Calibration with Epsilon and DSLR help

[tooth_dr]

This is my first time using the Nikon D800E for imaging, and I'm trying to work out where I'm going wrong here.  I've attached a master flat and bias, an uncalibrated and a calibrated sub, and the resultant (heavily stretched) stack.

You can see the light banding along the base of the calibrated subs, so something isnt working properly, and the flats dont appear to be correcting properly.  I took my flats using a tracing panel, diffused with a sheet of grey acrylic.  I'm using APP to stack the data.

Any help very much appreciated.  I dont like to tag people on here but if you have time, would you be able to offer some advice?  @mftoet @sharkmelley  My stars are not good shapes either so any input on that would also be great.

 

Thanks

Adam.

[tooth_dr]

I ran CCDI on 8 subs from above, giving this average:

[tooth_dr]

Another test in CCDI using a different set of images from a different target, do the curved corners indicate collimation issues or spacing issues?  I'm on a fixed spacer - NIKON-->TAK screw

 

 

 

Edited September 21, 2020 by tooth_dr

[sharkmelley]

I'm not clear exactly what you are saying here.  It's the first time you are using a D800E so does that mean the previous camera you used on your Tak Epsilon calibrated perfectly and had excellent shaped stars?  

Mark

 

[tooth_dr]

5 minutes ago, sharkmelley said:

I'm not clear exactly what you are saying here.  It's the first time you are using a D800E so does that mean the previous camera you used on your Tak Epsilon calibrated perfectly and had excellent shaped stars?  

Mark

 

I was using a KAF8300 mono Atik 383+ which has a much smaller sensor.  It calibrated 100%, and was more forgiving on the star shapes but I don’t think the scope has been perfectly collimated.  The D800 was already owned by me (not bought for astro) but it’s been suggested on here by a few images to start using it with the epsilon.  So now I’m here and struggling with it 🤷‍♂️
 

[sharkmelley]

Now it makes sense.  A bigger sensor is a much greater challenge.  The Tak Epsilon vignetting is quite severe on a large sensor so you need very accurate flats.  The collimation and sensor orthogonality is also far more critical, affecting star shapes.  For flats I strongly suggest using dusk sky flats at first until you are sure your tracing panel and diffuser are up to this very demanding task.  As for collimation and tilt, this can take a long time to get right.  Spacing should be fine because the wide T-mount adaptor screws directly into the reducer but collimation requires very accurate tools and a lot of patience.  Even when the collimation is perfect you may find (as I did) that you need to shim the wide T-mount adaptor to fix tilt in the adaptor.  I found the Tak Epsilon cheshire type collimator worked very well because it screws directly into the focuser in place of the reducer.

Mark

Edited September 21, 2020 by sharkmelley

[tooth_dr]

13 minutes ago, sharkmelley said:

Now it makes sense.  A bigger sensor is a much greater challenge.  The Tak Epsilon vignetting is quite severe on a large sensor so you need very accurate flats.  The collimation and sensor orthogonality is also far more critical, affecting star shapes.  For flats I strongly suggest using dusk sky flats at first until you are sure your tracing panel and diffuser are up to this very demanding task.  As for collimation and tilt, this can take a long time to get right.  Spacing should be fine because the wide T-mount adaptor screws directly into the reducer but collimation requires very accurate tools and a lot of patience.  Even when the collimation is perfect you may find (as I did) that you need to shim the wide T-mount adaptor to fix tilt in the adaptor.  I found the Tak Epsilon cheshire type collimator worked very well because it screws directly into the focuser in place of the reducer.

Mark

Thanks Mark. I have the screw on Tak collimating tool - that is attached below the threaded adapter and then I insert the eyepiece in the 1.25” hole.

There are quite deep indents on the adjustment side my secondary mirror, marred by the tips of the adjustment screws. These seem to be preventing the final fine adjustments from happening. I unscrewed and tried to lift out the secondary mirror this evening to sand off the marks but I can’t get it through the gap in the spider vanes.  I read that you have removed the primary cell. Can you confirm how you did this?  On mine the black rear cell appears to bolted on but there are nuts on the inside.  These nuts appear to be free to turn.  I’ve marked them in red below.  Is this correct?  The mirror is quite dirty - there is a fan on this scope and it looks like it has drawn in dust and it has stuck to the mirror when dew was on it.  I’d like to wash it before collimating the scope.

Sorry for the plethora of questions, I’ve had the scope a year now so time to get it working properly.

 

 

 

[tooth_dr]

I know pictures aren’t that useful but I think it’s fair to say I’m slightly out.  No matter how I tried to adjust the secondary it didn’t stay in position when tightened. And I think or hope that’s due to marring one the secondary back plate. 

[sharkmelley]

Don't touch the screws you circled.  Instead the 3 big bolts on the rear need to be undone - this doesn't affect the collimation.  The 3 grub screws on the rim also need to be slackened slightly.   The mirror can be removed, cleaned and put back without affecting the collimation.  It's a design masterpiece!

Mark

[tooth_dr]

25 minutes ago, sharkmelley said:

Don't touch the screws you circled.  Instead the 3 big bolts on the rear need to be undone - this doesn't affect the collimation.  The 3 grub screws on the rim also need to be slackened slightly.   The mirror can be removed, cleaned and put back without affecting the collimation.  It's a design masterpiece!

Mark

Thanks again Mark. I’m feeling a bit more positive about the whole thing now.  Just the slotted screw heads to release  

 

[sharkmelley]

2 hours ago, tooth_dr said:

Thanks again Mark. I’m feeling a bit more positive about the whole thing now.  Just the slotted screw heads to release  

That's right.

[mftoet]

The colour gradients in your stack are also common with the D810a I use. There's always some green to purple hue. I think this is sensor related. The gradients can be corrected with the light pollution removal tool in AstroPixelProcessor, but sometimes needs several runs and careful placement of the calculation boxes.

The issue with the bolts on the secondary mirror also sounds familiar. I think I've just finally managed to get the secondary mirror in the correct rotation opposed to the focuser (and also centred under the focuser). In my case, the dot on secondary has always been just slightly off center causing even worse star images in the corners than on your example image. Yesterday I've imaged with an ASI1600 4/3 format sensor and I no longer detect collimation error. I have to verify with the full frame D810a though.

I collimate the primary mirror with the tube pointed straight upwards (to the zenith) and only use the cheshire eyepiece (not the tube with cross hairs). As Mark pointed out, you can screw the 1.25" adapter directly on the ED-corrector. That will make it easier to judge if all circles are perfectly concentric.   

[tooth_dr]

I took the epsilon mirrors out last night to clean and recollimate.

There were indents on the metal back of the secondary that I rubbed out and then painted black.  I was finding that the secondary didn’t stay in the position I moved it too, and perhaps these marks were to blame.  It wasn’t that dirty but I cleaned it with water and fairy liquid.

The primary was really quite dirty, but cleaned up well.  I’m going to get a few heater for the primary and run the dehumidifier more frequently in the dome.

Collimation of the secondary was really very tedious to get the spot centred, but I managed it after about an hour of faffing around.

I am happy that it is all together looking forward to star tests later  

 

Edited September 23, 2020 by tooth_dr