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

First mirror, 8" F/3.5-ish

  • Please log in to reply
199 replies to this topic

#151 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 16 September 2020 - 12:25 PM

Okay, since my last post I made a new Foucault plate for my phone (slit + knife) - it was way easier than a pinhole.
Things continue to improve according to ronchis. I dilute cerium more and more to avoid going too fast or too rough.
A strict measuring session will be the next thing.

Mirror cell is in the scope, I’m finishing the spider and will have a look at Mars with the mirror !
Also found a nice 133lpi ronchi eyepiece secondhand for more tests.

Then another story will be the bath construction and continuing the mirror. It’s endless but joyful 😀
  • davidc135 likes this

#152 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 17 September 2020 - 02:38 AM

This morning’s short work & ronchi.

What’s new is the artifact showing in ronchis at the edge : can someone confirm the edge is coming back ?

E61A7DFF-3DE2-44DD-B12D-B8BFABC6A7DE.jpeg

3D28CD69-45CB-4F49-A96B-7E6E50E404A6.jpeg

I’m mainly working to smooth the bad curving at the 70% zone.


  • PrestonE likes this

#153 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 19 September 2020 - 07:37 AM

Hello,

I'm a bit embarrassed, but my laptop has been stolen and I'm unable to run Foucault Unmasked on my last image set. Other computers I have access to at home or my office cannot run it (Mac OS everything... and wine doesn't run it anymore.)

 

If someone has the time to try to run it (without fiddling with my pictures, if they're rejected, I'll find another way) on one of those datasets it'd be really nice.

 

Dataset 1 : 32 pictures, low exposure, 0.25mm increments, ROC 1445mm, lightsource moving, 196mm clear aperture.

Download link : https://github.com/c... increments.zip (8.1MB)

 

Dataset 2 : 16 pictures, higher exposure, 0.5mm increments, ROC 1445mm, lightsource moving, 196mm clear aperture.

Download link : https://github.com/c... increments.zip (7.5MB)

 

Clear skies (despite the smoke...) !

Attached Thumbnails

  • Capture d’écran 2020-09-19   14.31.15.png
  • Capture d’écran 2020-09-19   14.30.43.png

Edited by chantepierre, 19 September 2020 - 07:38 AM.


#154 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 21 September 2020 - 03:48 AM

Well here's what FU has to say :

 

First dataset (0.25mm increments, low exposure)

Screen Shot 2020-09-21 at 10.40.26.png

 

Second dataset (0.5mm increments, brighter pictures)

Screen Shot 2020-09-21 at 10.45.11.png

 

From a consistency perspective with the last ronchis I'd feel the first one is more accurate while being inaccurate. I also see that there'll be a lot of "feeling" until I'm able to star test & take interferograms.

Edge still lacks behind a lot. Yesterday I trained myself to use the ronchi eyepiece on another scope. It'll prove itself useful really soon.


Edited by chantepierre, 21 September 2020 - 03:49 AM.


#155 ckh

ckh

    Surveyor 1

  • -----
  • Posts: 1,994
  • Joined: 21 Feb 2015
  • Loc: Arizona

Posted 21 September 2020 - 12:21 PM

The edge in the first one is the result of the last few zones with a cubic spline extrapolation.  It's not necessarily real. On the second test there are not enough outer zones.so everything after zone 7 is extrapolation.

 

At this rough stage Ronchi is a more useful method because it's much faster.  When you get closer, FU will give you more detail.

 

Carl


  • chantepierre likes this

#156 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 21 September 2020 - 04:37 PM

Hello Carl,
Thank you for your insight and for your software.
It’s really useful.
My pictures are to blame, most of the time FU doesn’t find the contour of a circle. And indeed it’s really hard to discern.

Could you give information on the edge detection algorithm ? How can we optimize pictures to help FU ?

Thank you,
Lucas

#157 ckh

ckh

    Surveyor 1

  • -----
  • Posts: 1,994
  • Joined: 21 Feb 2015
  • Loc: Arizona

Posted 21 September 2020 - 07:42 PM

Make sure the background is as dark as possible and use a bright light source. Also make sure your exposures are bright. You can even overexpose the brightest areas a bit without any harm.

 

The algorithm starts at the top, bottom, left and right of the frame and scans toward the opposition side until it see a significant change in brightness. It does some extra stuff to find the center the diffraction ring, if present.

 

After finding those edges, it checks that the width and height of the mirror in pixels are about that same. If not, but there are three edges consistent with a circle, it will just compute the missing edge.

 

Carl



#158 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 22 September 2020 - 01:25 AM

That’s instructive, thank you !
I’ll try black paper or felt behind the mirror and will not fear exposing more. Most of my picture series were really dim to have smooth gradients.
Have a nice day
Lucas

#159 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 22 September 2020 - 04:34 PM

Okay, today’s lessons : I did not get visible work done because I didn’t put enough force while polishing (and my lap is quite lightweight)

I tried short sessions while pushing harder on the lap (not faster) and figure visibly changed with ronchis after each session.
1 session of Danny’s TDE fixing strokes (did not make them short enough).
2 sessions of TOT edge work.
After the TOT strokes with about 10lbs pressure (measured by feeling) the start of a diffraction ring appeared in following Foucault tests. That’s what I was aiming for tonight since I couldn’t do numerical analysis yet.

Knowing that I can make the figure change again is nice. The lap is in good condition and passes each night and day on the mirror. I trim the channels regularly and reshape them with the soldering iron every 3 days.

Also got the time to print the secondary holder and spider hub :-).

#160 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 24 September 2020 - 01:53 AM

Hello,
Here are pictures of the lap and a ronchi outside ROC.
There are « two competing curves » and I’m working TOT to get the edge back and push its curve towards center. It’s progressing.
The edge was so much behind that I did not think of anything else.

Channels are cut through with my soldering iron. Picture was taken just after brushing.
My plan is to continue working TOT to get an even shape on the whole mirror and have a good edge. Then we’ll see correction-wise, even if that seems to be a step backwards.

BBC2AD38-9F15-4098-B83C-D7021366C482.jpeg
71C331E3-DE1D-444E-8116-151060EB08A4.jpeg

Oh and the spider is done !
F1835A81-95BC-4032-95EE-93FAE2EF555F.jpeg

 

If someone reads this thread, I'd like feedback on the plan of pushing the curve from the edge to the center.
Have a nice day.

Attached Thumbnails

  • F1835A81-95BC-4032-95EE-93FAE2EF555F.jpeg

Edited by chantepierre, 24 September 2020 - 08:09 AM.

  • Augustus likes this

#161 Augustus

Augustus

    Fly Me To The Moon

  • *****
  • Posts: 10,043
  • Joined: 26 Dec 2015
  • Loc: Stamford, Connecticut

Posted 24 September 2020 - 12:06 PM

I might recommend moving the spring to be above the spider hub instead of between it and the secondary holder.


  • chantepierre likes this

#162 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 24 September 2020 - 03:22 PM

Thanks Zane !
Could you explain why ? I put it there to prevent the thing from unscrewing itself with vibrations in the car but I can certainly move it !

#163 mic1970

mic1970

    Apollo

  • -----
  • Posts: 1,225
  • Joined: 09 Aug 2018
  • Loc: St. Louis Metro Area

Posted 24 September 2020 - 03:43 PM

popcorn.gif popcorn.gif popcorn.gif popcorn.gif popcorn.gif



#164 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 24 September 2020 - 03:48 PM

Is the presence of a spring in a spider holder that controversial ? I didn’t think so :-D

#165 Augustus

Augustus

    Fly Me To The Moon

  • *****
  • Posts: 10,043
  • Joined: 26 Dec 2015
  • Loc: Stamford, Connecticut

Posted 24 September 2020 - 04:20 PM

Thanks Zane !
Could you explain why ? I put it there to prevent the thing from unscrewing itself with vibrations in the car but I can certainly move it !

I honestly forget how it was explained to me, but an ATM I met talked about something related to tension/loads, and it's how the spring is placed on my Destiny curved vane spiders.


  • chantepierre likes this

#166 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 25 September 2020 - 01:46 AM

Thank you. I’ll get a shorter one to put under the top screw instead of the one I’ve been using.

Secondary is glued, this weekend I should have first light if the weather cooperates. Then I’ll feel visually how far I am with the mirror 😬

#167 macleod

macleod

    Viking 1

  • -----
  • Posts: 886
  • Joined: 21 Sep 2014
  • Loc: Napier, New Zealand, 4110

Posted 25 September 2020 - 02:37 AM

Chantpierre - Augustus is correct.Put a small strong spring on TOP of you spider hub, with a washer between spring and wing nutTighten down your three adjusting screws until you feel the tension of the spring resisting them. This way the spring is "pulling" on your central bolt holding the digonal support, which then pulls  it back against your collimating screws. This way there will be stability.Make sure your 3x screws do not "dig holes" in the back of your diagonal support if it is wood or fibre. Place either a large fender washer or better a disc cut from those liter milk bottles on the back of holder for the screw tips to work on. Not sure if you have the same bulk milk containers as us - its a tough slightly bubbly almost transparent plastic smooth on inside.  Best wishes, Macleod                                                                                                                                                                                   


  • Augustus and chantepierre like this

#168 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 25 September 2020 - 03:34 AM

Hey Mac !
Thanks for the explanation. We don’t have those milk jugs but we have other things made of HDPE !
The support itself is PETG which is not far. I’ll cut a plastic ring and move the spring asap.

Anyone has a comment on the mirror itself ?
Have a nice day
Lucas

#169 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 27 September 2020 - 06:03 AM

Hello,

continued my work from the edge - it seems that now FU will have less trouble with my pictures.

34C4A23A-8045-4415-9A28-BE3124DE21D7.jpeg

 

Also cut the focuser hole and measured everything to install the cell and secondary. First light was ambitious but it’s getting closer.

 

have a nice day !


  • PrestonE likes this

#170 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 27 September 2020 - 06:58 PM

Nice to see you fixed the tde. But as you saw, these fast mirrors like to snap back to spherical, especially in the outer zones.

 

id stick with ronchi till you get a nice flow. The start taking measurements.

 

A medium narrow, but long w’s will correct the outer zones. Of course MoT


Edited by Pinbout, 27 September 2020 - 07:01 PM.


#171 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 28 September 2020 - 01:23 AM

Hey Danny, nice to hear from you !

As you said, this mirror wants to go back to a sphere... It’s not worrying me too much, that gives me time to take better control. I’ll follow with your advice.

Yesterday it was too cold, contact was weird, and I think I introduced some astig. Ronchis do not clock while going through ROC but Foucault show a yinyang pattern. Could that be bad tester alignment ? It stays at the same angle when I rotate the mirror.

Have a nice day
Lucas

#172 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 30 September 2020 - 03:08 PM

Took the time to finish the mirror cell (print the nylon support pads & clips) and the spider & secondary holder...

And got first light tonight ! I watched the moon, boy it was that much blurry ! Now that the scope is mostly complete I’ll be able to star test.

I feared having the focuser hole too close, but it was a bit too far and I just had to push the mirror cell towards the front.

Have clear skies
Lucas
  • LarsMalmgren likes this

#173 luxo II

luxo II

    Vanguard

  • -----
  • Posts: 2,490
  • Joined: 13 Jan 2017
  • Loc: Sydney, Australia

Posted 01 October 2020 - 01:22 AM

Get rid of the regular pattern on the tool.

 

Back when I was making my own mirrors, on my first attempt I made a regular grid like you have and achieved a similar effect. Big mistake. Stripped the pitch off and started again.

 

To put channels in the pitch I softened the pitch (in a tub of warm water) then laid thick string over it in a crazy random pattern, some plastic wrap over that, and pressed it with the mirror. The random pattern of grooves produced by the string made sure there was no periodic lumpy effect like that on your mirror, and the channels lasted quite a long time.


Edited by luxo II, 01 October 2020 - 01:25 AM.


#174 chantepierre

chantepierre

    Vostok 1

  • -----
  • topic starter
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 01 October 2020 - 02:50 AM

Hello luxo,
Thank you for the advice
For now I prefer to follow dogbiscuit’s advice regarding laps (even if I’m too symmetrical) because it’s quite straightforward and can be evaluated... with a nonstandard lap I couldn’t know if the lap or my strokes were at fault.

Danny, I did 5 mins of 1/3 wide 1/2 long Ws MOT and lines started to flow a bit from center to edge. I’ll continue doing that unless advice is given to stop.

 

In order, lap (before work) / inside / outside

 

1F6AB37D-470A-4F0F-BD65-9A97C6860673.jpeg A0A8EADA-7901-47BB-A4F9-EE8A88710F98.jpeg 270F8DAA-7FD9-4308-980D-010B8FE4BA75.jpeg

 

clear skies


Edited by chantepierre, 01 October 2020 - 02:53 AM.


#175 luxo II

luxo II

    Vanguard

  • -----
  • Posts: 2,490
  • Joined: 13 Jan 2017
  • Loc: Sydney, Australia

Posted 01 October 2020 - 04:02 AM

Eew... And just for fun central crater and TDE....

Take it back to a good sphere and start again.


Edited by luxo II, 01 October 2020 - 04:04 AM.



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