is almost as easy to tune as the S&S.
I also used the Keihin CVK 40 from our ’93 FXR test mule. The Keihin Constant Velocity or CV type carburetor uses a piston and a butterfly in combination. This CV is modified with the H-Flow elbow, EZ-Just idle needle and has a Dakota kid fuel pedestal installed. The slide is stock HD that has been drilled for faster response, lightly polished and the edges are chamfered in order to smooth flow.
We decided to run three tests on each carburetor. Each test would be repeated three times so we could get an average.
Test one; Open the throttle smoothly from idle to wide open simulating normal acceleration and gear changes.
Test two; Open the throttle much more rapidly from idle to wide open simulating a more aggressive acceleration and gear changes.
Test three; Open the throttle from idle to wide open as rapidly as possible simulating more of a ‘racing type’ start or gear changes.
In order to get some consistency in our tests, we established a baseline level for the idle and adjusted each carburetor to that baseline prior to running the tests. The baseline mark was 2” above the cabinet top; this gave us a repeatable starting point for each test.
The results were pretty interesting and did prove what many experts were already saying. The Keihin CV 40 used on Harley’s has the best throttle response because it maintains the velocity of the intake flow better than either the S&S or the Mikuni.
Here are the results of our tests.
S&S super E butterfly carburetor:
In test one the ball transitions smoothly from the 2” mark to the 4” mark and holds steady.
In test two the ball has a momentary drop below the 2” mark and then climbs rapidly to the 4” mark.
In test three the ball drops to or very near the cabinet base before rapidly climbing to the 4” mark. This rapid drop clearly indicates a substantial initial loss of velocity and a drop in flow (CFM).
Mikuni HSR 42 slide type carburetor:
In test one the results are the same as the S&S Super E.
In test two the ball climbs smoothly to the 4” mark but seemed to rise at a slightly slower rate than the S&S.
In test three the ball drops slightly below the 2” mark before smoothly climbing to the 4” mark.
In all three tests the ball is slightly lower than the S&S at the wide open throttle position. This may indicate that the Mikuni has a slightly lower total CFM than the S&S.
Keihin CVK 40 Constant Velocity carburetor
In test one the results are the same as both the S&S and the Mikuni.
In test two there was no momentary drop and the ball climbed smoothly to the 4” mark.
In test three the results were the same as in test two indicating no initial loss in velocity or CFM.
In all three tests the ball was slightly lower at the wide open throttle position than either the S&S or Mikuni indicating a slightly lower total CFM.
As we tested the CV we noticed how the slide would not move as rapidly as we expected. When the throttle was snapped open, the slide would open much slower and would occasionally flutter a bit. It clearly showed how the velocity was effecting the slides movement.
While not overly “scientific” the tests clearly illustrate the different ways each carburetor responded to rapid throttle changes.
Many thanks to Rod Tiner, one of the best damn engine tuners around and a true MoPar LA small block engine nut.