From HubertBecker@gmx.net Thu Jan 16 13:45:00 2003 Date: Mon, 9 Dec 2002 01:40:30 +0100 From: Hubert Becker To: Steve Snow Cc: ATLAS SCT LIST , peter.jenni , balazs.szeless , steinar.stapness , soergel Subject: Forward Modul Review Hi Steve,   I had a look at the material which you want to present to the Reviewers next Monday.   Maybe you remember my objections against having the Final Design Review (FDR) for the forward modules as early as in August 02  before essential issues like the thermal performance and integration problems (grounding!) were solved.   Now you want to satisfy the concerns of the Reviewers by presenting some new and good results, but –again- the good ones are not new and the new ones are not good!   First of all to the good results: you confirmed our htc measurements, done more than two years ago (Why didn’t you mention this in the references?). And you propose now to use the ‘fully split block’ which was introduced by presenting prototypes already at the meeting in Cartigny in Nov 98  (You have been struggling against this concept since then).   Now to the new results: they were characterized by you as follows: “it is difficult to make accurate temperature measurements” (if you had a single glance at our simple but effective measuring device two years ago, you could have learned how to perform relevant measurements, not only with simple heaters but with realistic thermal modules).   Let me remind you further of the situation two years ago. At the SCT week in Dec 00 the statement was given: the cooling properties of the baseline are sufficient and there is good agreement between measurements and simulations. I replied that this agreement is based on measurements done in a wrong way and with FEA with false parameters inserted. For your FEA you used for the most significant parameter (htc) a value about twice as high as the values measured by us and now confirmed by you. The result was that the predicted temperature rise is low and by that sufficient cooling was pretended. In my mail to you of     6. 11. 00 (see attachment) I complained that you were not willing to use the correct parameters. In the corresponding thermal measurements not even the simplest rules how to perform those were kept.   What about such an agreement nowadays? In the minutes of Nov 02 SSG  one can read that the latest thermal measurements with a real module on a split block “seem rather similar to the unsplit block. The current plan is to go ahead anyway with a split block since Steves simulations indicate this should be not the case”. So whatever the outcome is, pseudoagreement or disagreement, the show must go on…. For time schedule reasons the Forward SCT needs a PRR – there should be a FIRST Design Review before that! I would be surprised if the Referees accept such a unscientific proceeding.   One of the reasons why your measurements with the split block deliver such poor results compared to the ones performed by us two years ago is that the module design has to be adapted to the fully split block! (If you mount the body of a caravan onto the chassis of a sports car you will not get a decent car!). System optimisation is needed! Hopefully the Referees will request this.   The statement in my letter to the SCT from 15.12.00 (also attached)  >>THE FORWARD BASELINE WILL NOT WORK!<<   still holds! The main reason for this statement was insufficient thermal properties, although many improvements were achieved in the past by adopting some of our proposals - after long fights: (the arguments against are in brackets) to use C3F8 as coolant (temperature was claimed to be too low!) to use TPG (too brittle) to use CuNi pipes (too massive) to use C-C blocks (too expensive) Your proof now of the superiority of  fully split blocks after four years resistance is another big step in the right direction, and this should open the door for the SCT to take over the last two big items of our Alternative Concept: 1.) A second module cooling contact opposite to the hybrid is not necessary, if the split block is carried out in a proper way! As a consequence the module design can be simplified to a large extent, for example a complicated and expensive spine would not be necessary any more and the production of modules would be much simpler. The corresponding cooling blocks and pipes could be eliminated. 2.) Running the remaining cooling pipe circumferentially instead of wiggly has imminent advantages as described in  ATLAS-INDET- 2000-019.   As pointed out in my mail of  05.2.01 (see attachment) I was hoping that with my demission the Alternative Proposal would be accepted soon. But  re-inventions as done with the split block now are too time consuming and this has fatal consequences: The complicated module design does not allow simple assembly set-ups, as proposed and prototyped years ago. Therefore the production of the more than 2000 modules cannot be done in time. The disks situation is even worse. The FDR was held in Nov 01, but up to now not even a quarter of a disk with all services has been built and tested. Therefore the grounding problems are far from being solved finally. Nearly daily ‘improvements’ become necessary (for example now it seems to be necessary to increase the thickness of the CuNi pipe). Under normal circumstances a fully working prototype should be presented before a FDR!   My statement, repeated above, is still valid, only the arguments have slightly changed. Because of improvements for the thermal performance, cooling may be not be the problem for non-irradiated modules, but the measurements done are far from proving that this will be the case for the design period of ten years. The complexity of the Forward SCT, caused by all that numerous ‘improvements’ to cure the wrong concept will be the dominating factor! An apparatus which is not ready on time cannot be claimed to be a working unit !   The waste of money and resources still appears to be no issue for SCT.   There is a last important point. All this ‘improvements’ are connected with a dramatic increase of material. At the ATLAS Executive Board the LAr Project Leader emphasised that the material budget should be a factor to be considered explicitly in choosing the optimal solution. The ID PL was asked to present an overall summary update of the material situation of the ID. Was this done? Due to the calculations we did for the Alternative Cooling Concept, the baseline of the Forward has a 50% higher material budget and this has probably increased  furthermore by the use of a thick CuNi cooling pipe.     There is a lot of criticism, but can the situation be improved? A very good chance was missed one year ago: to change over to the KB module, which would match perfectly with our cooling concept. The reason given was ridiculous – because the services are so far advanced. I predicted this in my mail from Oct 2001 (also attached).   The last chance to avoid a failure is to take over the remaining two big issues of our proposal: circumferential pipe routing in connection to a single but fully split block and the module design adapted to that. I am aware that it is very late for such a change. But  again: if you are on a wrong path, the right moment to change is immediately! An extensive description of the benefits is given in our ATLAS INDET Note.   Detailed drawings exist as well as prototypes. In a crash program everything could be updated within three month. I know excellent technicians at MPI and Uni Geneva having the knowledge to do this. This time loss would be only a small fraction of the gain in production time, ease of assembly and tests, and by avoiding repair due to the increased robustness.     Sorry Steve that I have to attack you in such a way. There are more people in  the SCT Collaboration who, for selfish reasons, are not interested in the achievement of good results. But the real scandal is that these people are protected by a management not being able to lead such a big project.   Let me finish my reproaches with a comment to your envisaged presentation to the Reviewers, where you admit at the end that the agreement of measurements with simulations is not so good. You then mention several possible reasons, ending with the words  <>. I would like to go further in this direction with the request that the competence of some people should be ‘sectioned’ first!   Regards   Hubert Becker [ Part 2, Application/MSWORD 37KB. ] [ Unable to print this part. ] [ Part 3, Application/MSWORD 44KB. ] [ Unable to print this part. ] [ Part 4, Application/MSWORD 39KB. ] [ Unable to print this part. ] [ Part 5, Application/MSWORD 44KB. ] [ Unable to print this part. ]