The Design (Phase 3)

#43
Epoxy mouldings were taken from the underside rear of the hull + the underside of the wingtips & the side of the fin/rudder to produce bases for the various fairings to be built onto. This would ensure a perfect fit of the final mouldings when fitted to the flying model. From front to back we have the tailwheel fairing, the two underwing tip floats, the rudder control linkage fairing & finally at the front, the small 3 part fairing on the top of the wingtips that covers over the tip elevon hinge.

The elevon tip hinge fairing is made from Chemiwood. This is only around 4-5" long and will be used to take a silicone mould before casting in fast cure PU resin.

Phil Clark, FighterAces
 
#44
The forward fuselage having rivets applied.

IMG_2886
by Alex Jones, on Flickr

The finished result on the forward pattern section. FighterAces really have done an amazing job! Its worth noting this has been finished to a high gloss as this is the pattern not the flying model. The finished model will not be in a high gloss finish therefore the panel distortions will have a more subtle look as shown in the picture of this section in a satin primer.

_DSC9444
by Alex Jones, on Flickr

IMG_2915
by Alex Jones, on Flickr

IMG_2916
by Alex Jones, on Flickr

A picture of the full-size showing the distortions in the panels, this aircraft had a very unique structure more akin to a boat but one that had to survive doing 100kts plus on the waters surface and survive being dropped into the water with an equivalent force of 3G!

_DSC8213
by Alex Jones, on Flickr
 
#47
Our first day back was productive and very exciting. We have finally managed to get the gear geometry to work, this has been a major undertaking and a massive project in its own right with lots of ups and downs for all involved. It's a good feeling being able to move onto the surface modelling designs for the skis and oleos.

The pattern will be scanned using blue light technology at a resolution of around 29 microns so we can pick out every detail that the Fighteraces team have put so much effort into creating.

Soon we will see the tooling and layup designs finalised. We have some very talented composite specialists working on this project and its exciting to see technologies and techniques integrated into this model that have not to our knowledge been used on an RC jet.

The team at FighterAces have done an incredible job. I'm very much looking forward to visiting them this month and seeing the completed pattern!

IMG_3346
by Alex Jones, on Flickr

IMG_3220
by Alex Jones, on Flickr

IMG_3217
by Alex Jones, on Flickr

Some pictures of the exhaust pattern detail.

IMG_3335
by Alex Jones, on Flickr

IMG_3333
by Alex Jones, on Flickr

IMG_3331
by Alex Jones, on Flickr
 
#48
FighterAces have done an outstanding job, thanks Phil! I'm looking forward to collecting it next week and taking some photos of the finished pattern all together for the first time.

The pattern will now be scanned which will allow us to accurately design the internal structure. We aim to start tooling and manufacturing components ready for a basic prototype to be ground/water tested here in the UK late summer 2018. Lots of work to do!

A few photos of the finished forward fuselage section. I love it but then I am heavily biased considering the last two years have been dedicated to getting it this far.

IMG_3375
by Alex Jones, on Flickr

IMG_3368
by Alex Jones, on Flickr
 
#49
The pattern finally finished. Thank you to the team at Fighteraces who’ve dedicated 10 months of full time work and have really managed to capture the essence of the full-size which was no easy task. Great job guys and thank you! The pictures don't really show the scale or presence of this model... its huge!

Now onto the scanning....

A few pictures at FighterAces of the finished pattern ready for delivery and thankfully I managed to get it to the scanners without damage!

_DSC9623
by Alex Jones, on Flickr

_DSC9621
by Alex Jones, on Flickr

_DSC9618
by Alex Jones, on Flickr

_DSC9617
by Alex Jones, on Flickr
 
#50
The prototype once built can be evaluated here in the UK ready to send abroad for flight testing although we are still in talks with the appropriate aviation authorities as this model breaks almost every rule in the book in most countries. I consider myself lucky to have such a good large model scene here in the UK. This is run by the LMA on behalf of the CAA which allows us to do almost anything with the correct oversight, inspections and flight tests. So a quick thank you to everyone involved in the LMA.

With phase 1, 2 and half of phase 3 in the pattern design and build completed we can now move forward with the composite design. The next tranche of work won’t actually see anything physical being built except the tooling. The time will be spent in CAD designing the internal structures, ski mechanisms and lay up designs etc… To facilitate this we are in the process of having the pattern scanned using blue light technology. There are many advantages to this that justify the significant expense of the process especially considering the resolution required to capture every rivet and panel deformation in the pattern.

Ordinarily on a project like this you would start with a CAD model, CNC the pattern and take the tooling from this. We don’t have the luxury of a CAD model which is where the scanning comes in. With the scan data we can accurately design all the internal structure but not only this we can now accurately determine the weight of the model before its built. This will significantly speed up the design process as weight it critical in determining how the model presents on the water. CG, ballast tanks, component position and many other areas of uncertainty can be designed in from the outset. Undoubtedly there will be changes in the design as there nearly always is but this allows us to get much closer to a finished model first time out.

The team at PES Scanning gave me a quick work flow run through of how they will scan the pattern along with a short demonstration starting with the photogrammetry. The scanning team have been very helpful through the whole process and I’m looking forward to seeing the results.

A few pictures of the scanning process which took around five days using blue light to the very high resolution we specified.

_DSC9640
by Alex Jones, on Flickr

_DSC9639
by Alex Jones, on Flickr

_DSC9637
by Alex Jones, on Flickr

_DSC9628
by Alex Jones, on Flickr

_DSC9631
by Alex Jones, on Flickr
 
#51
We’ve been managing all the companies involved so we can all come together at the right time with the correct designs and parts allowing us to build the first prototype with as few issues as possible. I’ve lost count of the number of miles I have driven to various design meetings and touring possible manufacturers facilities.

We are now at a stage where I’m very happy with all the companies involved and think we can make something spectacular. The tooling and parts manufacturers are very enthusiastic about the project and to see that our design will be produced along side parts at the highest level of composite specification is very exciting. I would love to tell you what sort of things these guys are doing and for who but unfortunately almost all the work these companies do is based on non disclosure agreements but the level of detail and the quality of the composite parts we’ve seen being produced is staggering and far removed to what we are used to in our models.

PES Scanning have done a great job with the scan at what was a very high resolution. We have captured every panel deformation and every rivet. The resolution was so high that the scanner picked up the titanium dioxide spray but the team also encountered some noise in the data which had to be removed by hand. This is mainly due to the pattern having the worst two properties possible for scanning which is dark and glossy. Even so the results are beyond what I had expected. The below picture shows the front portion of the fuselage in the process of being cleaned. The file contains over 170 million triangles and is in the region of 22GB in size.

Here are some renders of the finished scan data which has been decimated substantially but still retains lots of detail.

SeaDart PATTERN 3
by Alex Jones, on Flickr

SeaDart PATTERN 2
by Alex Jones, on Flickr

SeaDart PATTERN 1
by Alex Jones, on Flickr
 
#52
Now the scan has been completed we can progress onto one of the most challenging areas of the design...

Ski retraction/extension mechanisms and the skis themselves.
 
#53
This is an area where we struggled for months. We had professional mechanical engineers all struggling to get the geometry to work. Frustratingly close the retraction mechanism lulled everyone in thinking we could resolve the last 10% but whatever we did just didn't work.

Our initial work came close and you can see here that we are designing this mechanism exactly as the original actuating it in a prototypical manner.

image003
by Alex Jones, on Flickr

One of the many issues is that we can get the mechanisms to work independently but when you tie them together with a ski it binds as during retraction the distance between the two anchor points on the ski changes but of course this has to be constrained. You can see how the distance reduces and the oleo wants to ride into its housing.

image012
by Alex Jones, on Flickr

This is how the gear retracts, sliding the main oleo forward and folding everything into a flat space.

image002
by Alex Jones, on Flickr
 
Last edited:
#57
Work continues on the skis and retraction mechanisms but now with all three sources of research, scan, original drawings and photos we were able to get much closer to a system that works. With confidence enough to start working on scale features and designing parts for manufacture.

PES-LJC-01-20- 11 SKI FWD ATTACH BRKT-PORT-iss1-3
by Alex Jones, on Flickr

R3640.006 Bearing installed-1
by Alex Jones, on Flickr

PES-LJC-01-20- 11 SKI FWD ATTACH BRKT-PORT-iss1-4
by Alex Jones, on Flickr

PES-LJC-01-20- 11 SKI FWD ATTACH BRKT-PORT-iss1-6
by Alex Jones, on Flickr
 
#58
The most critical area of the whole retraction system is the position of the main oleo slide bar and the angle at which the rear oleo anchors to the ski. With this in mind and still a worry that the mechanism doesn't work as we would like it was decided we needed some flexibility in the prototype to change these angles. So areas of adjustability have been designed into the mechanisms.

The design now allows us to change the angles of the slider and anchor point.

Current position of slider bar-3 SECTION THRO
by Alex Jones, on Flickr
 
Last edited:
#60
our attention now turned to designing the skis. Using the scan data we were able to cut the surfaces in CAD enabling us to design the skis and CNC tooling.

The skis are two PrePreg 1mm carbon components sandwiched between a carbon inner core. Initial designs had the core in Aluminium but we decided due to cost that an inner carbon component would be more appropriate.

Section thro ski showing fuse skin break-in aperture-5
by Alex Jones, on Flickr

Ski for weight analysis_x_t wb
by Alex Jones, on Flickr
 
Top