Wear and performance characteristics of some polymers in piston water pumps
Abstract
In this project, wear tests were performed on selected plastic materials to
determine their suitability for constructing low cost durable piston water pumps
to be used in Kenya. In the pump design, the cylinder was to be made from a
plastic pipe of acceptable wear resistance, and the piston seals from a plastic
material of acceptable wear rate when sliding against the chosen cylinder
material. The study was a follow-up of the author's MSc project carried out
between 1977 and 1981, with the same objective, where rigid Polyvinyl Chloride
(PVC) was considered for use as a cylinder material. Polytetrafluoroethylene
(PTFE), camphor wood soaked in linseed oil, PVC and Nylon 6-6 were tested as
piston ring materials.
The tests were performed in both dry and wet conditions usmg pm on
reciprocating strip, and pin on disc test rigs. One test included sand particles of
up to 2000 urn in the water. Nylon 6-6 was found much better than the other
materials for sliding on PVc. It was estimated that in clean water at 20°C, a
pump with a cylinder made from rigid PVC, and Nylon 6-6 piston rings could 11m
~or about 30 million cycles (20,000 working hours) before requiring the
replacement of paris, while in the presence of sand particles the wear rate would
be greatly increased.
This estimate was not exhaustive with regard to pumping borehole water. It did
not cover the effect of ground water temperature variations and compositions.
The suitability of other polymers for pump construction, and the load and speed
limits suitable for the end use had not been explored. Hence, the studies reported
in this thesis considered the production of cylinders from pye, Polyethylene
(PE) and Polypropylene (PP), because they are widely used in plastic pipe
.production. The candidate seal materials were Unfilled Nylon 6-6,PTFE, High
Density Polyethylene (HDPE) and Nylon 6-6 filled with Molybdenum
disulphide, because of their good wear resistance.
Theinitial basic tests were performed in tap water, using a pin on cylindrical disc
machine for about four months. Thereafter, prototype pumps were tested in tap
water and inunfiltered borehole water on aprototype test rig.
Thewear rates of PE, pye and PP were compared at 16°ewhen sliding against a
Nylon 6-6pin. Nylon 6-6 was used because of its outstanding performance in the
previoustests. The order of wear resistance in this comparison was PE, pye, PP,
with PPperfom1ing very poorly and pye being comparable to PE at low
load/speedcombinations. PE and pye were then tested, varying temperature up
to the material limits which were so'c for PE and 300e for pye. Their wear
performances were comparable at low load/speed combinations, but PE was
muchbetter in the severe wear regimeThe plastic pipes available in Kenya are made from P'v'C. Therefore, pye was
chosen to be the cylinder material. Since the end use was a low load/speed
application, it was tribologically suitable. Further tests were only performed on
PVe. The wear rates of the other pin materials when sliding on Pv C were
compared atzrrc, and the ranking was Unfilled Nylon 6-6, Nylon 6-6 filled with
Molybdenum Disulphide, HDPE and PTFE. Nylon 6-6 was therefore used in the
rest of the tests where the effects of bearing area and wear track length were
investigated.
Finally, prototype pumps with pye cylinders and Nylon 6-6 seals were tested for
20 million cycles in clean water and 5 million cycles in unfiltered borehole water.
From the tests, it was estimated that the pumps can run in real life situations for
over four years in clean water, and over one year in very turbid water, before
requiring service.