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

Who wants to make a Ross Null Lens?

  • Please log in to reply
174 replies to this topic

#151 Mark Harry

Mark Harry

    Cosmos

  • *****
  • Posts: 8,798
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 07 September 2020 - 11:26 AM

Just a guess, a concave fit, right?



#152 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,065
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 07 September 2020 - 12:10 PM

Just a guess, a concave fit, right?

No, a pcx lens "flat" side on a thin "flat." The surface is a thin, matt, black poster paper on top of a breadboard. It was made just to show how to make the fringes stand out compared to some other methods I have seen on Youtube.



#153 Arjan

Arjan

    Apollo

  • -----
  • Posts: 1,263
  • Joined: 21 Jan 2009
  • Loc: Netherlands

Posted 07 September 2020 - 12:31 PM

I found that using a diffused DPSS laser (532nm) gives very high contrast fringes:

 

300f8-39.jpg

 

Plenty good enough to use DFTFringe for analysis.


Edited by Arjan, 07 September 2020 - 12:32 PM.

  • brave_ulysses, MKV and Gleb1964 like this

#154 Mark Harry

Mark Harry

    Cosmos

  • *****
  • Posts: 8,798
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 07 September 2020 - 07:58 PM

No, a pcx lens "flat" side on a thin "flat." The surface is a thin, matt, black poster paper on top of a breadboard. It was made just to show how to make the fringes stand out compared to some other methods I have seen on Youtube.

I meant the FIT between the two surfaces is a CONCAVE fit, ie the edge of the piece is closer than the center. (It shows the direction to go if you want a truer flat.)



#155 duck

duck

    Sputnik

  • -----
  • Posts: 43
  • Joined: 11 Jun 2020
  • Loc: madera ca

Posted 18 September 2020 - 08:50 AM

Here's a question about RossNullXP...the lens focal length in the output is not = (N-1)*(1/R1 - 1/R2).  What is it?

 

This leads to another question.  Can a finished achromat of known fl be used as a Ross null lens? I presume YES, but need to raytrace (need all radii) to get the setup paramaters and null quality.

 

The curious (me and others) may ask why use an achromat.  The fl of the complete objective is longer than the crown element giving a better null.  If making a large primary mirror and guidescope, maybe make the guidescope objective and use it as a Ross null lens for the main scope's primary.



#156 Steve Dodds

Steve Dodds

    Owner - Nova Optical

  • *****
  • Vendors
  • Posts: 607
  • Joined: 02 Apr 2007
  • Loc: Utah

Posted 18 September 2020 - 09:10 AM

Here's a question about RossNullXP...the lens focal length in the output is not = (N-1)*(1/R1 - 1/R2).  What is it?

 

This leads to another question.  Can a finished achromat of known fl be used as a Ross null lens? I presume YES, but need to raytrace (need all radii) to get the setup paramaters and null quality.

 

The curious (me and others) may ask why use an achromat.  The fl of the complete objective is longer than the crown element giving a better null.  If making a large primary mirror and guidescope, maybe make the guidescope objective and use it as a Ross null lens for the main scope's primary.

A finished achromat won't work because it's designed to remove all spherical, however the front element should work, it doesn't need to be plano convex, but you do need to know the exact radius and glass type.


  • PrestonE likes this

#157 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,065
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 18 September 2020 - 11:23 PM

A finished achromat won't work because it's designed to remove all spherical, however the front element should work, it doesn't need to be plano convex, but you do need to know the exact radius and glass type.

Steve, an astronomical achromat is designed in monochrome light to be free of spherical aberration for objects at infinity. This means that the first conjugate (source) distance is at infinity and the second (image) distance is at the paraxial focus for infinity. If you change one conjugate everything changes, especially the paraxial focus and spherical aberration. 

 

The amount of spherical aberration needed to null the aberration of the test optic is a function of where the null optics conjugates are placed and not the type of optics itself. That's why you can use mirrors instead of lenses for null tests, such as the Maksutov null test, and that's why there is no reason  an achromat would not work as a Ross null lens. 

 

Mladen

 

PS

 

I did a raytrace analysis and the results confirm that an achromat will do just fine as a Ross null lens. 

 

ross null achro.png


Edited by MKV, 19 September 2020 - 03:51 PM.


#158 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,065
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 18 September 2020 - 11:31 PM

Here's a question about RossNullXP...the lens focal length in the output is not = (N-1)*(1/R1 - 1/R2).  What is it?

It is, but why do you think it isn't?



#159 duck

duck

    Sputnik

  • -----
  • Posts: 43
  • Joined: 11 Jun 2020
  • Loc: madera ca

Posted 19 September 2020 - 09:54 AM

I goofed.  Thanks Mladen!



#160 duck

duck

    Sputnik

  • -----
  • Posts: 43
  • Joined: 11 Jun 2020
  • Loc: madera ca

Posted 30 September 2020 - 11:43 AM

trying to get a quote on a 6" diameter BK7 blank, generated to 80" ROC biconvex, with test plate and lap substrates.

 

Doesn't seem like any glass suppliers are interested.  No quotes forthcoming.

 

So "WHO WANT TO MAKE A ROSS NULL LENS" can morph into "WHO CAN GET GLASS TO MAKE A ROSS NULL LENS"



#161 GLS

GLS

    Explorer 1

  • -----
  • topic starter
  • Posts: 56
  • Joined: 01 Aug 2020

Posted 30 September 2020 - 04:55 PM

trying to get a quote on a 6" diameter BK7 blank, generated to 80" ROC biconvex, with test plate and lap substrates.

 

Doesn't seem like any glass suppliers are interested.  No quotes forthcoming.

 

So "WHO WANT TO MAKE A ROSS NULL LENS" can morph into "WHO CAN GET GLASS TO MAKE A ROSS NULL LENS"

Have you considered ordering the glass edged with parallel surfaces without the curve generation?  80" ROC is not much glass removal to reach the required sags and you will have made matching test plates in the process.  Also you might have better luck with snail mail rather than email. I bet their email client is covered up with spam if that's how your inquiries have been made.  Your inquiries might not have been seen. Gil


Edited by GLS, 30 September 2020 - 04:56 PM.

  • MKV likes this

#162 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,065
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 30 September 2020 - 05:53 PM

trying to get a quote on a 6" diameter BK7 blank, generated to 80" ROC biconvex, with test plate and lap substrates.

 

Doesn't seem like any glass suppliers are interested.  No quotes forthcoming.

Why a 6" lens for a Ross null?

Why 80" RoC?

Why biconvex?


  • PrestonE likes this

#163 duck

duck

    Sputnik

  • -----
  • Posts: 43
  • Joined: 11 Jun 2020
  • Loc: madera ca

Posted 01 October 2020 - 09:44 AM

The quotation requests were sent by email and I got an initial reply.  Afterward nothing.  I asked for all the pieces to be generated, but I can drop that if that's an issue.

 

Mladen...The null it generates for a 24" f/4 paraboloid is 1/20 wave.  Mirror/lens spacing error of > 0.2" to drop the null to 1/8 wave.



#164 BGRE

BGRE

    Gemini

  • *****
  • Posts: 3,005
  • Joined: 21 Mar 2016
  • Loc: New Zealand

Posted 01 October 2020 - 04:53 PM

An equibiconvex lens only requires a single concave test plate keeping costs down.



#165 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,065
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 02 October 2020 - 02:22 AM

The null it generates for a 24" f/4 paraboloid is 1/20 wave.  Mirror/lens spacing error of > 0.2" to drop the null to 1/8 wave.

Thanks. Can you share with us how you arrived at those numbers? 

 

The reason I asked is that, for a bcx lens of 80 inch roc, the focal length is about 80 inches which makes for a very long test setup -- approximately 4.5 meters (about 178 inches). 



#166 duck

duck

    Sputnik

  • -----
  • Posts: 43
  • Joined: 11 Jun 2020
  • Loc: madera ca

Posted 02 October 2020 - 12:03 PM

Mladen,

 

This is the printout from RossNullXP

 

Using these Mirror Parameters

Diameter =  -> 24"
Mirror Radius of Curvature = -> 192"
Mirror Conic Constant =  -> -1 (Parabola)

Lens Parameters

Sign Convention: With mirror on right, ) = -ve, ( = +ve
Image Side Radius =  -> 80" Convex
Mirror Side Radius =  -> -80" Convex
Thickness =  -> 0.6"
Diameter =  -> 6"
Refractive Index =  -> 1.52077  = BK7 @ 510nm
Lens Focal Length =  -> 76.80934"

Results:
Object to Lens Distance           = 27.5393"
Lens to Mirror Distance           = 148.4196"
Working Diameter of Lens         = 5.4995"
70% Zone Spherical Aberration    = -0.0015"
Marginal Spherical Aberration    = 0.0001"
Quarter-wave Tolerence           = 0.0081"
Maximum Optical Path Difference  = 0.0493 Waves

*** Finding 1/8 wave OPD values for positive error in mirror - lens spacing

Results for lens mirror spacing error of +0.232"
Object to Lens Distance           = 27.4465"
Lens to Mirror Distance           = 148.6516"
Working Diameter of Lens         = 5.4708"
70% Zone Spherical Aberration    = -0.0035"
Marginal Spherical Aberration    = -0.0041"
Quarter-wave Tolerence           = 0.0081"
Maximum Optical Path Difference  = 0.1255 Waves

*** Finding 1/8 wave OPD values for positive error in mirror - lens spacing

Results for lens mirror spacing error of -0.218"
Object to Lens Distance           = 27.6263"
Lens to Mirror Distance           = 148.2016"
Working Diameter of Lens         = 5.527"
70% Zone Spherical Aberration    = 0.0004"
Marginal Spherical Aberration    = 0.004"
Quarter-wave Tolerence           = 0.0081"
Maximum Optical Path Difference  = 0.125 Waves



#167 MKV

MKV

    Cosmos

  • *****
  • Posts: 9,065
  • Joined: 20 Jan 2011
  • Loc: Southeast, USA

Posted 03 October 2020 - 03:40 AM

This is the printout from RossNullXP

Thanks, that's important to make sure everyone is using the same numbers. RossNullXP defaults to the industry standard helium-neon (HeNe) laser test signal wavelength of 632.8 nm which for BK7 gives a test refractive index of 1.51509. Apparently you changed the test wavelentgth to 510 nm and a working BK7 refractive index of 1.52007. If you hadn't posted your layout that would have caused some confusion and departure from your settings for anyone trying to sync your results. So, thanks again for providing the numbers for the sake of clarity.

 

Making a biconvex lens makes sense, as BGRE explained, but that's not necessarily the only thing that needs to be addressed. A 6" lens with two gently curved convex surfaces and a concave tool-test plate is a logical solution for testing and achieving precise radii of curvature and spherical figure on both sides. One thing to worry about is the working diameter (WD) of the lens. It should not exceed 90% of the clear aperture (CA).  

 

The problem with a large WD relative to CA, such as in your case, is that the lens will have to be finished to precision tolerances over more than 90% of its physical diameter.  I think you may wish a slightly bigger lens with the same parameters as the 6", and possibly slightly thicker, such as CA 8" and central thickness 3/4" or thicker. An 8" lens will still have a WD of about 5.5 " which is only 69% of the CA. This should somewhat relax the fabrication tolerance requirements

 

The other issue for a large precision lens to be considered is the inhomogeneity of the substrate. A highly homogenous substrate (which is normally required of precision lenses) equals high price simply because the company that will sell a highly homogenous blank has to cut it out of a much larger slab.

 

Here's an example of an 8" ross null lens with your specs, and a 1" central thickness for comparison.

 

Diameter = -> 24"

Mirror Radius of Curvature = -> 192"
Mirror Conic Constant =  -> -1 (Parabola)

 

Lens Parameters

 

Sign Convention: With mirror on right, ) = -ve, ( = +ve
 

Image Side Radius  = -> 80" Convex
Mirror Side Radius = -> -80" Convex
Thickness =  -> 1"
Diameter =  -> 8"
Refractive Index = -> 1.52077  = BK7 @ 510nm
Lens Focal Length =  -> 76.80934"

 

Results:
 

Object to Lens Distance          = 27.4737"
Lens to Mirror Distance          = 148.1463"
Working Diameter of Lens         = 5.5337"
70% Zone Spherical Aberration    = -0.0015"
Marginal Spherical Aberration    = 0"
Quarter-wave Tolerance           = 0.0081"
Maximum Optical Path Difference  = 0.0481 Waves

 

*** Finding 1/8 wave OPD values for positive error in mirror - lens spacing

 

Results for lens mirror spacing error of +0.228"

 

Object to Lens Distance          = 27.3826"
Lens to Mirror Distance          = 148.3743"
Working Diameter of Lens         = 5.5063"
70% Zone Spherical Aberration    = -0.0035"
Marginal Spherical Aberration    = -0.0041"
Quarter-wave Tolerance           = 0.0081"
Maximum Optical Path Difference  = 0.1254 Waves

 

*** Finding 1/8 wave OPD values for positive error in mirror - lens spacing

 

Results for lens mirror spacing error of -0.223"
 

Object to Lens Distance          = 27.5625"
Lens to Mirror Distance          = 147.9233"
Working Diameter of Lens         = 5.5626"
70% Zone Spherical Aberration    = 0.0004"
Marginal Spherical Aberration    = 0.004"
Quarter-wave Tolerance           = 0.008"
Maximum Optical Path Difference  = 0.1255 Waves

 

As you can see the layout hasn't changed much, and yes, you still have very generous tolerances on the lens-to-mirror separation, a whopping 1/5 inch, more or less.



#168 duck

duck

    Sputnik

  • -----
  • Posts: 43
  • Joined: 11 Jun 2020
  • Loc: madera ca

Posted 05 October 2020 - 04:32 PM

Thanks Mladen.  There are no responses to my inquiry from two reputable glass suppliers.  I believe when I stated "1 item" the interest in supplying "1 item" dropped to nil.



#169 Pinbout

Pinbout

    ISS

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

Posted 06 October 2020 - 07:40 PM

Thanks Mladen.  There are no responses to my inquiry from two reputable glass suppliers.  I believe when I stated "1 item" the interest in supplying "1 item" dropped to nil.

Who did you ask 



#170 duck

duck

    Sputnik

  • -----
  • Posts: 43
  • Joined: 11 Jun 2020
  • Loc: madera ca

Posted 07 October 2020 - 09:01 AM

United Lens and Advanced Glass Industries



#171 Pinbout

Pinbout

    ISS

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

Posted 07 October 2020 - 09:28 AM

United Lens and Advanced Glass Industries

try

 

https://www.bmvoptical.com/



#172 Oregon-raybender

Oregon-raybender

    Optical Research Engineer

  • *****
  • Vendors
  • Posts: 856
  • Joined: 13 May 2010
  • Loc: Oregon, South Western Coast

Posted 07 October 2020 - 12:36 PM

I have some Schlieren grade high quality BK7  used for the TV projection. I just have to find it. It's flat, needs to cored to size and ground to a radius.  It would be perfect for Ross lens.

 

I am not surprised at no response. low level request. Sorry, they get requests like this all the time. I did, I would try

Glass Fab in Rochester.

 

Contact me via PM

 

Starry Nightswaytogo.gif

 

http://www.glassfab.com/



#173 JohnH

JohnH

    Gemini

  • ****-
  • Posts: 3,269
  • Joined: 04 Oct 2005
  • Loc: Squamish BC Moved!!!!!

Posted 07 October 2020 - 10:03 PM

United Lens and Advanced Glass Industries

That was your mistake



#174 Steve Dodds

Steve Dodds

    Owner - Nova Optical

  • *****
  • Vendors
  • Posts: 607
  • Joined: 02 Apr 2007
  • Loc: Utah

Posted 08 October 2020 - 02:36 PM

I have never had any problem ordering 1 or 2 small parts from United Lens.  You just need to call and talk to a person, don't just send a email.



#175 Oregon-raybender

Oregon-raybender

    Optical Research Engineer

  • *****
  • Vendors
  • Posts: 856
  • Joined: 13 May 2010
  • Loc: Oregon, South Western Coast

Posted 09 October 2020 - 04:00 PM

United Lens will assist. They need to know you are really interested in purchasing. Like Steve, I was one of their customers for decades.

Like I said, I received several e-mails a week with simple (but complex) requests. "How much for a 4 inch achromat lens, just need one, must be 1/10 wave, color free, and need it in 3 weeks, Oh yes the cost should be <$100!" Sorry I gave great customer service, but how does one let someone down easy, I often did with a quick reply and an explanation.

 

Still looking for my glass.

 

Starry Nightswaytogo.gif




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