Ever wanted to get into or improve your aerobatics?

[Peter Jenkins]

Steve,

Thanks for your comments. I have also taken to writing down all the Tx set up detail as I had to do that when I changed Transmitters and there was no way I could work out to have these details transferred - there probably is but it was beyond me!

I also write down all the initial set up details for the model but I'm less good at returning to alter the figures once I start tweaking the trim settings! It's easy to see how many sticky weights you've stuck on, but if you have adjustable wing/tail incidence, once you start down the route of adjusting those then you need to go back and measure the new angles that you've set up. I'm not very good at closing that loop!

Peter

[graeme jones]

Peter - all noted re. incidences etc.

My observation re. servo arm length is a result of 50 years of experience, but I could still be wrong. I agree that there must be a point where increasing arm length causes the servo centring error to be greater than the linkage error. If memory serves correctly someone published the math on this in RCM&E many years ago, coming to the conclusion that very short arms and horns were not a good idea (think about what would happen if you could reduce the arm length to near zero. The linkage slack would then be a huge percentage of pushrod movement). I'll see if I can reproduce the math and come back to you.

My point about locked controls will make sense if you have ever tried to fly a helicopter with stiff ball links. 5mm ball in 5mm socket equals no control. 5mm ball in 5.1mm socket is much better for the nerves!!! There's a little tool to ease the sockets so that the thing will just pivot under it's own weight. I use the same for fixed wing. To put it another way, whether using ball links or Z-bends, there has to be some clearance to allow ANY movement. That clearance results in UNWANTED movement.

Graeme

PS What is the normal amount of servo centreing error? Manufacturers publish figures for speed and torque, but what about accuracy?

[Ultymate]

I'm not quite sure where this threads going, is it aimed at getting people to fly competition aerobatics or to just generally get people flying better aeros at their club field ? If the latter you're getting very deep into setup, for sure you need a reasonably well trimmed model with a good C/G, but beyond that a lot comes down to personal preference for the club flyer ie throws rates and expo etc. Apart from high end F3A models very few aircraft have adjustable incidences on wings and tail.

Once you have a well trimmed model with a good C/G and two or possibly three sets of rates it's more a case of get out there and practice manoeuvres and lines and when I say lines I mean practice flying straight lines both the right way up and inverted, vertical lines and 45*lines , all a lot harder than most people think particularly if flying in a crosswind. When practicing loops and bunts try to keep diameters the same and start and finish at the same height.Throttle control is very important as indeed is the use of rudder one of the most overlooked controls of the aircraft. Practice knife edge both canopy in and belly in and also do stall turns canopy in and belly in

[Peter Jenkins]

Hi Ultymate

If you look, I was answering a question raised by Graeme Jones on the wing/tail/engine alignment. It was not my intention to make this a competition pilot thread and start the discussion with these types of details. However, unless you set up your model correctly with the basics, it will not perform as it could. I alluded earlier to setting up a Wot4 and those and other standard airframes is what I am principally looking to help folks set up correctly. If subsequently, some go on to fly in competition that's a bonus. It's just as easy to set up an aircraft correctly and read the benefits as failing to set it up correctly and then finding that it's difficult to fly accurate aerobatics. I agree entirely with your comments in bold but we need to do the basics first otherwise you will always have difficulty flying lines accurately. I have found that out in practice.

By the way, you are wrong about only high end F3A models having adjustable incidences. The Vanquish, which is an entry level F3A airframe has adjustable incidences. Some high end machines don't as their designer believes that he knows best!

You are more than welcome to contribute, as you have done. I don't own this thread. I am merely trying to provide some guidance to those who want it and judging by the response to this thread so far, it appears to be what at least some people are looking for.

Peter

[Peter Jenkins]

Graeme

I've sent you a PM on your first point.

On your second point, I think we are at cross purposes. I accept that if you have a 5 mm ball with a 5 mm socket you are going to get a seized up feeling! The degree of clearance required is so small that for all intents and purposes I cannot see any lost movement. I use a bolt through ball joint that is already located in its socket housing - does that make sense to you? - and once bolted up the only movement is due to the residual servo backlash.

How good should a servo be at centering? Well, I won't name servos here, but a quite expensive servo I was using had its centre vary when returning from the two extremes of travel of 2 mm at the elevator TE. On a pattern ship it was deemed unacceptable so I changed to Futaba and they return to the same place every time. JR also have excellent centering.

Peter

[Max50]

Never thought about servo backlash.Had always put my pushrod on outer or third hole out and adjusted by d/rates.

Thanks for the tip. Haven't got to inverted flying in straight line yet, just rolls with elevator and no rudder,so 45 up-lines turning to inverted would be new to me.My outside loops are my best part.Big,circular and exit on the same line as i entered.Using a speed air with OS55AX.

Will be watching this thread with interest. As you can see , i'm a beginer to aerobatics. One thing i'm finding usefull is a servo tester to line up servos and control surfaces when setting up.

["L"Plate]

Something that is worth noting about servos. Most metal geared servos have a greater clearance machined into the gears than in the nylon one and so often have more backlash. I think that is why most F3A pilots use nylon geared servos.

My sevo set up, even when using ball links, is to attach to the outer hole on the horn and then move the ball in on the servo arm until I have the required movement (especially if using normal clevises) as this reduced the efect of any play in the linkage (but it has no effect on the servo backlash or servo load). I also use 100% movement for high rate and adjust down for normal flight (not normally below 80%). I only use expo to get a liniar feel to the control, the only exception to this is the rudder on high rate which I only use for spin and stall turns.

With 3D aircraft the need for Expo and control set up is a completly different issue to the way F3A and classic aerobatic machines would be set up. I have an Angle 30 and it was set up to fly 3D now closer to F3A but the horn and linkages do not allow me to switch from one flight type to the other without moving the clevises out on the servo arms a lot! Now I have to say I just keep it in F3A mode.

["L"Plate]

For those out there who assemble ARTF models, I have to say the bit that lets most of them down is the control linkages, for control precision "Z" bends, flexible horns, loose clevises, poor hinges and bends in control rods will just not do (no matter which type of aerobatics you intend to fly).

It is worth spending a few pounds on good quality "Ball Links".

A good way of getting stiff rods is to buy threaded rod and a carbon tube that just fits over it. Make up the linkage and with a ball link at each, check out your control movement and centre positions, then place the carbon tube next to the rod and mark it so that when cut you have a couple of threads showing at each end. Now disconnect one Ball Link and apply a little epoxy onto the threaded rod slide the carbon tube over it (turning it all the time to help keep the glue inside the tube). Wipe of excess epoxy and once dry refit the ball link and away you go, you still have some adjustment to the length but you also have a rigid yet light control rod at a reasonable price.

[Peter Jenkins]

OK, we’ve set up our aircraft, the CG is in the middle of the recommended range, the control throws are set, we have some lower rates set as well. Like Algy, I use 80% for the low rate to begin with. If you have the option of 3 rates then you can use an even lower rate for the third position. This can be particularly helpful when learning how to slow roll, of which more later.

The first thing you’ll do when you get your newly set up aircraft airborne is to trim for straight and level flight. You’ll want around 60%-70% power. If you are flying a pattern type aircraft with better than 1:1 power to weight ratio (i.e. it you release it at full power point vertically upwards it will accelerate away) you’ll only want 50% power.

There is another approach which is to trim the aircraft on the elevator so that you need the same amount of pull to maintain upright level flight as you need push to maintain inverted level flight. By all means give it a go, but I find it a lot easier to operate with the aircraft trimmed to fly hands off upright. To avoid confusion, when I say hands off, I mean release the sticks keeping your fingers in close proximity and watch what the aircraft does and trim accordingly. If you can adjust the trim steps on your Tx set them to 3 or 4 steps per click for the first flight and then set them to 1 step per click for the second flight to fine tune your setup.

The next step is to decide where you want your CG. Do this first as if you go through the full trimming process and then decide you don’t like the CG position and change it, you’ll have to re-do all the following trim steps. For aerobatic aircraft, you do not want high levels of stability in pitch but equally you don’t want an unstable aircraft. Some folks like a neutral aircraft but I prefer a slightly positive stability aircraft.

To establish how your CG affects your aircraft’s handling, climb to a safe height, pull to 45 degree climb, then roll inverted and see what happens when you relax your hold on the elevator. If the aircraft pitches downwards rapidly, your aircraft is in the too stable area and you’ll need to move the CG back. If it pitches up, your CG is rather too far aft (although for good 3D flying you might want it there). What you are looking for the aircraft to gradually pitch nose down.

Adjust the position of your battery pack to achieve the change in CG or else use sticky lead weights mounted on the extreme tail of the aircraft – but not on the flying surfaces! This will give you the greatest effect for least weight added.

Once you have the CG right you will find that you may have to adjust your elevator trim as well. If you’ve had to use a lot of trim to achieve level flight at your selected cruise power, you may wish to re-centre your servos and adjust the linkages to give you the right amount of control surface movement. Doing this will mean that you have an equal amount of control movement around your chosen straight and level position.

It’s worth spending time on getting your trim set up correctly as it can make a huge difference to the way an aircraft feels. My earlier comment about how a clubmate felt when he flew my Wot4 illustrates the point. He said, it was as if he was flying a completely different aircraft and that it was so much easier to fly than his Wot4. So, it really is worth doing this and getting it right before your move on to the next step.

Peter

[Peter Jenkins]

Hi Max 50

A Speed Air with an OS55AX sounds a good piece of kit. If you follow all the trimming steps I will be describing you'll have a great platform for aerobatics and something that will take you through to your B certificate if that's what you want. One side effect of getting comfortable with aerobatics is that your confidence in dealing with aircraft in unusual positions means that you stand less chance of crashing. The other tip is not to fly too low as your margin for error is then significantly reduced and you use up a lot more adrenaline!

["L"Plate]

Peter is right about CofG and aircraft trim set up. Most sports models have the CofG forward of the ideal point, this makes them stable in pitch and have a negative AOA (angle of attack) on the tail.

There are two problems with this: 1) In the down lines like after a stall turn, as the model builds speed it put towards the canopy, so needs the elevator pushing to keep it straight down. 2) With the elevators trimmed slightly up, to keep the nose up in level flight works against you when inverted and makes the model dive, it also reduces the amount of available down elevator for outside loop manoeuvres.

I like Peter like to fly most of the time with the model trimmed to fly hands off upright. I am lucky enough that I can set different flight modes: 1) Spin / Flick (High rates) elevator neutral to maintain hands off upright. 2) Normal 90% rates on Aileron, elevator and 80% rudder, elevator neutral to maintain hands off upright and 5% extra elevator down. 3) Roll 80% every thing and elevator neutral down slightly so I need to hold a little back stick upright and a little push inverted. 4) Landing with spoileron and elevator compensation as required. These are just how I like things and I have flown like this for a couple of years now, however, every pilot has their own set up that they are comfortable with and not all have radios that are this flexible (nor do you need it). Dual rates are all we really need and in the old days we even managed without them!

[graeme jones]

Peter - I've sent a reply to your PM. My question re, incidences etc. came about because, many years ago, I always set up my aerobatic aircraft at 0/0/0, on the basis that they spend as much time upside down as right way up. Trouble is that straight and level flight the right way up requires the same amount of up elevator as inverted needs down. Not nice to fly. Since reading trimming tips on by UKRCAA and, even better, by Brian Hebert (on triangulation trimming) I have gone to a similar setup to yours.

I too have a Wot4. It started life with an OS70FS and a large chunk of lead in the tail (about 100 gms) but was still so nose heavy it would not spin. Changed to an Irvine 40 and took the lead out resulting in an aircraft 300 gms lighter, which balanced in the right place and which will spin well.

Graeme

[Peter Jenkins]

Graeme, I think the 0/0/0 set up was appropriate because the aircraft of that era flew at a much higher speed than today's do. Ballistic is the word I think!

Re the Wot4, I recently fitted mine, which has an Irvine 53 Mk 2 (brilliant engine and lighter than the 40!) with a mini pipe. The performance has been transformed so that I can now fly almost any size loop I want with the only problem being to keep the beast in sight! I do have quite low rates for flying aeros with the Wot 4 and increase rates for spinning as otherwise you just get a spiral dive. I also use the higher rates for snaps but these are so fast you have to be lightening quick to do just one!

[Peter Jenkins]

Some of you might think I’m being very pedantic on this trimming issue. Well, in aerobatics, you are less concerned with just flying your model as you with flying specific shapes. To help you to do that, we try and make the actual flying part as easy as possible. A well set up aerobatic aircraft is incredibly easy to fly. That lets you fly your chosen shapes very easily. However, to fly those chosen shapes perfectly is very difficult. Luckily, you can have enormous fun flying imperfect shapes while you dream about being able to fly the perfect shape one day!

OK, our next task having settled on our CG is to sort out the engine/motor thrust line.

First, fly your aircraft at its cruising power setting (65% say) in a straight line parallel to the runway and into wind. As the aircraft comes abeam your position, smoothly advance the throttle to full power and watch to see if the model dives or climbs. If you are very lucky, it might do neither and just proceed in horizontal flight – if you’re lucky that is! So, if the model climbs, that tells us we have too much upthrust. Depending on how your engine is mounted (upright, inverted, sidewinder) you need to tilt the thrustline downwards. If you have an upright engine mounted on typical engine mounting beams, put a washer behind the two top attachment bolts of the engine mount. Then go and repeat the exercise above. Note, that you’ll need to re trim for straight and level flight as you’ve altered the downthrust.

If the model descends when you apply full power, then you need some upthrust so, for an upright mounted engine place a washer between the firewall and engine mount for the bottom two bolts.

Once you’ve got that sorted, and don’t worry if you cannot get it absolutely perfect so long as there is a substantial period of time when the aircraft flight path continues unchanged.

Next, we’ll sort out the side thrust. To do this fly into wind with the aircraft straight and level at cruising power. Then smoothly ease the aircraft into a vertical climb and at the same time smoothly advance the throttle to maximum and watch what the aircraft does. Do not make any adjustment to the rudder while you do this. If the flight path curves to the left, then you need to put more right thrust onto the engine/motor and if it goes to the right you need to reduce your right thrust or put some left side thrust on. For more right thrust, and again for a vertically mounted motor, put a washer between the firewall and the engine mount on the two left hand mounting bolts. For less right thrust, put a washer between the firewall and engine mount on the right hand 2 mounting bolts. Keep doing this until the aircraft goes straight up for a reasonable amount of time – certainly for as high as you would like to go for a stall turn.

You can mix some rudder to throttle or elevator to throttle to give you a similar effects, but my experience is that sorting sidethrust issues using mixing is not as good as moving the motor thrustline.

[Steve Colman]

Posted by Peter Jenkins on 02/12/2013 20:53:44:

Some of you might think I’m being very pedantic on this trimming issue. Well, in aerobatics, you are less concerned with just flying your model as you with flying specific shapes. To help you to do that, we try and make the actual flying part as easy as possible. A well set up aerobatic aircraft is incredibly easy to fly. That lets you fly your chosen shapes very easily. However, to fly those chosen shapes perfectly is very difficult. Luckily, you can have enormous fun flying imperfect shapes while you dream about being able to fly the perfect shape one day!

Peter,

Personally I don't think you're being pedantic at all. In my few years of regular flying, I've certainly noticed how much more nicely and easily a model will fly when properly set up and trimmed. And, it's not just when flying aerobatic/sport models either. Any model will respond better, fly better, and be more enjoyable. And in my limited and humble experience, it will be less likely to suffer a crash too.

Or......perhaps a pilot who takes the time and care to prep a model properly for flight will also be of a type less prone to risk taking when flying?

Steve.

[Max50]

Peter . I don't know if I'm jumping ahead. You've done C of G, but what about lateral balance. I've got a speed air with a thumping big exhaust, (slight exaggeration ) sticking out one side, which I assume would throw the balance out laterally.How do you do it. Do you just hang it up by a piece of string front and rear, or do you do something more accurate.

[Peter Jenkins]

Thanks for that Steve.

Now, if you are not used to using rudder when you fly, you will be surprised as just how important proper control of the rudder is. Indeed, proper control of the throttle is just as important. In fact, it is essential that you become used to using all 4 primary controls together pretty much all the time if you are to achieve the sort of control that will allow you to fly your aerobatic patterns regardless of wind direction and strength. There are, of course, limits to what wind speed even FAI competitions are flown in but for our purposes, I draw the line at 20 mph.

So, back to the rudder. Now you will recall that the secondary effect of applying rudder is that the aircraft is induced to roll in the direction of the rudder. In aerobatics, we don’t really want this as it really makes doing slow rolls quite difficult. A pattern ship is designed so that application of rudder produces very little secondary effect whereas something like a Wot4 has an almost instantaneous roll once you apply rudder. So, how do you eliminate this unwanted secondary effect? Mixing. You mix opposite aileron to the rudder.

Having made some cockups when first experimenting with mixing, I now make it a rule to put in huge amounts of mixing to begin with to make sure everything is moving in the right direction and then back off the mixing to a starting figure of 5%. If you have never used mixing then hunt out your Tx manual and look it up. Since each manufacturer has different ways of getting to the mixing function that’s your task to find out how. Once at the right screen, you set rudder as the Master and aileron as slave. Application of left rudder should produce application of right aileron. This is called crossed controls. Doing this when the aircraft is flying very slowly has the potential to cause the aircraft to “depart controlled flight”! It might spin on you! So, you could put the mix on a switch to allow you to control when it’s on and when off. The other good thing about putting it on a switch is that you can try it out when at a decent height – just in case you got the set up wrong!

The other thing that can happen when you apply rudder is that the nose drops. So, once you get the rudder and aileron mix correct, check to see if the nose is dropping when you apply rudder. If it does drop, then the fix is to mix rudder with elevator. Again rudder is the Master and elevator the Slave in the mixing screen. Follow the same process that I described for the rudder to aileron mix to get the rudder / elevator mix to work in both the correct sense and with just a few percent as your starting point. You can allocate a separate switch to control the rudder/elevator mix until you are happy that you have it right and then you can put both mixes on the same switch if you want to cut down on the number of switches you need to operate.

If you don't understand mixing or allocating switches to mixes please shout and I'll go into some more detail. However, as I have been using JR Txs for some time, I am really only familiar with setting them up with this type of programming. So, I might need to ask the Futaba, Spektrum, Hitec, Graupner, Jeti etc experts to chip in with how it's done using those manufacturer's Txs.

At the end of this process, you’ll be able to fly the aircraft in flat circles either to the left or the right using just the rudder. Why should you need this I hear you ask. Well, Ultimate mentioned earlier that you need to practice flying lines. By that he means flying on the line you will be using to carry out your aerobatic manoeuvres. This is usually, but not always, parallel to the runway in use. The wind rarely obliges us by being aligned with the runway in use so we need to ensure that as we fly along this line the aircraft is neither blown in or blown out by the cross wind component. In aerobatic flying you use the rudder to make fine adjustments to the line while keeping the wings level. Mixing out the unwanted secondary effects of rudder allows you to focus on keeping a constant distance out without having to juggle with aileron and elevator to keep the aircraft on an even keel. Ah, that’s cheating I hear some people say. To which my response is, if you are good enough to fly an accurate line without mixing please go ahead and show the rest of us up! There is nothing in the aerobatic code that prevents the use of mixing. What is not allowed is having any other aids, such as a gyro, to provide stability.

[Peter Jenkins]

Hi Max50. You are right - lateral balance will be along in the next day or so!

[Rick Tee]

When application of rudder produces some roll component (roll coupling) it can be positive (in the direction of the rudder) or negative (in the opposite direction of the rudder).

Trim chart: **LINK**

Some tips on flying aerobatics: **LINK**

[Peter Jenkins]

Rick

You are absolutely correct to point out that the secondary effect of rudder can be roll in either direction. However, I've never yet seen an aerobatic model aircraft exhibit this behaviour and I'd be really interested to hear from those who have.

The "how to trim your aerobat" link is absolutely first class. I read that earlier this year and thought it was an excellent contribution. Max50's request on dealing with lateral balance is well covered in that link and was the technique I was going to describe in my next post.

It was my intention to describe not just how to do an aerobatic manoeuvre but also how to deal with the effects of headwind, crosswind and aircraft characteristics. This is still some posts away but those who want to get on with it might find it a useful starting point.

["L"Plate]

Most aerobatic aircraft have been designed to remove coupling with rudder but some still have a small amount of pitch or roll coupling. In days gone by we needed to change the amount of dihedral and used anhedral on the tailplane to help get around this now with our new TX's we can use mixes to help get rid of these to some extent.

It is very dependent on the design and CofG how much we need to mix to make the aircraft pour in yaw.

[Martyn K]

A post to follow along. I have a feeling with the salient points extracted this would make a very good article.

M

["L"Plate]

Martyn you are probably right.

Don't forget that once you have adjusted you thrust line and re-trimmed you should go back and recheck your CofG. You need to do this as with any change in the elevator trimmed neutral is in effect altering the AOA of the tail and therefore the overall effective centre of pressure (or if you like the position from which the total life is acting) compared to the CofG.

As with nearly all the trimming stages you will end up going back to recheck from stage 1 all the way through to the next step and then back you go again. It is time consuming but it is very worth the effort.

I have had fairly new pilots who have just got their "A" fly one of my old pattern ship and been supprised at how easy it was to fly, as Peter says "Well, in aerobatics, you are less concerned with just flying your model as you with flying specific shapes. To help you to do that, we try and make the actual flying part as easy as possible."

[Masher]

Peter

Just a note to say that this is very good stuff, you are explaining things very well and I now feel confident enough to go and play (carefully) with mixing

Keep up the good work!

[Peter Jenkins]

Right, Max50 wanted to know about sorting out lateral balance. First question is why isn’t a static lateral balance satisfactory. Taking a further step back what is static lateral balance. Well, let’s say we have a low wing aerobat. We support the back end on the tail wheel or skid, and hold onto the spinner having first set the engine with the piston at bottom dead centre. This gives us an indication that the wings are roughly the same weight and any non-symmetric bits, such as side winder silencers or cylinder heads when using a side winder, are balanced out. Well, turns out that once you get airborne and start subjecting the model to aerobatic manoeuvres you find that the static balance hasn’t really helped as when you pull out of a dive one wing always drops.

The reason is quite simple. You may have achieved a balance but the weight distribution across the wing, or tailplane might mean that when you start pulling “G” slight variations in the position of the weight distribution between each side start to matter when the G reaches 10 or 15. The most agressive full size aerobatic aircraft are stressed at around +9G to -9G. There is no point in going higher as the pilot wouldn’t be able to cope. Indeed, without a G suit, or special training, the human body finds it difficult to cope with 4G and is far less tolerant of negative G. Our models don’t have that restraining feature so we just pull or push the elevator stick to our hearts content, and provided the wing doesn’t snap, we’re quite oblivious to the load factor (G) to which we are subjecting our aircraft. So, that’s why doing a static lateral balance is not always time well spent. It is sorting the dynamic lateral balance we really want to achieve.

Here’s a way that those of you who read the trimming link that Rick Tee gave us will recognise. I used to use the technique of pulling tight loops to see which way the aircraft was pulled by the heavier side rolling the aircraft in that direction but this method is simpler.

Ideally, you need someone to help you with this as it’s easier if there are two of you watching what happens. Climb high and then close the throttle completely and push the nose down to achieve a vertical dive. (I’ll give you a tip in a later post on how to achieve a true vertical hands off dive). Then pull out fairly quickly into level flight. What you and your helper are looking to answer is “which wing dropped”. You may need to do this several times and you also need to ensure that when you pull back on the stick you don’t inadvertently apply aileron which would completely negate the test results. Once you have identified which wing drops consistently you can then decide how much weight to add to the wing tip of the “lighter” side. This is very much an iterative process. Add a bit too much weight and the light wing becomes the heavy wing. Make small adjustments and re-fly the test until you find that you can pull out with wings level.