Welcome to dmradas.co.uk, detailing my amateur radio astronomy observatory. The pages show the location of the observatory, the equipment being used and my findings and analysis of pleine lune janvier.
The observatory is located in SE Wales at an elevation of ~240m at the top of a south facing gentle valley.
It is shielded from significant Radio Interference from the North, East and West by low hills but has a clear view down the valley to the South.
However, the City of Bristol lies in this direction and has VHF radio 88-108MHz, UHF TV 450 –800MHz, long range UHF surveillance Radar 555MHz, Cell phone installations 900& 1800MHz, industrial sites and general city electromagnetic noise. The broadcast UHT TV and the Pulsed Radar are the most troublesome ground based sources requiring careful filtering in the antenna head amplifiers.
Interference from GPS satellites can sometimes be problem when observing at 1420MHz if the filtering and amplifier IP3 characteristics are inadequate.
Running across the site from East to West is an 11kV single phase power transmission line (the pole is located to the right in the picture below).
Fortunately there are no near neighbours operating electrical equipment and the land area available for the location of antennas etc is around 2 Acres.
The weather at this location can be a problem. On the west facing side of the UK rain, mist and general dampness can be a problem in the Winter months and requires all outdoor equipment to be protected. It is not possible to leave antennas and head amplifiers out of doors without gradual decay and potential damage. This means that set-up times for observing sessions can be up to half an hour to set out the equipment.
In the Summer months the temperatures can reach the low +30s Celsius and head amplifiers can not be run without closed loop active cooling.
Day to night temperatures can fluctuate by as much as 30 degrees Celsius.
This observatory has been developed by David Morgan.
A physics graduate, I undertook a Masters in Radio Astronomy at Jodrell Bank with a thesis on Lunar Aperture Synthesis Radar at 408MHz.
This was followed by a Phd in Magneto Plasma Physics preparing and launching VLF receivers into the Earths’ Magnetosphere.
About 5 years was then spent in the Aerospace Industry developing and testing scientific spacecraft.
30 years beyond that have been spent on a variety of projects in the Aerospace sector leading to partial retirement at the end of 2003.
With more time, it then became possible to consider fulfilling an ambition from school days to design, build and use a radio telescope.
My efforts and the time spent on the 15 foot antenna at school were detrimental to my A Level studies , however the experience was useful in the formal interview for a place to study at the UK Radio Astronomy site of Manchester University at Jodrell Bank.
On taking up the subject again in 2005 it was fascinating to see how far electronic devices have improved since those days at Jodrell. Sensitive preamplifiers (PHEMT) costing a few tens of UK£s were now able to achieve levels of performance that only research groups using cryogenically cooled devices were able to do in the 70’s.
Starting again from ‘scratch’ much had to be reviewed and relearned. The Internet was of course a great source of knowledge and contacts. As I have gained so much from people – Amateurs & Professionals who publish their material on the Web I decided to make the effort and put some of my own observations into the public domain. I hope you find something of interest in these pages.
strad 11Ghz TVRO 60cm Antenna
Work on building this Radio Astronomy equipment began in Summer 2005.
Having ceased to work for five days a week there was now time to begin to fulfil a long held ambition since school days to observe the skies at Radio Frequencies. The first efforts were made with a 60 cm offset reflector and an LNB that had been used for receiving analogue satellite TV. A simple screw driven elevation arm was attached to the dish and the whole mounted on a 3600 azimuth bearing with a brake. The IF @ ~ 1GHz was fed to a receiver via 75W satellite grade coaxial cable. The typical spectrum is shown below – many analogue transmissions interspersed with digital signal from ‘Hotbird.
Choosing a frequency between the TV transmissions it was possible to process the signal from the receiver and observe a transit of the Sun through the antenna beam. This detection was made on 2/6/05 and showed that it was possible to detect a non terrestrial object without much difficulty.
First Detection of Sun @ 11Ghz
It soon became clear however that there was little signal from the Milky Way at these frequencies, partly due to the lack of synchrotron emissions and the limited spatial integration afforded by the few degrees beamwidth of the antenna.
The obvious next step was to operate at a lower frequency and use an antenna with a broader beam.
Thus the first low frequency experiments were begun with a high gain Yagi on 408MHz. This was the simplest and lowest cost way of building a basic receiving system thought to be capable of detecting the Milky Way
A suitable design for a Yagi was found on the Society of Amateur Radio Astronomers Web Site
A 408 MHz Quagi Receiving Antenna for Radio Astronomy by Hal Braschwitz and Jim Carroll . Jim Carroll demonstrated a prototype of this antenna at the 1990 SARA Conference. It was a version of an antenna Wayne Overbeck described in the April, 1977 issue of QST Magazine and was called a “Quagi Antenna.”
The Quagi antenna shown above was mounted on a wooden Az-El mount and supported on a non-conducting tripod to reduce the effect of mutual coupling to nearby conductors and thus minimizing any distortions of the predicted gain and beam pattern.
The 50 W coax used to connect the active element to the head amplifier was fitted with an inductive clamp balun. The total impedance to the out of balance sheath currents was around 500 W.
Early measurements of the antenna frequency response using a Spectrum Analyser + Tracking generator and an electrically small monopole transmitting antenna some 20m across an open field from the Quagi showed that the antenna had a good symmetrical response with the –3dB Bandwidth of ~40MHz.
Frequency Response of SARA Quagi Antenna
-3dB Bandwidth of SARA Quagi
The polar response of the Quagi was measured sometime later on a more sophisticate ‘Antenna Range’ and was found to be ~300… At this stage only simple MMIC block amplifiers were being used as head amps with typical gains of 18 to 22 dB and Noise figures of > 3dB. The next step was to purchase a very low noise amplifier and try to detect the Milky Way. The LNA chosen was the Radio Astronomy Supplies 406MHz WD5AGO design with an 18dB Gain and a 0.37dB Noise Figure.
Several other LNAs were available and details are obtainable on the SARA Web Site all priced at around $150US. Several large manufactures in Europe and the US also produce suitable components, however they tend to be built to higher commercial / industrial standards and have precision machined metal enclosures – but at a rather higher price.
With the LNA and the Quagi antenna it was now possible to attempt to detect the Milky Way. The equipment was set up and the antenna set to an elevation angle of ~500. The antenna was manually rotated from North to South passing through the region from Gemini to Orion. The whole set up and conduct of the measurement was very ad hoc. However after several azimuth sweeps had been carried out it was clear that a consistent response was being obtained. The presence of a person close to the antenna to rotate it had a significant impact - but never-the -less the signal showed a clear broad peak between Gemini and Orion. This is the emission from the Galactic Plane looking outward toward the edge. Even though the whole affair was rather crude and unreliable, the thrill of a first detection is something to be experienced.
Being in the UK of course, this called for an immediate cup of tea!
The records are shown below : all hand written and the data plotted out using a simple pen recorder.
This was the beginning of activity in Radio Astronomy at this site in Wales in 2005. Much has been done since using a variety of antennas, dish feeds and interferometers at the frequencies of 408MHz, 1420MHz, 4GHz and 11 GHz.
In the range 405 to416MHz depending on the type of antenna, direction of observation and sources of narrowband interference. SARA Quagi (Yagi) antennas are used at this frequency. In addition the 3m dia dish is used with a 3 element Quagi feed.
Radio Astronomy Supplies 408MHz 0.37dB NF LNAs are used in this band.
Both Total Power measurements and Interferometric measurements are made at this frequency
In the range 1419 to 1435MHz . 1419 to 1421 MHz covers the frequency spread of the Doppler shifted H line with a base frequency of 1420.4MHz
The principal antenna for these frequencies is a 3m dia dish with cylindrical waveguide feeds.
Radio Astronomy Supplies 1420MHz 0.37dB NF LNAs are used in this band.
A pyramidal horn of ~ 1m2 physical aperture feeding a waveguide to coax coupler is also used in this frequency range
Two helical antennas have been constructed to operate around 1GHz and 2GHz
Two C Band TV LNBs ((0.35dB NF) have been purchased and enclosed in temperature controlled enclosures as feeds for the 3m dia dish.
Several commercial TV LNBs are used as feeds for a 60cm and 80cm diameter offset dishes.
An Icom-R7000 communications receiver has been modified to deliver an output signal from the AGC level circuit that can be processed and recorded. Amplitude and frequency stability is maintained by an active temperature control system attached to the receiver.
The antennas used at this station are shown below:
Twin Quagi 1.1m Separation
3m Diameter Dish f/D=.28 with 408MHz Quagi Feed
1420MHz Pyramidal Horn
1GHz Helical Antenna
3m Dish with 1420MHz Cylindrical Waveguide Feed
3m Dish with C Band Feed
60cm TVRO 11GHz offset Dish
80cm TVRO 11 GHz offset Dish
The system below has a 30m separation between the Antennas along an E-W Baseline
West arm of 408MHz Interferometer
East arm of 408MHz Interferometer
Receivers and Signal Conditioning
There are a number of receivers and spectrum analysers that are used for making observations. The total frequency coverage available is from a few mHz to 18GHz. The primary receiver used for almost all observations is an Icom-R7000 shown below. An DC – LF output signal is derived from the receivers AGC circuit by linking to the diode that drives the S meter. The output impedance is kept high by a series 10kW resistor no minimise circuit loading.
A coax connection carries the signal to an analogue signal processing unit for amplification and filtering.
Icom - R7000 Communications Receiver
Analogue Signal Processing Unit
From the antennas, signals are either fed directly into the receiving room or passed through an 18dB Gain low noise wide band (10MHz – 2 GHz) line driver . RF signals pass through about 30m of low loss 50W URM 67 coaxial cable into the receiving room. A wideband variable attenuator (shown) is connected between the coax and the receiver.
The Icom receiver has a number of demodulation options, the AM (~4kHz BW) and the Wide FM (180kHz BW) are used most often. The difference in detection sensitivities are compensated for in the Signal Processing Unit (SPU). Using the wider detector bandwidth is the best option when increasing the signal to noise ratio if used in conjunction with a low bandwidth filter (long time constant) in SPU.
The SPU has a number of functions including amplification / attenuation, filtering / time constant, level shifting, signal inversion, logarithmic compression and audio signal level measurement with a proportional DC output. The output from the SPU is delivered to conventional meters both inside the room and sent back to the location of the antenna for remote reading. Another output is fed to an 11 channel 12 bit ADC (Pico ADC11/12) for data logging in a PC.
Data processing in the PC relies mainly on Excel with image making / mapping software from Stanford Graphics.
There are a number of Spectrum Analysers covering frequencies from 10 mHz to 18GHz and these are used mainly in developing equipment and antennas etc rather than for long run observations.
PC for Data logging & Analysis
[Temperature control equipment of head amplifiers is also shown]
A few examples of H line signals, Doppler & velocity profiles, extended Right Ascension repeated spectrum measurements and signal intensity distribution plots,
An example of a complex H Line spectrum showing various source components with different Doppler shifts / velocities
Repeated 1MHz wide frequency scans of H Lines as a function of Time (Right Ascension) [This method captures the peak level of any H Line signal irrespective of its exact frequency / Doppler shift]
A map of the H Line intensity in the region around the Cygnus Arm of the Milky Way
The measurements shown below are made at a frequency close to the 1420MHz H Line frequency [1453.5MHz in this case] . Received power is predominantly from synchrotron emissions – not from Hydrogen ground state spin transitions - and shows a less intense distribution than at the H Line frequencies.
The map shown below is generated from measurements at 409MHz using the 3m Dish and a Mk 2 Quagi feed.
The –3dB BW is about 170 as shown in the overlay on the map.
409MHz Intensity Map -10 to +60Dec & 17 to 22hrs R.A.
Mk2 408 Feed
The 3m dish and 408MHz Mk2 Feed used for Measurements
Observation of Cygnus A & Galactic Plane @ ~3.9GHz
This observation was made using the 3m Dish and the C Band Feed.
The signal intensity from Cygnus A and the Galactic Plane is much lower at these frequencies than at UHF or VHF. The graph above shows the signal level as a function of time / RA (reversed) plotted on top of a 408MHz radio map of this region of sky from ‘Radio Eyes’ software. The original measurements are due to Haslam et al. At almost 4GHz the meteorological conditions prevailing during the measurement affects the stability of the observation. Rain in particular will disturb the measurement.
Cygnus A & Galactic Plane (RA Reversed) - Radio Eyes Picture
Cygnus A Radio emission @ 21cm (Galaxy is at the centre)
Observations of The Moon
Click image for larger version
406MHz Interferometer Measurements
The measurements shown below are made using a 406MHz interferometer constructed from two Twin Quagi antennas mounted on towers separated by 30m. The output from this system is a fringe pattern with a period determined by the observing frequency, the baseline and the source Declination.
Measurements are made with the antennas pointing due South at an appropriate Declination for the source to transit the beam. At the present time ( Feb 08) the system measures the Total Power as well as the fringe signal from the point sources. Thus the gentle ‘bump’ in the graph on the left of the page shows the amplitude of the signal from the outer rim of the Galactic Plane. Hopefully this year I will build the electronics to turn this system into a Phase Switched Interferometer that will ignore the background signal from diffuse sources and produce only fringes for the point source of interest. This current interferometer has an angular resolution in the E-W plane of ~1.50 degrees.
Total Power Plot showing signals from Galactic Plane & Taurus A
Processed data showing only the Fringe Pattern from the point source TaurusA
Optical Photograph of the Radio Source Taurus A - NGC1952 3C144 CTA36 M1 [Crab Nebula] SNR 6,300Ly
This supernova remnant lies in the constellation of Taurus and can be found above the well known ‘Orion – The Hunter’ in the northern hemisphere.
Right Ascension: 05:34:30
Freq: 178 MHz Flux: 1420 Janskys
Freq: 960 MHz Flux: 1030 Janskys
Calculated Spectral Index: 0.19
Cassiopeia A is a another supernova remnant [SNR] often observed by amateur radio astronomers and is much brighter than Taurus A
Cygnus A is another bright ‘point source’ and it is not a supernova remnant in the Milky Way. It is a remote Galaxy undergoing violent change producing an immense radio (and other electromagnetic) output
Another extra galactic radio source is Virgo A. This is a giant elliptical galaxy in the constellation of Virgo. It too is a highly active and unusual galaxy producing jets of highly relativistic particles over very large distances that radiate at many wavelengths by the synchrotron process involving weak large scale magnetic fields. Al though it is not as strong a source as the others previously mentioned it can be observed by amateurs. Below is an example observation of this source.
This is the Total Power basic plot of signal strength for a transit of Virgo A [Note the rising curve as the beam starts to track into the inner part of the Galactic Plane]
Processed Data showing only the Fringe Pattern due to the ‘Point Source’ Virgo A [as Virgo is a point source in a ‘cold’ region of sky it can be used to calibrate system sensitivity]
Virgo A 3C274 M87 Right ascension 12:30:48 Declination 12:22:59 Freq: 408 MHz Flux: ~560 Janskys Calculated Spectral Index: 0.79
It gives a great deal of satisfaction to develop stable equipment that is capable of detecting and recording signal from galaxies such as Virgo A . It is amazing to realise that amateur efforts can be rewarded by a feeling of almost personal contact with something as large, remote and immensely powerful as a radio galaxy like Virgo A. It is through the efforts of technology that people can experience this ‘contact’.
A composite picture showing an Optical Image of M87 Virgo A with its relativistic Plasma Jet (click image to enlarge).
There are very stable synthesised signal generators from a few kHz to 2.6GHz. A number of spectrum analyser tracking generators (very useful) and general purpose signal and pulse generators also principally used in equipment development. A typical system connection diagram for 1420MHz using the 3m Dish and circular waveguide feed is shown below.
Click Image to Enlarge
One of the most difficult aspects of operating a very sensitive system of cascaded high gain amplifiers is that of temperature stability. The 40dB gain in the antenna head amplifiers and the 18dB gain in the RF line driver amplifier are out of doors and are sometimes subject to temperature variations in excess of 300C over the 24 hour period of some observations. Active temperature control of these critical gain stages is essential and is achieved using a closed loop temperature sensor I.C. and Peltier Cooler/Heater mounted on a heat conduction plate on to which the amplifiers are fixed.
A Temperature Controlled Head Amplifier for 1420MHz Radio Astronomy
This document describes the design and build of a temperature controlled head amplifier assembly for use in Radio Astronomy at 1420MHz. The design is compact so that the amplifier can be mounted close coupled to the dish antenna feed, in this case to a 6 inch diameter cylindrical waveguide with a coaxial output. The system uses a Peltier semiconductor cooling / heating unit driven remotely from a temperature control electronics unit which takes real time temperature readings from the cold plate on which the Peltier element and the head amplifiers are mounted. The system schematic is shown below:
The low noise amplifier is a 1420MHz unit with 28dB gain designed by W5AGO and available from Radio Astronomy Supplies http://www.radioastronomysupplies.com The filter is a 1200MHz high pass SMA type VHF-1200+ available from Mini Circuits http://www.mincircuits.com. . The line amplifier is a BGA616 MMIC gain block with a gain of 15dB @ 14220MHz mounted on a silvered PCB inside a die cast aluminium enclosure fitted with SMA connectors.
The gain of the combined amplifier / filter chain @ 1420MHz is 43dB. The purpose of the temperature stabilisation is to maintain this figure to a 0.1dB over long periods of time during observations. This requires that the closed loop temperature control is better than 10C for an outside temperature variation of 200C In order to achieve this the amplifiers are well thermally bonded to a thick aluminium cold plate with significant thermal capacity. The Peltier cooling element is bonded to the other side of this plate and to the heat sink from which the fan draws waste heat. The unit on the cold side of the plate is surrounded with expanded polystyrene insulation and the whole enclosed in bright aluminised Mylar foil to reflect heat from outside. Finally the entire assembly is fitted inside a white painted plastic rain cover.
The temperature sensor is an LM35DZ IC connected to a CA3140EZ Mosfet Operational Amplifier providing a gain of 10x and able to drive the 30m of twisted pair cable to the control electronics.
1420MHz Antenna and Temperature controlled Head Amplifiers
The thermo electric temperature unit is fitted on the back of the 6 inch waveguide feed as shown below. Both units are enclosed in white painted light weight weather proof plastic containers.
(Click for a larger image)
Below is the actual head amplifier assembly with the components indicated.
The Radio Astronomy Supplies’ LNA and the BGA616 line amplifier were mounted to the cold plate using heat sink compound to ensure good thermal conductivity and fast heat transfer to the active components. The Peltier cooling element cannot be seen in these pictures as it lies at the heart of the assembly between the cold plate and the multi finned heat sink. The temperature sensing IC is spring loaded onto the cold plate and faced with heat sink compound. The whole assembly is ~ 100mm diameter and fits inside the plastic weatherproof container shown below.
The graph above shows the operation of the temperature controlled head amplifier assembly. The light blue trace is the temperature of the cold plate as a function of time over a period of ~ 1 hour. The yellow trace is the temperature of the RAS LNA (which shows some thermal lag behind the cold plate). The cooling is switched on at the start of the trace and reduces the temperature from 180C to 70C in a few minutes. Once the required control temperature is reached the Peltier element drive current [dark blue trace] reduces from maximum drive and, following a short settling period, keeps the temperature stable. Twice during the hour long test a 1kW fan heater is used to warm up the outer skin of the head amplifier assembly (the second instance is marked on the graph).The Peltier drive current responds by increasing and thus keeps the Amp and Plate temperatures constant to within 0.20DegreesC
The evidence of the use of the heater can be seen also in the brown trace which is the air temperature in the control room where the test is being performed. The Temperature scale is 1V = 100DegreesC
The temperature control unit is a straightforward OP Amp based analogue system where the actual sensed temperature of the head amplifier is compared with the required temperature (determined by a front panel setting). The difference signal is amplified with sufficient current (up to + - 3A to quickly drive the Peltier semiconductor element in the correct direction to reduce the temperature difference to zero. Temperature control is better than 10C for air temperatures of –100DegreesC to +300DegreesC. Temperature control to this accuracy ensures that the noise level and gain of the head amplifier does not affect the stability of the measured astronomical signal over long periods of up to 24 hours.
A Temperature Controlled Head Amplifier for 408MHz Radio Astronomy
This document describes the design and build of a temperature controlled head amplifier assembly for use in Radio Astronomy at 408MHz. The design is compact so that the amplifier can be mounted close coupled to the Quagi antenna, in this case to a combiner which connects two Quagi antennas together. The system uses a Peltier semiconductor cooling / heating unit driven remotely from a temperature control electronics unit which takes real time temperature readings from the cold plate on which the Peltier element and the head amplifiers are mounted. The system schematic is shown below:
The low noise amplifier is a 408MHz unit with a nominal 18dB gain designed by W5AGO and available from Radio Astronomy Supplies http://www.radioastronomysupplies.com The filter is a 400MHz low pass SMA type VLF-400+ available from Mini Circuits http://www.mincircuits.com. . The line amplifier is a commercial MMIC gain block with a gain of 22dB @ 400MHz mounted in a machined aluminium enclosure fitted with SMA connectors.
The gain of the combined amplifier / filter chain @ 1420MHz is 44dB. The purpose of the temperature stabilisation is to maintain this figure to 0.1dB over long periods of time during observations. This requires that the closed loop temperature control is better than 10C for an outside temperature variation of 200C. In order to achieve this the amplifiers are well thermally bonded to a thick aluminium cold plate with significant thermal capacity. The Peltier cooling element is bonded to the other side of this plate and to the heat sink from which the fan draws waste heat. The unit on the cold side of the plate is surrounded with expanded polystyrene insulation and the whole enclosed in bright aluminised Mylar foil to reflect heat from outside. Finally the entire assembly is fitted inside a white painted plastic rain cover.
The temperature sensor is an LM35DZ IC connected to a CA3140EZ Mosfet Operational Amplifier providing a gain of 10x and able to drive the 30m of twisted pair cable to the control electronics.
Temperature Controlled weather proof 408MHz Head
The construction of the head amplifier is very compact. It needs to be small and as light weight as possible so that it can be mounted alongside and in-between the two Quagi (Yagi) antennas on a connecting boom. The assembly needs to be rain proof in all orientations of the antenna which is achieved by placing the power and RF output connectors on the underside of the assembly. The package is 200mm long by 100mm diameter.
(Above) Compact Head Unit and Casing
(Above) 408MHz Head Amplifier Assembly
The temperature control achieved is better than 0.50C for the normal temperature variations occurring at the observatory site in the . With this level of control it is possible to maintain a stable system gain to around 0.1dB which is sufficient for drift of long term observation baselines not to present a problem. Two of these head amplifiers have been constructed and used in a 408MHz Interferometer arrangement with a 30m antenna separation giving an East – West resolution of ~ 1.50. The dual channel temperature controller unit sits in the receiving room and is based on the same design as the 1420MHz controller. A single multi-core screened cable connects the controller to a distribution box outside, which then feeds each antenna head amplifier. The RF outputs are added in a Wilkinson Combiner and fed via URM 67 coax to the receiver.