Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Frustration with some kind of abberation

  • Please log in to reply
30 replies to this topic

#1 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 07:33 AM

Good morning, all.  I'm having a heck of a time with my 16" scope.  I've never gotten a good image out of it above like 50X power, and it's driving me nuts.  For reference, I am not (I hope!) a complete idiot.  On my Z8 I re-spotted my mirror, loosened the mounting clips, and re-assembled the back of the scope.  I purchased a new Antares secondary and mounted it to a Destiny curved spider with the three spots of glue.  On the Z8 I re-assembled from new components I get bang-on collimation and amazing planetary views.  It's only f6 and with a much stiffer mirror, but like, I know at least the basics, right??

 

I am not here to point fingers at any supplier, etc, but to try to solve my problem.  I have a 16" scope with an f/4 mirror, 1.5" thick, normal glass, from a small mirror maker but not Cadillac like Zambuto or Lockwood.  It's in an 18 point cell with teflon pointed triangles, and a cable sling, rear fan and two boundary fans.  I have a 4" Antares 1/14 secondary.  I originally had the telescope maker use a 4-vane Destiny curved spider, which was glued on by the three spots of glue again.

 

Since I've gotten the scope, I've had some kind of triangular distortion or triangular star images (See three sketches below).  I use a Visual Paracorr, and have tried everything with and without it.  My only TV lens is a 22NT4, so coma correction is not perfect for the XWs.

 

The top left is a sketch of a defocused star at high power.  One one side of focus I get a squashed triangle.  It's two very closely spaced bright lines on the outline, and then mostly black.  It's not symmetric inside and outside focus.

 

Mars at 200X shows three discs.  I get a very strong trefoil impression - it's three discs.  Both of these I've tried moving all over the visual field, to see if it's collimation and coma causing problems, but the shape holds up.  Massively defocusing a star (bottom sketch) shows a well centered secondary shadow.

 

I'm collimating with a self-aligned Zhumell laser that can't be off by more than a mm or two on the mirror.  I use this to aim the secondary at the primary mirror spot.  I then use a Barlow'd laser to align the primary.  I use an offset pupil Catseye autocollimator to fine tune the secondary, and re-check the Barlow laser again.  I'm cooling the mirror 1-2 hours before use, fans on.  I've checked the triangle pads, made sure they were moving freely and unstuck.

 

I'm going nuts!  This is a pretty severe abberation!  I'm not seeing less than ideal star images at 200X, I'm seeing unacceptably ugly stars at 50X+.

 

I thought maybe the glued on secondary was the problem.  I bought a secondary holder from Astrosystems, and mounted it in there.  I still had ugly stars.  I removed some of the batting, because I thought it was too tight.  Last night I went out and set it all up again, and I got the sketches I'm linking below.  There's something seriously wrong still occurring.

 

Is it collimation/alignment?  A bad secondary?  Bad secondary mount?  Bad primary?  Bad primary mount?  I don't think this is primarily coma, as it gets much worse with mag and I can't find a clear spot in the field where it goes away.  Again, I don't want to blame telescope maker / primary / secondary maker, so if you're reading this it's not a hit job, but I am here to solve a problem.  The scope maker said they had it out one night for use before I picked it up and it looked good on Jupiter.

Attached Thumbnails

  • sketches.png


#2 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 07:37 AM

Attaching images of the secondary mount as it is now:

 

 

Attached Thumbnails

  • secondary front.png


#3 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 07:38 AM

Secondary from behind:

 

That's the Astrosystems secondary holder.  I have a block of walnut behind it as a spacer, then a milk jug washer.  There's three long adjustment screws for tip/tilt, they protrude ~10 mm from their mounting plate to the milk jug washer.  The center is tensioned with a (very tightly compressed) spring around the central bolt.

Attached Thumbnails

  • secondary behind.png

Edited by TiSaph, 27 October 2020 - 07:40 AM.


#4 Darren Drake

Darren Drake

    Skylab

  • *****
  • Posts: 4,151
  • Joined: 09 Oct 2002
  • Loc: Chicagoland

Posted 27 October 2020 - 08:06 AM

Make a note of exactly the orientation of the triangular images in the star test inside or outside of focus.   Then rotate the primary mirror 30 degrees or so and retest.  If the triangular pattern rotated with it then the problem is definitely in the primary mirror.   If not then your secondary or secondary support is suspect.  After that we would look into primary mirror support issues..

 


Edited by Darren Drake, 27 October 2020 - 08:10 AM.

  • Jon Isaacs and stubeeef like this

#5 happylimpet

happylimpet

    Fly Me to the Moon

  • *****
  • Posts: 5,213
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 27 October 2020 - 08:08 AM

I think the finger of suspicion points very heavily at the primary mirror cell - got any pics from behind?

 

Can you be sure its not being pushed down by the clips? That would be my first check.


Edited by happylimpet, 27 October 2020 - 08:08 AM.

  • Asbytec likes this

#6 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 08:14 AM

I'll get a picture of the mirror cell.  I've examined it extensively and it seems fine??  It has no clips - there's three retaining rods, but they do not make contact with the mirror at all.  It's on a thin cable sling.

 

I'll have to figure out how to rotate in the field.  Two sticky pads with velcro anchor the cable on the mirror edge, and that will be tricky to manipulate in the dark without dropping the mirror.  Maybe remove them and find a star near zenith?  I'll still have to tip the telescope to get access to the rear of the mirror to rotate / manipulate it.



#7 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 08:17 AM

Here's the report from the original mirror

 

150 line Ronchi outside of focus

Attached Thumbnails

  • primary test.png


#8 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 08:18 AM

Another test:

Attached Thumbnails

  • primary test 2.png


#9 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 08:54 AM

Oh snap - I wonder if this has been the problem all along?  I think the cable isn't tight enough (and there's no room left on the turnbuckles).  It's hard to tell because of the velcro pads, but the mirror may be resting on the mirror retaining clip posts:

 

 

Attached Thumbnails

  • cable 1.png


#10 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 08:55 AM

Other side:

 

 

Attached Thumbnails

  • cable 2.png


#11 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 08:56 AM

Turnbuckles, not any room left to tighten:

Attached Thumbnails

  • turnbuckle 2.png


#12 happylimpet

happylimpet

    Fly Me to the Moon

  • *****
  • Posts: 5,213
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 27 October 2020 - 08:59 AM

blimey, could well be.....



#13 KBHornblower

KBHornblower

    Messenger

  • -----
  • Posts: 445
  • Joined: 01 Jul 2020
  • Loc: Falls Church, VA (Washington DC suburb)

Posted 27 October 2020 - 09:23 AM

I would try disassembling the turnbuckle and sawing off about 1/4 inch of each screw.  That will allow it to take up about another 1/2 inch of slack.


  • Jon Isaacs and wrnchhead like this

#14 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 27 October 2020 - 09:31 AM

I might just try to get another screw of the same threading, but shorter.  I need to take the mirror out to wash it anyway, it's filthy.  Gonna be a bit scary.



#15 wrnchhead

wrnchhead

    Mercury-Atlas

  • *****
  • Posts: 2,626
  • Joined: 28 Aug 2017
  • Loc: NE Kansas

Posted 27 October 2020 - 09:59 AM

What is going on at the top of the turnbuckle, is that another eyelet end? Could maybe do something there.

#16 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 48,246
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 27 October 2020 - 02:10 PM

Clearance between mirror and bolts is a bit too tight.

The turnbuckle screws need to be shorter.

The mirror should not rest on the bumpers.

The three-cornered hat astigmatism is normally caused by clips, but it could be in something contacting the edge of the mirror.

Is the mirror glued in?


  • Jon Isaacs likes this

#17 jagg3d

jagg3d

    Lift Off

  • -----
  • Posts: 21
  • Joined: 03 Oct 2017

Posted 28 October 2020 - 10:55 AM

I have the same issue with my 8 Inch Orion SkyView Pro reflector. I swapped the secondary for a 1.83 Antares secondary to increase planetary performance. When focusing on Mars the image comes together from three separate discs and then splits again into separate disc as you pass through focus but the split changes direction 90 degrees.

 

I have the original orion mirror cell. I did tighten the clips since one wasn't very snug and the mirror could move. Could the clips be bending the mirror ever so slightly?

 

The contrast seems to be good otherwise, I zoomed in to 450x on the moon last night and it looked great, Mars I can see good detail on as well,  but the edges are diffused because of the splitting into three discs.

 

I've spent a lot of time collimating the scope and it all lines up nicely, just the issue at focus like TiSaph has.



#18 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 48,246
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 28 October 2020 - 11:05 AM

Yes, the pressure of the clips can and will distort the mirror.  There should be no pressure at all on the mirror from the clips.

It's generally recommended they don't touch the mirror at all.

But, that's in a dob, where the mirror doesn't slide back and forth.

In an equatorially-mounted scope, the only ways to avoid the distortion from clip pressure are:

--glue the mirror to the cell.  This, too, can cause distortion of the wavefront, but it's less of an issue on a small mirror or a very thick mirror.

--use a series of 6 small plastic screws at intervals around the perimeter of the mirror that touch the edge of the mirror but apply no pressure

to keep the otherwise loose mirror from sliding around.  They must press against the mirror at the front/rear center of gravity point to avoid astigmatism.


  • jagg3d likes this

#19 TiSaph

TiSaph

    Messenger

  • *****
  • topic starter
  • Posts: 490
  • Joined: 24 Jan 2017

Posted 28 October 2020 - 11:59 AM

I'm almost certain the issue comes from the mirror resting on the retaining posts.  I've read Mike Lockwood's article on mirror support quite a bit over the years, and two randomly placed hard posts are NOT what you want for edge support!  In a few days I'll pull the mirror out for cleaning and get shorter screws for the turnbuckles, so the cable sling takes the weight of the mirror and lifts it off the posts.  It's a perfect explanation - I just REALLY hope it's correct!


  • happylimpet, wrnchhead and jagg3d like this

#20 sixela

sixela

    Hubble

  • *****
  • Posts: 15,703
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 28 October 2020 - 04:55 PM

--use a series of 6 small plastic screws


3 are actually enough, of course.

#21 Starman1

Starman1

    Vendor (EyepiecesEtc.com)

  • *****
  • Vendors
  • Posts: 48,246
  • Joined: 23 Jun 2003
  • Loc: Los Angeles

Posted 28 October 2020 - 06:12 PM

3 are actually enough, of course.

To hold the mirror, yes. But then a rotation of the tube results in the mirror resting on one screw. With 6 screws, the mirror will almost always rest on at least 2 screws. This lowers the force on the screw and should calculate to a lower wavefront error from the edge support.
  • Jon Isaacs likes this

#22 sixela

sixela

    Hubble

  • *****
  • Posts: 15,703
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 29 October 2020 - 03:09 AM

Don, for each screw there is always one rotation at which the mirror is balanced on that one screw, when the COG is directly up from the screw. There are three more of those when you have six screws. When you're not in one of those metastable situations, the mirror always rests on two screws whether you have three or six (if you got everything just right the two lower screws, but if you have six screws very often not - think of making sure a six legged stool rests on all legs, which is a lot harder than for a three legged stool ).

I'll grant you that the weight is going to distribute over two screws faster if you have six screws and if you're lucky enough to get your six-legged stool correct (which is harder if you want no play at all), but that's at the expense of more potato-chipping even in the optimal case.

Supposing that all screws are on a circle that is just larger than the mirror (which means the mirror transfers weight on the two lower screws without actually moving and changing collimation) , four screws is actually better than three if you look at average deformation errors over all rotations (with the scope pointed at the horizon).

Six screws is actually worse, since even in the best case the mirror rests on two screws at +-30° from the vertical axis rather than +-45°. You're always oscillating between the worst case scenario (the weight over one screw) to a less bad one. With 6 screws you're oscillating faster (as I said the weight will redistribute onto two screws faster), but the worst case is the same and the best case is not as good.

Unsurprisingly, on the few GSOs on equatorial mounts where I fixed the edge support, I drilled four holes in the cell ring. Some of them were imaging scopes where the user almost always lets the focuser point up as much as possible during the imaging session (to have the imaging train over the tube as much as possible, rather than dangling at the side), and then it's not hard to put them in an optimal orientation (which, with four screws, happens to also be optimal when the user switches to visual observation with the focuser horizontal).

I confess, I haven't done the math for five screws.

Edited by sixela, 29 October 2020 - 04:01 AM.


#23 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 88,225
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 29 October 2020 - 07:31 AM

Supposing that all screws are on a circle that is just larger than the mirror (which means the mirror transfers weight on the two lower screws without actually moving and changing collimation) , four screws is actually better than three if you look at average deformation errors over all rotations (with the scope pointed at the horizon).

 

Six screws is actually worse, since even in the best case the mirror rests on two screws at +-30° from the vertical axis rather than +-45°. You're always oscillating between the worst case scenario (the weight over one screw) to a less bad one. With 6 screws you're oscillating faster (as I said the weight will redistribute onto two screws faster), but the worst case is the same and the best case is not as good.

 

 

I am still using the mirror cell from the 16 inch Starfinder Dob. It seems to use the same cell as the EQ version used.  It has either 8 or 9 screws, I don't remember which.  I have removed some of the upper ones for Dob use.  

 

In my mind, the assumption that the screws are in a circle that is just larger than the mirror is questionable.  I adjust the screws so they are all taking some load.  This means that one screw is never taking the entire load, it's always distributed. 

 

Jon



#24 sixela

sixela

    Hubble

  • *****
  • Posts: 15,703
  • Joined: 23 Dec 2004
  • Loc: Boechout, Belgium

Posted 29 October 2020 - 07:39 AM

If the screws are not in a circle, then if the screws are set with little play enough you'll tend to end up with the mirror always resting on the three screws that define the smallest circle.

BTW: if there's really almost no play, and screws are all set so that they just touch the mirror without pressing on it (you can do whatever you want with two screws, but all the next ones that are tightened must not press to hard on the mirror), if the mirror's edge is circular then by construction they are on a circle.

You never control what screws "take up some of the load". The geometry of the problem does, together with the mass of the telescope and gravity ;-). You definitely want the points arranged in such a way that load is always on the two lowest screws, otherwise your six or nine points are a figment of your imagination, rather than something real.

If you mean they all press on the mirror enough to cause friction, that's BAD (and yes, I'm shouting). Once the mirror cannot be freely rotated in its cell, you're pretty sure to introduce forces normal to the surface, which is far worse than any edge support without out-of-plane forces, even one with all the weight supported by one screw...

Edited by sixela, 29 October 2020 - 07:54 AM.


#25 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 88,225
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 29 October 2020 - 08:01 AM

You never control what screws "take up some of the load". The geometry of the problem does, together with the mass of the telescope and gravity ;-). You definitely want the points arranged in such a way that load is always on the two lowest screws, otherwise your six or nine points are a figment of your imagination, rather than something real.

 

It's worth remembering that I am using this cell in a Dob so the configuration is set and only the lower units have any load.  

 

In the EQ configuration, I have never set one of these cells up but the screws and clips are supported by aluminum angle brackets that probably have a tiny amount of flex so they maybe designed to equalize the loads between the screws.

 

One thing about the scope that's the topic of this thread, a 4 inch secondary on a curved vane spider would not seem to contribute to the current problem but it might be an issue when this gets sorted out.

 

Jon




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics