Friday 21 October 2011

Designing a transmitter setup for F3F

F3F models are amongst the most demanding in terms of programming. Over the years, I've programmed a variety of radios for this task. These include the Futaba FF7 Super, MPX 3030, MPX 4000, MPX Royal Evo, MPX Cockpit SX, Futaba 12FG and FrSky X9D+ running OpenTx. More recently, on a FrSky X20 running Ethos.

In this post I'll review the main features to look for when purchasing a transmitter. I'll also describe my approach to designing a setup, with particular emphasis on ergonomics.

Simon Thornton prepares his model at the 2015 Welsh Open F3F


Goals for a great setup

In my experience there are three main goals for a great F3F setup:

  • The transmitter must have all the mixing required
  • The setup must be simple to operate in flight
  • Key settings should be adjustable in flight, for quick trimming

Okay, so let's start off by reviewing the mixers, flight modes and other features required for F3F. 

Mixers

F3F is pretty demanding in terms of mixing. I will define a 'mixer' in this context to mean a single interaction between a stick and a pair of control surfaces.

Using this definition, eight mixes are needed to drive the wing servos on an F3F model:

  • for Roll input:
    • Aileron stick -> aileron servo
    • Aileron stick -> flap servo
  • for Crow input:
    • Crow stick -> aileron servo
    • Crow stick -> flap servo
  • for Elevator input (elevator to camber mix, commonly known as Snapflap):
    • Elevator stick -> aileron servo
    • Elevator stick -> flap servo
  • for Flap input (direct camber adjustment):
    • Flap lever -> aileron servo
    • Flap lever -> flap servo

Similarly five mixes are needed to drive the v-tail servos:

  • Elevator stick -> tail servo (pitch)
  • Rudder stick -> tail servo (yaw)
  • Crow stick -> tail servo (trim compensation)
  • Flap lever ->tail servo (trim compensation) [useful but not essential]
  • Aileron stick -> tail servo (for 'coupled ailerons and rudder' aka. 'combi')

Check through the documentation of your radio to see how it handles these mixes. If a radio is short on mixers, one option is to discard the less important mixes. Or else it may be time to upgrade!

Right, so that's mixers covered, let's move on to...

Flight modes

At their very simplest, flight modes allow you to switch between different sets of trim settings, in particular for elevator trim. However to be really useful, flight modes should also be able to activate the particular mixers and control rates.

Only one flight mode can be active at any one time. For this reason, flight modes should be based on temporal phases of flight. The usual ones for F3F are 'Launch', 'Normal', and 'Landing'. These conveniently map to a 3-position switch.

A fourth 'Reflex' flight mode is often used in high lift conditions as an alternative to Normal.

Using this scheme, we can derive a simple workflow for flight:

  • Before launch: pre-select Normal or Reflex mode for the actual run, depending on the conditions.
  • On launch: activate Launch mode
  • Before entering the course: activate Normal/Reflex 
  • At the end of the run: activate Landing mode

Mixer interlocks

Each flight mode requires a particular combination of mixers, shown in bold/italic below:

  • Landing
    • crow enabled
    • snapflap off
    • camber preset off
    • reflex off
  • Normal
    • crow off
    • snapflap enabled
    • camber preset off
    • reflex off
  • Reflex
    • crow off
    • snapflap enabled
    • camber preset off
    • reflex enabled
  • Launch
    • crow off
    • snapflap off
    • camber preset enabled
    • reflex off

Landing mode

Having a dedicated flight mode for crow brakes ensures a consistent aerodynamic configuration.  I call this 'Landing' mode.  In Landing mode, camber or reflex should be inhibited, so that the brakes always start from a zero camber baseline.

On older radios, a switch may be needed to activate Landing mode. With OpenTX and other modern operating systems, Landing mode may be activated automatically as the crow lever is moved form idle. However this increases the risk of accidental deployment if your landing areas are rough (like the Bwlch in Wales). Personally, I prefer to use a switch.

Aileron Differential Suppression

Aileron differential is used to counter adverse yaw when applying roll commands. However differential is a two edged sword: when full crow brake is deployed, any differential will reduce the roll response. 

To counter this, diff should be suprressed gradually as the brakes are deployed.  Some radios have diff suppression built in (Multiplex Royal Evo/Pro/Profi). Other radios (e.g. Profi 4000, OpenTx) require it to be programmed explicitly. 

Note also that only aileron differential should be suppressed. Flap differential should remain constant.

To further increase the roll response, you may wish to incorporate a 'reverse differential' mix, where the downgoing aileron’s movement is further increased as full crow is approached.

In flight adjustment

It's tempting to make adjustments in the programming menus whilst flying, however it's not really safe. A better method is to land the model first, however that's not very convenient. 

The best solution is to use spare transmitter controls - not only is it safest, but you can make comparisons between settings in real time.

Some candidates for in-flight adjustment:

  • Aileron differential
    Being able to adjust diff on the fly is really useful. I use the rudder trim buttons for this, since moulded models don't need rudder to be trimmable.

  • Snapflap volume
    Snapflap is an essential adjustment. If possible assign it to your throttle trim or other 'sticky' control. Or use a switch to provide two or three levels of snapflap.

  • Camber
    Camber is used in Launch mode. Some pilots like to adjust camber actively, I prefer to have it on a knob - I find a good setting and leave it.

  • Crow compensation
    An adjuster for crow compensation is essential to ensure a safe landing during that all important first flight. The mix should also incorporate a compensation curve to handle non-linear trim change.

  • Snapflap Expo
    In flight adjustment of snapflap expo is useful though not essential. I have programmed this on my Profi 4000 as well as the Taranis, and I understand that it can be cooked up on the Futaba 12FG as well.

Spec for an F3F setup

Putting all this together, here's a typical specification for an F3F setup:

  • Four flight modes: Normal, Landing, Launch, Reflex.
  • In-flight adjustable aileron diff, snapflap volumecrow compensation and camber.

Some goals to keep in mind for a good user interface:

  1. Minimise the number of switch operations
  2. Place the main flight switch so it's easily accessible
  3. Use the appropriate type of control (switch, slider or knob) for each adjuster.

Here are the layouts for my MPX 4000 and OpenTx transmitters:


F3F layout on Multiplex mc4000 

F3F layout: FrSky X9D (OpenTx).

F3F on mid-range systems

These days the power is in the software, not the hardware. Some of the cheaper sets are well suited for F3F - the difference is in the amount of customisation available.

FrSky Q X7

The FrSky Q X7 is a £100 radio running OpenTx. It makes an excellent F3F radio, when loaded with my F3F template.

Cockpit SX

The original Multiplex Cockpit SX has all the basic mixers required for F3F. However it inevitably has some limitations: you can't activate flight modes and mixers via the same switch, you can't make adjustments except via the programming knob, you can't disable the crow, and you can't define your own mixers.

Royal Pro

The Multiplex Royal Pro sits somewhere in between the Cockpit SX and the Multiplex 4000. You can adjust up to two settings in flight. You can also define your own mixers and curves. There are some annoyances though, chief of which are the horrid 'global' mixers, and the limited programmability of the 3-position switches.

Futaba 12FG

The Futaba 12FG implements all the features of our ideal setup, except that snapflap expo is tricky to program (there is a discussion on the BARCS forum).

Summary

Hopefully this article will get you thinking about the wider challenges, as well as possible solutions, involved in programming an F3F model effectively.

Links

No comments: