Thesis (MScEng)--University of Stellenbosch, 2001. / ENGLISH ABSTRACT: The use of cylindrical heat pipes for the thermal control and management of casting moulds
have been investigated. Heat pipes are tubes that possess a high capability to transfer heat,
up to a thousand times or more than an equivalent solid copper rod. The heat pipes used in
this thesis are copper tubed, use water as working fluid and have (phosphor-bronze) screen
mesh wicks. Experiments relating to practical casting situations in industry were designed
and performed, using pure tin as the casting metal. Three cases pertaining to the
requirements of an industrial casting mould were considered. The first case considered
heating of a mould through heat pipes, in order to keep it at a specific temperature. The
second case relates to the situation where metal is cast around a core, and the core is
cooled by a heat pipe connected to a heat sink. The heat sink in this case was an air cooled
fin. The third case is representative of the situation where molten metal is cast into an
external mould and the mould heats up due to the energy flowing in from the casting. In order
to cool the mould, heat pipes are used to transport the heat to a water cooled heat sink.
These three cases were modeled theoretically, which included using a standard finite
element method (FEM) computer package, NASTRAN 2.0 for Windows. For the FEM
simulations, the heat pipes are modeled using an equivalent conductivity approach.
Theoretical and experimental results are to within ± 30% of each other, but better results
could possibly be achieved using a better finite element model for the heat pipes.
A simulation case was performed to compare the use of an uncooled mould with a heat pipe
cooled mould, and a two and a half time improvement of production rate was achieved.
In support of the above mentioned casting related experiments, experiments have also been
performed on a specially designed cylindrical heat pipe to determine the evaporator and
condenser heat transfer coefficients. It was found that the heat pipe can transfer more than
500 W for vertical operation and around 160 W for horizontal operation. The heat transfer
coefficients of the condenser and evaporator ends are in the order of 1800 to 2000 W/mK. Experiments were also performed on the fins used as the heat sink in the experiment where
core cooling is investigated, to compare the experimentally determined fin heat transfer
coefficient with the theoretically predicted coefficient.
A theoretical study was also performed for an inclined ammonia thermosyphon in order to
compare the theory to a set of previously determined experimental results. The theory
produced accurate results for vertical operation, but it is clearly limited for inclined operation,
and can lead to inaccurate results.
A special correlation factor, the splashing factor, was defined to analyse the deviation
between the theoretical and experimental results. The splashing factor can be used in two
ways. Firstly, it can be used as a design correction factor and secondly, it can be processed
to indicate which operational variables have the highest impact on the discrepancy between
the theory and the experimental data.
It is recommended that further research into the use of heat pipes for the thermal control of
moulds be considered, based on the results achieved in this thesis. Furthermore, a finite
element model for a heat pipe can also be considered. It is also recommended that the use
of the splashing factor be considered for the analysis of thermosyphons. / AFRIKAANSE OPSOMMING: Die moontlikheid om hittepype te gebruik in die termiese beheer van gietvorms is ondersoek.
Hittepype is buise wat oor 'n baie goeie warmte-oordragsvermoë beskik, 'n duisend maal of
beter as 'n ekwivalente soliede koper staaf. Die hittepype wat gebruik is in die tesis is
gesëelde koperbuise, wat water gebruik as werksvloeier en ook 'n (fosfor-brons) sifdraad
pitmateriaal bevat. Eksperimente wat verband hou met industriële gietprosesse is ontwerp en
uitgevoer. Suiwer tin is gebruik as die gietmateriaal. Drie giet gevalle is ondersoek. Die
eerste geval het die verhitting van 'n gietvorm met hittepype behels. Die tweede geval hou
verband met die situasie waar metaalom 'n kern gegiet word en die kern word afgekoel deur
middle van 'n hittepyp wat gekoppel is aan 'n hitteput, wat in die geval 'n lugverkoelde fin is.
Die derde geval hou verband met die situasie waar gesmelte metal gegiet word in 'n
eksterne gietvorm en die gietvorm verhit as gevolg van die energie wat vanaf die gietstuk
invloei.
Hierdie drie gevalle is teoreties gemodelleer, wat die gebruik van 'n eindige element analise
(EEA) rekenaarpakket insluit (NASTRAN 2.0 for Windows). Tydens die EEA simulasies is die
hittepype gemodelleer met behulp van die ekwivalente geleidingskoëffisiënt metode.
Teoretiese en eksperimentele resultate is binne .± 30% van mekaar. Beter resultate kan
moontlik verkry word as 'n verbeterde eindige element model vir die hittepype ontwikkel kan
word.
'n Simulasie geval is uitgevoer om die produksietempo van 'n onverkoelde gietvorm te
vergelyk met 'n hittepyp-verkoelde gietvorm, en 'n verbetering van twee en 'n half maal is
gevind vir die verkoelde gietvorm.
Ter ondersteuning van die bogenoemde gietverwante eksperimente en teoretiese modelle, is
eksperimente ook op 'n spesiaalontwerpte silindriese hittepyp uitgevoer om die kondeser en
verdamper hitte-oordragskoëffisiënte te bepaal. Daar is bevind dat die hittepyp meer as 500
W kan oordra tydens vertikale gebruik en ongeveer 160W tydens horisontale gebruik. Die hitte-oordragskoëffisiënte vir die kondenser en verdamper is in die orde van 1800 tot 2000
W/m2K. Eksperimente is ook uitgevoer op die finne wat gebruik is as die hitteput in die geval
waar die kern verkoeling ondersoek is, om die eksperimenteel bepaalde fin hitteoordragskoëffisiënte
te vergelyk met die teoretiese koëffisiënt.
'n Teoretiese studie is ook uitgevoer vir 'n skuins termoheuwel sodat die teorie vergelyk kan
word met In stel bestaande resultate. Die teorie gee akkurate voorspellings vir vertikale
gebruik, maar is duidelik beperk en kan lei tot onakkurate resultate vir skuins gebruik.
'n Spesiale faktor (splashing factor) is gedefiniëer om die verskil tussen die teoretiese en
eksperimentele resultate mee te analiseer. Hierdie factor kan op twee maniere gebruik word.
Eerstens kan dit gebruik word as 'n korreksiefaktor en tweedens kan dit geprosesseer word
om aan te dui watter veranderlikes die hoogste impak het op die verskil in eksperimentele en
teoretiese resultate.
Dit word aanbeveel dat verdere navorsing gedoen word op die gebruik van hittepype vir die
termiese beheer van gietvorms, gebasseer op die resultate wat verkry is uit die tesis. Verder
kan 'n eindige element model vir 'n hittepyp ontwikkel word. Dit word ook aanbeveel dat die
"splashing factor" oorweeg word in die analise van termohewels.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/52563 |
| Date | 04 1900 |
| Creators | Groenewald, Abraham |
| Contributors | Dobson, R. T., Stellenbosch University. Faculty ofEngineering. Dept. of Mechanical and Mechatronic Engineering. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
| Format | 1 v. (various foliations) : ill. |
| Rights | Stellenbosch University |
Page generated in 0.0058 seconds