Thread Tools
This thread is privately moderated by BAF23, who may elect to delete unwanted replies.
Jan 08, 2018, 04:04 AM
The sky is the limit
BAF23's Avatar
Thread OP
Build Log

Ka2b scale 1:3 repairs after major crash (part one)


This is the second thread about my Ka2b saga, if interested in the earlier restoration, use following link: https://www.rcgroups.com/forums/show...albe-scale-1-3

This thread contains 4 consecutive parts:
part 1 covers the crash, the assessment and the fuselage repair
Part 2 covers the port wing repair
Part 3 covers the starboard wing repair
Part 4 covers the final assembly and the horizontal stabilizer change



Disaster in Pottes

Last week of July 2017 BiGGS held its week-long gathering near the French border in Pottes/Celles. As I was the event coordinator I was kept busy all week and flew little. The almost constant force 3 to 4 crosswind also made things tricky and restricted the type of gliders that could be flown. Saturday had to be the highlight with the celebration of 40 years of the Waloon model association AAM with official speeches by representatives. Although I was reluctant I got a bit pushed to fly my historic Ka2b for the public event and assembled it for its 5th flight. This would be a short one because at only about 50 meters during the climb, the nylon snapped in the hook and I declared an emergency, joined a low downwind and settled down in a smooth landing.


I immediately joined the cue and rapidly found myself behind a new German tow-pilot who had lots of challenging practice that week. At around 200 meters he messed up a turn by turning too sharp but I was able to follow. At that time he spoke something to me in German but as I didn’t understand clearly, I took no chances and disconnected because that might have been what he was asking me. He was surprised because he then made it clear he just had told me to hang on because he had everything under control. My reply was that unless anything was wrong, he better not speak to me in a language I’m not so familiar with. I couldn’t pick up thermals and as my earphone was inoperative I had no vario neither altitude info and flew the old way. The second landing was with serious headwind and was remarkably short and smooth, what an elegant machine. I joined the queue again because I wanted to perform at least one normal tow and flight that day, after which I was going to stop and leave the Ka2b for static admiration by the dignitary and visitors.


When it was my turn again I found myself behind the French designer of the Bidule. As we had a NOTAM for 900 meters AGL that weekend, he took me up to 500 meters before we separated. This eighth flight was the first one during which I could relax and enjoy the nice and steady gliding characteristics of my flagship. The traffic pattern was standard and the final leg was steady and lined up for a landing with little spoilers. I got surprised by a first gust just short of the threshold and pushed down a tad late to bring it back towards the glide slope. I touched with full up elevator a bit more firmly than desired and with full left aileron to counter a minimal right bank. I afterwards realized that I should have better used left rudder but that is easy when watching the slow images in a chair.


I thought I had it made when shortly after the touch, a wind gust lifted the port wing (remember it had 2° more incidence than the starboard one) and at that bank rate there was nothing else I could do than jump away for my own safety. The starboard wing caught the ground causing the glider to cartwheel with the right wing breaking off at the root and the left wing at about half span. Forces on the tail also caused cracks on the left and right fuselage sides under the horizontal tail. The starboard wing also caused the canopy to split at the root, then it slammed the nose but luckily only caused superficial damage there. Nobody had seen it coming, everybody thought it would be a normal landing again. I was just unfortunate to have encountered a serious wind gust at the wrong place and time.

Ka2 scale model crash at Pottes july 2017 (0 min 45 sec)


I had some help to quickly move the model away from the flying activities and few guys stayed at the scene but none made pictures (I hate disaster tourists at such times). I quickly disassembled the model and had sufficient official business to do during the rest of the day so I wouldn’t think too much about the crash and the long repairs ahead. Here follows a list of the chronological sequence I later made before and during the restoration.


assessment and repair sequence

-Repair tail cracks
-Repair and paint minor nose damage
-produce new port wing aft attach metal +6mm
-dismantle wing turtle deck
-make new canopy top window
-cut port wing joint balsa capping to allow new ones to be glued shoulder to shoulder on ribs
-produce port wing carbon wing box junction
-reproduce missing or broken wing ribs
-remove main spar capping top and bottom
-install plywood leading edge overlap basic start
-install balsa trailing edge wedge
-join inner and outer port wing parts and measure incidence angles all along wing
-repair trailing edge plywood planking
-reconstruct leading edge with balsa and carbon reinforcement
-cap D-box and sand flush
-re cover with white Oratex and paint
-extract stbd wing key and smoothen junction area
-solder loose stbd mpx wires
-produce new stbd wing key block
-reinforce port aft inner wing rib attachment
-shape stbd wing key block and transfer metal attachment to it
-attach stbd wing key to wing with bolts and glue
-glue stbd inner rib to wing after bolts are tightened
-cut stbd balsa capping straight to allow new ones to be glued shoulder to shoulder on ribs
-cap stbd wing root with 3mm balsa and sand
-reproduction of lighter fixed tailplane and paint
-trial fit both wings on fuselage and measure incidence angles, also of horizontal tail
-shave aft fuselage as required to obtain correct stab incidence (-1,5° initially?)
-check fitting of canopy along wing leading edges
-if incidences ok, paint and completely fix port wing aft metal supports, if not make new supports
-fill intrados wing junction with fuselage and modify turtle deck to mate wing and aft fuselage
-eventually decrease tail incidence and/or side angle by taking away fuselage wood
-weight and balance of separate wings, then of complete glider assembly
-Modify compression metal between aft wings, holes 2mm closer to each other at 107mm
-Mount receiver (plus airspeed sensor?) and bind,
-change washout of outer port wing from +1 to -0,5°.
-Check and CA inner aft corner of stbd wing.
-check and modify wing support in camper
-find stronger new 5mm wing bolts for aft wing support
-order and install new powerbox
-replace rudder servo arm after discovery of skipping wheels
-modify (widen) house cradle
-repair fuselage where cradle pinched during assembly
-final assembly test
-nulify all controls (new elevator link)
-long check all electronics
-range check
-perform test and adjustment flights
-set fail safe after final trimming


fuselage repair


Only at the end of the flying season did I find the courage and time to undertake the major task of repairing my Ka2b. I opted to start with the fuselage and with the help of my father moved it into my hobbyroom for analysis. Inspection in the glider’s nose showed no interior damage at all except a bent ground hook and some paint scuffs on the outside. The heavy lead had perfectly withstood the impact forces and was still firmly held into position by the nylon screw assemblies. Things looked worse on the aft fuselage under the tailplane. I suppose torsion from that tailplane on the fuselage when hitting the ground caused a serious crack running 3/4 of the fuselage round the tail. On the port side a long crack indicated serious weakness.

Pic4370




On the starboard side the crack was not that long but on following picture it shows the crack to continue all around the top of the fuselage.


Pic4367





After removal of the former rudder servo access panel I had a first glimpse inside and was happy to see none of the wooden stringers had cracks. To be sure everything was OK all around the inside I performed an endoscopy using flexible inspection camera with light at the end. I was very happy to note that the cracks on the outside were just that, cracks in the very thin plywood skin material, none of the adjacent frames nor stringers showed any sign of damage. I noted the position of the adjacent frames in order to gain better access without further weakening the structure.


With such flimsy fuselage shell plywood it was obvious that repairs had to be strengthened by using reinforcement plates on the inside. To gain better access and allowing clamps to be used during drying, I cut a hole in the surface where the tailplane rested. I had to take into account the position and stresses of the 3 screws that kept the tailplane in position, the position of the frames, and the thickness and width of the clamps. To my surprise I needed the use of my circular Dremel saw because that solid plate was 7mm thick plywood! I then made a cardboard pattern of the fuselage side at the crack, before carving out an inside and outside foam block of the same shape. The reinforcement plates had to be strong to absorb fore-aft forces that could lead to separation at the crack, but capable of being bent in order to hug the inside curves of the fuselage. I bought 0,8mm birch aviation grade plywood that fulfilled those requirements and cut 3 pieces that fit exactly between the frames and stringers where the cracks were. I then used a sharp knife to eliminate the split wood in the cracks, allowing both sides to line up with each other with minimal gaps. One at the time, the 3 plates were coated in PU wood glue before being pressed against the crack by the custom foam shape that was flat on the other side for the clamp.


Pic 4391



The expanding glue having no other place to go filled the gap and could not expand further to the outside due to the outside foam block held tight by the same clamp. The minimal access hole restricted it to a single clamp so after one crack area had dried for 4 hours, the clamp was removed, the foam cut loose at the crack, and the same procedure got applied to the two other crack areas (above and below the central stringer). When one side was finished I applied lightweight putty to full the depressions, and re positioned the clamp to do the other side. I slightly Dremeled out the crack on top of the fuselage and filled the void with PU wood glue. This produced a very light yet strong repair with minimal filling before priming and painting.


In the meantime I cut another hole adjacent to the port aft wing attachment hardware. The port wing incidence had to be reduced by 2° and the easiest way to do that was to raise the attachment point by 6mm by making longer plates. Removal of the plates required unscrewing 3 solid bolts from the inside and therefore a new hole was cut to allow access for the tools. On the old plates and measuring the angles I was able to mark the direction in which the new plates had to be extended and the new hole drilled out 6mm higher.

Pic4385



This modification will require the turtle deck to be modified and a filler strip to be made between the above deck and the intrados of the port wing. Those things will have to be done later after the longer plates are in position and the wing can be trial fitted. The canopy brass tubular frame was slightly deformed and I delicately separated the airfoil shaped cracked clear acetate from the canopy shaped thicker clear material. Both had been joined by a combination of clear silicon and canopy glue, also on the brass tubing.


Pic4377



I definitely needed the exact shape of this flimsy 0,5mm material in order to cut a new one so I could bend the tubing back to their approximate shape without the possibility to mount the wings yet (that were still broken at that stage). The bent tubing caused some buckling of the remaining clear canopy at both aft parts. After separation of those aft vertical windows I was able to bend the frame into shape again and then attached and glued the windows again. I made marks at the extreme nose part of the wing plates and the places where the sharp leading edge radius began on top and bottom, this would later help produce the exact leading edge shape on the new material. Aging of the acetate caused it to very easily crack when flattening it for reproduction. A pattern of this piece was then made of cardboard and transferred on new clear polyester sheets of 0,5mm. I use the plural form because I needed 3 attempts before I got the shape exactly right.


Pic4392



The new piece was extremely difficult to fit within the existing cutout and I decided to use small screws to initially hold it in position. Getting new small screws seemed an easy task but in my city of 75000 inhabitants I was unable to find them, and no store could even order them, even if I took 200. I finally ordered them through the international Conrad online shop (together with 3 other sizes of self tapping small screws to have some stock for the future). In these times of mass consumption it is getting nearly impossible to find basic parts in a local shop anymore, most buy a complete product and throw it away after some use, repairs seem something of the past.

Parts 2, 3 and 4 will be posted hereunder
Last edited by BAF23; Jan 25, 2018 at 03:43 AM.
Sign up now
to remove ads between posts
Jan 08, 2018, 04:48 AM
The sky is the limit
BAF23's Avatar
Thread OP

Part two


Port wing repair:

The port wing had been torn apart about half span, close to the aileron cutout. The trailing edge had been pulled in pieces but the sections were complete and usable with minimal damage to the ribs. The leading edge was a mess, with substantial parts of the 3mm D-box either missing or cracked beyond repair, and that included 5 nose ribs and the leading edge itself. It thus was clear that the port wing also cracked due to the fuselage stopping abruptly during the cartwheel to the right, the outer wing just wanting to continue forward by inertia. Both remaining wing halves were still in perfect condition, including the complete root with its complex but unique spar attachment. As I had collected most of the broken smaller parts on the field, it was easy to put the puzzle back together to make cardboard templates. Although many of my modeller friends wonder why I didn’t make a set of new wings from scratch, I took up the challenge to restore the original wings to their former glory, it just makes me feel like a woodworm again and that boosts my confidence to tackle the complete assembly of other traditionally built models. Here is a picture of the outer wing remains.

Foto 4428


And a picture of the inner wing at the crack

Foto 4429


It should be noted that the main wing spar all along the wing is a buildup of vertical plywood sides with two wooden beams in between, it were the latter that broke. The plywood beams are not solid, but contain large vertical slits into which the front and back wing ribs are glued in a staggered manner, that makes up for many weak point all along the span. Except for a 3mm leading edge and ridiculously thin plywood trailing edge, there are no other transverse beams in the wing. Most of the (torsional) strength comes from the 3mm balsa planking over the top and bottom of the complete so called D-box nose assembly.

First job was to cut out the broken 3mm balsa D-caps from the good structure. Being so important for strength I decided to make the replacements in one piece and anchor them firmly on the leading edge and the main spar instead of just gluing them shoulder to shoulder with the remaining bits. This thus necessitated ample delicate cutting and sanding to expose and prepare the old structure to receive the new wood. Where old and new join at the ribs, the 3mm multiplex wooden ribs were wide enough so that I could cut the old balsa along the length of the rib and have old and new join completely over the rib structure. Remaining good fabric was ironed on solid to the ribs before the torn fabric was cut off.

The wing tapering both in depth and thickness all along its span resulted in no two ribs being identical so cardboard was used to draw and adjust individual nose rib patterns before cutting their intricate shapes from 3mm plywood. Having a correct shape of the outer missing/broken ribs allowed me to sandwich the new middle ones in between and hand adjust/file/sand the new ribs so wing continuity would be assured. Lightening the ribs by taking the excess middle material away as per original was tedious and required 60 holes to be drilled and 20 times refitting the blade of the figure saw through them. I had decided to keep the existing forward rib remains that were solidly glued into the wing spar and just glue the new complete ribs to the spar and remaining rib bits, to recreate the shapes and keep essential strength. These nose ribs later being capped by balsa would be invisible so having them displaced a few millimeters was no problem.

The old leading edge was exposed by cutting away the upper and lower balsa fillers that had been glued between the ribs, and cut neatly at a 30 degree angle just as the new leading edge, so as to have more glue area during the reassembly. The leading edge passes through the front of those 5 new ribs so careful cutting and preparations were necessary to ensure correct alignment during the final mating. Here is a picture of all the old and fabricated parts prior to reassembly.

Foto 4430


All nose ribs were then glued to the outer wing remains, together with the new leading edge, and balsa upper and lower leading edge cap supports glued, shaped and aligned. The aft rib was glued to the inner wing and the trailing edge joints reinforced by balsa trailing edge shapes inserted in between the flimsy plywood sides, plus a carbon strip inserted at the aft part. Both wings were then made ready for final alignment and permanent mating. A carbon rod of 20cm long by 1,2cm wide was then fabricated by club friends and sanded in tapered shape to be inserted within the wing spar of both wing halves. Many dry runs were made to ensure all was ok and no wing twist nor bend would ensue. On following picture you can see both wing halves ready for mating and the 67gram hollow fabricated carbon key on top of the spar it was to be glued into.

Foto 4434


The more than 2,5 meter long wing had to be kept in place during the long hours of drying process of the expanding wood PU glue, but I didn’t dare to do that on the floor because during the alignment I wanted to be able to look along the leading and trailing edges to ensure everything remained absolutely straight. My table in the living room was much shorter but was suitable because I could eyeball everything from both sides just by kneeling. Liberal amounts of PU glue was plastered on the areas to be joined but much was expected to be scraped away during the carbon being pushed into the wing boxes, or dripping off during drying expansion. The whole mating operation therefore was done over a plastic garbage bag onto which the PU glue didn’t grab. A heavy weight and some clamps were used to keep everything aligned during the drying and a metal straight edge was used to correct any small deviations. 24 hours later I just lifted the wing from the plasic on the table and used a knife to remove the excess glue that had dripped and settled all around the joining area.

Foto 4436


The wing being straight as an arrow, I didn’t lose time and immediately started capping the nose section, starting with the relatively flat bottom. I used a single spanwise wooden 3mm balsa plank but as the maximum width of such planks is 10cm, I needed a second one to be glued to it to cover the complete nose gap. After a first rough sanding I turned the wing around to do the top side. The top side had a pronounced radius that a 3mm balsa straight plank could not hug without risk of warping the wing during the dry-out. My solution was to immerse the balsa in the bath for half an hour, letting it soak the water and for the rest of the evening I just watched television whilst continuously using my thumbs to force the balsa to take the exact shape of the nose extrados and stay that way by using elastic bands. Next morning after the balsa was let free, it still was sufficiently wet to allow me to refine the shape before gluing it on the ribs and using rubbers to ensure it dried that way. Here is a picture of the shaped 3mm balsa plank before being glued. You can also clearly seen the leading edge buildup that was geinforced by carbon strips at the old/new junctions, both on the plywood as on the balsa.

Foto 4443


The two irregularities to the right were cracks in the old paint that did not warrant replacement of the complete panel because the integrity was still guarantied. I just used a knife to cut away the paint and produce a channel in which I allowed thin CA to flow whilst pushing the wood with the thumbs from below. That opened the crack sufficiently to allow the thin CA to fill the nonexistent wood gap and keep the wood together forever. The channel was subsequently filled with lightweight putty and sanded. Being over-enthusiast during the sanding of the balsa capping I noticed the planking to become extremely thin at the lower front. I wasn’t through it but did not want to take risks with vegetation tearing my capping apart in case of accidental landing in crops. I thus glued an additional 1mm balsa planking over the area I had thinned out to much, and kept it tight with rubber so no air would be trapped when I sanded the edges away after filling the junctions with putty.

Pic 4445



Pore filler was then applied to the planking and more sanding performed before getting the surfaces join completely flush with the intact portions. Before applying natural Oratex I figured that the few remaining outer panels before the aileron servo box could better be cutout as well so overlap paint would be less visible and the Oratex could at least be attached more strongly by using the inside of the aileron box instead of only the narrow rib.


Pic 4453



In order for the Oratex to better adhere to the mix of new balsa, old painted Oratex and filled plywood trailing edge, I first applied “Oracover iron-on heat activated adhesive” to the contact surfaces. This allows for a better grip during subsequent heat shrinking. With all this being used on both top and bottom of the wing, the tedious repair job got completed with the wing still straight as an arrow but structurally even stronger than before the crash. For the painting I opened the spoilers by extracting the spoiler servo from its box, also allowing me to remove the 92gr lead I had placed there for balancing the wings after my maiden flight.

Pic 4452


After masking the blue trim line, I applied several layers of white acryl paint, sanding them in between over the wood (not the Oratex) to obtain a uniform finish all over the span, thereby leaving only the very attentive bystander to see traces of a major repair having taken place. After reassembly of the spoiler and aileron, I put it on the scales where it tipped 3062gr and balanced at 92cm out of the root/100cm out of the junction bolts. The complete repair only added 120gr, including the 67gr carbon spar reinforcement.
Last edited by BAF23; Jan 25, 2018 at 03:45 AM.
Jan 08, 2018, 05:53 AM
The sky is the limit
BAF23's Avatar
Thread OP

Part 3


Starboard wing repair:

Analysis revealed that at the end of the crash, the wing swung forward due to the inertia of the model cartwheeling. That tore the inner rib free from the wing aft of the main spar, putting all the strain on the wing junction spar that cracked on the root. The wing swung more forward thereby compressing the canopy’s upper aft assembly before ending up against the nose. The rest of the wing appeared to have no other damage, but repairing that highly visible beautiful wing junction vital structural beam was challenging because I had no idea of how the builder modified the Schneider plans to build this scale looking wing-fitting into the rest of the wing.

Pic4396



First task was to cut into the wing and try to find out how this transverse wing attachment had been integrated into the forward swept main wing spar. I started by removing the bottom wing planking to gain a better insight about the design, but that still did not reveal a sufficiently clear picture of how the multiplex assembly was fastened to the built-up main spar.

Pic4401



A smaller opening on the top of the wing didn’t make me any wiser, and I started using sharp long knives to cut the glue between the wing attachment multiplex and the top and bottom hardwood of the main wing spar box. Removing the reinforcement triangles at the front was no problem, and separating the nose ribs also went smooth. At the broken other wing I had seen how those hardwood beams were sandwiched between long tapering plywood plates to form a hollow but solid main spar over the complete span of the wing. I tried drilling horizontally to find out if the spar was hollow at the root but that did not seem the case, how was it filled and glued then? As I cut deeper and deeper from both top and bottom, it became clear that at the root this was a solid assembly, but it took me much work to get completely through it (breaking many knife blades in the process). When I finally was able to separate the multiplex beam I was surprised to see that the builder somewhere shaped it to integrate it at an angle within the main spar beams, no wonder I had so much trouble cutting through it.

Pic4404



I was then left with the dilemma of either Dremeling the remaining multiplex from in between the 3 sides of the main spar, or make an exact copy of the part I extracted and glue plus bolt it to the very strong remaining wing assembly. I choose for the latter and luckily I have some modeller friends very capable of tackling complex jobs with other than wooden materials, and who can think out of the box to produce complex stuff that can never be found on the shelves. In this case the three of us worked in concert during an entire weekend to fabricate an exact copy of the broken part, not in multiplex but in built-up strong carbon. The idea of alternating lighter wood with heavier carbon to produce a laminate was abandoned because it would complicate the sanding and ultimate attachment of the complex shaped piece to the existing metal and retained wood.


“Carbon project manager” Bert had used pictures and measurements to draw the dimensions of the various carbon pieces that had to be glued solidly together to form the exact shape avoiding expensive surplus material. Harry made the cuts that were needed to produce a 6cm wide key out of 5cm carbon strips. He then used nail into a solid wooden base so all the carbon could be assembled in a kind of jig, being pressed together to squeeze out the excess two-component glue during the drying process under a heat lamp.

Pic 1713



Next day when the assembly was dry we were surprised the total weight was a hefty 513gr. Bert used a sharp tool to mark, plus a tool to duplicate angles to indicate what material had to be removed. From following picture it is clear that it would take time to reproduce that complex shape that tapered in all directions to position the metal that was critical for the wing dihedral and forward sweep.

Pic 4357crop



I spent the rest of the afternoon in an icy wind exposed shed pressing that piece of carbon in various directions and angles against a belt sander that had been mounted inverted just for that purpose. Everything had to be done in small steps because we only had one piece, but just before dark I was proud to present them the key that had lost weight to 364gr. That was still 200gr heavier than the wooden one but the shape was essential and even if it was an overkill for strength, it had to have those dimensions to be able to replace the original in the inside wing as well as on the highly visible scale wing assembly under the canopy. As from the start we already decided to limit the number of 3mm bolts that would be used to effectively join the metal to the carbon. The position of those fewer holes was critical for all the wing angles and drilling them was a bit of a nightmare because of the metal plate alignment front and back, we even twice broke a drill in the carbon trying to get everything aligned. On that Sunday it was 10pm before the 3 of us called it the day and left for our respective homes, but the new key with temporary attached metal (weighing a total of 494gr) was ready for further trial fitting.

Pic 4948



Back home some minor adjustments had to be made before the new piece could be aligned with the one in the left wing but it looked promising. After re positioning and tightening the bolts using a piece of wood to simulate the flat top of the deck, it seemed that that angles were correct and the carbon was ready for further fitting in the starboard wing.

Pic4455



The victory cry came too soon, the carbon part didn’t fit in the wing (yet) and had to be sanded more precisely to fit between the ribs, and offer that perfectly smooth surface towards the wood it had to be glued to. That extra weekend we also drilled additional holes in the carbon so we got at least 4 bolts keeping the extremities of the metal tight against the carbon. The number of original bolts were thereby reduced from 18 to 9, primarily in order not to weaken the carbon block. The nine functional bolts were now in strong 12.9 steel and the heads of the old bolts were Dremeled off and glued as fake bolt-heads onto the holes in only the front metal plate. A couple of 4mm holes and a 5mm hole were then drilled through the carbon and old wing main spar so both parts could be pressed during the curing of the PU, hopefully holding them together for the rest of the flying years. I preferred 3 bolts so even if one would break or come lose, the wing probably would just flap a bit but not bent skywards. Some more sanding was done to accommodate the span-wise wing taper and just before mating in the wing, the carbon/metal assembly including all bolts tipped the scale at 480gr. That is a very acceptable 120gr extra considering the much more solid yet scale appearance wing key.

Pic4458



Liberal amounts of PU wood glue were used between the main spar and the carbon key. The top and bottom hardwood was nicely flat against the carbon, but the old cutoff middle inside wood could not be made 100% flush and I counted on the trapped PU expanding to fill the uneven voids. No way to check it for sure, but using flat roundels and nylon inset locking bolts ensured everything would hold together tight during the drying process and future flights. The large triangular plywood nose reinforcement plates were then glued back in their former positions and an additional small hole was made on the aft forward side to allow a nut to be used on the forward bolt without weakening the plywood too much.

Pic 4462



This picture was made after I just applied a liberal amount of PU glue to the bottom of the wing spars so the balsa 3mm capping would help produce a solid torsion box at the root. The balsa caps were first cut at custom dimensions and immersed in water to allow them to be shaped before gluing on the rounded ribs.

Pic 4460



A reinforcement in the wing trailing edge void was inserted to produce a more solid anchor point for the vital first aft rib. With the single metal attachment checking fore and aft forces plus wing incidence angle, it was essential that this aft rib part (which had been torn off during the crash) was reattached in a way that the (torsional) forces on it would be spread over a larger part of the wing root. I therefore also used carbon dowels through the rib and into the reinforced trailing edge to keep everything tight. Shaping, cutting, fitting of that inner rib so as to have maximum contact with the rest of the wing took 3 hours whilst capping the top of the wing a mere half hour. Lightweight putty was then applied to smooth everything out and produce that continuously curved airfoil profile all along the root area. All the wear and crash caused nicks and dents in the surface were also filled out in a few passes and sanded before the wing was ready for priming. The aft wing corner also had to be taken care off because it had been lost in the field.

Pic 4468



Further towards the tip I discovered that where planking had been joined (twice due to the wide span) during the initial build , it had cracked on the nose extrados but not on the intrados. A less than subtle forcing showed it was not only the paint, but also the balsa but the rest of the wing structure could be slightly separated just enough to allow thin CA glue to flow in between. After release and use of a weight on top those former cracks, they were solid as a rock and minimal putty was used to close the minimal gaps.

Pic 4464



The thick primer was then carefully sanded so no dark surfaces showed through and a perfectly smooth surface without traces of repair would emerge after a few roller paint applications. It then was possible to weigh both wings and compare them to the situation before the crash. The point is that not only the net weight augmented on both wings, but due to the location of the weight increases the result was bound to show serious differences. I therefore put it all down on a table and also noted the weight as measured under each tip. I figured that would probably affect the Ka2 flight trimming most, but that was miraculously within 1 gram. Only flight testing will reveal if that thinking proves correct, and I further hope to reduce the weight of the horizontal stab (and nose lead) to compensate the 357gr increase in total wing weight.

wing / Before crash /repaired / diff / At root/ At tip / Moment before/ Moment after
port/ 2943g -99cm /3067g/100cm/+124g/ 1961g / 1106g /291357 / 306700
stbd /3042g-97cm/ 3275g/94cm / +233g / 2168g /1107g /295074 / 307850
Last edited by BAF23; Jan 24, 2018 at 12:50 PM.
Jan 08, 2018, 06:38 AM
The sky is the limit
BAF23's Avatar
Thread OP

Part four


Wing to fuselage mating and measuring:

Halfway December 2017 I attempted the first wing to fuselage assembly in my apartment and was terribly disappointed by the result. Either I could join the wings together but not with the fuselage, or each wing with the fuselage but not with each other anymore. Fitting the various bolts through the tight holes had already been difficult prior to the crash, but now the angles were off even more, some due to the elevated port wing attachment and some due to the angles of the metal fitting on the new wing key and/or the alignment of wing key itself, the latter mostly because I forgot to take into account the thickness of the wood that had been eliminated when cutting/sawing/separating/sanding the broken part of the key from the intact part of the main wing spar. After forcing some of the bolts “almost through” I was able to see, photograph, measure and mark the shortcomings on the various metal fittings.

Pic4474



With the wings very near to their final positions I was able to see that a serious horizontal gap existed at the front of the starboard wing, and a serious vertical gap at the aft half of the port wing. A rough positioning of the turtle deck at this stage allowed me to see where the vertical attachment hole had to be drilled in the carbon wing key and how much material had to be taken away from the turtle hump for it to fit correctly over the now much higher port wing trailing edge.


The only good news was that the wing incidences were within half a degree of each other and no significant wing warping could be measured. This on the other hand also meant that there is no real washout, the port wing even has an undesirable half degree wash-in. These measurements confirmed that the new port wing attachment plates were correct and could be bolted firmly to the fuselage. On the other hand, trying to fit the canopy between the wings showed totally unacceptable alignment with the wing leading edges so the new transparency was cut loose and the leading edge guides bent at a new angle. Final adjustment and definitive gluing of the canopy sub-assembly parts got postponed till the wings fit correctly to the fuselage. Look at the grossly misaligned canopy fairing versus the resulting changes in wing positioning after the modifications/repairs in following picture.

Pic4470



With the pictures it was easy to find out how the fuselage holders for the wings could be adapted to obtain better alignment of the bolts and the inner ribs. Luckily the compression bars in between the aft wing attach points provided some alignment, whatever changes were made to the fuselage holders. The lengthened plates holding the port aft wing were bolted tight and the upper access hole closed again. The fuselage had to be heightened a bit to close the serious vertical gap with the port wing intrados.

Pic4473


Horizontal stab change:

My friend Eric was so kind as to produce a new fixed horizontal stab hoping it to be lighter than the original. That stab was delivered with only the bottom 2mm planking in place because the positioning of the 3 attachment bolts had to confirmed and made solid, plus the position and angle of the multiplex connector. I mounted the connector in a proper sleeve so there was no way it could inadvertently be pushed further in the stab during field assembly. I also drilled the holes for the 5mm nylon bolts and in order to avoid compression during field assembly I inserted carbon sleeves that also guided the bolts so they would always engage correctly into the existing fuselage mounted bolts. Using squares and ropes for measuring distances between tips, fuselage and vertical fin, I adjusted the seating and bolt channels to obtain a correct positioning of the fixed stab.

Pic4476



After curving them in water I then applied the top planking and also used thin ply on the trailing edges to reduce the gap between stab and elevator. The first fitting looked right. I then glued the leading edge which is of a more solid wood and then carved and sanded it to produce a smooth shape all around.

Pic 4475



I knew that the incidence change of the wing would affect the stab so I removed former filler material from the forward stab seating on the fuselage. Unfortunately when I brought the stab home, I forgot the plans so there was no way I could figure out why the complete leading edge wood stood in the way from the frame where the stab seating began. My solution had been to cut out the complete lower half away and had the top half saddle the aft fuselage. With the new so called lightweight stab completed I cut the turtle deck free from its base plate, a first red mark showing what had to be cut away before it could be adapted to lay flush between both wings and the aft fuselage.

Because the purpose of the new stab was to loose weight, I meticulously noted down the weight at all intermediate stages, including the finish that my friend first didn’t want to delegate because he was afraid that my acryl paint would add too much weight.
-Stab original: fixed incl servo: 478gr, elev: 192gr , together 670gr.
-Stab light: fixed incl servo after sanding: 345gr, plus partial primer: 357gr, after sanding 351gr,
After Oratex over the complete fixed stab: 382gr, after 1st paint coat: 390gr, after 2nd coat :395gr , after 3rd coat: 396 gr, all finished: 397 gr. The paint was much lighter than ever imagined.
-Complete with elevator 595gr, total a weight save of 80gr versus the original.


Second assembly:


The day after Christmas my friend was here and we spent most of the day trying to make all parts fit for a smoother field assembly. We started with both wings and that required serious thinking and hole ovalisation before both even could be mounted to each-other and on the fuselage. Most of what I had prepared had been in the good direction, but it was obvious that a lot more had to be done because the starboard aft attachment was misaligned by almost a full centimeter (neither had been touched upon during the repair) and forcing it would result in the trailing edge of the wing crushing into the fuselage. The cause could only lay into the new wing key and its bracket mounting during the repair. Forward wing sweep also was incorrect (measured between fuselage nose and wing tips). Correction of this complex problem was best achieved achieved by ovalisation of some of the 8mm wing key holes and the starboard aft wing attach bracket. Care was taken only to enlarge one of both adjacent holes in order not to allow any play in either direction after bolting. Here you can see that eroding the holes of the stbd aft wing attach to allow the wing to move forward doesn’t allow play because neither the wing pen nor the compression bar holes were altered, so everything bolts together rock-solid.

Pic 4486



After a few attempts we got it right without unwanted horizontal play but the field assembly will have to be made in a different order than before the crash. It then was time to measure the wing incidences and I was relieved both were identical, with just the port wing still having some positive wing warp at the tip. An attempt was made to correct this by applying maximum manual torsion on the wing whilst the other ran the heat-gun over and under the painted Oratex but a day later the wing had just resumed its former shape.

The canopy was put in place and the leading edge angles further corrected. It was obvious that closing the gaps with the wings could only be done by gluing 3mm balsa on the canopy sides. We also discovered some “wing flapping” that could easily be eliminated by gluing some thin “blech” under the inner port wing key above the fuselage. Fitting the turtle deck required much more material elimination on the raised port wing side and when we got it right we first used point CA and PU to glue it on the base plate in it’s curved position on between the wing. Final gluing, filling out and painting could only take place after curing and removal from the wing. Following picture illustrates the clear acetate having been adapted to the newly twisted leading edge angles and bolted/glued back into a solid unit before 3mm balsa “ear muffs” were glued to the sides to close the gaps with the wing leading edges.

Pic 4481



We then mounted the complete horizontal stabilizer on the fuselage and measured the incidence. Despite the cutting away at the front, the complete stab still had an unacceptable +2° incidence. We first thought about raising the aft part of the stab (by about 5mm) but that meant even more weight plus necessitating an adaptation of the electrical connection. Cutting away the material from under the middle weakened the structure of the entire tail so after carving away half of the thick plywood I glued 3mm ply planking underneath till the fuselage walls. These also formed a solid base for the plaque I later PU’d to close the hole I had made to gain access to the aft fuselage. Lowering the front of the stab wasn’t easy because my friend somewhere messed up with the leading edge shape and dimensions so that it clashed with the plywood former it rested against. That meant I had to cut away the complete leading edge hardwood in order to allow it ti fit just aft the frame. It wasn’t a dream solution but measuring it on the fuselage/canopy wing chord indicates about -2° stab angle which sounds like a much better start point. Furthermore we took the opportunity to measure how square the stab was versus the vertical tail (twisted fuselage build) and wings. The advantage of this solution was that the bolts still line up well and minimal spacer/filler was necessary, and the existing MPX connectors also fitted without having to be extended. Here is a picture of the works as described but still to be final shaped.

Pic4480



Final assembly plus weight and balance:


All those unforeseen tasks prevented me from completing the Ka2 before the end of 2017. The 1st of January 2018 I assembled the model with my friend again but mounting the wings to each-other and on the fuselage still required serious twisting and unacceptable force. Putting our heads together we figured the only possible way to get everything aligned properly and fitting without the use of brutal force, nor ending up with lateral or vertical play, was to ovalize one more hole in the new aft port bracket and produce a new single (instead of double) compression bar with the exact 5mm round holes 2mm closer to each-other.


With the wings in place I tried to fit the canopy between the leading edges but that required a lot of sanding from the balsa ear muffs.Furthermore I still had to bend the port leading edge a bit more down, a delicate move because I noted previous bending already having half-opened of he soldering on the crucial copper tube joint. With that in place I also had to ovalize one of the turtle-deck holes to allow it to be bolted on, only to discover the bent down canopy leading edges produced a 3 and 4mm gap between the canopy and turtle deck.


The elevator was then bolted into place and the incidence meter put to work. It rapidly became clear that the port wing washout hadn’t held, it again indicated +1°. The rigid wood structure of the wing would have to be twisted over a longer period instead of just trying to use a heat gun on the fabric. Over the next two weeks the port wing thus stayed twisted and was regularly heated to settle with a washout. Heating it a bit too much on the bottom resulted in the Oratex starting to pull away from the D-box so I quit the heating and filled and repainted that undesirable gap. Using a variety of household weights to keep the trailing edge of the inner half wing down and a relatively soft shape around the tip and forcing the leading edge 5° down for a couple of weeks seemed to only remaining method because the Oratex and paint layers did not allow vapor or water to moisten the wood for easier deformation.

Pic 4489



After two weeks I removed the contraption and measured the tip having a -1° washout. This of course was just after the release and another measurement was taken half week later when the wood and fabric had completely settled. The washout was about minus one quarter of a degree, not much but a lot better than the +1° it originaly had. The stab mods resulted in -1° angle versus the main wing structures but the tilt angle perpendicular to the twisted tail didn’t look good and had to be further modified to fall halfway between the wings and twisted tail. That glider will never look straight/right from any angle and I slowly lost my courage finishing it. The only motivation was that after 4 months I finally saw the end of the repair nearing.


We then lifted the model onto the bar stools and drew the 50mm lines on the underside of the wing. Previous flight had proved that was an acceptable CoG so I decided to balance it about the same spot. I use the word about because I again had to use the rubber rounded sanding blocks to divide the massive 15kg+ weight over a reasonable amount of 3mm balsa planking under the leading edge of the wings. I was able to remove 2 lead blocks from the nose. 275gr from the nose plus the 80gr from the lighter tail and 95gr from the port wing meant that even with the extra weight of the repairs, the model still had lost 200gr in total, mainly from the dead lead in the nose.


Following days I further ovalized some holes with the drill, produced a hardwood extension for the turtle deck and a 1mm thick strip over the port stab seating. I then smeared filler and primer to smooth out all canopy, turtle-deck and stab seating transitions. I used CA to strengthen the stab leading edge mods and canopy glue to finalize its shape versus the wing leading edge. After sanding I applied several coats of white paint because the darker wood kept on peeping through.

When I installed the receiver I noted that my powerbox didn’t work anymore. I remember after the crash I couldn’t switch it off the normal way but forgot about that problem. Now even with the switches off, it twice lets the 6,6v through from the LiFe’s, unacceptable for my servo’s. An identical powerbox would cost me 90euro plus shipping so I decided to replace it with a similar Graupner PRX 5A HV powerbox that costs only 47euro and was easy to install and connect in the old gliderhardware. After powering it and binding the receiver I noted that my rudder wasn’t centered. I quickly found out that rudder and servo were not rigidly connected anymore. The metal gear servo worked perfectly, but the nylon servo arm was skipping teeth over the servowhen loaded. I don’t think that had anything to do with the crash but transport rash probably slowly degraded the grip of rhe nylon. I took a stronger plastic arm from another Hitec receiver and installed it while also tightening the bowden cables and zero out the mechanical neutral position.

End January I finally collected all missing or modified hardware and assembled the model again. This time all fittings and bolts went in without force after mounting the starboard wing first while resting the tip on a chair. Having not done that previously had resulted in minor cracks in the fuselage planking (at their joints) where they had leaned excessively against the cradle arms. I thus also widened those arms and redid the gluing of the spongy foam against the wood or hard foam, the construction expanding spray did not hold well and all was scraped away and PU glue used. This operation was also repeated on the camper transport cradle. It was a mess that I undertook during the few days in between the swap of my living room furniture. I also measured the tailplane and it is perfectly aligned with the wings and front of the fuselage, but the slightly tilted vertical tail is now more visible.

With the model under power I was able to mechanically set all flight controls and the spoilers to a completely neutral setting. To do that I had to change the angle of the elevator servo and make a new pushrod (that was lighter and could be better adjusted). All that’s necessary to assemble the model is now kept together in one bag to facilitate field assembly. After four long months of intense labour my Ka2b is finally ready to take the sky again (when it gets warmer) and test flying has to be redone from scratch due to the many modifications, but I have confidence it will fly much better than ever.
Last edited by BAF23; Feb 05, 2018 at 10:26 AM.


Quick Reply
Message:
Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Help! Crashed heli, and now have reduced lift after repair. Help, please. interestingfello Electric Heli Talk 20 Feb 02, 2016 07:57 PM
Discussion Need major help with my Blade 450x after a crash. Noodles46 Beginner Training Area (Heli-Electric) 18 Oct 18, 2015 07:08 PM
Discussion repairing after big crash, advice on mods instead of factory parts/settings colesdad Coaxial Helicopters 2 Dec 17, 2013 09:56 AM
Discussion How to repair Park Zone Ultra-Micro J-3 Cub after bad crash jrutter Micro Ready-to-Fly 15 Aug 06, 2010 05:12 PM