Collider Scaling and Cost Estimation

  • R. B. Palmer
Part of the Ettore Majorana International Science Series book series (EMISS, volume 29)


The primary motivation for high energy physicists to study new acceleration mechanisms is to find a way to build colliders at energies above the SSC at costs less than the SSC. Cost considerations are, unfortunately, crucial. It is simply not useful to know how to build an accelerator that would cost 100 billion dollars. Although it is difficult to make cost estimates without knowing the technology, I believe the attempt is useful.


Peak Power Free Electron Laser Linear Collider Radio Frequency Quadrupole Bunch Length 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References and Notes

  1. 1.
    B. Richter, “Requirements for Very High Energy Accelerators,” SLAC-PUB-3630 (1985).Google Scholar
  2. 2.
    N. Kroll, “Surface Heating by Short Bunches of Radiation,” ‘Laser Acceleration of Particles,’ AIP Conference Proceedings #130, p. 296 (1985).ADSCrossRefGoogle Scholar
  3. 3.
    P. B. Wilson, ‘Linear Accelerators for TeV Colliders’; IBID, p. 560, and also SLAC AAS-Note 2.Google Scholar
  4. 4.
    For the purposes of this analysis I have taken the cost of a SLAC klystron to be $100,000 and its modulator and feeders to cost $200,000. I have further taken the klystron cost to be associated with peak power need, and divided the modulator cost into $100,000 for average power and $100,000 for stored energy. I was guided in this by conversations with Greg Loew of SLAC. The performance assumed, including SLED, was: stored energy 120 Joules, pulse length.8 μsec, and peak power 150 M watts. This yields $800/Joule (rounded to 1000), 7×10-4 dollars per peak watt, and $4 /average watt. Without SLED the cost per average watts is $2.Google Scholar
  5. 5.
    For induction linac driven FEL costs I have had to rely on conversations with members of the Two Beam Accelerator group who have demonstrated such a source. I have assumed that a peak power of 1 GW can be obtained (as reported in the Wall Street Journal). I have assumed a pulse length of 20 nsec and thus a stored energy of 20 Joules. I have assumed that such a system could be made to cycle at 1 kHz and that it would cost 1.4 million dollars. If I divide this cost into.2 million dollars for stored energy,.4 millon dollars for average power and.8 million dollars for peak power; then I obtain costs of $10/joule, $20/ave. watt, and.8 × 10-3 dollar/peak watt. Despite the Wall Street Journal publication the performance of the system is classified and thus no reference is available for the 1 GW operation. A reference to a lower performance is: T. J. Orzechowski et al., Phys. Rev. Lett. 54 p. 889 (1985). See also J. S. Wurtele, p. 305, (1985).ADSCrossRefGoogle Scholar
  6. 6a.
    For average power costs I used some numbers quoted by D. D. Lowenthal of Spectra Technology Inc. from a study by that firm for another application. Values as low as $50 per watt and as high as $1,000 per watt were given depending on optimism and time scale. I have taken $100/watt. For peak power I note that Corkum achieved, for 2 psec pulses, power densities of the order of 1012 watts/cm2. Assuming one tenth of this density, and taking $100,000 as the cost of a 1 cm2 final amplifier, I obtain 10-6 dollar/watt for peak power. The stored energy cost of $10,000 per Joule is obtained by considering the cost of lasers designed for high storage capacity. It must again be emphasized that these estimates have errors of a factor three or so either up or down. The Corkum reference is: P. B. Corkum “High Power, sub-psec ten μm Pulse Generation”, Optics Letters 8, 514 (1983). See alsoADSCrossRefGoogle Scholar
  7. 6b.
    D. Lowenthal and J. Slater, ‘Laser Acceleration of Particles,’ AIP Conference Proceedings #130, p. 818 (1985).CrossRefGoogle Scholar
  8. 7.
    Z. D. Farkas, “Binary Power Multiplier” SLAC-PUB-3694.Google Scholar
  9. 8.
    The ‘switched power linac’ was proposed first by W. Willis, ‘Laser Acceleration of Particles’ AIP Conference Proceedings #130, p. 421 (1985); see also Proceedings of the ECFA/INFN Workshop, CERN 85/07 (1985). F. Villar has a similar idea: SLAC-PUB-3804. In general, these papers concern a single pulse of radiation generated by a pulse of current from a wire photocathode to a high voltage anode. The idea can be extended to the excitation of a resonant standing wave in a small cavity by fast pulsing a photocathode wire within such a cavity. This idea, which I refer to as a “microlasertron” may be compared with a conventional lasertron in which a photocathode is again pulsed, but in which the electrons produced are focused into a bunched beam and energy extracted as in a klystron by ring cavities about the beam. See E. L. Garwin et al., “An Experimental Program to Build a Multimegawatt Lasertron for Super Linear Colliders”, 1985 Particle Accelerator Conference (to be published), IEEE Trans. Nucl. Sci. NS-32; also SLAC-PUB-3650.ADSCrossRefGoogle Scholar
  10. 9.
    P. B. Wilson, “High Energy Electron Linacs” SLAC-PUB-2884 (1982); also Ref. 3.Google Scholar
  11. 10.
    M. Bassetti and M. Gygi-Hanney, LEP-Note-221, CERN, Geneva (1980).Google Scholar
  12. 11.
    T. Erber and G. B. Baumgartner Jr., Proc. 12th International Conference on High Energy Accelerators (Fermilab, August 1983), p. 372; T. Himel and J. Siegrist, “Quantum Effects in Linear Collider Scaling Laws,” ‘Laser Acceleration of Particles,’ AIP Conference Proceedings #130, p. 602 (1985). See also Ref. 3.ADSCrossRefGoogle Scholar
  13. 12.
    Roger Erickson, “Final Focus”, SLAC AAS Note #6 (1985).Google Scholar
  14. 13.
    R. Palmer “Super Disruption”, SLAC-PUB-3688.Google Scholar

Copyright information

© Plenum Press, New York 1987

Authors and Affiliations

  • R. B. Palmer
    • 1
    • 2
  1. 1.Stanford Linear Accelerator CenterStanford UniversityStanfordUSA
  2. 2.Brookhaven National LaboratoryUSA

Personalised recommendations