XMM Optical Monitor Mullard Space Science Laboratory University College London Author: M. Cropper ---------------------------------------------------------------------- STATUS OF EOB-1 INTEGRATION AT 27 MAY 1993 Document: XMM-OM/MSSL/PR/0037.01 27 May 1993 We have reached the end of the second month of the EOB-1 integration and there is a further one month to go. The second cycle of support by US colleages has just ended with Cheng Ho returning to Los Alamos this morning. It is an appropriate time to take stock of the progress so far, and to make corrections where we have done less well while there is still time to make them. The EOB-1 integration had 6 goals, each of which built on its predecessor in an increasing level of ambition. The details can be found in XMM-OM/MSSL/NT/0019.01 but in summary these were:- 1) verify electrical interfaces between blue detector, DPU and ICU/EGSE; 2) verify software protocols between blue detector, DPU and ICU/EGSE; 3) integrate ICU/EGSE and SDT; 4) capture an image end-to-end from the blue detector to the SDT; 5) demonstrate field acquisition and tracking; 6) demonstrate a full cycle of exposures in a pointing. Where have we got to so far? Broadly speaking, we are somewhere in the middle of goal 4. We're somewhere in the middle, because we have transmitted images through all three subsystems to the SDT, but not the final accumulated image, only the individual `ping' and `pong' frames. We almost got to the final accumulated image but just ran out of time before Cheng left. Nevertheless, getting an image from the blue detector all the way through to the SDT was a very satisfying achievement. In the process, we have exercised the electrical interfaces and software protocols to a considerable extent and we appear to have recified all the initial difficulties we experienced. Is this rate of progress satisfactory? Probably. We'd hoped to reach the end of goal (4) but we haven't. It is clear that in order to reach a satisfactory level of progress by the end of June, a further visit will be required from Jim and Cheng at least. This has been agreed. In order to get this far we've had to run the DPU in a manual way with a high level of intervention from us via the ICU/EGSE. This allows us to rectify any erroneous data from the DPU code at each stage before proceeding. We're clear on what needs to be modified as far as we've got, but we're not sure what difficulties we could still run into before we reach the end of goal 4. We will need also to automate the operation of the DPU in such a way as to make it operate as it should. Will we be able to complete the programme as envisaged? No. There are several respects in which we will not be able to achieve the goals we set ourselves. It is unlikely that we will be able to include Fast Mode. We possibly will not be able to include the absolute field acquisition tasks, which if so will eliminate part of goal 5 and all of goal 6. Have we been too ambitious? We have been ambitious in this EOB-1 in that our goals for the integration have forced a high level of software development early in the programme. I consider that this is very wise given the overall software task for the XMM-OM and I think it has paid off. In the light of developments, however, we will need to scale down some of our ambitions. Fast Mode will have to be dropped. We will try to keep the field acquisition module as a requirement, but if we can't, then we'll drop it to ensure we meet the core requirement of the EOB-1 testing: the demonstration of end-to-end image capture and the demonstration of the tracking algorithms (i.e. part way through goal 5). We cannot reduce our requirements further than this both for the health of our own programme and because this is the level required by ESA for our EOB testing. I have confidence that we will at least meet this core requirement by the end of the integration process. In order to ensure this, Cheng, Jim and perhaps Fred will make another visit currently scheduled from 6 to 11 June. I am concerned that this will not be quite sufficient. It is only 4 working days, of which at least 2 will be taken up with getting back to where we currently are. Cheng and Jim suggested a scheme they called `keyboardisation' which would enable the EEPROMs to be reprogrammed via the ICU/EGSE. This will be extremely important for the final phase of the integration, and will continue to be important for minor modifications after the US team leaves. However, the checkout of this new software will also take some fraction of the 4 working days. In order to ensure that we meet our core requirements as discussed above:- 1) those items already identified as faults in the DPU code should be rectified; 2) `keyboardisation' should be implemented; 3) there is a strong requirement that Cheng remains for a few days (preferably a week) after Jim (and Fred) return - this has to be balanced against Cheng's personal/family constraints, but we request that he consider it if at all feasible. In the meantime at MSSL, hardware faults in the MIC addressing leading to hairlines between detector windows will be rectified, software will be modified slightly to set the windows correctly, the ICU/EGSE FTP protocol will be changed to NFS and the command/display interface from VxWorks will be enhanced so that outputs from the DPU such as heartbeats do not interfere with the inputting of commands.