Hi Folks We're looking to implemented a linear rail with one end embedded in concrete sitting on a shallow river bed, with a passive sled that moves up and down. The sled will have an arm attached to a small floating pontoon that houses water quality analysis probes. So the pontoon will be tethered and able to move up and down as the water level rises and falls. The rail, or perhaps pole with attached rail, will have an electronics enclosure, antenna, and solar panel at the top. The arrangement needs to be as simple as possible. I was thinking about using an extruded aluminium section as the pole with a wheeled sled. One key question though is whether V wheels or similar, are available with stainless steel bearings? Any views on this or other ideas would be appreciated. Cheers Ron Wellington, New Zealand
Let's start with the simplest of limiting factors, aluminum should never be embedded in concrete. Concrete is extremely corrosive to aluminum. Galvanized steel pipe is probably your best bet. You can either sleeve the pipe for the float or use chain link fence rolling gate rollers to guide the float up and down the pipe.
Rick, thanks for the reply, very helpful. I hadn't considered the effect of concrete on aluminium, great catch! Roller gate rollers sound like they could do the job. Obviously the are made for external use. One further issue though is that the 'sled' mustn't rotate around the pole as it needs to keep the pontoon in a fixed position, which is why I was thinking of a rail type design. Will need to investigate whether there are suitable types of roller gate mechanism. So if the pole isn't itself effectively a rail, one may need to be fixed to the pole. Another possibility is rather than embedding the 'pole' in concrete an aluminium section could be bolted-on using a flange, perhaps to a pre-cast paving stone. The latter would save some effort, particularly if we want to easily reproduce the entire thing, which is the eventual aim. Cheers, Ron.
First thing that pops into my head: Cast a concrete block, say 18Wx12Dx10H (or whatever's appropriate, that general aspect ratio though) with anchor holes and bolt a pair of hot-dip galvanized poles to it via flanges through the anchor holes (I'm assuming "galvanized 2" pole with pre-welded end flange and bolt pattern" is readily available, seems like it should be). Dig a hole in the river bed, drive steel stakes through additional anchor holes in the block to set your poles upright, and then embed the entire thing with hydraulic cement (eg. Rapid Set Cement-All, it's good stuff). Metal-to-concrete contact of any flavour should really be kept to a minimum, and embedding anything in it is usually a bad idea in the long run. You've now defined a plane with the poles in which your motion can be constrained. I wouldn't try to do any kind of bearings or wheels in a river bed open to the elements; I'd do oversized HDPE bushings that slide up and down the poles with maybe 1/4" or so wiggle room for particulates, hydraulic expansion, rust, whatever may become an issue. Enough to avoid binding problems, but shouldn't be enough to cause cosine errors on the data acquisition (I'm assuming "pushed towards sensor" and "drifting past sensor" are statistically distinguishable experimental arrangements?), at least, not ones that wouldn't be an order of magnitude smaller than other likely variables. You then attach some kind of screw-on caps to the top of the posts so that the pontoon can't conveniently be stolen or washed/blown away. The HDPE bushings are bolted to the pontoon, and can be easily and cheaply replaced as necessary (probably quite infrequently, stuff should last years).
