1968
DOI: 10.1016/0017-9310(68)90086-0
|View full text |Cite
|
Sign up to set email alerts
|

Turbulent heat transfer in annuli with small cores

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

1970
1970
2022
2022

Publication Types

Select...
5
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 12 publications
(3 citation statements)
references
References 5 publications
0
3
0
Order By: Relevance
“…The correlations are given in Table 2-1, adapted from [4]. The predictions they evaluated included the works of Davis (1943), McAdams (1954), Foust and Christian (1940), Monrad and Pelton (1942), Wiegand et al (1945), Kays and Leung (1963), Petukhov and Roizen (1964), Dittus -Boelter (1930), Stein and Begell (1958) and Crookston et al (1968). Some of the works mentioned above are more applicable to this study than the rest and will be considered more extensive below.…”
Section: Work From the 1900'smentioning
confidence: 99%
See 1 more Smart Citation
“…The correlations are given in Table 2-1, adapted from [4]. The predictions they evaluated included the works of Davis (1943), McAdams (1954), Foust and Christian (1940), Monrad and Pelton (1942), Wiegand et al (1945), Kays and Leung (1963), Petukhov and Roizen (1964), Dittus -Boelter (1930), Stein and Begell (1958) and Crookston et al (1968). Some of the works mentioned above are more applicable to this study than the rest and will be considered more extensive below.…”
Section: Work From the 1900'smentioning
confidence: 99%
“…To reduce the uncertainty (random error) in the measurement of temperature at a certain point, more than one thermocouple was used: the entry and exit ports of the inner tube and annulus each had 4 thermocouples, while the inner tube and annular walls each had 2 thermocouples at a measuring station. The precision for each of these measuring stations was thus calculated by , = 2 × √ (5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19) where N is the number of thermocouples at a station and is the standard deviation of the averaged data set of N number of data sets for the measuring station. The results for the inlet/outlet ports and tube walls uncertainty measurements are shown in Table 5-2.…”
Section: Measuring Equipment Uncertaintiesmentioning
confidence: 99%
“…For comparison, numerous results in the literature concerning the position of maximum velocity, are plotted with our results in figure 8. Besides the values referred to in the survey of the literature, the experimental results of Crookston (Crookston 1966;Crookston, Rothfus & Kermode 1968) are shown for radius ratios a = 0.0318, 0.0627 and 0-1047, respectively. The experimental results are fitted quite well by the Kays-Leung relation (n = 0*343), the line of the zero shear stress (n = 0.386) being definitely lower than the line for the maximum velocity.…”
Section: Zero Shear Stress and Maximum Velocitymentioning
confidence: 99%