Explore Astronomy

Astronomy Picture of the Day

JBOMD - Latest updates and system notices

Update: 05/07/2011 - The system is temporarily down while we re-tune the filter for a different transmitter.

Update: 04/01/2011 - While the system is still not optimised for 55 MHz (the current operational frequency), JBOMD is picking up Quadrantid activity nicely.

Update: 17/12/2010 - changed back to 55.245 MHz (following a receiver reset). Seeing some low level activity, but the antenna is still not tuned for this frequency and we're still using the old 48.25 MHz filter.

Update: August 2010 - changed to 55.25 MHz, but antenna is not tuned for this frequency (also outside the passband of the filter). Some low level activity being detected. Plan to retune antenna and modify the filter.

Update: 29/04/2010 - The transmitter we have been monitoring has finally gone offline in the great switch to digital TV in Europe. There will be no data until we can change frequency.

Update: 07/12/2009 - In an attempt to resolve the periodic SpectrumLab issues, upgraded to version v2.74 B7.

Update: 03/10/2009 - SpectrumLab fell over again on September 30th, so stopped monitoring for a few days. Fixed on 03/10/2009 at 0115UT.

Updates: 11/08/2009 - JBOMD is now up and running with a new computer, just in time for the Perseids. Both the 5-minute and hourly images are updating again, and being archived on the host machine. Pings are being counted, but automatic logging and RMOB updates are not yet functioning.

Update: 17/06/2009 - While trying to add a new hard drive to the computer, we encountered a few issues. The system will be offline for at least 24 hours so we can solve the problems.

Update: 21/04/2009 - Pierre has updated his Colorgramme software to upload to a new server to try and avoid the clashes that were occurring previously. JBOMD is now using v2.3 and uploading to the new server.

Update: 26/09/2008 - As we've nearly got a complete year of data at 48.25 MHz, we've been looking at the data. Below is a plot of the data so far.

Year one

The data are recorded in bins of one hour in length and the colour scale in this plot indicates the number of reflections detected in each bin. Large black areas are where there were problems causing the system to be offline for a significant amount of time. The diurnal effect of the Earth's rotation can clearly be seen. Strong activity can be seen during the Leonids in November 2007 and the Perseids are visible in August 2008. Periods of interference, or where ther transmitter was drifting, are also visible for short periods throughout the year.

Update: 11/09/2008 - Reported in various meteor mailing lists yesterday, the September Perseids provided a brief but noticable burst of fireballs in the early hours of September 9th. Reported by several observers using video systems, radio records showed an increase in counts in some cases. Others, like ours, showed no increase in counts over the period of activity (around 5am to 11am UT), but an increase in long-duration echoes due to fireballs. Hourly plots for this time period from JBOMD are available over here. There are also several reports of results from other observers on the meteorobs mailing list.

Update: 04/09/2008 - Upload problem fixed. Transfers resumed. (Thanks Bob!)

Update: 03/09/2008 - Due to the same login problem, live transfer of files to the web server has been suspended again.

Update: 29/08/2008 - Upload problem fixed. Transfers resumed.

Update: 28/08/2008 - Due to a login problem, live transfer of files to the web server has been suspended for the time being.

Update: 04/08/2008 - A strong RFI signal at 1305 MHz has been causing problems. The hourly reset has been causing the system to lock to the strong RFI instead of the meteors. Altered starting bandwidth from 650 - 1500 MHZ down to 600 - 1300 MHz. This should put the RFI outside the band so that the script no longer locks to it, and shifts the whole band slightly lower due to long-term changes in the transmitter we're listening to.

Update: 12/06/2008 - Well, by a rather convoluted method, the pc is back up and running for the time being. The problem was (as suspected) a lack of hard drive space. The old data have now been backed up on another machine freeing up several GB of space which should let it run for some time.

Update: 06/06/2008 - As of May 2008, the system ceased operations as a result of a problem with the PC. Until the problem is fixed, the system will be offline. Sorry, I'd fix it if I could!

Update: 21/05/2008 - For the forseable future, the system will not be regularly maintained (not because I don't want to, it's just not currently possible).

Update: 31/01/2008 - It appears my script thinks there are only 30 days in January.... fix will be implemented next week.

Update: 29/01/2008 - No fixes this week, if it falls over it will stay down until Saturday.

Update: 25/01/2008 - Reset script worked. There was a glitch just after 5am this week and the system only lost a few minutes of data.

Update: 21/01/2008 - Scanner reset script now polls the scanner every 10 minutes and only resets if it finds a discrepancy with the receiver's setup.

Update: 14/01/2008 - Scanner reset now operating. A short perl script running once every half hour resets the scanner back to the correct frequency (uses the Win32::SerialPort module) to try and avoid problems due to power glitches.

Update: 09/01/2008 - It looks like there was a power glitch at some point yesterday afternoon (1911UT, restored 1913UT). The receiver has been off frequency for a while. Reset at 10.50am this morning.

Update: 07/01/2008 - More RFI problems. There's a strong signal which appeared last night at about 9pm and has been present ever since. It wanders around in frequency on short timescales and is strong enough to fool the script. Unfortunately, it's not constant in amplitude, so the auto-reset is not working as planned either. The source of the signal is uncertain at the moment.

Update: 04/01/2008 - The Quadrantids put on a good show today, but it proved to be too much for the system - there were so many reflections that they often blended together, hence the counts are lower than they should be. Another item for the ToDo list.

Update: 30/12/2007 - Power glitch yesterday evening at 1712. Receiver came back on 0.3kHz from where it was. Not a problem until the transmitter drfited back to it's nominal frequency this morning - below the bottom of the band being sampled at the time. Scanner reset at 1020 this morning. Samples from 9 and 10 today have been deleted.

Update: 16/12/2007 - Added proper initialisation values and a reset command to the script in case it locks to the wrong signal.

Update: 20/12/2007 - The script now seems to be following the signal nicely. There are a couple of resets built in now - it resets the band wide once an hour just in case it accidentaly locks to an RFI signal, and it also resets if no pings are detected after five minutes. The five minutes plots should now follow the frequency range currently being sampled, the frequency width of the plot is 300 Hz and the sampling bandwidth is 100 Hz.

Update: 15/12/2007 - Reset back to the original settings. Implemented a simple following algorithm using the conditional actions within SpectrumLab. Basically, set a reasonable threshold to avoid too much noise, measure the frequency of the peak signal each time it rises above this threshold, average the past five such measurements and use that to set the frequency range for the next measurement. It's clunky (using more timers than I think are actually needed), but we'll see how well it works over the next few days.

Update: 13/12/2007 - Transmitter issues again! Due to the channel we were listening to going offline yesterday (it did come back on this morning) I've shifted the band slightly. The receiver is currently set 0.6 kHz higher.

Update: 06/12/2007 - Very low counts for 20071205.1300 to 20071206.1600 due to a power glitch which reset the receiver. Working on a way to fix this automatically...

Update: 28/11/2007 - Only a few days after we started logging echoes from the second transmitter, it went offline. Consequently, the second RMOB files are now no longer being updated.

Update: 20/11/2007 - Now sampling a second channel (RMOB). Also updated menus to include links to current and archived data. All our RMOB files are archived to the web, the directory includes a readme file describing the contents of the directory and the format of the data.

Update: 18/11/2007 - Back online after a brief bit of downtime for the Leonid star party in the Visitor Centre.

Update 16/11/2007 - The counts we record are now being used by the Colorgramme software to upload plots to http://radio.data.free.fr/.

Update 15/11/2007 - Aside from a bit of frequency drift which the transmitter seems to be suffering from, we now have a script running which counts reflections per hour. You can see the output from this script here: jbomd_latest_rmob.txt. Very low, or zero, counts are likely to be either due to me restarting the script after an adjustment or the transmitter drifting out of the band over which the script is sampling.

RMOB test1, RMOB test2.

Update 08/09/2007 - It seems that Eddie's modifications have solved the problems with the amp - it's been running continuously for two weeks now!

Update 31/08/2007 - We did eventually acquire another computer, thanks to the IT staff, and the receiver is now back online and updating this page. It's also now on a UPS, so we should keep running even during power glitches.

Update 07/07/2007 - The amplifiers have only been lasting for about five days of continuous use before they break. Eddie built a new amp box with a few modifications and when it was installed it worked fine. Trouble is, now the computer which processes the audio signal from the scanner is refusing to boot. It appears to be a terminal problem, so the receiver is offline until we can scrounge another one.

Update 16/05/2007 - Back online after the amp was repaired and the bias box modified. (Stopped again on the 19th at about 9am.)

Update 05/06/2007 - After a week of operation, the amp died again last weekend. On the plus side, the computer now recovers automatically from power glitches (the power fluctuates a fair bit out here).

Update 14/05/2007 - The system appears to be having some trouble after operating for a few days. The signal level dropped drastically again over the weekend, possibly due to the amplifier oscilating?

Update 07/05/2007 - The signal level dropped drastically on Monday. A quick investigation shows that the amplifier connected to the antenna appears to be the cause...

Update 06/05/2007 - Attempted to pick up the eta Aquarids (during one of our Star Parties) but we're getting no meteors overnight, is the transmitter off? When tranmission resumes, there is a lot of activity.

Meteor receiver

30/04/2007: Six months on from starting, and on Wednesday last week we finally got permission to install the antenna for the meteor experiment on the roof. The image on the left shows the antenna, in the background you can see the Lovell telescope. The building is the corner of the control room, and the antenna is the silver-looking T bolted to the brickwork near the roof. It may not look like much, but it took a lot of effort to put together. The cables run down into the cellar (passing the signal through an amplifier and filter on the way) where there is a computer to process the audio signal from a frequency scanner, linked to the PC via an audio cable.

The next stage of this project will be to add an extra step in the process where we can measure meteor counts. Then we can start to do some statistics, both on meteor showers and the sporadic background, and compare with the results of other observers around the World.

There are lots of people doing this around the World, some of them you can find on the web. One is Andy Smith who also has a very nice live display at two diffierent frequencies, another in David Entwistle who's site contains lots of useful information. Lots of details about the technique of forward scatter can be found at the International Meteor Organisation and the software we're using is the excellent Spectrum Lab.

Lots of people helped get this up and running, notably: Eddie (who built the system), Phil, Jim, Ian (it was all his idea!), and assorted other people.