The EQ3 DSO Challenge

[Art Gecko]

1 hour ago, Stub Mandrel said:

I'm learning to hate the Crescent Nebula, it's cursed!

I think it should be called the "brain-ache" nebula.

57 minutes ago, mikey2000 said:

Art Gecko - I think you caught some good data for your crescent but I'm not convinced Photoshop is the best software for the initial stretch.  I use PS for final tweaks but I highly recommend a free trial of astrotools.  Feed it a autosave.fits file from DSS and have a play.  Or if you like, dropbox a .fits file and I can have a try for you (for comparison purposes)

Mikey, I appreciate the advice, but to be fair I don't think my problem is down to my using photoshop for the full process.... seriously, if you saw the gear I'm using to get these images you'd be as amazed as I am that I can get anything at all!!.... My camera and my EQ3 mount are the best bits of equipment I own! (and the camera is technically my girlfriends)... my scope has been crudely botched for imaging and can just about gain focus before the draw tube pivots on the cog putting the sensor out of axis, and I can't use the camera in RAW because my 10 year old mac-mini hasn't got the memory to stack RAW files and DSS literally takes 6 days to do just a handful of subs.... I just use a planetary imaging software called Lynkeos to stack all of my jpeg images, since that's the only other software I can get working on my mac.... So... when I get over all these other issues, I'll try out astrotools, cheers

[mikey2000]

Art, if I've learned one thing about astronomy, it's that there are some people who have superhuman abilities to extract things out of their equipment that everyone else said was impossible.  I think you might be part of that tribe :-)

 

Now, as this community gives so much, I'd be more than willing to give something back.  So, if you shoot raw+jpg and manage to find a way to get the raw files into some dropbox/onedrive place, I'd be happy to send you a raw/stacked tiff to play with. (although I'd totally understand if you want to do your own thing with the data - I'm not entirely sure how I'd feel about someone else playing with my data.  Part of this journey is in the processing, not just the photography....)

 

[Art Gecko]

Hahaha thanks @mikey2000 I learned how to stretch something from nothing through iPhone imaging... I'm just tenacious enough to try stupid things like that! managed to get this M51 with an iPhone 6 when I got the EQ3:

That's when my girlfriend felt sorry for me I think and bought the DSLR... anything is an improvement from there!!

I do want to start shooting RAW and at some point I will get round to upgrading to a windows PC, so thanks for the offer but for now I'm quite happy pulling off "small miracles"

Art

[Stub Mandrel]

The sawtooth ramp behaviour is exactly what I'm seeing as well, the image on page 12 of this incredibly useful tutorial is the same, only the other way up.

Analyzing_PHD2_Guide_Logs.pdf

I suggest anyone trying PHD2 takes a look at it.

[bobro]

In an effort to have goto I've been playing with a CG5 (sacrilege I hear you say!) using AstroEQ and found the same sawtooth using single direction guiding a couple of nights ago. You can see from the graph the mount wants to go in the same direction as guiding, which is wrong as the mount should be pushing against guiding and the slack taken up before guiding starts. In any case I wouldn't use single direction guiding as dithering is such a useful feature (used with my EQ2) and requires bi-directional guiding. So I've done my best to remove the slack in DEC waiting for a retest.

What I would rather do is put the stepper motors on the EQ2 as its guiding performance isn't much different to the CG5 and much more pleasant to carry around (I'm learning more about guiding and can make the EQ2 guide better as it wasn't being controlled enough). Putting a motor on RA is straightforward but DEC only has a limited range of movement. 

Can anyone offer a suggestion of how to modify the EQ2 DEC to allow 360 degree control with a stepper motor?

[Stub Mandrel]

53 minutes ago, bobro said:

I've been playing with a CG5

Get thee hence to a place of contemplation!

[Stub Mandrel]

Well, thanks to kappa-sigma stacking, this image isn't as bas as I expected, but all the stars are pear shaped as the DEC excursions have given them a little point on top.

But anyway, here it is my first proper guided image taken with a cooled DSLR.

90% best of 29 5-minute subs at 800 iso. The cooling (probably down to about -10 or 15) seems to have worked (no darks used), so that's something to celebrate.

And one with stars rounded using the 'one weird tip' that works in Photopaint but I can't see how to do in PS:

Anyway, the bearings have come (only one washer in each pack, thankfully it look like that will be enough) so off to fit them and document the process. Also a Skywatcher 9x50 finderscope (with a WO end plug!) from the SGL classifieds

 

Edited August 12, 2017 by Stub Mandrel
I thought I'd got rid of the 23MB version!

[Stub Mandrel]

OK, I've done the thrust bearing mod and it looks promising. this isn't a tutorial as (a) I have no idea if this will work and (b) I have reservations about the way I did it.

First, remove the DEC work drive by removing four screws and lifting it off to reveal the worm wheel. Look how small it is, the small size is the main cause of all our woes!

Now remove the nut inside the polarscope hole, I plugged the hole with felt to catch the nut and washer so they didn't fall down the hole and scratch my polarscope. A skinny spanner helps, but a pair of pliers should do it:

Pull the DEC assembly out. If it leaves a red fibre washer behind, remove it:

The washer may remain here instead, junk it, fibre washers aren't designed to give free movement. See how we don't have a continuous bearing surface. Clean up the grease on all of both parts:

Drop on the big bearing thrust washer and big bearing:

Lovely jubbly! Now grease your ralph bearings including the shaft (this will still be running in a close fitting sleeve bearing) with some nice grease. This is much nicer around scopes than lithium grease:

ideally a roller bearing shoudl be only 1/3 full of grease, to make sure the rollers don't 'aquaplane' on the grease. Now slip it back in place - the needle roller bearing will run on the (hopefully) unsullied top surface of the worm gear. Bear in mind our rates of <0.1rpm are not going to cause high wear! Feed tehw orm gear with some clean grease too:

See how it's m,oved up slightly, but about 2mm? Now at the lower end fit the following onto the screw - the original big steel washer (best looking face downwards), small bearing, small bearing thrust washer. Discard any small black washer that was originally present:

Now fit the nut, (I used tweezers to fit get bearing, washer and nut in place). Tighten it until there is no shake, then about 1/8 of a turn more. You want a small amount of preload but not enough to cause drag. You should be able to turn the DEC assembly by hand with virtually no feeling of resistance now. Problem - the nut doesn't go on far enough to engage the nylock part. Solution - I used a blob of silicone sealant/adhesive to help keep things in place. I will watch this and if it loosens put some more up the end of the nut. Why silicone? It sticks, but is easy to undo and less likely to drip on my polarscope than threadlock:

Now replace the worm assembly. I fitted two washers to help compensate for the extra depth of the bearing. Unless you lengthen the holes, you won't be able to restore original alignment. I didn't stress over this, as my worm wheel looks worn and a new alignment will use a different part of the gear, although engagement may not be perfect. use the two screws either side and the central screw to take up as much backlash as possible while keeping the worm gear free. this is a mission as using eth two long screws from below to lock it once adjusted, affects the alignment. Just keep tweaking and zero in on a good alignment.

My sophisticated test lab showed I hadn't really reduced the backlash, but the unloaded movement of the setup seemed to show much less 'stick slip' at low speeds. Good, that's what this was meant to cure:

The other change I made was to swap out the old (printer salvage) stepper for a NEMA 17 one (also printer salvage). That was a good move, the old stepper could be stalled by finger pressure on the wooden stick, with the NEMA 17 stepper I could pull quite hard without stalling. I can now unbalance the mount to reduce backlash - and hopefully more accurate GOTO moves now as well.

All I need to do now is test to check the stepper moves the right way then recharge my battery!

 

 

[Art Gecko]

Looks like you've started to lift that curse!

[Stub Mandrel]

1 minute ago, Art Gecko said:

Looks like you've started to lift that curse!

On no, it should be way better than that, but if I can get my DEC issue started I will try 5-minute Ha subs on it.

[Art Gecko]

Look forward to seeing it

[Stub Mandrel]

So do I! If I can get a really good result, all this will have been worth it.

[Art Gecko]

1 hour ago, Stub Mandrel said:

And one with stars rounded using the 'one weird tip' that works in Photopaint but I can't see how to do in PS:

 

To take those little tails off in Ps, blow up your image 4 times, then duplicate the layer, set blend mode to "darker" then move your top layer one or two pixels in the opposite direction to the tail of the star, just enough to remove the tail, then flatten the image and restore to its original size.

HTH Art

[Stub Mandrel]

That's what I do in Photopaint using 'if darker'. it doesn't seem to work as well in PS, but I only have PS2.

[Art Gecko]

4 hours ago, Stub Mandrel said:

That's what I do in Photopaint using 'if darker'. it doesn't seem to work as well in PS, but I only have PS2.

I don't like to use the method to be fair, I've tried it a couple of times but I am never happy with the results either and usually just ignore the misshaped stars or simply don't publish it and try for a better version... I never posted my dumbell nebula because I missed the focus point and pivoted the sensor making my stars turn into - shapes.... I could fix it with the above technique but the nebulosity also gets affected, then you have to mask it off... but then the stars within the nebula are still - shaped....... It's a headache!! Best to just try again and see if you get better results next time

[Stub Mandrel]

Well struggling again tonight. some improvement, but still a way to go.

I don't seem to be able to get the PA far enough out! It insists on drifting north when I want it to drift south so the guide pulses are against the imbalanced OTA.

I think my polarscope is out of whack.

[Stub Mandrel]

After improved, but less than perfect, guiding on the NAN, I've swung lower to the bubble nebula. Watched it like a hawk for my first 5-minute sub: 0.73" in RA 0.72" in DEC!

Well buffer me shindigs!

Clearly sticking the mount straight up isn't ideal.

But I'm formulating a plan to fit a belt drive to DEC by making a new spindle and dovetail mounting plate.

[Stub Mandrel]

Finally something that starts to look like success  - please excuse double post in the deep sky forum as I want feedback as its my first Ha+RGB (or is that HA+OSC?)image.

 

 

[alacant]

15 minutes ago, Stub Mandrel said:

starts to look like success

Excellent. And not an ed80 on heq5 in sight!

[Stub Mandrel]

And the North America Nebula + Pelican:

This one suffers from only being 12 subs as well as poor alignment matching! I think it has some fixed pattern noise, too, so maybe I need some supercooled darks!

[mikey2000]

Result!  If you can get .7" RMS guiding, that's probably keeping the jitters inside just one or two pixels on your camera sensor.  I get similar results from my guider set up and by the time I down-rez the 24megapixel (APS-C) files by 50%, I'd probably not really notice any difference if I got guiding down to .2" RMS.  I have a feeling the best I might ever achieve with this mount is a total RMS of 0.5".....

 

Also, thanks for the DEC bearing write up.  I'm not sure if I'll try it

[mikey2000]

Talking of bearings, how easy is everyone finding RA balancing the eq3?  Mine is quite hard - the RA axis is a little sticky.  It moves smoothly enough but not totally freely.  If it's way off balance, one end will drop but when it's near the middle, it's rather tricky (ie impossible) to judge where the exact balance point is because the friction in the axis is sufficient to overcome the out-of-balance.   Hence, judging east-heavy offset is hard too.

 

Just wondering...

[Stub Mandrel]

I wasn't that good all night! Funnily enough I figured out it was that good when DEC was pointing the OTA straight up i.e. perfectly balanced, and the guiding got worse as it moved west.

Soo... next time I will try with the OTA balanced in DEC, not unbalanced. This is sooo confusing!

In RA I have it quite unbalanced, surprisingly the RA guiding seems to be consistently good.

Edited August 13, 2017 by Stub Mandrel

[Nigel G]

22 hours ago, Stub Mandrel said:

Finally something that starts to look like success  - please excuse double post in the deep sky forum as I want feedback as its my first Ha+RGB (or is that HA+OSC?)image.

 

 

Very nice, I found it very hard and failed to get the complete Bubble, Well done  

Nige.

[Nigel G]

A clear spell last night so I tried my luck on M16.

Unfortunately the clear spell only lasted a short while as cloud started to roll in, I then switched to Cygnus and managed to get 2 x 4 minute subs before that also hid behind clouds.

M16 17x120s ISO 800 with dark, flat & bias. Modified 1200D, Equinox 80, EQ3. As the Eagle neb is low in the sky above heavy LP, @800 ISO 120s is max exposure for me, still managed to capture some detail though.

Nige.