Unistellar and FITS

[Grant Privett]

I was chatting with a friend recently who had purchased a Unistellar and was happily using it to take pictures from a quite light polluted site. It has gone down very well and he was enjoying using it. However, he mentioned, almost in passing, that he was not aware of any Unistellar provision for producing a FITS format file (or 16 bit TIFF or png ie something other than a jpg) which could be used to to do accurate photometry or astrometry.

They mention scientific mode on their website, so I assume you must be able get a FITS file from your stacked images.

There also seems to be a 4second exposure limit – I am curious as to whether that because the sensor is uncooled, the drive PE is severe or perhaps due to scene rotation on an alt az mount. Unless the field is very wide and pixels very small, I would have thought you could get away with 20secs or so before significant trailling occurred at the field edge.

I won’t be rushing out to buy one, but they do look quite fun – though a bit pricey for me.

• This topic was modified 1 year, 7 months ago by Grant Privett.

[Dr Paul Leyland]

I don’t have a Unistellar or the like and have no interest in getting one. Consequently no great research into your question has been done.

However, https://www.aavso.org/unistellar-evscope might prove an interesting and/or useful starting point.

Paul

[Grant Privett]

Yeah, I read some of the AAVSO pages on this. Sounds like they are not in a hurry to provide FITS files – which is kind of weird given how its a trivial bit of coding. A doddle in C and trivial in Python.

Sounded like the way the Citizen Science is done is taking observations and sending them to someone else who does the data reduction… Sounds a bit dull to me. Data reduction is what cloudy nights are for.

Some mention of the code being open sources and then of release of an API, but nothing yet – and promises are cheap of course.

[Mark Fairfax]

You can request the FITS data from Unistellar support and with Apple iOS devices the file is saved as PNG and not JPG. Unistellar have always seemed rather ‘guarded’ around the data and have said it will be available at some point but have said this for some time now, (there was talk of accessing via a Cloud shared area). This is a shame as their main competitor Vaonis Stellina allow easy access to the data in FITS & TIFF formats.

I can confirm exposure times are between 1ms & 4secs – I’m unsure if this is a limitation of hardware/software or if it is intentional.

The Citizen Science is definitely not dull – very informative & engaging with a great deal of communication & collaboration worldwide via Slack app groups. Just check out the Unistellar Citizen Science sections on their website
https://www.unistellar.com/citizen-science/
Unistellar collaborate with SETI and recently NASA-funded Unistellar exoplanet program UNITE.

I absolutely love my Unistellar eVscope and wouldn’t hesitate to get the next version down the line, (with a bit of saving up mind you).

Mark Fairfax

[Grant Privett]

Oh, I think I missed something there….

You have to ask them to give you a copy of your data as FITS? Theres no SD card or something where all your session data is saved that is accessible to you? Does this mean you don’t have anything else other than a jpg/png they serve up to you live? I’m surprised at what you said but, clearly, they have found a market.

Just to be clear, as I have never used one, what is the bit depth on the png provided? I think that both 8 and 16 bit are legal png formats, though some software doesn’t like 16 bit pngs or TIFFs.

If the png is 8 bit (or even 3x 8bit) then they have rather constrained you to the Citizen Science route, as that makes it harder to do any on your own. But, I don’t know the png file format well enough to judge if its metadata content could be used to populate a FITS header.

• This reply was modified 1 year, 6 months ago by Grant Privett.

[Mark Fairfax]

Unfortunately no access to the internal SD card (storage capacity 12GB on eVscope 1 and 64GB on eQuinox & eVscope 2) – you upload data to the Unistellar network via Wi-Fi. I think the market is for people who either don’t have the time or technical expertise (let’s face it Astrophotography has a steep learning-curve) and to improve the accessibility of Astronomy & Astrophotography to a wider audience. The on-board computer stacks and processes the frames (dark and background removal, shift-adding and stacking) to produce an improved image in real time. Personally there are times I’d like to get down ‘n’ dirty with the RAW data being an ex-Systems Analyst & data monkey.

I’m by no means an expert but opening an image in Affinity Photo 2 showed the following info for an example PNG from my eVscope:
2560 x 1920px, 4.92MP, RGBA/8 – sRGB IEC61966-2.1
Apple dev state RGBA8 as a 32-bit-per-pixel, fixed point pixel format in which the red, green and blue colour components precede the alpha value.
However, the Sony IMX224LQR image sensor in my eVscope 1 is 12-bit (improved Sony IMX347 on eVscope 2).

Thanks for making me look into this subject a bit further… everyday’s a learning day for me. Plus, it nudged me into sending a reminder to Unistellar about their promises around open-access to the RAW data.

Mark Fairfax

[Author]

Posts