Friday 23 October 2015

Glass Slipper (5)

The wing. I spent quite a lot of time wondering how to build the wing. At first sight it looks quite straightforward, however, the curved LE and spars will generate lots of strange stresses. This means that the wing need to be assembled and sheeted while still pinned to the building board.

Before I start though, I need a full set of wing ribs. These were supplied by LaserCraft Services for a reasonable fee, cut from DXF files that I supplied



The only straight edge that I can work from is the hinge line for the ailerons or just in front of it.

To make sure that it stayed straight, a wooden straight edge was screwed to the building board and  the wing was assembled from that point.

So with the position of the false TE established, lines were drawn representing the rib positions.



A 1.5mm spruce false TE was cut and the ribs glued into place pushing back against the false TE. This was aligned to the upper edge of the ribs like this


Assemble the wing as you normally would


Note the false LE held in place by suitable weights and curving back as the ribs are fitted. It took a couple of sessions to get them all in place



The spar is laminated from 4 of 3mm square spruce strips at the root, tapering to 2 strips at the tip. A slight change from the plan. I originally intended to taper each spar backward - in the end, I used a stepped taper to get the correct fit, adjusting the spar slot in the rib until I get a nice snug fit.


This was the junction from 4 to 3 laminations with a slight taper on the outer spar


After the spar was fitted it looked like this, small 1.5mm gussets added at the rib/TE junction and then the wing was sheeted over - while the wing was still pinned down.


Left to dry/stabilise for 24 hours and then removed from the board and very quickly adding the lower spar and shear webs for the length of the panel


Don't forget to fit the servos. These are built in with  no hatches, if a servo fails it will be surgery to replace it.

The servo bearers are cut from 10mm square pine (B&Q) and simply glued to the upper sheet. The servos then are located in a lite-ply frame and glued and screwed into place. Align the arms before you proceed, there is no easy return path if you this wrong.

The wing joining tube is fitted and thoroughly glued into place.  The tube is glued against the webs - this makes a nice load transition from the aluminium tube to the spruce spars. The end stop is a 3mm off-cut of the joining tube epoxied into place


Finally, a small hardwood ply plate to hold the wing retaining screws


Add the lower sheeting (after removing the building tabs) and when dry, trim back to the false LE

The real LE comprises of a balsa laminate capped with a spruce laminate


and held in place with bands while the glue dries


The wing tips are laminated from 2 x 3/16 sheet balsa with a 1/32 ply core and glued into place



and the joint reinforced with a strip of Nylon - well soaked in PVA


The process was then repeated for the other wing half..



Glass Slipper (4)

The fuselage close to completion



Getting the last few planks in was quite tricky, just lots of patience required.


The canopy was acquired from Vortex Vacforms - CN39 Glider canopy - and extensively cut down to size. I am really pleased how well it fits and importantly how the fuselage lines have been maintained. Rather better than I could have hoped for.



The fin and tailplane came next.

The fin is a thick sheet balsa outline with a support for the tailplane


All very straightforward and all using very soft/lightweight wood.

It was simply glued into place checking for vertical - using the joiner tube as a datum in this case.


The area between the fin and the rear fuselage is simply blocked in using light 12mm sheet balsa.

Finally, the tailplane was added - laminated from 2 sheets of 3/16" balsa to get the desired thickness


Extensively carved, planed and sanded before it gets glued into place

Now with the rear block carved and sanded, it's hopefully getting its elegance back again


The tailplane is sandwiched between 2 x 12mm blocks and then sheeted in - not forgetting to fit the elevator and rudder push-rods first!


Like this


Which completes the fuselage assembly and we can start on the wing





Flair Meteor

I bought this 35 year old kit at a recent UKCAA event with strict instructions that it had be be built not stored. I remember the kit from a long time ago although I was (and still am) having problems dating it. Estimates are that it was produced from the late 1970's into the early 80's

The kit is a  fairly basic constructional specification, foam wing and a very distinctively shaped glass fibre fuselage. It's quite a small aeroplane - about 52" span and originally designed for a 40. The planned modern 46 will make it very fast flying.


 The original box was rather large and was cluttering up my workshop so this was a big incentive to get started. Additionally, I had spare moments available during the Glass Slipper build and the opportunity was too great to ignore it.


 The photo on the front - it looks very 1970ish


I was quite fortunate that the previous owner had bought some new foam wing cores. The rest of the materials were as original with some beautiful quality balsa - which was very dry.


One of the earlier owners had made a start - and a bit of a mess gluing the bulkhead in place. The surplus resin was removed with a Dremel and sanding drum and the former re-drilled to suit an engine mount that fitted my chosen power plant  - an SC46 from my recently departed Gangster 63. You can also see the ply shim added to provide a little right thrust.


 The engine has recently been serviced with new main bearings, has good compression and has so far been very reliable.


The engine mount held in place with cap head bolts and blind star nuts. The quality of the piano wire was excellent - black - high carbon steel - far better than the rubbish we get nowadays


The tailplane was fitted - it was very simple to align as there is no real option - the moulded slot simply fitted and then I added the full length elevator hinge blocks, using Kevlar cloth as the hinge material. The only critical part was getting it centred and also triangulated to the centre at the nose to make sure it wasn't skewed.


Likewise, the fin, rudder and rudder post simply slotted in and was almost guaranteed to be vertical. Slow setting epoxy was used to hold it all together. By modern standards, the rudder is quite small but I resisted the temptation to increase its size. (For the time being).


 The wings were glued together using Gorilla Glue  before 170gm fibre glass cloth was used as the bandage. Slots for  the U/C bearers were carefully cut and the blocks inserted and glued using Gorilla Glue.


The ailerons were cut out using a band saw with the wing jigged to make sure that the cuts were vertical. The middle piece was removed as this is where the hinge blocks will be fitted.


Sanded down ready for glassing


The wing retaining bolt platform is from the kit and comprises of a small length of TE stock and a plywood face. I have replaced the original 2 x M4 plastic bolts with a single M5 steel cap headed bolt


Min servos fitted into the wing and slightly recessed. This may cause some consternation, but the power output of these small servos is far higher than standard servos of the 1980's. 3kg/cm should be more than enough for a single aileron.


Once again, Gorilla Glue was used to assemble the ailerons, hinge blocks and wing. The ailerons and wing were capped with lite-ply  to keep a nice sharp edge.


Fitting the belly pan was the only real tricky part of the build. I eventually hand drew a symmetrical hour glass shaped waist  and carefully removed the excess on the bandsaw. I was either a very good judge or very lucky because very little additional fitting was required.


It took a couple of evenings to cover the model, A mix of film from different manufacturers was used - basically, what I had in stock.


I didn't bother with the outline canopy, I prefer its nakedness and raw elegance.


Underneath a different scheme was used - similar to that I used on the Concept. I hope this model lasts a bit longer.

The covering has also been sealed using SolarLac - it helps prevent fuel/oil ingress under the film and keeps it properly adhered. This is quite important around the Glass Fibre to film junctions. In my experience, film does adhere particularly well to Glass Fibre or resin.


AUW - Dry 2.35kg. About 200g less than suggested in the manufacturers instruction leaflet.

Additional parts I had to supply:

1. 3 x Wheels (2.25")
2. 10oz SLEC square fuel tank (Yellow)
3. 10x8 prop. I would normally fit an 11" prop but there isn't much ground clearance. Hopefully the 8" pitch will keep the revs (and noise) down.
4. Snakes' clevises, saddle clamps etc,
5. Servos - Futaba S148 for Ruder and Elevator and E-max E5363 for the ailerons
6. FrSky X6R Receiver with telemetry

Looking forward to seeing this fly.

Tuesday 13 October 2015

Replacement Aerial for a Taranis X9D

On Saturday, while packing up, I broke the aerial off my Taranis, I have to admit the method of assembly isn't particularly great - the aerial simply clips into the case which means to replace it you have to de-solder the coax lead off the XJT Transmitter board. This is the second aerial I have broken.
I have decided to make a proper repair - something that I think FrSky should have done in the first place.
I bought the bits from T9 Hobbysport
This was the aerial I chose to buy.

The pigtail with SMA connector (250mm)

I also bought the high gain 5dB Tx aerial

I haven't stripped it, but it looks like its a 5/8 over 5/8 collinear.

Assembly was fairly straightforward.
De-solder the old aerial using a clean and very fine pointed soldering bit
Remove the old aerial by unclipping from inside and pushing it out
As you can see, the old and new aerials are identical except for the different bases

dscn2326.jpg
Fortuitously, the SMA connector is a perfect coaxial fit inside the hole vacated by the old aerial
dscn2323.jpg

The nut is only used to centre the connector. Screw the aerial fully into place and then screw the nut up so that it 1/2 thread adrift
When happy with the fit, solder the shortened pig tail to the Tx board. You will need to cut to length, strip the outer sheath and then solder/tin the coax outer. Cut the inner so that the dielectric is about 1-2mm longer then another 0.5-1mm of inner which is very carefully tinned and soldered to the track

dscn2319.jpg
Not brilliant soldering, but before proceeding check that there are no shorts. There are 2 ground pads in the photo both are linked to the ground plane. The original lead is soldered between them, its slightly easier just to use one as I have done.
Check for connectivity for the inner to the pin in the pigtail SMA. If all OK proceed.
I simply epoxied the connector into the Tx casing hole
dscn2324.jpg
and checked that the aerial will screw in fully..
dscn2327.jpg

It does!
Obviously, I need to test it but I am not expecting any problems or change in directivity.
The high gain aerial will be interesting. To get gain, you need to sacrifice directivity. When aerials are stacked - if this is a collinear, lobes will form and also associated nulls. So in addition to the normal nulls off the end of the aerial you will get nulls out in the general direction (usually about 30 degrees out) of maximum radiation.
Not sure I'll use it, but it will be worth a test or two.

Footnote:

The aerial has been tested out to several hundred metres flying a T240 with a FrSky Delta Receiver. It has also been range checked with no discernible difference between this and the original aerial.


Monday 12 October 2015

Kwik Fli IV (5)

The Final Part...

Back on the cowl.

The engine has been temporarily fitted (and subsequently replaced with a SC61)

dscn1784.jpg

Hardwood blocks added to locate the cowl

dscn1785.jpg

Wrong sized spinner but currently it fits like this

dscn1786.jpg

Managed to chip the glass when trimming down to size but that has subsequently been redone
Its not too bad - weighs 80g.

The next job was to get the wing to fit.
The chord is 1.5" wider at the root than the standard Kwik Fli III wing. The 2 pager out of RCM&E doesn't actually say how to make the change. The root wing section is also somewhat different to the mk III as well so a certain amount of trial and error and sanding etc. to get the wing to fit at the same time at getting the wing level with the tailplane.

dscn1788.jpg

A thick ply wing plate has been fitted - using slow dry epoxy and the wing drilled for 2 M5 Nylon bolts. I then realised I cant measure.. and managed to displace one of the bolts by 5mm 

dscn1791.jpg

I am so cross with myself over that stupid error. Its not too noticeable with the wing fitted though.
This morning, I started fitted out for the radio gear. 3mm Lite ply plates on 10mm sq Pine runners and cross beams.

dscn1795.jpg

After the usual faffing around

dscn1796.jpg

The receiver was replaced with an X8R


dscn1805.jpg

So - this evening, the model assembled on the stand

dscn1806.jpg

And another one, I like it when they are like this...

However, I need to check the wing incidence

dscn1807.jpg

We already established that the tailplane is parallel with the datum. so put the phone on the tailplane and press the zero button

dscn1808.jpg

Put the phone onto the incidence measuring jig platform

dscn1811.jpg

0.4 degrees positive. The plan says for the MkIII it should be +0.5, so that is pretty close. Certainly close enough to risk flying it and then assess. To sanity check, reverse the jig. If the jig is accurate, the incidence should be exactly the same (it was).

dscn1812.jpg

It looks aligned as well. No obvious tail tilt

dscn1814.jpg

And it weighs this much so far. Probably looking at an AUW of 3Kg

The last couple of days have been spent adding detail, sanding down smooth and making a better job of the wing seat. Not much in the way of photos, so yesterday I decided to cover the model and now it looks like this.
dscn1815.jpg

and

dscn1816.jpg

A contrasting (and much simpler) colour scheme underneath

dscn1817.jpg

It's been Tangoed!

This morning I completed the installation of the Radio Gear, everything appear to work in the right direction but it will get final checks before its let loose.

Conventional (if very short) 8G torque legs have been bent up. 8G is about 4mm dia, however, the orifice in the top of the oleo is 5mm so the leg (about 15mm long) was padded out with a length of 0.5mm wall brass tube, hammered into place. The lower set screw location now located in a slot ground into the leg. It seems quite sturdy, time will tell

dscn1828.jpg

And that is the undercarriage done. Its actually not too heavy as the oleo and leg are manufactured from aluminium. Probably a bit more draggy though. If it doesn't work, its very easy to revert to conventional 8G piano wire legs. (Postscript - conventional legs have since been used, I couldn't get these to track consistently)

The cowl has been painted - KlassKote - thanks Andy Green - I painted (sprayed) it and the 'Magic' this morning (the white bits). Its a 2 pack epoxy paint and sets hard in about 6 hours on a warm day.
So now, with the undercarriage done, the cowl fitted and spinner fitted its ready for its maiden. I have a Taranis X8R receiver on order although currently it has a V8 fitted, I have decided to wait for the receiver that I want to use.

CG is a guess as no CG shown on the mods diagram. Currently set about 20mm in front of the main wheels.
The Canopy came from Vortex and sprayed Matt black on the inside after it was cut to fit. A small balsa former cut to the inside front and rear and this was used to glue it to the fuselage. The seam was sealed with Canopy Glue. Looks quite a neat installation.


dscn1822.jpg

and

dscn1823.jpg

and

dscn1824.jpg

It's Ready to Crash.. Aurora 60 next....
3 weeks and 3 days since the first part was cut.. It really is an easy build

Postscript - the model flies well, the cowl seems to have survived but the tailplane needs additional reinforcement. It failed doing a snap roll - folding slightly - causing down elevator and a rather rapid decent. All repairable but the tailplane now has a spruce spar extending about 150mm into each half.