2008 Amundsen EM302 Upgrade
Kongsberg Representative: Erik Aase
ArcticNet Representative: Doug Cartwright
Summary
From October 10 to 17, 2008, the Kongsberg EM300 multibeam echosounder was upgraded to an EM302. This required the installation of a new transceiver and a system calibration. A Kongsberg representative also replaced a significant number of boards in the new transceiver box.
The system appears to be working, however it was not possible to run complete tests due to the program commitments of the CCGS Amundsen. While the system does appear to be working normally (with some concerns) it will not pass two of it's critical internal diagnostic tests. This, in combination with the inability to fully test the capabilities, resulted in the system not currently being accepted. Kongsberg is investigating the BIST test issue, which may be due to software rather than a system hardware problem.
Specifications
| EM300 | EM302 | |
| max swath width | > 5000m* | > 8000m* |
| # of beams per swath | 135 | 288 |
| # of soundings per swath | 135 | 432 (high density mode) |
| # swaths per ping | 1 | 1 or 2 |
| pulse options | 0.7,2,15ms CW | 0.7,2,5ms CW and FM chirp |
| water column logging | none | yes |
* not tested
Installation
Installation went smoothly with the new transceiver box being installed in the place of the EM300 model. Brackets were modified but not removed so that the EM300 could be reinstalled if there are issues with the new transceiver. The cables were somewhat short for the new installation requiring them to run to the side of the transceiver meaning that the cover meant to protect the rear of the box was not installed. It might be prudent to change the installation to avoid this situation, however this will require significant modification, including welding on the side of a fuel tank. After installation, all the TX RIO boards, the BSP RIO board and capacitor boards were replaced with new equipment brought by the Kongsberg Representative
After installation the BIST self tests were run and the TX and RX tests failed. Kongsberg has indicated that the software assumes that the EM302 transceiver will be working with the EM302 transducers. As this is not the case with the Amundsen installation the test would not pass. We await a revised version of software to confirm this.
A bad stave was detected in the "Stave Display" in SIS. switching RX RIO boards caused the problem to shift to another stave, indicating that the problem was not with the transducers or cabling. Kongsberg will be providing a replacement RX RIO board.
The ability to use the attitude velocities from the version 3 POSMV was confirmed by setting up a temporary connection whereby the POSMV UDP data could be received by the new transceiver. The SIS software indicated that the attitude velocity values were being received. This should mean that the EM302 will be able to use FM chirp in deeper waters, but this has not been tested. The Amundsen equipment manager will be installing a dedicated network cable for this input.
Courses
Data was collected during a transit down the coast of Labrador, Canada. The following examples of uncleaned EM300 (figure 1.) and EM302 data (figure 2.) were measured in the same area with similar sea state. The quality of the EM302 data appears similar to the EM300, with the exception of the nadir area. In the figures below the central line shows the mean surface, the solid line on either side show the standard deviation of the soundings at 95% while the dotted lines represent allowable errors for Special Order, Order 1,2 and 3. This would indicate that the EM302 currently might have accuracy issues under certain conditions.
figure 1. EM300 data (uncleaned)
figure 2. EM302 data (uncleaned)
A similar issue was frequently observed at the transition from one transmit sector to another as shown in the figure 4. This was most commonly observed on the starboard side which may indicate that it is related to the use of the ice protection windows.
figure 3. Artifact between sectors in EM302 data
Overall, the EM302 data was much more work to clean than the EM300 data. This is of course partly due to the doubling of data with the multi ping mode, however the bulk of the cleaning is to remove these artifacts. Figures 4. and 5. show a section of data that show both the nadir and the sector error (in this case on the port side). Even with aggressive cleaning there is still a visible artifact.
figure 4. Uncleaned EM302 data
figure 5. Cleaned EM302 data
Figure 6. shows a section of profiles from this area showing both errors along with the associated standard deviations.
figure 6. Errors in EM302 data
Summary
The EM302 has been installed and is functioning. There are certain issues that have arisen that need to be addressed or investigated further, however it should be reiterated that it was not possible to do an exhaustive test of the system due to the assignment of The Amundsen to other duties. This report should be considered preliminary and open to further investigation.