Joseph Shepherd
Professor of Aeronautics
Caltech | Galcit | Explosion Dynamics Lab   
Biography
Contact
Research
Publications
Teaching
 Home >> Research >> ASCI >> Vibrometer
2005 Measurements on Elastic Vibrations of Tubes
This page archives strain data for elastic vibrations excited in tubes by propagating detonations. The facility is described in the thesis of Tong Wa Chao. The experiments were carried out by Florian Pintgen using an optical vibrometer in addition to strain gages. The key results from these tests were
  • Under repeated dynamic loading with strain reversal conditions, the strain gage response degraded rapidly, the data were not repeatable, and also showed significant artifacts.
  • The vibrometer data were very repeatable.
  • The tube wall thickness varied significantly with angle and this was quantified by destructive testing.

Presentations
The experimental setup, data from strain and pressure, and conclusions about repeatability are given in the slides presented at LLNL in June. Some revisions were made to update the plots describing the tube thickness variation effect.
  • June 2005 LLNL 'roadshow' presentation (revised 8/23/2005) (pdf)
  • Detonation-driven fracture experiments: Flexural wave validation exercise (pdf)

Setup

Schematic of test rig

The specimen tube is 6061T6 alumminum, the same type as used in previous studies. Outer diameter is 41.28 mm (1.625 in). The thickness of the tube is nominally 0.89 mm (0.035 in) but actual dimensions vary by +/- 10% due to the manufacturing technique. See the data at the bottom of this page for the measured tube thickness of this specimen.

Data

Description summary file

All shots used an identical mixture: 3O2+C2H4. Two initial pressures were used: 1 bar and 0.5 bar. The CJ conditions and parameters for the loading model of Beltman and Shepherd (see Appendix B, Journal of Sound and Vibration, 252(4):617-655, 2002) are:
P_0U_CJ P_CJP_3T
(bar)(m/s)(bar)(bar)(ms)
1.0237533.311.650.240
0.5234216.25.670.243

The decay time constant T was computed using a distance of 1.9 m between the orgin (ignition location) of the detonation wave and the measurement location. The time is given in units of millisecond (ms), 1 ms = .001 s.

The measured velocity is about 1% lower than the CJ values in the last half of the detonation tube. The corresponding peak pressure (P_2 in the model) will be about 2% lower than the CJ value. You will get slightly better agreement with the vibrometer data by adjusting the peak pressure and velocity downward to account for this. The actual values at the measurement location might even be slightly lower, as much as 1.5 to 2% for the velocity and 3-4% for the pressure.

Strain Gage location in degrees
GageDeg
S10
S2120
S3240

Strain Gage axial location and effective tube length

The gage axial location and position position relative to the fixture were slightly different than in Tony's thesis. The effective length of the tube is defined by the total distance between collets that grip the tube. In the present experiments, the collets were spaced about 420 mm apart. The collets very tightly grip the tube and are rigidly attached to the stiff fixture frame; this makes a good approximation to a built-in boundary condition at these loations.

The strain gages S1, S2, S3 and the vibrometer were located half-way between the collets, 210 mm from the first collet (at the ignition end of the fixture). The gage S4 was located 20 mm ahead (closer to the ignition end) of the tube midpoint and the gage S5 was located 20 mm behind (closer to the tail end of the fixture) the midpoint.

Shot conditions, vibrometer location, and data files

The data files are in asci format with space separators. The columns in the vibrometer data files are time (s), velocity (m/s), displacement (mm), and strain. The columns in the stain gage data file (labeled by location as given above) are time (s) and strain. Encapsulated postscript (eps) files of the strain time history are given for each gage. Only a limited number of oscillation cycles (about 8-10) are given for each test. Only one set of strain gage data is given. The vibrometer data was very repeatable and high quality, the strain gages less so.
shotP_0 (bar)Vib Loc (deg)Vib DataStrain Data
141225(dat) (eps)(S3) (eps)
151225(dat) (eps)
161225(dat) (eps)
170.5225(dat) (eps)
180.5225(dat) (eps)
190.5225(dat) (eps)
20115(dat) (eps)
21115(dat) (eps)
22115(dat) (eps)

Peak strains from vibrometer measurements are given in the text files in this directory The text files contain columnar asci data, space separated. The first column is time (s), the second column is normalized time (s), and the third column is the peak strain. The normalized time is the time relative to first peak; both positive and negative peaks are given.

Tube thickness measurements (file) (eps)