COLLAPSE OF TEXAS TOWER NO. 4 143 



Mr. Kendall. Is that the letter I have ah^eady asked you to put 

 into the record ? 



Commander Foster. I thmk so, sir. 



(Note. — The letter m which neAv criteria was given, reads as 

 follows:) 



New York, N.Y., July 3, 1958. 

 Subj : Movement and stresses in Texas towers. 

 Eef : OinCC letter April 18 transmitting Report on Measured Oscillations of 



TT-4. 

 Officer in Charge of Construction, 

 Texas Tmver NOy Contracts, Navy Building, 495 Summer Street, Boston, Mass. 



Dear Sir : We had hoped that the oscillation frequency count made on TT-4 

 and submitted to us with the referenced letter might furnish some information 

 as to the general behavior of the tower, and a possible clue as to the functioning 

 of the new diagonal struts which were placed on the tower in the upper panel 

 of the A-B side to replace those lost in transit during construction. We thought 

 that perhaps, by comparing the period as measured, with the theoretical period 

 of the elastic structure, we could definitely state that the braces were effective. 

 Unfortunately the field measurements were not made in severe weather as the 

 maximum reported height of swells during the observations was 15 feet and the 

 wind velocity only 2.5 knots. The period of the movement reported at the same 

 time was between 1.5 and 18 cycles per minute and there seemed to be no definite 

 correlation between the pei'iod and the wave or wind condition. Our analysis 

 shows that, with the bracing working, the period of oscillation should be be- 

 tween 26 and 46 cycles per minute and without the bracing being effective, 

 rigidity would be reduced so that the period should be between 16 and 19 cycles 

 per minute. The ranges mentioned cover various assumed ranges of compressi- 

 t)ility of the sand foundation. 



It can be seen that the measured frequencies agreed pretty well with the cal- 

 culated frequencies for the condition of the bracing not being effective and it 

 might be at first assumed that such was the case. However, as before stated, 

 the conditions were not very severe and, for such as they were, the total move- 

 ment was probably not enough to take up the specified tolerance in the fit of 

 the pins in the pinholes so that, although the braces may not have been effec- 

 tive during the conditions prevailing when the oscillation count was made, more 

 severe weather conditions with a larger potential sway might cause the pins to 

 engage and result in a shorter period. Thus the frequencies measured proved 

 nothing conclusively. Similar measurements during a severe storm would pos- 

 sibly be more fruitful. For the record, a very condensed summary is attached. 

 Figure 1, which shows average measured frequency plotted against wind velocity. 

 Table 1 shows conditions assumed for computations of theoretical frequencies 

 and a summary of our calculations. In the above discussion, torsional oscilla- 

 tions were not considered, as the periods for this type of motion are completely 

 out of the significant range. 



As a result of your verbal request we also investigated the ability of the tower 

 legs and bracing to withstand natural forces in case the replacement braces 

 were loose at the lower connections or rendered useless for other reasons. The 

 results of these computations are shown in table 2. It will be noted that two 

 sets of computations were made: (1) to see how much of a wave could accom- 

 pany 125-mile-per-hour wind and (2) to see what magnitude of wind and wave 

 could be tolerated if combined in accordance with the formula shov/n and de- 

 fined at the head of table 2. It will be seen from an examination of the table 

 that, for both sets of conditions, the weakest link in the entire system is in the 

 legs at elevation minus 75 which is at the top of the highest panel of bracing 

 assumed (for purposes of this investigation) as working. The most pertinent 

 figures of this table are emphasized with an extra box around them. The fol- 

 lowing facts are to be commented on : 



(1) Without any wave and without exceeding the basic stress of 20,000 p.s.l., 

 the tower should withstand a wind of 96 miles per hour. This is too favorable 

 a condition to be expected over a long period of time. 



(2) With the steel permitted to go to the yield point the tower should with- 

 stand forces from two combinations, viz : 



(a) A 125-mile-per-hour wind with a 36-foot wave. 



