Cheap Lasers Aid in Alignment of Sailplane Wings

During Assembly for Flight


Hugh Helmick





Installing and removing glider wings many times over the years, I have sometimes remarked to myself how very easily assembly and disassembly can be on those rare occasions where things are really aligned perfectly.  (It never happens when it's extremely hot, or when someone is watching.)  I had always assumed that perfect support and alignment was the cause of this - on those occasions, the pins simply slide in or out with no pressure at all.  I had read how useful laser beams can be for measurement and alignment tasks.  I asked myself:  Why couldn't lasers help in alignment of glider wings?  In recent years, inexpensive red diode lasers have become available.  Conveniently, they are compact, and are widely available as pointers for office presentations.  I looked at some red lasers, and figured that they would not cast an alignment spot that would be visible in daylight.  I understood that green laser pointers which would be much more visible to the human eye, and are becoming available, though still at very high cost.  Finally, an idea struck me while using the one-man assembly system described elsewhere in this HP Sailplanes Homepage:  It shouldn't be necessary to see a projected laser spot in direct daylight, and in fact, with the one-man assembly dolly position controls located under the wings, and with laser targets under the wings in shadow, the idea would still be workable.  The below pictures and captions show what has turned out to be an extremely repeatable alignment method which allows glider assembly every time, with the very least adjustment troubles that one can imagine.  Further developments seem possible, including sailplanes manufactured with factory supplied lasers integral to the airframe, even using small windows in the fuselage sides. 


Limitations to the Present Design:


DISCLAIMER / WARNING:  Laser light directly impinging on the eye is potentially damaging to the eye, and must be avoided, even with the low power (<5 Milliwatt) lasers used here.  I do not know of any hazard to the eye caused by these low power red lasers while REFLECTED FROM A DIFFUSE SURFACE, NOT INCLUDING A MIRRORED SURFACE.  EVEN A SMALL MIRROR-LIKE SURFACE MIGHT POSSIBLY CREATE HARMFUL EFFECT SIMILAR TO LOOKING DIRECTLY AT A LASER, AND BOTH SHOULD BE AVOIDED.  ONE MUST PRECLUDE CHILDREN OR UN-SUSPECTING ADULTS FROM HAVING EYES DIRECTLY EXPOSED.


The pictured installation on my HP-18 is not ideal - I did not want to pierce the fuselage for exit of the laser beams in order to allow the beams to project parallel to the wing spars.  I installed lasers on the fuselage just forward of the front drag spar, angled aft slightly, under the wing leading edge.  As a result, this installation has some parallax, but the angles and lengths involved are such that this is not a noticeable effect.  The beams produce spots on the alignment targets under the wings and have been quite visible in every situation to date.  The laser target and red dot are convenient to the position adjustments on my wing support dolly.  Location of the lasers at the front of the forward drag-spar has been very satisfactory, and coincides nicely with the removal of the turtle-deck during wings installation and removal.  Initially I considered the installation to be temporary, using epoxy to attach the lasers to bent-up brackets after they are attached by existing bolts.  Initial alignment of the lasers was accomplished by bonding them in place while supported with small wood wedges so that their beams strike near the center of a target fastened rigidly to the wing cuff.  (The cuffs are accurately placed on the wings by aligning them with ink marks on the wings.)  Later, finding that the .065 steel brackets were very slightly flexible, I bonded them with some low-strength epoxy (5-minute variety) in order to add some more stability to their mounting.  Cross marks added to the targets with a marker pen.  Results during daily assembly / disassembly have been extremely satisfying:  No more ambiguity which way that wing needs to be moved, less running back and forth to adjust, and pins just slide in usually. 





This idea really seems to reduce effort and wear and tear to close fitting glider parts.


This particular installation is not well adapted to assembly using manually held wing-tips.  A person at the wing tip cannot see the red dot underneath the wing and potentially eye hazard might result, anyway.  In the setup shown, notice that one looks AWAY from the laser source.


Obviously, consistency in the location of targets and support of the wings and fuselage is central to success of this or any similar installation.    It turns out that during assembly, the fuselage needs to be rigidly held in roll axis for consistently easy assembly  (the main gear is retracted with the fuselage dolly elevated to max height).


During disassembly, it's desirable to allow the fuselage to freely roll SLIGHTLY as both wings are lifted upward.  This prevents any preloading of the fuselage relative to either wing, and allows for some wing flexure while wing support forces are increased.  My fuselage dolly is designed to still provide very slight roll freedom in the fuselage when in the dolly is in lowered position with the main gear extended.  This fuselage dolly can later lift the main tire just clear of the ground for wheel retract before going inside the trailer.  


I note that the system has not been demonstrated on an all-composite glider, most of which have wings more flexible than the HP-18 here.  Still, I should think the technique would be of value on any glider, especially those that are really difficult to assemble due to large size and weight.  Let me know if you try this and have any ideas for further improvements..........  




Laser pointer mounted to the forward left drag-spar fitting on fuselage.  The small piano-wire clips rotate over the laser switch buttons, to hold the switches "on", as needed.  I slotted the end caps for easy replacement of batteries in this tight location.  A magnet is handy here for removing depleted batteries.  Power from the glider battery might be a desirable option.




In this case, the one target is attached to the wing support dolly, since it is very consistently located with the wing cuffs when supporting the wings.  It is made of a moderately stiff white plastic sheet obtained as notebook dividers, available at stationery stores.  This material will not do any damage when touching the fuselage, etc, when removing and returning to the trailer.  The "x" marks are added to the corresponding sides of the target sheet when wings are supported as a first step to remove wings from the ship.  (In initial setup, first support the wings such that all wing pins are in a sliding fit situation.  This needs to be done in the same manner that normally will be used in supporting wings for installation and removal.)



In a later experiment I tried targets which could be plugged into tiedown attachment holes, as shown here.