Hi Elton, I've got the model of the Tagger hall and the Dump enclosure running. The model plan is shown in the figure, it is a combination of the GEANT3 model supplied by Richard, and my old BSY Dump model (modified). The design parameters for the first run are taken "arbitrary" with the goal to suppress the background in the Tagger hall as much as possible, and also to provide wall thickness to take care of the ground water activation. The inner dimensions of the dump vault are taken as approx. 10 by 3.5 m, with the height equal to the height of the Tagger hall (4.5 m). The Dump proper is positioned inside of the iron block - filled far end of the stub (shown by the red hatch). The thickness of the side concrete walls is taken as 36 inch, 43 inch for the front wall, and the thickness of the labyrinth walls is 24 inch concrete. The vacuum beam pipe from the tagger building is extended into the dump vault as shown, it penetrates through all of the labyrinth walls. No tagger magnetic field is present. 12 GeV, 3 uA beam was originated at the beginning of the dump, and the outgoing products were traced through the walls and into the Tagger hall, using cascade multiplication techniques to increase statistics. Small red circles show positions of all scored hits in the ground, and also at the control volume close to the tagger scintillator counter plane. The results at this preliminary stage look as follows. 1. I evaluate the counting rate at the tagger counters as 10-50 Hz of events above 1 MeV threshold per 100 cm^3 of the scintillator, per 36 kW 12 GeV beam in the dump. 2. The max. neutron (Ekin > 20 MeV) flux entering the ground outside the dump vault is about 2000 n/cm^2/sec in the areas where the iron plug is not working at full strength (backward scattering from the dump). According to Erik, we have to bring this number down factor 10 to feel safe about the ground water activation. This converts roughly to extra 200 g/cm^2 of shielding on the sides of the dump vault (about a meter of concrete, or 30 cm of iron). The iron layer probably can be set up inside the vault along the walls, and the ceiling and floor can be made thicker. The question then, is this model realistic and satisfactory from the viewpoint of physics and civil engineering, and what should be the next iteration? Regards, Pavel Elton Smith wrote: Hi Pavel, Thanks for your estimates. I have a couple of questions/comments. First I assume the rates are for an electron current of 3 microA (nominal current at high intensity of 10^8 s-1). The estimated rate in a 100 cm2 counter is ~10^7 s-1. Our counters are approximately smaller by a factor or 10 (10 cm2), so the rate would be reduced to ~10^6 s-1. The calculated bremstrahlung rate at this current is of the order of ~10^7 s-1, so the S/B ~ 0.1. Based on this estimate we would conclude that the background rate needs to be reduced by an order-of-magnitude. Please check my interpretation of your numbers. Questions: 1. we assume that the scraping of the bremstrahlung tail is negligible in this calculation. Can this be estimated? 2. What is the energy spectrum of the background? What threshold is used? If the spectrum is is predominately at lower energies, it might be possible to increase the effective thickness of the detectors by mounting them at an angle relative to the scattered electrons. On your speculations 1. We have thought about the option of using one of the permanent magnets in the exit line to deflect the electron beam away from direct line of sight to the scintillators. Clearly the scraping on the edges will have to be considered. Thanks, Elton. Elton Smith Jefferson Lab elton@jlab.org (757) 269-7625 On Mon, 13 Mar 2006, Pavel Degtiarenko wrote: Dear Will, The crude estimate for the count rates of the tagger scintillators follows. Assumptions: 1. forward tagger scintillators, the closest to the beam line, are in direct view of the beam dump face through the dumpline itself. 2. the dump is a copper cylinder, diam. 12 cm, length 30 cm 3. all radiation flux coming out of the dump is shielded, except the back flux streaming along the dumpline, no labyrinth, etc. 4. the sample scintillation detector is 10 by 10 cm area, 1 cm thick. Result : The flux of particles with energy above 1 MeV coming at the detector consists mostly of electrons (~ 4.0 10^8 s^-1 msr^-1), photons (2.0 10^10 s^-1 msr^-1), and neutrons (3.0 10^9 s^-1 msr^-1). Assuming roughly photon and neutron detector efficiency as 1 percent for 1 cm scintillator thickness, the rate is expected to be about (6.3 10^8 s^-1 msr^-1). For the 10 by 10 cm detector at a distance D (meters) the trigger rate is expected to be (6.3 10^9)/D^2, that is, 6.3 10^7 s^-1 at 10 meters, and about 1-2 10^7 s^-1 for the present design with about 10 m distance from the tagger to the entrance of the dump tunnel plus 10 m distance to the dump inside the tunnel. The count rate of real detectors can be evaluated when the tagger drawings are available, generally the rate will scale as the detector's volume. Multiply the result by the volume in cm^3 and divide by 100. Speculations: If we can possibly collimate the back flux such that it won't irradiate the counters directly at the line-of-sight, the rate should go down. The labyrinth openings would increase the flux, but not dramatically compared to the present numbers. Lead shielding of the scintillators from the back flux would help to stop lower energy electrons and photons (but it will produce cascades from higher energy photons, increasing efficiency to them, so the net effect might be small). Best regards, Pavel Will Brooks wrote: Dear Erik, At that meeting I inquired as to whether a simple estimate had ever been performed for the amount of backscattered radiation, i.e. based on simple ideas like neutron evaporation, scaling from previous experience, etc. I was told that no simple estimate had ever been performed, and my recollection is that Pavel said that such an estimate would take 3-4 days to carry out. I think that the right way to proceed is to do a simple calculation first. If it is inaccurate by two orders of magnitude, but the answer is that the rate would be ok even if it were two orders of magnitude larger, then this is good enough to choose the separation distance of the building. I do support doing the more elaborate calculation on a slower time scale, as was written at the bottom of the prioritized list of items, since that information will be useful in thinking about shielding and in defending the design at the reviews. - Will abkemeier wrote: Allison, (et. al.) RadCon appreciates the prioritized list of due items, and completely understands the critical nature of timely answers. Having said that, I must emphasize that some of the due dates are not realistic, most notably Priority 1, because of the time intensive process of modeling. RadCon met with Hall D members (Will Brooks present), and decided that this must be a collaborative effort involving transferal of GEANT coding that Hall D had already performed for the tagger building. Once this is provided (which it hasn't been yet), it will take RadCon 3-4 weeks to get a decent answer, with the earliest "rough estimate" available in the neighborhood of 2 weeks. The good news is that some of the coding legwork, once completed, will ease determination of some of the other questions, such that some of the later due dates may be more realistic. Thanks! Erik ------------------------------------------------------------------------ Subject: [Fwd: 12 GeV and Radcon Support] From: Craig Ferguson Date: Thu, 09 Mar 2006 17:23:38 -0500 To: Erik Abkemeier , Pavel Degtiarenko To: Erik Abkemeier , Pavel Degtiarenko Gentlemen, To help with clear scope and action/due dates Allison has provided the attached. I assume that this was the topic of your meetings, but I wanted to see the precise request. Can you support this in the time frame requested? Craig ------------------------------------------------------------------------ Subject: 12 GeV and Radcon Support From: Allison Lung Date: Thu, 09 Mar 2006 17:09:21 -0500 To: Craig Ferguson To: Craig Ferguson CC: Rebecca Yasky , Will Brooks , Leigh Harwood , Delvin Whitlock Dear Craig, Attached is a prioritized list of near-term tasks in support of 12 GeV which requires the expertise of the Radcon staff. Due dates have been included. Will Brooks is your contact person for item #1. Rebecca Yasky is your contact person for the remaining items. Tasks marked as Personnel (PSS) or Environmental are items/issues that have been raised by the Radcon personnel themselves. Since this information is necessary to complete the 35% design of the Hall D Complex, the due dates are quite close. We greatly appreciate your support, and the support and expertise of your Radcon staff. Regards, Allison