TITLE: Attaching threaded bar to rebar
DATE: 2020-02-16
AUTHOR: John L. Godlee
====================================================================


Last year when I did terrestrial laser scanning (TLS) in Bicuar
National Park in Angola, I used lengths of threaded bar which can be
driven into the ground to mount the reflective targets which are
needed to align multiple scans. Conventionally each reflective
target is mounted on its own tripod, which allows flexibility in the
height of the target, but it was unfeasible to carry 6 tripods on
the plane. The lengths of threaded bar I used had been manufactured
previously by a colleague who had also used them to mount reflective
targets for laser scanning on a tidal mudflat. I found however, that
although the threaded bar was a step in the right direction, they
were still very heavy, weighing about 8 kg each. This year I
designed an even more minimal system to hold the reflective targets,
which utilises local materials where possible. My chosen method
might sound like its very simple but it took me a long time thinking
about this to achieve the optimal setup.

  {IMAGE}


  {IMAGE}


While shopping around I had noticed that it was possible to buy
shorter lengths of threaded bar with the thread pitch I needed for
the targets (5/8" Whitworth - BSW, 11 threads per inch), commonly 1’
or 8". I decided that I could buy one of these small pieces of bar
for each target and then find some way of attaching them to a pole,
which I could buy in Tanzania, rather than taking the entire pole on
the plane.

  {IMAGE}


The local material I decided to use for the pole was steel
reinforcing bar (re-bar), which is used in poured concrete
construction to strengthen walls and foundations. Re-bar is found
easily around southern Africa. As well as being used for building,
it’s commonly used as a welding material to make gates, fences,
construct carts and repair vehicle chassis. In my work we often used
re-bar to mark plot corners, which can be partially buried at the
end of the plot census and recovered with a metal detector at a
later date.

  {IMAGE}


Now I had to figure out how to attach my short lengths of threaded
bar to a longer length of re-bar. In construction, a common way to
attach lengths of re-bar is to tie them with an overlap using wire.
Unfortunately when I tried this there was too much wiggle. I need
the target to remain completely still during the measurements so
that the GNSS measurement lines up with the location of the centre
of the reflective target. Continuing my research I found a company
called UniTec, who make lots of re-bar jointing methods. One that
seemed promising was called a mechanically bolted coupler, which has
a long clamp that you put both pieces of bar inside, then tigthen
screws to clamp the bar tight. I had planned to buy these until I
realised that a) each coupler weighed over half a kg, b) they only
ship from the USA, and c) they cost about £25 each. I imagine they
are so heavy and expensive because they have to adhere to a number
of regulations surrounding the weight-bearing potential for the
construction of very large buildings. Another issue that made the
coupler less than ideal was that I would be relying on finding
re-bar that fit the coupler. I did some research into common sizes
of re-bar but the standard sizes differ a lot between countries. I
could have used bits of wire as shims to pad out the coupler, but
this would likely have introduced wiggle.

  {IMAGE}


I looked briefly at manufacturing a set of metal backing plates
which I could attach wire rope grips to, sort of like a DIY
mechanically bolted coupler, but it would have taken a lot of work
to get the metal backing plates manufactured to the correct
specification and I was running out of time at that point.

  {IMAGE}


I looked also at buying boss head clamp stands like they have in
chemistry labs, but this was too expensive, as I would have needed
at least 10 clamps, two for each target to avoid lateral wiggle. It
might also have been difficult to get them tight enough. An
advantage to this method however is that the clamps are easily
removable and adjustable.

  {IMAGE}


One option which I probably didn’t give enough thought to was
whether I could weld the threaded bar to the re-bar. I could have
found out what material the threaded bar is to see if it could be
welded to re-bar. My worry with that method is that the weld may
have broken due to poor workmanship, and it wouldn’t be repairable
in the field.

In the end, I was wandering around B&Q trying to find a hammer and I
came across a set of hose clamps (aka jubilee clips, worm clamps)
which are used to attach a rubber hosepipe to a tap fitting. They
are adjuatable with a flathead screwdriver and accommodate many
different sizes of re-bar, they are also very lightweight. I chose
18-25 mm diameter clamps. In the UK I performed a test using two
hose clamps to join a piece of threaded bar to a piece of re-bar,
which I was kindly given by the local scrapyard. It creates a very
tight connection, the only compromise being that it is slightly more
difficult to hammer the re-bar into the ground as it is easy to
accidentally strike the threaded bar, which should be avoided in
case the thread gets messed up. To solve this problem I bought a
spare short length of re-bar about 8" long that can be used with the
hammer to transfer all the force to the re-bar as it is hammered.

  {IMAGE}


There are a couple of advantages to the hose clamp method over the
original large lengths of threaded bar. Because the threaded bar
lengths are considerably shorter I can take 10 rather than five,
allowing me to leapfrog between subplots, scanning the first while I
GNSS the next, this could make the work considerably faster once I
get on a roll. Also because the re-bar is cut near to the
fieldsites, I can get many different lengths cut, allowing me to
achieve the variation in target height which is desirable for
calibrating the scan rotation, and also allowing me to place the
targets well above the thick long grass, allowing them to be more
easily identified by the software in the scan alignment stage.

  {IMAGE}