An RC Depron Douglas X3 Stiletto

[Simon Chaddock]

I mentioned I was thinking about this in the "winter builds" topic.

The first question is why on earth attempt to build such a thing? It was pretty much a failure in full size. designed for Mach2 but could only exceed Mach 1 in a lomg dive. It also had the highest take off and landing speed on any plane of the period.

Well the answer is with such an exotic layout I suspect it will only be able to fly at all with the benefit of a truly super lightweight Depron structure which the challenge I like and with such a small wing span it wont take up too much room hanging vertically on the wall. 😉

Obviously the wing are easy to design from a 3 view but despite scouring the internet I have been unable to find any fuselage cross sections. The problem is the fuselage is what the Stiletto is all about so I want to get is right.

I eventually bought a 1/48 plastic kit so I at least can "eyeball" its various shapes. Apart from the final bit of the nose spike being circular it has no simple shapes anywhere else.  

As I fly from a rather rough grass field it will have to be a hand launch and a very careful belly land. It crash resistance will be zero if not negative!

My usual first task it to take a 3 view into MS Paint and clean it up removing any the excess detail and cleaning up all the lines so they ar both black and just one pixel wide so it can be blown up to the size I want without going 'blurry'. This tends to be a rather slow eye testing process.

I started with this.

And ended up with this

 

My intention is to use a single 50 mm EDF which means of course both a bifurcated inlet and exhaust with the inlet higher than the exhaust the duct is not going to be a simple shape. Not ideal from an EDF efficiency point of view. It will likely be as underpowered as the original. 😟

This could take some time.  

 

[Futura57]

You're a brave man. Good luck getting it to fly.

[mightypeesh]

This will be interesting! Good luck 🤞

[Simon Chaddock]

The first task is to get an idea of the where the duct has to go . The immediate problem is the exhaust outlets are quite near the wings so to incorporate a bifurcated exhaust places the EDF even further forward. This in turn indicates the battery cannot go vary far forward and be clear of the inlets,

So a bit more work with MS Paint to position the EDF and battery and an outline of the ducting to connect them.

 

It looks possible but not ideal. Still thinking! 

Scaling up this view to a 50mm EDF the Stiletto will be 1497mm (58.9")long with a wing span of just 481mm (18.9").😲

My guess is it must not weigh more than 350g (12oz) to stand any chance of flying let alone a hand launch! 

Edited January 12 by Simon Chaddock

[Simon Chaddock]

As in the above view the EDF interferes with the single piece wing and the bifurcated inlet is very long (two small ducts are less efficient than a single of the same area) to give space for the battery I arrived at a Mk2 duct layout.

The inclined EDF clears the wing and allows  the battery to placed under the duct significantly reducing the length of the inlet bifurcation  It also 'frees up' the possible location of the battery to achieve the required CofG.

It all looked to give a significant duct improvement so I started to use the Free Cad "loft" process to create duct sections to give a smooth transitions form one shape to another.

 

The EDF is inclined upward at 15 degrees meaning the two exhaust ducts also have incorporate a 15 degree "bend" to meet the EDF. No problem for a Free Cad loft however when you export a loft into CURA it will only correctly slice it if each end is parallel to each other, which of course the Mk2 exhaust ducts are not. 😒

I do like the proportions of the Mk2 duct so a bit more thinking is required.

To really frighten myself I printed out the above 3 view to the full size of the model.

 

If we assume the CofG is at 25% of the wing chord the side area ahead would be larger (twice?( that behind it, particularly with that tiny fin. I am sure the layout is fundamentally stable however I bet if the yaw exceeds a small angle it would become un controllable. What might be acceptable for an "X" plane I am sure for an RC plane a bigger fin is essential and possibly a larger tail plane as well.

I think a simple 'profile' chuck glider might be a good idea.😉  

 

[Shaun Walsh]

"I am sure the layout is fundamentally stable however ". The full size X3 suffered from roll inertia coupling........

"Walker made an abrupt left roll at Mach 0.92 and an altitude of 30,000 feet (9,100 metres). The X-3 rolled as expected, but also pitched up 20° and yawed 16°. The aircraft gyrated for five seconds before Walker was able to get it back under control. He then set up for the next test point. Walker put the X-3 into a dive, accelerating to Mach 1.154 at 32,356 ft (9,862 m), where he made an abrupt left roll. The aircraft pitched down and recorded an acceleration of -6.7 g (-66 m/s²), then pitched upwards to +7 g (69 m/s²). At the same time, the X-3 side-slipped, resulting in a loading of 2 g (20 m/s²). Walker managed to bring the X-3 under control and successfully landed."... courtesy of Wikipedia.

[Trevor]

Full marks for a challenging subject! A chuck glider is definitely called for but  I'd be tempted to fit a gyro from day one anyway, if that's not against your principles. Also, although I know you love you like your bifurcated ducts, are you sure it wouldn't be easier to use twin edfs on this one?

 

Trevor 

Edited January 19 by Trevor

[Simon Chaddock]

Shaun

I was aware of its roll inertia coupling but I was lead to believe it is a high speed issue rather than a simple aerodynamic effect. My concern is the X-3 lateral stability at low speed as it appears to rely on the fin creating a yaw correcting force greater than the opposing force generated from the forward fuselage area. I suspect the fin does this but only for a low yaw angle. I am not sure a remote RC pilot can ever keep the "ball in the middle" as well as an on board pilot or electronics.

 

Trevor

I have no concerns using a gyro and do fit them particularly when the plane has any 'out of the ordinary' configuration. A gyro is definitely intended for the X-3.

At the size I intended to build it twin EDFs would have to be small say 30mm. Efficiency at this size is low meaning the battery weight for a given thrust is even higher than it is for a 50 mm. I felt the extra weight from two EDFs and more battery was really not what the X-3 needed.

Now if I doubled the size, it would be nearly 9 feet long, then twin 50mm EDFs would make sense. Maybe  next time! 

 

From my previous experience printing LW-PLA ducts I anticipate the weight of the complex duct will be not be that significant.

I did print the two parts of the exhaust duct of my first duct layout as a test.

 It weighs 4.2g. Suggesting the complete duct will be in the order of 25g or 1/3 the weight of the 50mm EDF. A viable option as long as I don't loose too much thrust using it.😉

Still only paper and electronic design at this stage so nothing committed....yet.  

 

      

 

Edited January 19 by Simon Chaddock

[Simon Chaddock]

This is my solution to the duct each section having to have perpendicular ends.

Duct Mk3.

The EDF is  moved forward slightly and raised to be inline with the inlet duct. The exhaust thus not only bifurcates but "S" bends down to match the exhaust outlets.

The hope is the benefit from a straight inlet will compensate for the losses from the complex exhaust path.  The plan view of the duct is virtually unchanged.

The nest task is to create the many fuselage formers using the plastic kit fuselage as a guide. Much will have has to be hand drawn pixel at a time in FREE CAD. It will be a slow process I suspect once the fuselage building starts there will be a fair bit of 'adjustment' to the formers as it progresses.😉

Of course with the heavy EDF well above the equally heavy battery there could well be some form of roll/pitch coupling. 😟

A construction "go ahead" is slowly getting closer. 

[Simon Chaddock]

Still working on the design of the ducts.

I deeded to see what the duct actually might look and feel like when printed in LW-PLA and as I still don't have the 50 mmm EDF that is coming on the slow boat from China printed a dummy EDF in PLA.

This is the printed Mk3 duct.

The duct is very light at just 16g or about 25% of the weight of the 50mm EDF. It will rely heavily on close fitting fuselage formers to reduce the risk of the inlet duct collapsing under negative pressure 

It does fit into the proposed size fuselage.

I really do need to try the duct with the EDF to give me an idea of the thrust it delivers.

The static thrust figure obtained will rather confirm or reject the viability of actually building the plane.😟

 

    

[Gordon Whitehead 1]

On 18/01/2024 at 22:20, Simon Chaddock said:

As in the above view the EDF interferes with the single piece wing and the bifurcated inlet is very long (two small ducts are less efficient than a single of the same area) to give space for the battery I arrived at a Mk2 duct layout.

The inclined EDF clears the wing and allows  the battery to placed under the duct significantly reducing the length of the inlet bifurcation  It also 'frees up' the possible location of the battery to achieve the required CofG.

It all looked to give a significant duct improvement so I started to use the Free Cad "loft" process to create duct sections to give a smooth transitions form one shape to another.

 

The EDF is inclined upward at 15 degrees meaning the two exhaust ducts also have incorporate a 15 degree "bend" to meet the EDF. No problem for a Free Cad loft however when you export a loft into CURA it will only correctly slice it if each end is parallel to each other, which of course the Mk2 exhaust ducts are not. 😒

I do like the proportions of the Mk2 duct so a bit more thinking is required.

To really frighten myself I printed out the above 3 view to the full size of the model.

 

If we assume the CofG is at 25% of the wing chord the side area ahead would be larger (twice?( that behind it, particularly with that tiny fin. I am sure the layout is fundamentally stable however I bet if the yaw exceeds a small angle it would become un controllable. What might be acceptable for an "X" plane I am sure for an RC plane a bigger fin is essential and possibly a larger tail plane as well.

I think a simple 'profile' chuck glider might be a good idea.😉  

 

Bearing in mind that the CG must be ahead of the main gear location for it to stand on all 3 wheels when on the ground, a 25% CG is unlikely to be stable Simon.

[Simon Chaddock]

Gordon

I agree but I was only using such a CofG position to indicate the very limited fin area compared to the fuselage area.

Establishing a workable CofG was one reason for a simple 'profile' chuck glider although it may also require a profile fuselage in plan as well to replicate the effect of the forward fuselage area.  

 

I was certainly surprised at the 8% CofG position of my Bachem Natter which also had a very substantial forward fuselage area. It did actually fly with a more conventional CofG position but if the AoA exceeded a rather limited value return to stable flight was no longer possible.

 

I fully accept that if and when it gets built the chances of a fully successful maiden are not great. 😉 

  

[Simon Chaddock]

I mad a boo boo! I miscalculated the correct area of the exhausts.

I had intended each of the bifurcated exhausts would be 50% of the FSA except for the final "nozzle" which would reduce to 45% FSA but after printing it I judges it looke too small so I redid the calculations only to find each duct was nearly 50% too small. All the exhaust duct had to be reprinted

The old and revised. The correct is on the right.

 

A short video of the first test.

https://www.youtube.com/watch?v=RqqtZASn8xI

So far so good. At least it blew the door shut!

   

[PDB]

9 minutes ago, Simon Chaddock said:

I mad a boo boo! I miscalculated the correct area of the exhausts.

I had intended each of the bifurcated exhausts would be 50% of the FSA except for the final "nozzle" which would reduce to 45% FSA but after printing it I judges it looke too small so I redid the calculations only to find each duct was nearly 50% too small. All the exhaust duct had to be reprinted

The old and revised. The correct is on the right.

 

A short video of the first test.

https://www.youtube.com/watch?v=RqqtZASn8xI

So far so good. At least it blew the door shut!

   

Are they printed in LW-PLA? Do you print the sections in vase mode?

[Erfolg]

I am very interested how this goes.

 

My own thoughts were very much about placing the DF right at the back, pretty much adjacent to the out let. I also thought of increasing the inlet size. I was also thinking that I would have a cheater hole.

 

Due to my lack of experience of DF, that is why I started with a simpler DF model (so I thought). I could not decide where the Lipo would probably need to go.

 

I have drawn up the body shape. Also it compares with the TSR 2 in general proportions.

 

 

[Simon Chaddock]

PDB

Yes in LE-PLA and in vase mode.

I actually print it with the exhaust outlet on the bed and with a 3mm brim to ensure it sticks to the bed and to provide some stiffness to keep the nozzle circular when it is lifted.

 

Erfolg

It is generally agreed that the published thrust figures for EDFs relate to it being tested in "free air", with a bell mouth and no ducting at all.

There is an argument to suggest that the inlet ducting is usually bigger than the fan swept area so has slower moving air in it thus the losses are lower. It follows that for the most efficient EDF installation it should be placed right at the back with no exhaust duct at all!

Difficult to prove as any benefit is likely to be quite small bit it certainly works.

The EDF at the back of my Fairey Delta 2.

In this case the weight of the EDF is an advantage as it meant the battery had to go ahead of the twin inlets meaning no inlet duct could be of the required 1.2 times the FSA and absolutely straight up to the intakes in fact it was built around a piece of gutter down pipe which was only removed when sufficient fuselage  outer skin was in place.

It used a lower power 6 blade 55mm EDF so overall efficiency was a concern.   

      

[Erfolg]

Simon, this is as far as I got, before chickening out, that is get some experience before going for the X3.

 

My model is a +60 year old model built prior to leaving school. The books are my references. As to being absolute scale no way, the drawings are tiny, who knows how accurate the drawing are anyway.

 

I stopped at the rear duct, which I was making a round to oblong, to correspond roughly with the two scale out lets.

 

It may interest you in that I have seen the D558? at Pensacola, and a sky rocket somewhere in Nevada (Reno?) they are both tiny, the pilots must have been small.

 

Another model I have in the past considered is the Leduc at Paris (Le Bourget?), again the pilot must have ben very very small, who did not mind dying. 

 

 

[Simon Chaddock]

I found an usual result whilst testing the latest duct.

The RH nozzle gives less thrust than the LH one and by quite a margin about 12%. 🤕

The two halves of the bifurcated duct are mathematically identical and the EDF is exactly in the middle of the duct feeding the bifurcation.

A 12% thrust difference is likely to be significant so I spent some time thinking about why this could happen. The only possibility I came up with is the airflow is spinning as it leaves the fan so the fact the ducts descend downwards at about 30 degrees be a benefit to one duct and a disadvantage to the other.

To test this hypothesis I printed the ducts with only a side off set and none downward. Then added to the full duct to test.

Holding the duct over a small digital scale showed the thrust from each exhaust was virtually identical to within a gram or two.

However this layout would mean the EDF is 'inline' with the wings. It would likely require a substantial "over and under" spar going around the duct. Perfectly possible but it would be relatively heavy.

Some more thinking required.

 

Edited January 28 by Simon Chaddock
Spelling

[Zflyer]

Have you considered altering the thrust angle of the edf as is done with conventional engines/motors.

Please bear in mind i have no expertise in this area but look at it from outside the bowl.

 

[Simon Chaddock]

Zflyer

Certainly a possibility but my concern would be that although the thrust may be equalised at each nozzle there might well be an overall loss in thrust. With this plane I am already struggling to arrive at a workable solution of thrust to weight to wing area. Even a potential loss in thrust I can do without.

 

I think I may have a potential solution - make the plane 20% bigger!

It would keep the same EDF but the increased fuselage dimensions would enable an "inline" duct to avoid the EDF and inlet duct "interfering" with a single piece wing. The extra wing area would be useful too.

As all the "heavy" bits would be the same and the airframe still made from the same 3mm Depron the increase in weight should be a bit less than the increase in wing area. It could actually reduce the wing loading. compared to the first design.

On paper at least a win/win!

 

[J D 8]

  Optimism is Simons game👍 You only fail when you stop trying. Thomas Edison.

   Waiting to see this one go.

[john davidson 1]

How far could the wing area be increased and still look scale? Did that with a TSR2 but still flew like a brick, also needs a lot of thrust.

,

[Simon Chaddock]

Using the inline exhaust duct which give virtually equal thrust on each nozzle this is what it would look like over the elevation of the "bigger" X3  Bigger means it will be 6ft long with a 2ft span!

However efficient the 'inline' exhaust duct may be note how close the EDF ends up to the bottom of the fuselage. In addition the inlet duct seriously interferes with the concept of a one piece wing. Some sort of over and under spar(s) design will have to be included with the duct.

According to the literature this EDF gives 850g thrust on a 4s. Although not quoted the efficiency of converting Watts into thrust (g/W) rises significantly at slower fan speeds. As the LiPo is going to be the single heaviest item of the whole plane it makes sense to make use of lower power efficiency and use a smaller lighter 3s LiPo.

My guess is the EDF will give 550 g thrust on 3s.

I have to expect to loose up to 40% with e ducting giving 330g thrust.

My guess of the all up weight is 360g.

It should fly but with not much to spare.

Once I have the full duct  completed I can test to see if the thrust it creates is at least in the right ball park.

What ever the thrust is the X3 will have to be seriously light.😉

[Erfolg]

An in passing comment, although the X3 never came anywhere near its intended performance, this is a consequence of the intended engines being cancelled. No similiar engines (that is thrust) was available at the time. The solution was the use of engines of substantially lower thrust.

 

The X3 was not unique, many aircraft have been cancelled or underperformed due to engines that were intended, either being cancelled, or never coming close to performance that was promised sometimes so unreliable that the reputation of the airframe was destroyed.

[Simon Chaddock]

Erfolg

Yes indeed about the X3. Perhaps the moral for aircraft designers is that engine development takes longer than the design & build of experimental airframes.

Would the X3 have succeeded with more powerful engines? The example of the similar simple inlet geometry of the Fairey Delta 2 suggests that Mach 2 could have been achieved but with the extra weight of bigger engines its already record breaking take off and landing speeds would have been even higher.

 

The fifth iteration of the duct for the bigger X3.

 

It is 27.5" (700mm) long and in LW-PLA it weighs 25g just 1/3 that of the EDF..  .

The 20% bigger airframe also means the scale inlets are now closer to the ideal 1.2 times the FSA. Hopefully any benefit in thrust this gives will compensate for the extra length of the duct.

A test run using a substantial 3s.

https://www.youtube.com/watch?v=czHn-uf2NS4

This test at least proved that the inlets could withstand the reduced pressure without relying on any support from the fuselage formers.

It also passed my simple "blow the door shut" test I used on my other bifurcated duct foam planes.

I will however have to reprint the whole inlet duct as I made a small but significant error. it is 10 mm too long!

Nearly ready to make a start on actually building the hX3.

Edited February 6 by Simon Chaddock