Slide Carbs and Bumble Bees!

You may have heard the old argument, "According to scientist, Bumble Bees shouldn't be capable of flight". Of course, we all know they can, and the whole argument casts serious doubt on the merits of the so-called "scientists". The reality is that a so-called "scientist" (taken to mean one who isn't very well versed in micro-aerodynamics) just can't seem to make sense of a Bumble Bee. They think they've got it all figured out, and that's dangerous! You see, a laymen doesn't give a second thought to Bumble Bees; they fly, and so long as the bees don't fly around their heads, there's no problem! A skilled aerodynamicist has a competent understanding of the mechanics of Bumble Bee flight, and doesn't get confused about Bumble Bees. But the "scientist" is different. They have just enough understanding of the underlying principles to be dangerous. They can through around some technical terms, like 'Coefficient of Lift', 'wingspan', 'wing beat rate', 'gross weight', and make a pretty convincing argument that Bumble Bees absolutely, positively can't fly! They can be very vocal about their opinions, and sometimes try very hard to pass them off as fact.

So what does this have to do with slide carburetors? As an exercise, try researching Posa on Google. Chances are you'll find lots of dated material (which is understandable, since Posa is no longer in business), and strong opinions that slide carbs are low tech, primitive, and near worthless. You may see references or implications that the underlying problem is that the carb operates in a non-linear fashion, and analogies of "a square peg in a round hole". This sounds like a valid observation. The throat of the carb is round, and as the slide is moved, the carb throat opens, thereby increasing the area available in which to pass air into the engine. A basic understanding of high school geometry tells us that the area of a circle which is partially obscured isn't linear. The area starts out small (or zero), then increases, slowly at first, then more rapidly, finally slowing down again as the circle is nearly fully exposed. It's not hard to see the "non-linearness" of that arrangement.

Examining the fuel metering needle shows that the needle is a cylindrical rod, with a taper cut or ground into it. The tapered rod (needle) moves with the slide, and as the slide moves out, so does the needle. Because of the taper, more fuel is allowed to flow into the engine because less needle is in the way blocking it! The taper on the needle is typically linear, and that *could* imply that fuel is metered in a linear fashion as well.

Thus, with that info in hand, the "scientist" concludes that a non-linear air opening matched with a linear fuel opening means that the design of the slide carb is inherently faulty, and thus deficient. This in turn explains why tuning them for a proper mixture across the entire rpm range is so difficult.

Well, the "scientist" is only partly right. Tuning a slide carb can be difficult, but not necessarily for the reasons they believe. They've got just enough understanding here of geometry to be dangerous.

Let's look at what is really happening inside your slide carb. First we'll examine the air intake side of the equation. As we saw earlier, the carb air intake is typically a circular cross section. The slide blocks a portion of the circle to adjust the area available to pass air. This in turn allows control of the power produced by the engine.

The tricky part comes with understanding the geometry of exactly how the area changes as the slide is withdrawn. To calculate the unblocked area for a given slide position, you'll need to brush up on the method for calculating the area of overlapping circles. The details of this procedure are in your geometry texts, and I'll leave this as an exercise to the reader. Should you work through the math for a variety of slide positions, from fully closed to fully open, then graph the results, it would look similar to the graph below.

You can see by examination that the graph is not linear. However, once you get over the initial portion, it gets *close* to linear. I've added a line as a reference to illustrate this.

Looking at the fuel metering needle, the results are similar. The geometry is a bit different, but the principles are nearly the same. In this case, instead of overlapping circles, we've using sectors of a circle to calculate the area. Running the math and graphing the results produces the following graph.

Note that the last portion of the graph flattens out, giving the whole graph an "S-curve" appearance. Again, it's not linear, but there is a portion of the graph that is approximately linear. Again, I've included a linear reference to illustrate this. There is a major difference when comparing the fuel graph to the air graph: the fuel needle only uses a small portion of the taper. In other words, when the carb is fully open (full throttle), the fuel needle is tapered only to about 30% of its diameter. I've marked the typical range used on the graph below.

The next step is to directly compare the two graphs. I've overlaid the air graph and the fuel graph, but only using the normal range of the fuel graph. Additionally, the scales were adjusted to correspond with throttle travel, so everything lines up properly. Studying the graph below, we can make a few conclusions.

1. The curves may not be linear, but they track each other pretty darn close!
2. Using the graphs, we might predict the carb to be running rich at low power, pretty good at medium power, and slightly lean at high power. In this case, the theory matches pretty well with actual test results.
3. Low power gets fairly rich, fairly quickly. The needle taper here is already pretty low, so fixing this rich-running behavior might be difficult.

The graph also suggests that a compromise might be had if we make a VERY minor change to the needle profile. We need to allow a bit more fuel when the throttle is above half open. If we re-taper the needle *very slightly* from about 50% open to full open, we can in turn make the profiles match a bit better. More on re-tapering the needle will be described in another page.

Lastly, I want to point out that the geometry only suggests the underlying relationship. There are a number of other factors that can have a significant impact on mixture. These other factors could be from the construction of the carb itself, such as surface finish or irregularities, or due to flow dynamics in the intake pipes, or any number of other things. All these factors are important, and if you're having problems, any one of them could be the culprit, but I simply wanted to explain the underlying principles of the slide carb. It's not as horrible as people have made it out to be.

• http://mathworld.wolfram.com/CircularSegment.html
• http://www.analyzemath.com/Geometry/circles_problems.html