robin_astro

Hi Vlaiv, A quick check suggests at 100 lyr a main sequence G star would be Vmag ~7.5 an M dwarf would be ~13.5. https://sites.uni.edu/morgans/astro/course/Notes/section2/spectraltemps.html http://www.calctool.org/CALC/phys/astronomy/star_magnitude The magnitude is not so much the problem though. It should be possible to get the required SNR (2000 to detect 0.1% at 95% confidence) in 5 min total exposure for a mag 11 star and 200mm aperture http://spiff.rit.edu/richmond/signal.shtml The main problem is likely to be identifying transits against background systematic variations at the mmag level with the same timescales as the suspected transits, particularly where potential events extend over different observers. Cheers Robin

Can you supply evidence of this ? Detecting a 0.1% transit with a known timing using an amateur setup from the ground with a reasonable confidence level is already (surprisingly) impressive. I would want to see evidence though that detecting a transit with unknown timing of this depth from potentially months of background data collected from different sites is likely to be possible with a reasonable confidence level.

I routinely measure velocities spectroscopically over 10^4 km/s from a few days after explosion so yes the initial velocities are high. Here is an example of my measured velocities. http://www.spectro-aras.com/forum/viewtopic.php?f=38&t=2308&start=30#p12713 (though in this particular case it is a thermal runaway type Ia supernova rather than a core collapse type II) Also if we take the crab nebula for example which was a core collapse supernova which exploded 965 years ago, we can measure the current expansion rate directly and spectroscopically to be currently ~ 1000km/s eg http://spiff.rit.edu/classes/phys231/crab/crab.html Here is an example of an amateur spectroscopic measurement of the crab nebula expansion rate by Christian Buil et al http://www.astrosurf.com/buil/us/mission2/mission2.htm Cheers Robin

I believe there has been some somewhat speculative work done in this area in the past. This is the particular paper in question (I wish these on line articles would include the reference) https://arxiv.org/abs/1803.08692 Robin

This is for pulsars at ~70cm though. For Hydrogen there is an interesting practical comparison of a yagi (actually a helical) v a small dish early in the Hydrogen line thread. The dish won out in this case. Cheers Robin (G8DVW)

This isn't a half wave dipole either 😉

There are two amplifiers shown in the block diagram but the one connected to the back of the ground plane appears to be the first one "LNA 03-70cm from US4ICI (NF = 0.3dB, commercial solution)" visible in the image here rather than the second "LNA FOR ALL" shown in the block diagram at the other end of the feeder before the SDR dongle Of course matching to 50/70 ohms is not super critical for receiving but it would need to be tuned to resonance for efficient operation and with that box there it is uncertain where the length of the element is referenced from Cheers Robin

Actually a half wave dipole in free space has an impedance at the feed point of 73 ohms. though that depends on the height above ground. Adding parasitic elements also reduces the impedance. In the article I referenced above, the 72 ohm impedance of the 3D corner antenna similar to this one with a 3/4 wavelength element was reduced to 50 ohm by adding an additional parasitic element, though this design does not have this. Robin

Yes I suspect some kind of matching arrangement otherwise why is it there rather than mounting it directly to the ground plane ? Perhaps the element just slides within the box to tune the length to resonance (The 53cm long element is an electrical 3/4 wavelength long to resonate at 70cm when mounted on the ground plane.) I would want to know before building this though Cheers Robin

The chicken wire should be connected along all edges and to the ground plane plate and outer of the coax connector. See page 20 of this paper for example for tips on building a 3D corner antenna https://www.qsl.net/yu1aw/ANT_VHF/shorten3dcra.pdf

What is inside the box the antenna element is mounted on ? What is inside the box ?

The ratio of antenna length to thickness affects bandwidth. A thicker antenna element increases bandwidth

any multiple of 1/4 wavelength will resonate

The distant objects are not moving through space. The distance between them and us is increasing because space is expanding. The measurement of time and distance are not absolute but are dependent on the motion of the observer (special relativity) and distribution of mass ie gravitational field (general relativity)

Yes this is the latest in a series of workshops and other initiatives over the past few years by the BAA to support members wanting to get into spectroscopy and produce useful scientific data. You can see the presentations from the first workshop which kicked off this initiative in 2015 here https://www.britastro.org/downloads/15701 The BAA now also have a spectroscopy database, to complement their variable star database. Good quality spectra of any object can be submitted for use by researchers. Over 4000 spectra have been added in the 2.5 years it has been running https://britastro.org/specdb/ Earlier this month the BAA also took part in a Pro-Am session at the professional UK National Astronomy Meeting where I gave a short paper on amateur spectroscopy, https://nam2019.org/schedule-by-session/details/27/325 a follow up to one I gave at the same meeting 10 years earlier. http://adsabs.harvard.edu/abs/2009JBAA..119..121L The number of amateurs doing spectroscopy at a Pro-Am level has increased significantly over the past 10 years. Robin

Your plan is to detect 1 mmag transits. This level of precision sounds tough to me. Have you checked with any of the experienced amateurs already doing this sort of work if this is achieveable?

There are a lot of deep optical surveys looking for transients already running. If there are bright fast optical transients around it would be surprising if someone had not picked up on these already. Provided images are not undersampled, I dont really see cosmic ray hits producing false positives for fast optical transients. Optical transients and cosmic ray hits etc are easily distinguished from their PSF so you don't need synchronised telescopes to distinguish them, just a second follow up image. An independent corroborating image showing the transient would be useful to convince sceptics that the detection is real of course but a first step could be to check some existing sub images looking for potential transients which are not cosmic ray hits to see how many candidates there might be.

The problem is cosmic ray hits are not the same as detecting " a couple of photons" though. A cosmic ray hit produces typically tens of thousands of electrons in the detector so are extremely strong signals. I think you would have a problem detecting the correlation of a weak signal between amateur images. I have sometimes wondered though about using synchronised telescopes to do high speeed photometry, removing scintillation noise by cross correlation and detecting high speed variability brightness in targets. Robin

The loss of light from internal reflection is insignifcant compared to other losses but that is not the problem with the front surface mirror you are proposing to use. It is important that reflective coated surface with the slit engraved in it is on the face towards the spectrograph not towards the telescope (ie mirror slits are back surface, not front surface mirrors) otherwise you risk ghost spectra from the internal reflections. The back coated slit gives weak ghosts in the guider image but as Ken pointed out these are not a significant problem. Cheers Robin

It is a dumb question anyway. Using the figure of 0.5 arcsec (presumably as measured from earth using an object outside the atmosphere), the distance to the hypothetical book held in front of the telescope in part A is (spoiler alert) much less than the depth of the atmosphere, even from Mauna Kea so we would need to know the resolution as a function of atmospheric thickness. I sometimes wonder if brighter students who actually think, fail at the first hurdle because of this sort of "spherical cow" type question. Here is another one

How does it look from the back? (Mirror slits are normally back surfaced so you dont get ghost spectra from internal reflections reflected into the spectrograph) Cheers Robin

Hi Eric, That's a nice build and first light results. You can get to impressively high resolution to with the 1800 l/mm grating. There are some recent results with this grating on the VdS forum, for example splitting the binary components of Mizar in the H alpha line https://forum.vdsastro.de/viewtopic.php?f=24&t=5685 Cheers Robin

This is just speculation as I dont have the figures but I suspect these USB sticks are deliberately designed with low gain to avoid problems with overloading so are pretty deaf (They have an unfiltered front end and are just designed to receive strong signals from commerical stations eg TV in the presence of other strong signals.) The additional pre-amp plus filtering just turns them into a typical weak signal receiver configuration to give a measurable output . The second pre-amp could possibly be relocated to the other end of the feeder without affecting performance. Cheers Robin

https://en.wikipedia.org/wiki/Electrical_length the probe needs to be 1/4 wavelength long but this is not the same distance as in free space The correction factor depends on several factors which are difficult to predict accurately, hence the need to trim the antenna element to length or use the figure established for a given design. For example the particular Seti feed design posted has a probe length of 4.6cm compared with your calculated free air 1/4 wavelength of 5.25cm From wikipedia The electrical length of an antenna element is, in general, different from its physical length[better source needed] [4] [5][6] For example, increasing the diameter of the conductor, or the presence of nearby metal objects, will decrease the velocity of the waves in the element, increasing the electrical length.[7][8]

Are you planning to tune the antenna by trimming the length? If not, 5.25cm will not give the correct resonant frequency. It will be too long. What you need is the electrical length to be 1/4 wavelength which will be shorter than this, depending on a number of factors. Unless you are planning on tuning to resonance I would advise constructing a feed to the exact specification and dimensions of a known design. Robin