HobbyKing Bonsai Review – Low stress fun on a budget

Low stress fun on a budget – just leave it in the car

My model flying has gone from a weekly event to an almost daily event thanks to the Hobby King Bonsai. It just stays in my car and is a great little model to just chuck up on spaces like the cricket pitch next to my house – spaces where it would be just impossible or dangerous to fly any of my other models.

I met a couple of guys while camping at the Weston Park Model airshow and one of them, who was heavily influenced by Josh Bixler of Flite Test I think, thrust a transmitter into my hands while he was flying his Bonsai. I loved it immediately and ordered one the next day. Delivery took about 7 weeks but it was well worth it.

So what’s so good about the Bonsai ?

=> At only 24” it sits on the parcel shelf of my car and goes everywhere. My wife loves it – she no longer lives in fear of damaging some precious model as we pack the car. It is almost indestructible and takes up very little space. It also looks cheap so there is no fear of it getting stolen.

=> At only 24” and being a wing you can fly it from a pretty tight space. I have flown it from cricket pitches, football pitches and even my mum’s back garden ( although 2 trees had to be climbed on that last occasion ). I guess this is “park fly” territory but that class of model never existed when I started in the hobby.

=> It is almost indestructible and feels low risk. It has been bounced, cartwheeled and crashed into walls at full speed with very little damage to itself or to others. At 180g ( or less ) and with a pushed prop it has much less potential to cause harm. Superglue works really well to repair it.

=> It pretty cheap. The wing itself is £14, the motor is £7, 2 servos are £3 each, esc ( Turnigy 10 plush amp ) £6, and the batteries are £4 each. The light weight 4ch receiver for my Taranis was about £15.

=> It’s the first time I have had a model with a prop-saver. They are worth their weight in gold.

=> The manual is surprisingly good for a Hobby King model

=> It is just great, care free fun 🙂

So what are the downsides ?

=> It is not suitable for an absolute beginner. Being a wing it is very easy to become disorientated if the wing ends up inverted – which It can do fairly easily.

=> Being a wing the flight characteristics are a bit different from conventional models but it is pretty robust.

=> Being small it can get blown about a bit. The manual says not to fly when “wind strong enough to make the tees rustle”.  This is good if maybe a little over cautious advice. The best fun I had with it was flying it off the top of a mountain in a very strong breeze that required full throttle just to stay still. It was an absolute blast.

=> The quoted stock motor ( 1612 ) does not seem to be available. I used an 1811 2900kv motor which works well. The size of prop is pretty much limited by the design ( unless you mod it ) so you need to find a motor that will do 3000 kvm on a 2s or 2000 kvm on a 3s. I was wary of needing to stock 2s batteries but tbh they are so cheap that it is not a problem. The connectors on these models also tend to be smaller jst form factor.

=> I had to disassemble one of the HK 3g servos and fix it when it jammed after 30 seconds. I would highly recommend using a servo tester to stress test the servos before use.

=> There were no self tapping screws included with the wing or the motor

Other useful stuff

=> I used extreme packing tape on the leading edge to make it a bit more durable

=> I used the same tape to cover the esc and receiver

=> I added 2 carbon rods to the nose to stiffen the nose up a bit after the first few crashes.

=> I also used an 800mah battery instead of the suggested 500 mah unit. This meant that I had to fit 20g of lead to the motor mount to get the balance right. In hind sight I would go for either the 1811 motor with a 500mah battery or a larger motor and the 800 mah battery.

=> Don’t forget to order some JST battery pigtails leads to connect the esc to the battery.

=> Don’t forget to order spare props and prop saver bands – possibly these

The Verdict

Would I buy one again ? I already have and it is sitting there in its box for when this one dies 🙂

Aileron Differential / Adverse Yaw and a touch too much cleverness – Hard Lesson Learned and



Update – while everything below still stands the cause of the problem was worse than I thought. In introducing a curve for the aileron differential I had reversed the servos to make it work BUT I had not tested “basic or panic” mode afterwards – and so the aileron controls were reversed in that flight mode – which is kind of ironic. This has made me even more convinced that a basic / panic mode with no mixes is not necessarily a good idea or needs to be done and (re)tested with as much care as the other modes.

I read an excellent article in the BMFA journal ( issue 121 , unfortunately not on-line ) about Aileron Differential / Adverse Yaw but unfortunately never got around to doing much about it until this week. Essentially this is where symmetrical deflection of the ailerons causes too much drag on the lower aileron – just like dropping a flap on one side of the model only.

Last week when I set up my new Hobbyking Phoenix 2000 ( review coming soon )  with quite large aileron throws so that I could use crow braking / spoilers. When I looked at the amount of downward deflection of the ailerons it was very significant and alarm bells rang in my head. I used curves on the Taranis to reduce the downward deflection to 40% and this seemed to work really well.


In my Taranis setups I usually have a basic mode with no mixes as a fallback in case I have programmed something wrong. I have always been quite proud of this feature – until today.

As I circled around for landing I accidently knocked the switch into this basic mode. I wasn’t too worried as it was a simple landing and I didn’t need engine or braking. Unfortunately during the slow turn the model stalled, dropped a wing and dived into the long grass.

What had happened was that the 40% limit on the downward aileron deflection had been removed and the excessive downward deflection on the inside wing during the turn had created so much drag that the wing had stalled.

Thinking back to my first slope session a couple of weeks ago when a crashed another plane the symptoms were very very similar.

So the lessons learned :

1) aileron differential is very important

2) you can be too clever when programming transmitters sometimes

I will update the 6 channel Taranis eepe this week

V2 FRSky Taranis 6 Channel Configuration eepe with trainer, flaps, crow spoilers, reflex and instructions

Update – see this important article on Adverse Yaw. I hope to add this feature soon.

markedup v2

Here is version 2 of my 6 Taranis Channel setup for planes such as the Bixler. New additions include volume control, black box recording ( telemetry, switches and sticks ) and a timer.

This file is strictly for OpenTX 2.x and will not work for OpenTX 1.x

Files => eepe file and instructions

I have had my FRSky Taranis for a few months now and am loving the software side of it. My son has a 6 servo bixler and I have set up a configuration on the Taranis to support trainer ( buddy ), flaps, Crow Spoilers and Reflex – mostly just to experiment. You can download the eepe file, the sounds and the instruction pdf below. I will probably blog various sections too.  The latest version ( what ever that currently is ) can be found here => http://www.seancull.co.uk/taranis  The key features I wanted were :

  • A Motor Safety Switch
  • A trainer function so that I could connect the Taranis to another buddy box via a buddy / trainer lead. I wanted to have two modes – a partial mode where I retained throttle control and a full mode where I could pass across all 4 primary functions.
  • High, Medium and Low Rates with Expo
  • Rudder and Aileron mixing to smooth out turns
  • Flaps with elevator compensation
  • Spoilers / Air Brakes / Opposing Ailerons and Flaps or Crow as the glider folks call it. Also requiring elevator compensation
  • Reflex – positioning the flaps in a slightly up position – supposedly makes for more speed – I just like experimenting
  • A basic “panic mode” that removes all mixes
  • Dangerous functions such as flaps or spoilers should have repeating warnings to remind you they are active.
  • Occasional functions should have a once off voice alert to remind you what they are – e.g. the partial trainer mode selection.
  • Timer
  • Volume Control

If you are specifically after an F3 glider set up see Mike Shellims Configuration

Taranis / OpenTx 2.x – the difference between raw stick values and Inputs for rates

A quick note on something I have just realised.

When you look at the mixes screen in OpenTX 2.0 there are now options in the mixes to use Ail or [I4] Ail ( the number may vary )

These are VERY different.

[I4]Ail is the aileron stick value after the rate lines have been applied.

Ail is the raw stick value. If you use Ail then your Rates Switch will not make any difference – you will always be on Full Rates. So check these settings carefully when you upgrade or if you upload eepe files from other people.

There is a useful forum post here


Taranis Smart Trainer Mode using Sticky Switches in OpenTX2.x

Note : this scheme has the instructor using the buddy box


A better Tx with a better Balance Bar

My son is still struggling to get good control of his plane and I reckon it is partly to do with his old buddy box which has poor sticks and no balance bar.

I thought I would make the Taranis the student transmitter and the old 35Mhz Futaba buddy box the instructor “transmitter” – but the Taranis will be transmitting.

I also wanted to be able control the handover from the buddy box and to be able to put my instructor Tx on the ground unattended to hand launch my son’s plane.

Refer to this article about the basics of Taranis and buddy box / trainer mode

OpenTX 2.x and TR1 – 16

This can now be done with OpenTX 2.x because the buddy box / trainer is always transmitting signals TR1 – 16 to the Taranis even if the switch controlling the student transfer is not active – this in effect means that a buddy box switch can be used to control the hand off from instructor to student => the instructor can use the buddy box.

Smart Trainer Switch ( sort of )

I initially tried using the channel 5 switch on TR5 and this worked well but I wanted to try something along the lines of the “Smart Trainer Switch”. This forum post scheme is to transfer control to the instructor when there is a difference in the inputs between the student and instructor. This would allow the instructor to “follow along” with the student but in my ( fairly limited ) experience there will almost always be a difference – for example the instructor may use rudder input whereas the student probably won’t.

So I settled for a scheme where a significant input of the elevator, aileron or rudder will transfer control to the Instructor and it will remain with them until switch 5 on the buddy box is cycled.

Sticky Switch

One of the new features in OpenTX2.x ( maybe 1.9x ) is the Sticky Switch. This is a Logical Switch which is latched into position by one action and then unlatched by a different action. In the example below moving to SA↑ latches the switch and moving to SB↑ will unlatch it. The latching action will last for 3 seconds and will only start 4 seconds after SA entered the SA↑ state.

No latching will take place unless switch SC is in the SC- position


Note that the sticky switch is set by the transition of the trigger and release values into the required state. So for example if SA↑ is left in place and SB↑ is applied then the latch will be reset and SA↑ needs to be removed and re-applied before the latch will set again.

The Scheme

Step 1

Add TR1,2, 4 ( your channels may vary ) to a new virtual channel as shown below. These represent the trainer elevator, aileron and rudder. The throttle is deliberately left out of the equation. Using a channel like this means you can just use one formula to detect stick movement.

In my case Channel 14 now represents an indication of how far the instructor sticks  ( in my case on the buddy box ) are away from their centres.


In my case I only want this feature to work in “Normal Flight Mode”

Step 2

There is no option to use the TR5 signal in the logical switches so it must be mapped to a channel that can in turn be referred to. In my case I am using channel 13



Step 3


Create a Logical Switch ( L9 for me ) that detects movement by the instructor. I selected a value of 25% as all 3 axis are added together and you may have some trim in. Its a matter of trail and error. I also only want this to be active with SG↑ as I use SG to determine the trainer mode I am using.

Then create a Logical Switch ( L10 ) that detects that the TR5 switch ( Channel 13 for me ) has been cycled. In my case it is either +100% or -100% so I used < 10% as a trigger.

Finally create a Sticky Logical Switch that is set by L9 ( Instructor movement ) and reset by L10 ( TR5 switch cycling )

The Result

If the instructor moves their sticks in “Normal Mode” with SG↑ then they get control. If they cycle switch 5 on the buddy box with the sticks centred then control reverts to the student using the Taranis

I was a bit worried that if the buddy lead as pulled out you could be left with no control but it seems to revert back to the Taranis. You should test this for yourself.