The Lowspec spectrometer

[Thalestris24]

Tried it out with the 500mm lens again - fov calc gives 33' x 24.75' and 1.55"/pix. Of course, I plan to put the Lowspec on the 115mm scope which will give 20.62' x 15.46' and 0.97"/pix.  Here is a quick tryout on a street lamp (unlit!) with the 500mm lens:

I used EZPlanetary to take that (more control over exposure), and the view is 75% of full. Very difficult to focus via the reflective surface but I guess you don't need sharp focus for guiding. I'm thinking the guide image will be fuzzy at 0.97"/pixel anyway.

I'm wondering how you manage if you can't avoid having more than one star on the slit? I suppose you have to just go to a different star...

I'll try the Lowspec out on the proper scope soon-ish .

Louise

[drjolo]

1 hour ago, Thalestris24 said:

I'm wondering how you manage if you can't avoid having more than one star on the slit? I suppose you have to just go to a different star...

In this case I think you will just have two spectra that you can process separately. You may select both stars regions and analyze separately.

Today I designed and printed simple Relco SC480 starter holder to LowSpec - I attached the models and the source file. It requires to drill 12mm diameter hole in the LowSpec cover - in the optical axis, near to telescope inlet. Then holding element needs to be glued into this hole ( I used superglue ). The limiting resistor for starter is placed in the AC plug. Here are some photos:

  

  

Starter can be put into LowSpec and secured with nut. Then after calibration it should be removed, there is another nut to cover the hole, and also safety enclosure for starter. I tested calibrating of Sun spectrum using SC480 in this holder, and I was able to measure Sun spectral lines positions with 0.15nm accuracy (600 l/mm grating, about 0.4nm FWHM). This starter should also help to focus main camera, when real target is very faint.

Sun spectrum calibrated with SC480 starter in BASSProject software:

Relco SC480 starter spectrum:

3D models to print and source file (SCAD). It should be printed with layer height 0.12-0.15mm: relco-starter-holder.zip

 

Edited January 11, 2020 by drjolo

[Thalestris24]

7 minutes ago, drjolo said:

In this case I think you will just have two spectra that you can process separately. You may select both stars regions and analyze separately.

Today I designed and printed simple Relco SC480 starter holder to LowSpec - I attached the models and the source file. It requires to drill 12mm diameter hole in the LowSpec cover - in the optical axis, near to telescope inlet. Then holding element needs to be glued into this hole ( I used superglue ). The limiting resistor for starter is placed in the AC plug. Here are some photos:

  

  

Starter can be put into LowSpec and secured with nut. Then after calibration it should be removed, there is another nut to cover the hole, and also safety enclosure for starter. I tested calibrating of Sun spectrum using SC480 in this holder, and I was able to measure Sun spectral lines positions with 0.15nm accuracy (600 l/mm grating, about 0.4nm FWHM). This starter should also help to focus main camera, when real target is very faint.

 

Hiya

I didn't realise the Relco bulb was so small! The spectrum looks similar to the (bigger) Sylvania (for 70W tube) bulb I recently tried - plenty of peaks at the blue end . However, I don't think the Relco is available in the UK anymore - I didn't come across any when I was searching. I have some 70-125W bulbs to try out. I kept my holder a simple push fit but I'll have to see how it performs in normal use. I thought that as it only has to be in for a short while the placing and removing of it should be fairly quick and easy. Anyways, a minor point. With the 12V DC - AC inverter I'm using, you can just as well keep the resistor in the box with the inverter but I had also originally thought about putting it in the mains plug as you've done. Are you going to use the Lowspec with your Meade F10? Can I ask what the screws are for that I can see looking through the hole at the scope end? I'm assuming you've done a modification?

Louise

[Merlin66]

Eric,

I use the original Starlight Lodestar camera...

 

[Merlin66]

Louise,

It's normally much easier to orientate the slit horizontal or vertically in the guide camera, this makes it easier to judge which way to nudge stars onto the slit and maintains the RA and Dec direction for acquisition.

The slit gaps are 6mm long.

The guide FOV will be constrained by the size of the visible slit plate and the focal length of the scope being used. You can roughly estimate your guide FOV by measuring the slit length on the guide image to get a scale mm/pixel. In my case due to the slit cover plate I have an opening/ FOV of 8mm x 6mm. 

Using CCDCalc (see image) this gives 9.8 x 7.4 arc min coverage.

You want to achieve as good a focus as you can on the slit gap... then when positioning a target star it also should be in good focus (using the scope focuser).

 

[Merlin66]

Louise,

Re stars in slit gap....

You don't have an option to change target stars if your doing a specific observation. If there's a second star in the slit, as Lucas says you can just not select it for processing.

 

[Paul Gerlach]

Louise,

Just as a reference; I made a quick snapshot of the view of a QHY 5-II camera on the slit.

Paul

[Merlin66]

Lucas,

Your results look very promising...well done!

With the reference lamp, change the exposure (depending on the wavelength area of interest) to ensure you don't saturate the lines - you can't easily use saturated lines for calibration....

[Thalestris24]

2 hours ago, Merlin66 said:

Louise,

It's normally much easier to orientate the slit horizontal or vertically in the guide camera, this makes it easier to judge which way to nudge stars onto the slit and maintains the RA and Dec direction for acquisition.

The slit gaps are 6mm long.

The guide FOV will be constrained by the size of the visible slit plate and the focal length of the scope being used. You can roughly estimate your guide FOV by measuring the slit length on the guide image to get a scale mm/pixel. In my case due to the slit cover plate I have an opening/ FOV of 8mm x 6mm. 

Using CCDCalc (see image) this gives 9.8 x 7.4 arc min coverage.

You want to achieve as good a focus as you can on the slit gap... then when positioning a target star it also should be in good focus (using the scope focuser).

 

Hi Ken

Yeah, re slit orientation,  you said before. I was just trying to see if I could get the whole of the slit in the guide cam fov, but couldn't. Then I was using the 500mm mirror lens but couldn't get a sharper focus than the screen grab of the street lamp. It should be slightly better with the proper scope as it's fitted with a usb focuser. Mind you, I've never been able to get a definite, sharp focus of any target. I think that's a combination of low target altitude (not much more than 30-35 deg), building thermals, and usually poor transparency and poor seeing. Of course, I'll do my best when the time comes!

I think I'll get confused if I have more than one star on the slit so I'll just try and choose stars where that won't happen. Mind you, I suppose the majority of low magnitude stars' spectra are catalogued anyway. At the same time, imagers do keep imaging the same targets over and over...

Cheers

Louise

 

[Merlin66]

Louise,

Spectroscopy 201.....moving from a transmission grating to a slit spectrograph....

Practise, practise, ......

When you start to record stellar spectra you'll find a new world of issues/ problems/ concerns etc etc etc.

The main thing at the moment is to get comfortable with acquisition and processing.....

[Merlin66]

Paul,

Your guide coverage looks very similar to mine - approx 8.6 mm x 6.8 mm.

 

[Thalestris24]

6 minutes ago, Merlin66 said:

Louise,

Spectroscopy 201.....moving from a transmission grating to a slit spectrograph....

Practise, practise, ......

When you start to record stellar spectra you'll find a new world of issues/ problems/ concerns etc etc etc.

The main thing at the moment is to get comfortable with acquisition and processing.....

Oh dear, that doesn't sound good... If it gets a lot harder than with a Star Analyzer then I'll probably lose patience and abandon it, I'm afraid.. For me, it's just a hobby, something interesting to do to pass the time. I've already pretty much given up on dso imaging - I've not even done any live stacking/eeva since the beginning of September. I think I've lost my imaging mojo I will give the star spectroscopy a go though.

Cheers

Louise

[Merlin66]

Louise,

Don't get too discouraged, you've come a long way....

remember what Kaler said "Pictures of stars show us what they look like - a spectrum reveals their souls"

 

 

[Thalestris24]

I managed to get the Travel Scope mostly sorted yesterday and today. Had to do some guide ring musical chairs and ended up with a pair which fit the body of the scope but are rather on the tight side... However, it looks like it's possible to more or less centre the two fields of view using a tv aerial on a nearby roof. The hard part will be getting it down to the window of the qhy5l-ii and keeping it there. At the moment I just have a Canon 1100d mounted on the 115mm scope. I've managed to roughly centre the Travel Scope + 178m in the dslr fov. I need to tighten the bolts holding the guide rings. I'll swap the Canon for a qhy5l-ii tomorrow and see what I can do. Trying to get it spot on will probably drive me nuts...

Louise

[Paul Gerlach]

Louise,

It's been a while for me that I did some actual spectroscopy on stars. Hopefully will be up and running again this year.

I used a e-finder made up off a QHY 5-II with a 1:1.4 25mm c-mount camera lens from Ebay and mounted that on the rail the telescope was mounted on. That way I was sure that there was no flex.
During day-time I aligned the e-finder roughly  with the telescope. At night I slewed the telescope at a bright star and centred it on the slit. I used EQMOD together with Chart du Ciel so I did a 'SYNC' on that star. Then I realigned the e-finder (with the help of the electronic cross hair of PHD) at the star. Checked if the star was still on the slit and repeated the process until everything was aligned.

When slewing to a faint target I first 'hopped' to a nearby bright star and did a SYNC. Or even three stars surrounding the target and synced on those (creating a correction map for the mount) before slewing to the target which I then identified with the star map from Chart du Ciel.

 

Paul

[Thalestris24]

1 hour ago, Paul Gerlach said:

Louise,

It's been a while for me that I did some actual spectroscopy on stars. Hopefully will be up and running again this year.

I used a e-finder made up off a QHY 5-II with a 1:1.4 25mm c-mount camera lens from Ebay and mounted that on the rail the telescope was mounted on. That way I was sure that there was no flex.
During day-time I aligned the e-finder roughly  with the telescope. At night I slewed the telescope at a bright star and centred it on the slit. I used EQMOD together with Chart du Ciel so I did a 'SYNC' on that star. Then I realigned the e-finder (with the help of the electronic cross hair of PHD) at the star. Checked if the star was still on the slit and repeated the process until everything was aligned.

When slewing to a faint target I first 'hopped' to a nearby bright star and did a SYNC. Or even three stars surrounding the target and synced on those (creating a correction map for the mount) before slewing to the target which I then identified with the star map from Chart du Ciel.

 

Paul

Hi Paul

I have a gpcam3 178 on a Travel Scope 70 (fl =400mm) which I've previously used as a guide scope though the 178 is new. Using plate solving I expect to be able to get it close to centred on a target. But it will have to be precisely aligned with the centre of fov of the imaging scope (fl=800mm), and the target will have to be visible in the fov of the Lowspec guide cam in order to be able to nudge it on to the slit. The mount is an AVX so no eqmod. I can't properly do a star align as I only have a limited view of the sky in the East via an open window. I've usually managed to find dso targets ok with plate solving but there is always an initial positioning error. Getting a particular star on the slit will be challenging but hopefully doable. The AVX isn't the best mount for doing this procedure but I'll see how it goes.

Cheers

Louise

[Thalestris24]

On 12/01/2020 at 17:17, Thalestris24 said:

I managed to get the Travel Scope mostly sorted yesterday and today. Had to do some guide ring musical chairs and ended up with a pair which fit the body of the scope but are rather on the tight side... However, it looks like it's possible to more or less centre the two fields of view using a tv aerial on a nearby roof. The hard part will be getting it down to the window of the qhy5l-ii and keeping it there. At the moment I just have a Canon 1100d mounted on the 115mm scope. I've managed to roughly centre the Travel Scope + 178m in the dslr fov. I need to tighten the bolts holding the guide rings. I'll swap the Canon for a qhy5l-ii tomorrow and see what I can do. Trying to get it spot on will probably drive me nuts...

Louise

Well, in the end I decided the guide rings weren't big or stable enough so I dug out a pair which were rather too big... But I substituted one of the adjustment knobs with a long bolt on each ring. Seems to work a lot better now and I've been able to centre the guide scope fov with that of the imaging scope. Also found out that by using mono8 rather than mono14 in SharpCap I was able to access the image flip which is useful. It won't matter for plate solving but it would have been nice to be able to flip the 14 bit image. I don't know why it's not available in SharpCap but I posted a query on the forum.

Louise 

[Thalestris24]

On 11/01/2020 at 22:14, Paul Gerlach said:

Louise,

Just as a reference; I made a quick snapshot of the view of a QHY 5-II camera on the slit.

Paul

 

 

I just readjusted my guide cam to get the slit horizontal. As mentioned before, I can't fit all of it in the fov using a qhy5l-iim. This is about the best I could get (excuse the sensor dust!):

 

That's a 20um slit. The position of the slit isn't stable - the rotation mechanism isn't working properly. It looks like the Ovio slit isn't necessarily moving in tune with the selector i.e. it's slipping. Annoyingly, it was fine before I reprinted the moving part but now the movement is sticky rather than smooth. I guess the lock ring isn't gripping the slit properly but I haven't been able to get it to properly screw in - had to snap it in place. It looks like I'll have to print yet another lock ring... It's not been coming out right even though I did the extrusion calibration. The first few layers print bigger than the rest so parts have a miniature skirt ('skirt' is not set in Cura). If anybody knows what's causing this printing problem I'd be grateful for any suggestions/help!

Cheers

Edited January 14, 2020 by Thalestris24

[Merlin66]

Louise,

Can't help with the printer problems....

I'm a little surprised to see the differences between your slit image and Paul's image....taken with the same QHY5 II camera.

I assume you both use the same focal length guider lens???

 

[Thalestris24]

3 minutes ago, Merlin66 said:

Louise,

Can't help with the printer problems....

I'm a little surprised to see the differences between your slit image and Paul's image....taken with the same QHY5 II camera.

I assume you both use the same focal length guider lens???

 

Hiya - not the same. Paul has qhy5-ii, I have a qhy5l-ii - smaller pixels.

Louise

[Merlin66]

Ahhh, OK.

The QHY5 has a 5.3 x 6.7 mm chip and the QHY5L II a 4.8 x 3.85 mm chip.

I also had to add a UV-IR filter to my guide cameras to improve the image.......

 

Edited January 14, 2020 by Merlin66

[Thalestris24]

2 hours ago, Thalestris24 said:

It looks like I'll have to print yet another lock ring... It's not been coming out right even though I did the extrusion calibration. The first few layers print bigger than the rest so parts have a miniature skirt ('skirt' is not set in Cura). If anybody knows what's causing this printing problem I'd be grateful for any suggestions/help!

Cheers

I wonder if the printing problem is a Cura setting, something to do with pla, or something to do with the Ender 3. I noticed Paul mentioned he prints threads with polycarbonate but he has a different printer also. 

Louise

[Paul Gerlach]

Louise,

Try to print at a layer height of 150 micron (0.15 mm) and set the first layer also to that height (or 100%). Set the first layer 'extrusion width' or 'layer width also to 100%. Print slow (say 30-40 mm/sec).
If you still have no success try printing with a raft. Set the raft to two layers and three top layers at 100% infill set the distance above the raft (the first layer of the actual object that's being printed) to 0.15 mm. Again set the layer height of the print to 150 microm (0.15 mm).

Yes I do have a different printer. A Felix Pro 3 that has multi extrusion. Meaning I can print with two materials. I can print with material A and use material B as support. Or use two different colours. It can create great prints but at a cost of about 2700 euro...

Paul

Edited January 15, 2020 by Paul Gerlach

[Thalestris24]

10 minutes ago, Paul Gerlach said:

Louise,

Try to print at a layer height of 150 micron (0.15 mm) and set the first layer also to that height (or 100%). Set the first layer 'extrusion width' or 'layer width also to 100%. Print slow (say 30-40 mm/sec).
If you still have no success try printing with a raft. Set the raft to two layers and three top layers at 100% infill set the distance above the raft (the first layer of the actual object that's being printed) to 0.15 mm. Again set the layer height of the print to 150 microm (0.15 mm).

Paul

Hi Paul

Thanks for the suggestions. The first paragraph is what I do except at 0.12 layer height. Is there a reason for setting it to .15?

I'll try the raft option but I don't understand why it's been printing the way that it has. I feel like maybe I've been doing something stupid...

Louise

[Paul Gerlach]

You know what? Turn that ring upside down. Print it 180 degrees rotated so that the side you you are going to thread in is on top.