Step 4. Prepare a preliminary (scaled) drawing of the cockpit and fuselage.

The above picture represents an increase of the fuselage length just aft of the wing trailing edge of one full bulkhead. Plus nose is longer than the old position of the spinner. All up taking the length from 5m to 6m, at this stage the tail fin ruder and the all moving horizontal stabiliser have not been increased in size.  Further examination is still required.

The main wheels have also been moved aft an inch or two to allow for clearance to the longer (in span) central box spar. Increased from 2m to 3m.

Step 3. Select the type of configuration to be designed

Given it is nearly always desirable to place the fuel, payload and empty weight C of G at the same longitudinal location. Doing this limits the C of G travel, resulting in a configuration with less wetted area, due to less need for trim control.


Using the definitions of outlined in Chapter 3.3 “Configuration Possibilities” on page 95 of part II, the following description applies given the mission statement. This aeroplane is land based, conventional, twin tractor engines mounted in wing nacelles, a cantilever low wing, zero sweep, full span flaps with droop ailerons and a tricycle retractable undercarriage.

Given this is a development of a known design; many of the choices are already made in terms of configuration layout. The fewer changes from the original design configuration, the better!

Step 2. Perform a comparative study of simular aeroplanes

I have found other examples accross the internet of light multi engine aircraft.  Unfortunatly no have been anywhere near my megar budget.


1. Company: Zenith Aircraft. (USA)

Model: Gemini CH620

Type: Kit development

Website: www.zenithair.com/gemini/gem-what.htm

Note: 2 Jabiru engines, fixed pitch and fixed gear, development is on hold.

2. Company: Tecnam. (Italy)

Model: P2006T

Type: Certified aircraft

Website: www.tecnamaircraft.com/Tecnam_P2006T.htm

Note: 2 Rotax engines, 4 seater, retractable gear. As at 2008 cost EUR $280,000 for a basic model

3. Company: Aeroprakt (Russian)

Model: A28 and A36

Type: Complete aircraft.

Website: www.aeroprakt.kiev.ua/eng_html/main.html

Note: A36 built on request and A28 still under development.  I did find the A36 listed on a US disributors web site for $175,000 USD

If anybody finds any more, please leave me a comment.

Step 1. Review Mission Statement

This step does seem straight forward.  No changes yet and work done thus far is inline with the original concept. 

Preliminary Design Sequence I

According to Part II, there are 16 steps in this sequence of design. They are referred to as class one methods. For a summary of each of these steps can be found on page 11 of chapter 2. Note these sizing methods only have an accuracy of ± 10%.   So step by step I will try and follow them and post the results here. Some of these steps look complicated, oh well...

Sizing to cruise speed requirements

From Chapter 1, part 3, page 162 thru 165 it can be seen that formula for cruise speed is;

V = 77.3{h_p(W/S)/σC_D(W/P)}^1/3                     note: V is in mph

or

V = (550SHPh_p / 0.5ρSC_D) ^1/3

	Sea Level		5000 ft
	mph	kts	mph	kts
Vcr @ 50% power =	123	107	130	113
Vcr @ 75% power =	141	123	148	129
Vcr @ 100% power =	155	135	163	142

Note: with 5 hours of usable fuel on board, at 75% horse power then range is 600 Nm at sea level.

Sizing to climb gradient requirements

From Chapter 1, part 3, page 132 that it is possible to combine equations (3.29) and (3.30) to yield the following formula for climb gradient;

CGR = 18.97h_pσ^1/2/ (W/P) (W/S)^1/2C_L^1/2-(L/D)^-1

where: L/D = C_Lclimb / C_Dclimb

            C_Lclimb = C_Lmax – 0.2

            C_Dclimb = C_D0 + C_Lclimb² / πAe

		Climb Gradient  1 as to:
	FAR 23	Sea Level	5000 ft
Take off climb AEO (All Engines Operating)	1 : 12	3.6	4.0
Balked landing climb AEO	1 : 30	4.8	5.6
Take off climb OEI (One Engine Inoperative)	~	3.6	4.0
Balked landing climb OEI	~	4.9	5.7

Sizing to rate of climb requirements

From Chapter 1, part 3, page 129 through 131, it can also be seen that rate of climb (RC) can be determined by;

RC = dh/_dt = 33,000 x RCP   (ft/min)

Where: RCP     = Rate of Climb Parameter

= [h_p / (W/P) – {(W/S)^1/2/ 19(C_L^3/2/C_D)σ ^1/2}]

To maximise RC, it is evidently necessary to make C_L^3/2/C_D as large as possible.

(C_L^3/2/C_D)_max = 1.345(Ae)^3/4/C_D0^1/4

For normally aspirated engines, assumes a reduction in sea level horse power of a factor of 0.85. (i.e. 85hp/engine * 0.85 = 72hp/engine at 5000 ft)

From Chapter 1, part 3, page 129 through 152, it can be seen the formula for best rate of climb speed is;

Vy = [2(W/S)/ρ(C_D0πAe)^1/2}]^1/2

Best rate of climb and speeds	FAR 23	Sea Level		5000 ft
	ft/min	ft/min	kts	ft/min	kts
Take off climb AEO	300	1710	82	1315	88
Balked landing climb AEO	~	1461	58	1049	63
Take off climb OEI	steady	527	79	304	85
Balked landing climb OEI	~	293	57	52	62

What is the FAR 23 Climb Requirements?

From Chapter 1, part 3, page 129 through 131, it can be seen that a twin engine aeroplane with a take off weight less than 6000 Lbs and a stall aped of less than 61 kts is required to meet the following requirements, for FAR 23 certification.

Take off climb AEO (All Engines Operating)

• Minimum climb rate of 300 ft/min at sea level

• A steady climb angle of at least 1:12

Configuration: gear up, take off flap, max cont. power all engines.

Balked landing climb AEO

• A steady climb angle of at least 1:30

Configuration: gear gown, landing flap, take-off power all engines.

Take off climb OEI (One Engine Inoperative)

For multi engine (reciprocating type) aeroplanes with a MTOW < 6,000 Lbs and with a V_so < 61 kts, the requirement is only that a steady climb rate at 5000 ft must be determined. 

Configuration: gear up, take off flap, max cont. power all engines.

Balked landing climb OEI

Note a positive climb performance is not required!

Configuration: gear gown, landing flap, take-off power all engines.

Sizing to landing distance requirements

FAR 23 certification requires an approach be made 1.3 times the stall, over a 50’ obstacle. The landing ground run (S_LG) is related to the square if the stall speed, in the landing configuration, as is the total distance required (S_L) As per Part 1, Chapter 3 page 108 as;

Ground roll (S_LG) = 0.265Vso²

Total distance over 50’ obstacle (S_L) = 0.5136Vso²

	Sea Level				5000 ft
	SL		SLG		SL		SLG
	ft	m	ft	m	ft	m	ft	m
Full flap	1394	425	719	219	1617	493	834	254
Flapless	1788	545	922	281	2078	633	1072	327

Sizing to take-off distance requirements

For FAR 23 certified, propeller driven aeroplanes it can be observed that the distance (S_to) required to clear a 50’ obstacle is 1.66 greater than the ground run. The ground run (S_toG) is a function of wing loading, power loading, & take off co-efficient of lift.

Utilising 2 * 85 Hp Jabiru 2200 engines, then as per part 1, chapter 3, page 95 as;

S_toG = (W/S)_to(W/P)_to/σCL_to    

where:  CL_to = CLmax/1.21 

Sea Level				5000 ft
StoG		Sto		StoG		Sto
ft	m	ft	m	ft	m	ft	m
167	51	277	84	194	59	321	98

Sizing to stall speed requirements

FAR 23 certified multiengine aeroplanes with W_to less than 6000 Lbs must have a stall speed of no more than 61 knots. The power off stall speed maybe determined from part 1, chapter 3, page 90 as:

Vs = {2(W/S)/ρC_Lmax}^½   (answer is in feet / second)

Given the following details from the single engine design are already known; Wing profile NACA 747A415, a highly laminar flow wing with low drag profile but generates poor lift at slow speeds so C_Lmax (clean)   = 1.4 and C_Lmax (flap)   = 1.8 With a Mean Aerodynamic Chord (MAC)    = 1300 mm and a proposed increase in span to 9 meters then;

(S) surface area = 120 sq.ft

(W/S) = 16.5 Lbs/sq.ft

	Sea Level	5000 ft
Stall Clean  (Vs)	59.0 kts	63.6 kts
Stall Flap     (Vso)	52.1 kts	56.1 kts

The single engine aeroplane starting point

I found my single engine staring point!

It is a homebuilt, available as plans only, no kits.  This is good for me as I will need lots of the small details that a set of plans can provide.  I have chosen the Cherry BX-2  Here is the link http://www.bx-2.de/e/html/cherry_bx-2.html

Why this aeroplane?  Well....

1. Two seats, side by side 
2. Low wing design
3. Designed low horsepower requirement (only 65 HP)
4. Tricycle retractable undercarriage
5. Detachable wings, for road transport
6. Full span flaps with drupe ailerons
7. Composite construction

In short many of the features I want in the twin have already been worked out by the designer and given the numbers of these aircraft that are flying world wide, it is safe to say that these features have proven themselves reliable.  If I do not have to re-invent the wheel then all the better, as there is less scope for me to muck things up.  

Preliminary Sizing

Using the information on the engine, for 75% power cruise fuel burn. The previously discussed payload weight and that weight range as a starting guess, we can then use the fuel fraction method as outlined in the Roskam text.  Plus using Breguet's range equation, the total weight of fuel can be estimated to be 271.2 Lbs.  This includes 45 mins reserve.  

 From this we can estimate empty weight.  As the text explains a liner relationship between log of the empty weight and the log of the take-off weight, using the regression line constants for multi engine aeroplanes of composite construction, we can find an allowable value for empty weight.  Comparison of these two methods, showed they matched within 0.04%.  Thus with some confidence I can say the following;

W take-off    =    1984 Lbs   900 kgs

W empty      =    1152 Lbs   522 kgs

W fuel          =     272 Lbs    123 kgs 

Why choose an aircraft of composite construction?  Well....  I think I have found my base model single engine aeroplane for conversion into a twin.  More about that later in the next post.

Payload

Research suggests that standard passenger weights are about 80kg per person, so lets be generous and increase that somewhat.

Pilot          =  100 kg
Passenger =  100 kg
Baggage    =  50 kg

Given most airlines around the world only allow 20 kg per passenger as a checked in baggage allowance, 50 kg for two people is also generous.

If I design the VLT to carry these weights, then the limiting factor may well become the physical size of the compartment.  This would become preferable as it would simplify things, assuming weight and balance is OK, then if you can fit it in, then you can go.

Engine choice

Looking towards using two(2) x Jabiru 2200 engines. They are however rated at 80 HP

Check out their web site https://jabiru.net.au/engines/

Why this one? oh maybe....size, weight, power, fuel consumption.  As an added bonus, I have flown behind one before, they are relatively common and I kinda would like to give Australian made a go.

So until my VLT design evolves to a point where a larger engine is required, I will set about using the specifications for this engine in the design process.  Now we can go forward from this starting point, in terms of  fuel consumption, horse power, weight and physical dimensions.

How Light is a VLT?

Running with the idea of converting a light single to a light twin.  Its almost self explanatory that the lighter the starting point the lighter the finished conversion is going to be and consequently the smaller the engines (and smaller running costs) I can use.

In the growing sport aviation and recreational markets there are a few striking efficient and surprisingly advanced designs out there.  If you go back a generation (or two) in performance, I have discovered a whole class of two seater, typically older designs that where engineered to be flown behind a 65hp continental engine.  Many of these where light enough to be classed as ultralights "back in the day".

I am thinking I need to find a low wing design (easier to convert into a twin), based upon an airframe that could be pulled through the sky by a 65 hp motor.  These aircraft often have max weights in the order 450 - 550 kg.  Allowing for the weight of an extra engine, more fuel, beefed up structure and larger control surfaces; then perhaps two 85 hp engines on an airframe with a max weight in the order of 800 - 900 kg would be possible.

How about some instructions?

The problem now is of course, the not so little problem of exactly how does one go about it? There is the aerodynamics and structural considerations.  Searching the Internet, some good reference texts and examples turn up.  I especially like the free links to resources on certain uni websites.

Many years ago when I taught flying, one of my then students was at Sydney Uni studying aeronautical engineering.  They were using a series of books collectively called "Airplane Design" by Jan Roskam.  Apparently it was common for the students, to refer to these texts as "the cook books".  It would seem that the formulas in the text were based upon the empirical data collected from groups and classes of certified aeroplanes, developing statistical norms for these groups and such the development of formulas that closely predict performance based on those norms.

My goal for the VLT, is to have stable, standardised flight characteristics, in line with every other light twin I have flown.  I do not want to end up with a design configuration that requires any level of extra special pilot ability to control.  Using the Roskam texts should encourage a conservative design, consistent with my goals.

An idea to start with

How and where to begin? That is the question, I have more ideas than actual plans at this stage.

The leading idea I have kicking around, to take a leaf out the big manufacturers handbook.  Look at Piper or Beechcraft where they used their larger single engine aeroplane models as the baseline for conversion into their small twin engine aeroplane models.  Some examples are the Piper Arrow became the Seminole, Comanche became Twin Comanche, Cherokee Six became the Seneca. The Beechcraft Bonanza  evolved into both the Travel Air and Twin Bonanza, the Musketeer became the Duchess. Interestingly Cessna who favours high wing designs in their singles, did not evolve any high wing singles into twins, Cessna appears to have had to design their light twins from scratch.

Clearly I need to find a light single engine, low wing design to convert into my small twin engine design. I need to find a design that has many of the features that I want in my VLT, but already in a single.  If the base single design already has most of the features worked out. Then there is less work for me in the designing process. Why re invent the wheel? The less I have to do the less I can muck up!

Now the search is on for my starting point, using those requirements from the mission statement as a guide, what singles out there fit the bill?

Why a Twin ?

If a single engine aeroplane is cheaper to build, maintain and run. Why would you want a twin?  Seems pretty conclusive so there must be other reasons to take on this project.

There is the whole safety argument of having as spare engine if one stops.  The counter argument to this is, all the good engine will do, is to get you to the crash site sooner.  True only if the pilot does not maintain their currency in engine failure after take-off drills.  Having my own plane, flying it regularly, maintaining my emergency procedure currency, then having a second engine should be a positive.

If I make best advantage of the rules that govern, designing, building, maintaining, (and yes flying too) an experimental aircraft, I hope that many of the associated costs (relativity speaking) can be kept down to within budget.  So compared to other experimental aircraft, yes it could be more expensive, but compared to even single engine certified aeroplanes, it would be cheaper.  Let alone any comparison to a certified twin, it would win hands down on cost.

The real reason to have a twin is to log multi-engine flight time in the pilot's log book.  It is the Australian experience, that employers value multi-engine time over single engine time.  Enshrined in the Civil Aviation Orders is a rule that requires pilot to have 500 hrs PIC multi-engine time prior to upgrading from a first officer to captain in a low capacity regular public transport operation.  For the Australian market you need those 500 hrs PIC multi before you even get a look in at a first officer position.  So if one day my kids did want to learn to fly, this would be the way to go, so all their flying is multi-engine time.

Hello and welcome

Hi,

I have been thinking about building my own aeroplane, actually I have been thinking about building an aeroplane on and off for a long time.  But what to build?  So many choices and so many conflicting specifications.  Actually defining what it was I wanted from the finished aeroplane, helped to shorten the list considerably.  The problem was once I identified my specifications, I could not find a supplier of either a kits or plans that satisfied me.  So...  


What to Do?

In the true spirit of the experimental aircraft category, lets just design and build it myself. Easier said than done, but why not? The rules allow it.  Could I not investigate this possibility? This idea has taken seed, and this blog is intended to follow the design of a light twin engine aeroplane. Why a twin, well that will be the subject for another post.

As I journey down this road I may well find the task is beyond my abilities or time constrains, however regardless of the result, I do expect to expand my own understanding of aerodynamics and structures. This can only be a good thing.