EXPERIMENTAL STUDY OF HEAT LEAK_\GE. 743 



used because it was the simplest form of apparatus immediately avail- 

 able at the time by which internal lagging could be provided. It was 

 found that although the regeneration effect noticed with the V-type 

 of case was absent, or practically so, a considerable error due to heat 

 leak in the cross-channel, and through the talc lining of the tube T', 

 persisted. This seems to have been due mainly to the depression of 

 temperature experienced by the steam in passing through the baffling 

 chambers. The most important bearing which the runs made with 

 this apparatus have is an incidental one concerned with the question 

 of the dryness of the steam; this is discussed later (see IV, 2, g). 



d. Straight-away type of plug case. 



The final form of plug case adopted, and the one by means of which 

 the various heat-leak difficulties are believed to have been solved, 

 is illustrated in Fig. 4. It will be seen that the annular lagging (that 

 concentric with the plug) is quite heavy; it is nearly two inches thick. 

 The slight axial taper given to its inner surface is intended to prevent 

 variation in the velocity of axial flow along the outside of the plug. 

 The apparatus is designed throughout to preserve a nearly constant 

 velocity of flow, equal to that in the secondary superheater. The 

 end-lagging (that at the open end of the plug) is also thicker than in 

 previous apparatus, being somewhat over one inch in thickness. All 

 lagging is of asbestos wood. 



The two thermometer bulbs are placed almost as close together as 

 it is possible to get them. Their distance of separation may be varied, 

 if desired, by using longer bushings at B, B. 



The plug case is designed so that it may also be used for an axial 

 flow plug, if desired. The change is effected by replacing the radial 

 flow plug and its lagging with a suitably lagged axial flow plug. The 

 low side cap is removed and replaced with a coupling, into which is 

 screwed a cylindrical chamber of about the same dimensions as that 

 which, in Fig. 4, is shown containing the radial flow plug and its lag- 

 ging. The low side cap is then screwed onto the other end of this new 

 chamber, which contains the heavily-lagged low-side thermometer. 

 No runs have yet been made with this axial flow set-up. 



III. OTHER APPARATUS. 



K general view of the essential parts of the whole apparatus is given 

 in Fig. 5 (plan, with end elevation showing manometer connections) 



