What is the brain for man is the control system for our control line airplanes. Both of them tell the working parts what to do and how to do it. We can expect satisfying results only if those systems work properly. In case of the brain we don’t have much influence on the original design. But we do have in case of the control system. Since this is of much simpler construction there’s hope that we can make it work more reliably. In another article there is detailed information about proportions and relationships between control components. Here we try to develop the system right from the beginning and hope to find correct dimensions.
Since the FIREBALL age we use the same basic system, and since NOBLER years no radical change has occured. When building from a kit, or when using commercial control components, we need not trouble our brain. As soon as we deal with own designs however, we might find ourselves in dear need for special parts which we have to construct - and to design - ourselves. If we want to change something we need to know how the system works and what we have to do. Several aspects have to be considered. Let’s start right at our hand.

The hand
Try this exercise: hold your arm horizontally and fully move your wrist UP and DOWN. If you watch your movements carefully you’ll notice that you can move your hand more down than up. The amount can differ at different individuals, but this is the general way our wrist works ( I’ll come back to this later ). If you want to get an exact picture, hold a short stick in your hand, hold the hand close to a white cardboard, have a light throw a shadow on this cardboard, and draw the outlines of the shadow.
. You’ll be surprised about the painting! It’s shown in sketch 1 . Now we normally don’t use full wrist movement for flying our airplanes. But it’s soothing to know that we have some “reserve travel” should we ever get into trouble.
  The handle
In theory if the vertical line distance is ( say ) 100 Millimeter, the total line travel would be again 100 mm. But our wrist allows only part of this dimension. Since it’s different with each individual I'll not give exact dimensions; let me just call it “Travel B” ( see sketch 2 ). By the way, since most bellcranks have the same dimension ( 100 mm ) the bellcrank moves exactly as much as our hand does. With the help of our “hand sketch” we can also find “Angle A” at which our hand moves.
. The amount of this angle may be helpful when we draw the bellcrank size and location into the fuselage top view. I like to find this location by using an old unused bell crank, laying it on the plan, and playing with it by giving up and down control input ( it’s fun! ).
Bell crank installation
Usually the wing centre ribs are exactly located in the place where the fuselage sides meet the wing and are glued to it. This makes for a strong connection between both. However at the same time these centre ribs somewhat limit bellcrank movement. Using that newly found Angle A and/or Travel B on my play bellcrank, I try to find a position where the bellcrank can have full movement without needing excessive holes cut into the centre ribs ( see sketch 3 ).
Also I like to mount the bellcrank more inwards. If the pivot point is in or close to the wing centre, we have a situation as seen in sketch 3 : the bellcrank output arm is far off centre. Since the flap horn needs to be placed close to the wing centre there will be a problem to connect bellcrank and horn. At least we may encounter some asymmetry in control deflections. So in the end bellcrank location is a compromise and depends largely on our chosen airplane ( wing ) design.    
Component size
In order to avoid above mentioned problems one might be tempted to use a smaller bellcrank ( as often supplied in model kits ). This is not a clever solution. The reason is very simple: the majority of pilots want to use the full deflection of their wrist to control the airplane. Only this way is it possible to control precisely and to give tiny control inputs with subtle movements of the hand, and only this way can we have a smooth flying yet fully manoeuvrable airplane. With a small bellcrank we would give away this advantage since we would have less travel at the bellcrank, thus giving away part of the travel we have available at the handle. Two more arguments: with a smaller bellcrank we have smaller moment arms. Even if we have very little play - and we mostly have - in those bellcrank to pushrod connections, this play ( = bellcrank can move a certain angle without moving the pushrod ) is just smaller if the moment arm is longer ( see sketch 4 ).
  Another advantage: as physics tell us we need less force if we use longer moment arms. So with smaller bellcranks and horns we have higher loads on the pushrods. Apart from the problem of bending pushrods, we surely have much more wear in the bellcrank and horn holes on smaller components, at least on those more simple constructed ones.
Now we have got the right handle, a reasonable sized bellcrank, and we know Travel B ( sketch 3 ). What we are missing is the size of the control horns. To find these we’ve got to make some drawings. For reference see sketch 5. I usually start with the centre line L , with the bellcrank pivot point on it. Also we need two parallel lines P on each side of L with a distance of Travel B. The bellcrank hole on the input arm is moved from one line P to the other ( we could also use Angle A, but this method is not as precise and easy to do ). Of course this is exactly Angle A, but I’ll call it Angle C here. If we carefully draw Angle C on the pivot point, the circle arc of the output hole will cross the angle lines. The distance between these two cross points is Travel D. This is exactly the pushrod travel available to control the elevator and/or the flaps, depending on our choice.

We need this Travel D to find out the horn length. This time we use the opposite proceeding. It’s shown in sketch 6 . Again we need a centre line L with a pivot point R . Again we need two parallels P. These are drawn at the distance D ( = Travel D ). Now we draw a circle arc around point R. This arc will cross the two parallels, thereby creating two cross points. Connecting one of the two cross points with pivot point R will produce distance F. We’ve made it !    
Control systems
For simple airplanes only elevator control is used. So we can get on with bellcrank, one pushrod, and elevator horn. If we want to use flaps we have three different ways to choose from. One version uses a bellcrank with two holes on the output arm. Usually the hole with the longer moment arm controls the elevator. The smaller moment arm is used for the flaps since some flyers prefer less deflection at the flaps. If a 1 : 1 ratio is desired ( = equal deflection on flaps and elevator ) an additional advice can be fixed to the pushrod and controls the flaps.
. A more elegant solution is using two pushrods: one from bellcrank to flap horn, the other from flap horn to elevator horn. For this version a special flap horn is required , allowing for two pushrods to be connected. This method is needed for take apart airplanes. Finally we can use two parallel horns on the flap axle. This version allows maximum freedom of adjustability.
It doesn't make sense to change the size of the handle drastically because the space for the bellcrank is limited, and thus bellcrank size is more or less given. As a result the pushrod travel doesn't vary much. So we have to work with this number. We just decide on the desired control surface deflection, and with the description given above we can find the horn length. Whatever system is used, the method to find out bellcrank size, control surface deflection, and horn length is basically the same. We can also reverse the route and find the deflection for a given horn length. Or the required moment arm for a needed pushrod travel. Or - what do I know.

I said I will tell more about hand position. There is so much discussion about the so called "handle bias". This term describes a seemingly forward tilted handle but with the line connectors still in the same vertical plain. A good example is the famous Jim Walker U-Reely handle, and I build my handles the same way, as seen in the picture. I think the answer to this question is very simple. Just do the same trick with the stick in your hand. Hold your arm exactly the same way as you do when flying your airplane. This can mean a fully stretched arm or a bent elbow.
Watch the position of the stick and you have the answer. With a stretched arm and your hand held in a relaxed position the top of the stick will be tilted slightly forward, thus has some bias. This hand holding is your neutral setting with equal movement for up and down. If however you fly with a bent elbow ( many top flyers recommend this ) you will see that the stick in your hand stands vertical. Your handle should be shaped accordingly : no bias.
If however the bias is in our brain, things tend to get somewhat complicated. Sorry, I cannot comment on this. The bias, you know !

If you want to read more about control system components and construction you may visit Claudia Kehnen's website. There's detailed information about all essential parts, including a few drawings. Click on that little "bellcrank" below and see.