Vapour-compression water chilling machines of large capacity (megawatts
of refrigeration) are extensively utilised for process cooling in the
manufacturing, chemical and food industries, and for cooling hot mines. If
these machines under-perform, not only may the costs of consequent
losses in production be considerable, but products or working conditions
may degrade or become unsafe. Performance monitoring instrumentation,
including water flow meters, fitted to machines on site may be infrequently
calibrated, though. Therefore, the apparent performance of a machine,
indicated by the measurements in its water circuits, should be verified
through independent, confirming measurements in its refrigerant circuit.
For custom-built machines with screw compressors, though, there is a
difficulty with so verifying apparent performance. The oil cooling load of
screw compressors is large, and so must be taken into account in such
verification. If the oil is cooled by water, it is easy to measure oil cooling
load, but not if the oil is cooled by condensed liquid refrigerant in a
thermosyphon ‘loop’, because of the two-phase flow of the refrigerant
there.
This research report describes a study done on a water chilling machine
with a screw compressor, using ammonia as refrigerant, at KDC East
Mine, Westonaria, South Africa. This study was of a proposed, alternative
way of determining the oil cooling load – by measuring flows and
properties in the oil circuit. The oil cooling load so determined could be
independently verified by measurements in the water circuit of this
machine’s oil-to-water cooler.
The chief problem in determining oil cooing load from measurements in
the oil circuit is that the ‘oil’ is not pure lubricant, but a solution of oil and refrigerant. Therefore the most accurate available methods of predicting
the thermophysical properties of oil-ammonia solutions were identified and
employed.
The oil cooling load hence determined agreed within 0.43 per cent with
that determined on the water side of the oil cooler. It was hence used to
correct the coefficient of performance (COP) calculated from
measurements in the machine’s refrigerant circuit. This corrected
refrigerant-circuit COP, with the estimate of total mechanical power input
to the machine, were then used to indirectly estimate the machine’s water
chilling load. This indirectly estimated load compared very closely, within
1.12 per cent, to the apparent water chilling load determined from the
measurements in the machine’s water circuits.
It is concluded that in this study, oil cooling load could be determined with satisfactory accuracy from measurements in the oil circuit. Hence, this
proposed technique offers promise of being a useful and practicable
development, enabling apparent performance of such machines to be more conveniently verified than hitherto.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/14100 |
| Date | 11 March 2014 |
| Creators | Munyika, Nicholas |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | application/pdf, application/pdf |
Page generated in 0.0022 seconds