Topic: Fairey Delta 2 70mm EDF build

Those of you who followed my P1091 build in July will know that I thought it looked pretty similar to the FD2.  I read up on the aircraft and played around with a 3-view in ACAD to check areas, etc for a 70mm fan and with a little licence wrt the shape of the intakes and a slightly over-scale wing everything could be based on the P1091 structure.  I wanted to emulate the speed of the original and had a fairly high power fan/motor combo loafing on my box of stuff so the FD2 was added to the wish list to build after the Viggen and Vulcan.  However last month when I read of the passing of Peter Twiss on 31 August I decided to crack on with the FD2 now as a sort of tribute.  The wing design was finalised using Profili, but this time with 2 x full depth 3mm ply spars and extra stringers to stiffen the upper surface (lesson learnt).  I CNC cut the wing parts before I had to go away with work mid-September, however whilst I was at sea I was able to check the C of G/moment arms and completely finish the computer work for the fuselage.  I got back last Sunday and cut the remaining parts on Tuesday evening.  Seeing the parts “in the flesh” I think I might have over done it with the nose structure, but it will have to support 535g of 3000mAh 40C 6S battery (same packs I bought for the P1121).

I am going to use a 70mm Velocity RC WM400 pusher fan with the ARC 28-58-1 (same motor as I have in the P1121) and probably a Pentium 80Amp OPTO ESC (although I have 2 other ESC options available).  The EDF will go right at the end of the fuselage and be balanced by the 6S pack.  I will be using 2 Atlas DS-095FMD digital servos and they will be fully checked for centring this time before building into the wing.  Length excluding the long pitot is 1000mm (1150 with), wingspan 620mm, wing area 19.4dm2 and I’m aiming for an all up weight of 1600g.  Intakes from the WW Hunter, but will definitely need carbon tow/hysol stiffening.  Remainder of the intake duct rolled 1/64” ply.  The minimal cockpit will be carved from soft block.

This evening I cleared away the work bench and ran up the fan.  I’ll do a proper test with the Watt meter tomorrow, but 3300 30C pack gave 1805g static thrust and 204mph efflux (no intake lip) and the 3000 40C pack 1820g and 211 mph efflux.  Am not going to get too excited yet because the intake ducting will drag those figures down…………….but I can hope!

Gluing starts tomorrow after the GP.  Any input and comment would be very welcome.
Pete

Strangely enough I have Peter Twiss's obituary on the desk next to me - salvaged from last weekend's paper. I was thinking the FD2 would make an interesting edf project so looking forward to seeing how it goes 

Good choice of prototype IMO - Peter Twiss was one of my boyhood heroes.

I was watching the Discovery programme about the Gannet the other day, so sad he's gone.

Bert

If you haven't read it already, I recommend Peter's autobiography "Faster Than The Sun", a fascinating life.
 
I decided to wait until my CC ICE ESC arrives back from being upgraded later this week before properly testing the fan/motor combo as the data logging function will be very useful.  I did make the fin this afternoon though.  Laser cut internal framework assembled over the plan then skinned with 1/16” medium balsa sheet.  Some lines drawn on the skin helped to keep the profile even when sanding with a long permagrit bar.
Although the wing loading will be higher than the P1091 I think the FD2 will still be capable of hand launching and being a delta should have pretty docile flying characteristics. 
Pete

Yep - definately interested in this one 

Working late last night and in early today = afternoon off 
I needed to re-cut the intake lips in 4mm ply so stopped off on the way home to do that and buy some M3 fittings + stud to make the control rods.  Here are all the bits ready to go – I forgot to include the plans and the EDF is still bolted in the test stand.
Good news: my CC 120HV ICE ESCs were delivered this morning, so after walking the hound I’ll get soldering.  It will be a bit anti-social to fire it up tonight, but I should have time tomorrow afternoon.

Why the oversize wing? I would have thought that this design would be fine on a scale wing. So what made you decide on upping it a bit?

Slightly baffled

James

Hi James
To reduce the wing loading a tad to tip the balance towards hand launching (based on other models).  It has 30mm greater wing span and 25mm longer chord than scale which gives approx 9% increase in area - if I hadn't mentioned it I very much doubt anybody would notice. Other errors in the wing being too thick (6% vice 4%), wrong shape ailerons, no flaps, exhaust too big (no after burning EDF available) iand because I'm lazy I'm using ready-made Hunter intakes (the wrong shape, wrong area, wrong angle between the ducts).
 

Fair enough.
What wing loading are you anticipating?

I can't remember what my TSR2 worked out at now I think it's about 26 oz/ft^2 it is around there anyway and that hand launches fine with just about 1:1 thrust to weight ratio. Then again it is only about 1200 grams with a 4500 mah 3s Lipo on board. However it isn't in your class of build as it is EPP, and thrown together in my usual terrible way.
I have to say I am mightily impressed with your builds  I keep promising myself to go back to balsa. Indeed my first ever design and build keeps whispering to me to build it a new 72" wing and tail feathers after I dumb thumbed and cartwheeled it at just about full chat across the field 22 years ago or thereabouts.

Told a fib last time, my TSR2 worked out at 22.35 Oz/ft^2 Not sure where I came up with 26 from. Must have been the wine with supper.

If I can keep it to 1600g all up the wing loading will be 82gdm² (~27ozft²).
Thanks for your comments re my builds - but I'm very lucky in having the use of a CNC cutter.


Scared myself this evening when I checked the WM400 on a 3000 40C 6S pack.  Everything was higher than I expected ~90 amps, 1750W and 1850g static thrust .   Maybe I didn't have the right setting on the ESC or maybe the CC120 has a more efficient through-put, but regardless the 80Amp Pentium won't be good enough so I'll have to find a 100amp ESC.
Pete

Not had much spare time for modelling this weekend so have only been able to start the nose section.  Built as left and right sides over the plan then joined to skin as a single assembly later.  The keels are 3mm birch ply with 2mm light ply formers.  All a good tight fit locked together dry then glued with thin cyano wicked into the joints.  There are a few tweaks to make to the drawing before another one is cut.  When I edited the photos for posting I realised that I have made a basic error and made the nose upside down    – not the end of the world though and I’ll be able to work around that tomorrow. 

I glued a 3M blind nut in the tip of the nose for a static scale pitot arrangement – short balsa tip for flight.  I used a programme called “Cone Layout” to print templates for each section.  I checked the fit then cut the panels from med-soft 1/16” balsa sheet and applied with cyano.  I should have used thicker balsa to leave more “meat” for sanding to shape as there are still a few flat spots left.  Think I’ll glass the nose cone next then fill the low spots with plastic padding.

I decided to sort out the shape of the nose first then glass it. The sheeting on the first two bays got very thin and needed supporting inside so I sprayed some water down the nose and trickled some Gorilla glue into the tip.  After an hour it had expanded and set allowing me to achieve a fairly decent finish.  Glass cloth applied with Zap finishing resin.  Just in case I ever need another set of intake ducts I made castings of the set Peter kindly let me have last week.  One of the castings would need a little work to fill some bubbles before vac-forming, but for now they can all go into storage and wait until needed.

Not had much spare time, but did cut the battery hatch free and finish the static and flying nose tips.  A little more work with the plastic padding is required when I get home on Sunday to properly blend in the nose parts and also (building on the P1091 success) bed-in the hatch to minimise gaps.  The 3000 40C 6S Zippy pack fits in well and according to my C of G tracking spreadsheet everything is working out well.  I have assembled the intake vac-formings and now need some Hysol to apply carbon tow re-enforcing to add resistance to collapsing. 

Spent a therapeutic relaxing hour or so cutting the fuselage formers free and cleaning them up ready to build the fuselage halves this evening.  Plastic padding going off and I’ll blend in the nose tips whilst I watch the GP recording over lunch.  Now the MEK has done it’s thing bonding the intake ducts they have turned out to be stiffer/stronger than I remembered on the P1091 so I’m going to hold off applying carbon tow/Hysol strengthening until it is fully installed in the fuselage frame and I try a test run.

The fuselage halves built in my now standard way over the plan with enough stringers to hold straight before joining.  I left the intake ducts floating free until after the fuselage halves were glued together then ran MEK down the joints.  I haven’t glued the last former or the fan bearers on yet: these will be glued in after I fit the fan which will requires the ply intake tube to be in place first.

I used the Cone Layout programme again to print a template on thin card.  Did a check fit then I transferred the shape to a sheet of 1/64” ply and cut out with a pair of scissors.  After gluing in place I trimmed the back end flush with the former and tidied up the lip to make a good seat for the fan shroud – a few tweaks required to the CAD drawing it that area if there is to be another one cut.  Inside the duct some work needed to trim back the excess plastic and smooth the join line.  Am now happy with the way the tips blend in with the nose.  Whilst the plastic padding was out I seated the battery hatch.  Have now stalled on the fuselage because I want to get some base EDF test stand figures before I install the fan and sheet the fuz and to do that I need the 100A ESC (should arrive this week sometime).

I don’t want a repeat of the servo centring problem I had with the P1091 so drew a mount/hatch in ACAD and CNC cut it this evening.  Designed for an exact fit between wing ribs 2 & 3: if I hadn’t already cut the ribs I would have used some tabs and slots for accurate location and strength – another tweak for Mk2.

Not had much spare time recently, but have cut all the wing parts free from the ply sheets and cleaned them up ready to assemble.  I didn’t like the look of the wing root around the intakes so I re-designed the area to be closer to the FD2 (still based on the triangular hunter vacformings though) and cut the replacement parts earlier in the week.  The new ESC arrived last week and I spent last Sunday soldering up connectors, extending motor wires and doing a bench tests with and without the exhaust nozzle to compare with installed figures idc.  To strengthen the intake ducts against collapsing I have wrapped them with carbon fibre tow stuck down with Hysol.  Looks messy (it is!) but it works.  It was a bit awkward fitting the fan, but it is now installed and has been left overnight for the sealant and glue to dry.  I need to build and fit the wings next (so the intake lips are fully supported) then I can fully check out the ducts under full power.

I was going to try something different with the wings this time and use support tabs on the ribs this time to effectively jig the wings and ensure the 2 degree anhedral and 3 degrees of wash out are accurately built.  Unfortunately (for some reason I not sure of yet) nothing lined up properly, so I removed the tabs and have built the wing internal structure flat on the board over the plan.  Sighting down the leading edge it looks as if the washout has been maintained and as the root ribs are correctly angled by the tabs on the main spar I seem to have got away with it.

The wing tips will be rather thin so I added some 1/16” stiff balsa sheet and small 2mm ply tip ribs.  Got to sheet the underside next, being careful to keep the wing shape.  Then the tips will be strong enough to sand the top section to shape and construct the ailerons.

1/16” med/soft balsa edge-joined with cyano and sanded flat.  Both panels were cut to shape (leaving approx 6mm excess all round) and openings cut for the servo mounts.  They were glued with pva and held in place with sticky tape.  Both panels weighed down flat on the board and have been left until omorrow for the glue to dry. 

I was going to try something different with the wings this time and use support tabs on the ribs this time to effectively jig the wings and ensure the 2 degree anhedral and 3 degrees of wash out are accurately built. 

Why are you putting washout on a low-aspect delta? It ain't going to tipstall, and all you're doing is biasing the lift production away from the tips. This won't do wonders for the drag, but even ignoring that it could move the Cp forwards to the point where your calculated CG would be way too far back.

PDR

Why are you putting washout on a low-aspect delta? It ain't going to tipstall, and all you're doing is biasing the lift production away from the tips. This won't do wonders for the drag, but even ignoring that it could move the Cp forwards to the point where your calculated CG would be way too far back.

PDR

Because the 3-view GIF I based my CAD stuff on and several side photographs indicate that is what the real one has.
Dizz

The bottom skins were offered up with the servo mounts tight up against the rear spar and glued to the ribs with thin cyano a few inches at a time to avoid distorting the wing shape.  Both wings are now quite stiff and I’ll be able to add the aileron internal structure and blocks for the hinges without any worries.  The TE has been trimmed back to the correct line and sanded to create a tapered edge ready to accept the top sheeting.  I had a couple of hours free at sea today so I spent them drawing the cockpit “canopy” in ACAD, changing it to a 3-D model in Rhino, adding a surface and finally generating the .nc code for the CNC router to make a vac-forming plug later.

Aileron LE marked through the bottom skin from the inside and highlighted with a fibre-tip pen to show where to cut idc.  Aileron LE from medium Ľ” balsa, scrap soft balsa blocks for Robart point hinges and a 2mm ply aileron tip rib. Finally everything was sanded back flush with the ribs using a long Permagrit bar and rubber sanding blocks.

Top sheeting edge joined, outside surfaces sanded smooth whilst flat on the bench then glued to the wing structure with medium cyano.  The LE was sanded square on my belt sander and a balsa leading edge glued on.  When the glue was dry the LE was roughly shaped with a razor plane and sanded flush with the skin.  I decided to test fit the control horns and make up the push rods before cutting the ailerons free.  Then I got carried away and started glassing the wings this evening 

Both wings and the fin have now been glassed.  I used Zap finishing epoxy to fix the cloth in 2 sessions.  The nose section was glued on and I fitted the battery tray; I will have to make some sort of support for the ESC and incorporate some cooling vents for the ESC later.  The ailerons have been cut free and I removed a part of the wing root at the LE to fit around the intakes.  Point hinges fitted then the aileron LE was sanded to shape (drilling jig is more accurate used on a square edge).  The wings have a slot in the main spar which fits over a stub spar on a fuselage former.  A test fit allowed me to glue the intake formers in place.  The 3mm ply intake lips were not exactly the right shape (a square cut on angled ducts), so it took a bit of persuasion from some Permagrit tools and the Dremel to get them to fit snugly.  With the duct fully supported I was able to give the fan a test run to check the ducts wouldn’t collapse…………..and they easily withstood full power on a fresh battery.  I got approx 1670W and 1630g thrust, so at the moment running installed (and with a square lip) I have lost 150W and 260g thrust from the test stand due to the ducting.  I weighed everything required to finish the structure and that came to 1580g ins the 4S pack - there is glass cloth, resin and paint need on top of that - so maybe 1700g all up is a more realistic target weight.

I noticed that with my gash building skills I had stuck the nose section on canted off about 5mm, so I cut it off and re-attached correctly then started sheeting the rest of the fuz with 1/16” balsa.  I made platform for the ESC from 1/64” ply and 3m balsa scrap.  The motor wires have been tidied up and wire for the servo extensions fed through under the battery tray and out through the fuselage sides.  I fitted sockets then pulled the spare wire back through the fuselage leaving enough slack at the wing end to work with the servo mountings (excess has been coiled up and stuffed in the nose until I see where the rx has to go).  With all the wiring done I finished off sheeting, supporting over the intakes with more scrap balsa - also glued to the vac-forming which will help provide more duct stiffness.  A 2mm ply ring was been fitted to the exhaust nozzle and the last bay sheeted with 6 x 1/16” balsa “petals” cut using the laser cutter.  Everything sanded smooth with a coarse Permagrit bar, filled where needed and finished off with a 240grit sanding block.  The fuselage is now being glassed.

Hi Pete,
Just catching up on your progress - very nice too!  A couple of questions. Are the wing panels to be permanently attached and are you ducting any air across the top of the ESC's heatsink?
Paul W.

Fuselage bottom glassed during the GP yesterday.  After cleaning up the edges and a quick rub down this evening I glued the wings on with 30 min epoxy and the assembly was left jigged in alignment for the glue to cure.  Cocktail stick locating dowels for the fin.  3/16” balsa rounded intake lips have to be added and there will be some filling around the wing root joints required before the flow coat goes on, but I had to see what it looks together first though.  Current weight of all parts is 1132g (this includes a load of servo wire coiled in the nose cone) + 534g for the battery = 1666g.  Am considering a Stumax powered version for the future.

Hi Pete,
Just catching up on your progress - very nice too!  A couple of questions. Are the wing panels to be permanently attached and are you ducting any air across the top of the ESC's heatsink?
Paul W.

Yep, wings glued on Paul.  I have now signed up to the "keep the intake duct sealed" school to build dynamic pressure ahead of the fan and will be fitting some simple air scoops over the ESC, like the P1091.

Hi Pete,
Esc cooling and wing panel attachment are stages I've just been going through here, hence my interest.  The scoops look like an effective solution - and the model looks like an eleven hundred and thirty-two mph aircraft too..!
Paul W.

Those scoops are very expensive....................... they cost a medium latte each.
(disposable plastic spoons direct from Sainsbury's cafe  )


Once again I'm torn between 2 colour schemes - either the magenta it was painted in after gaining the record or the blue as seen at Cosford.  Both dead easy to do.
When I get the intake lips done I'll do another static power test and see what difference there is over the squared off lips when I checked the intake duct for collapsing.  Then when it finally flies and I can Doppler it I'll have a complete set of figures from start to finish which will hopefully help with future performance estimates.  Based on previous (incomplete) experience with a few of my other EDFs I'm anticipating an easy hand launch and a S&L speed close to 150mph.   
Pete

Based on previous (incomplete) experience with a few of my other EDFs I'm anticipating an easy hand launch and a S&L speed close to 150mph.

Sounds like the Magenta then?  Not to mention some large dayglo lettering on the topsides - something like 'this way up'?!  Although this being you, I am supposing that you would not want to apply the legend 'use no hooks' despite the intended hand-launch!
Re-using everyday stuff is very satisfying - I found that the rocker-covers on a large Mercury cowling could be accurately reproduced using a 5ml med. spoon to produce the nine resin blisters for the mould. 
Paul W.

Didn’t do anything to the model itself this evening, but did have a 3 hours in the workshop on the CNC machine and vac-former.  We made the cockpit canopy/lump and 2 semi-scale control rod covers from 1” balsa block then hardened the surface with several coats of  medium cyano, rubbing down with 800 wet and dry between each one to get a really smooth finish.  We pulled 1 styrene and 2 decent clear pvc canopies on the vac-former – 2 to use and one to go into storage to use as a mould for casting compound if I ever need to make more.  The pvc sheet is the thickest we have used so far (0.7mm) and it took 2 attempts to get the heating time correct.  It doesn’t like tight internal corners, but the push rod covers are still useable.  Whilst the CNC machine was working I made the parachute housing to fit at he base of the fin.

Yes Paul, I’m definitely leaning towards a gloss magenta finish, but when I checked today Humbrol do not seem to make a suitable colour, so I would be faced with mixing my own - but from what?  Then on the drive home I heard an advert for a local decorating supplier and it occurred to me that I could have something mixed to a standard paint chip chart colour.  I have checked the Crown interior gloss range and I think I have found something suitable (Z0.36.39) over pale lilac undercoat.  I know undercoat dries hard and can be sanded to provide a very smooth base for the top coat plus both paints could be simply brushed on.  Down side is the potential weight gain.
Thoughts anyone?

Hi Pete,
The paint idea sounds like a goer - normal emulsion and mixed with a little screen-wash perhaps?  I've just done a test on to film-covering using such a brew and the best adhesion was obtained having rubbed the film's surface with vinegar and then cleaned it with IPA. 
An alternative is Halfords: find a car that used the colour that you want and an approximate year (without this data, they appear to be unable to provide paint...) and they will then mix it into a bottle or into a rattle-can.  Just gone through this process on the Short Sealand, with satisfactory results.
Paul W.

Hi Pete,
The paint idea sounds like a goer - normal emulsion and mixed with a little screen-wash perhaps?  I've just done a test on to film-covering using such a brew and the best adhesion was obtained having rubbed the film's surface with vinegar and then cleaned it with IPA. 
An alternative is Halfords: find a car that used the colour that you want and an approximate year (without this data, they appear to be unable to provide paint...) and they will then mix it into a bottle or into a rattle-can.  Just gone through this process on the Short Sealand, with satisfactory results.
Paul W.

I had checked standard rattle can car paint but couldn't find the right colour again - didn't make the step to think they could do a bespoke mix so thanks for the tip.  I'll check out Halfords over the w/e.  Just as I get on a roll with this one it is all change at work and after a late start this afternoon I'll be away until Friday morning.
CD done, will post on the way in after lunch.
Pete

Hi Pete,
Thanks for that and good luck finding the colour - have you a swatch that can be scanned or visually compared to a paint-book?
Paul W.