Jump to content


Photo

Secondary Size vs Reflectivity

  • Please log in to reply
25 replies to this topic

#1 Mark Peterman

Mark Peterman

    Vostok 1

  • -----
  • Posts: 105
  • Joined: 07 Aug 2012
  • Loc: Texas

Posted 04 February 2013 - 10:26 AM

I should be receiving my new Protostar spider and secondary holder this month which means I need to make a decision on which secondary to install.

The scope is a 12" f/4.9 and the primary is coated with standard aluminum/SiO2 overcoat giving ~90% reflectivity (%R).

The choice of secondary mirrors that I have are:

1 - The stock 2.75" that is probably BK-7 glass, an unknown wave rating and coated with a %R of ~92-93%.

2 - Antares Optics 2.6" made of Pyrex, better than 1/20 wave and coated with a %R of ~96-98%.

When considering the diameter of the 100% illumination area, the two largest field stop sizes that I have are 1.52" (35mm Panoptic), and 1.08" (20mm Nagler).

The 2.75 will give a larger 100% illuminated area but a lower overall reflectivity while the 2.6 would have the opposite results (smaller 100% area, but higher reflectivity).

I know this is close and most might say that the difference is so small that I will not be able to tell the difference. However, I am thinking the higher reflectivity of the 2.6 might show up in other ways that the larger 100% illumination area of the 2.75 would not.

Thoughts or opinions?

Thanks,

Mark

#2 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 04 February 2013 - 10:42 AM

My suggestion:

Don't worry about reflectivity or field illumination, worry about the quality of the mirror. The quality of the secondary is nearly as important as the quality of the primary so a high quality secondary can make a significant difference in the scope's performance.

Jon

#3 GlennLeDrew  Happy Birthday!

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10643
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 04 February 2013 - 10:44 AM

You will not see the difference in reflectivity.

Your primary consideration now should be the balance between secondary obstruction size and the field illumination.

For visual purposes, even for 2"' eyepieces, a circle of full illumination 0.5-0 75"' is quite adequate. An illumination drop-off to 50% at the field edge is hard for even experienced observers to perceive.

#4 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1900
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 04 February 2013 - 02:30 PM

... The quality of the secondary is nearly (emphasis added) as important as the quality of the primary...


Since the imaged is "processed" by both mirrors why would the secondary not be equally as important?

#5 killdabuddha

killdabuddha

    Apollo

  • *****
  • Posts: 1137
  • Joined: 26 Aug 2011

Posted 04 February 2013 - 02:44 PM

Secondary quality is a point of contention and confusion. You'll even find opticians flustered

http://www.rfroyce.com/diagsize.htm

There have been threads here, and the consensus resolved to a 1/20th wave secondary as bein more than adequate, and hence the safe bet. As for, "Since the imaged is "processed" by both mirrors why would the secondary not be equally as important?" I thought the same thing. Others said no way. When I asked our optician, he said he wouldn't worry too much about secondary quality, yet you'll also find Royce (above link) sayin the same thing that you've suggested (which I quoted). So I think that we all tend to err on the side of caution.

#6 Mark Peterman

Mark Peterman

    Vostok 1

  • -----
  • Posts: 105
  • Joined: 07 Aug 2012
  • Loc: Texas

Posted 04 February 2013 - 03:41 PM

Your primary consideration consideration now should be the balance between secondary obstruction size and the field illumination.


It sounds like the known quality of the secondary should be a big part of the reason for going with the 2.6".

The difference in obstruction is only 1%, so really a non-factor in that regard. Here is how the field illumination looks with the 35mm Panoptic.

Posted Image

Here's the rest of the specs using the raw numbers:

Posted Image

* Before someone points it out, I'm not taking the beveled edge or the lip on the holder into consideration. Just using the raw MA size on each mirror to keep it apples to apples.

#7 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 05 February 2013 - 04:17 PM

... The quality of the secondary is nearly (emphasis added) as important as the quality of the primary...


Since the imaged is "processed" by both mirrors why would the secondary not be equally as important?


The reason I used "nearly" as important is that the primary mirror reflects the light back on itself so the surface error is doubled whereas the secondary reflects it at a 45 degree angle so the effect of a surface error is reduced, there's a square root of two in there somewhere.

Jon

#8 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1900
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 05 February 2013 - 09:44 PM

Assuming then that the wave front errors combine by RMS summation, a secondary that matched the error of the primary would make the the overall error only 22.5% worse.

This suggests that having equal accuracy for the secondary as the primary should serve pretty well.

Of course the "accuracy cost" in a secondary is much lower to get a 1/15 or 1/20 wave mirror so there isn't much cost pressure in a scope budget to <i>not</i> go better than your primary.

If it is twice as good then the error increase is only 6.1%.

Antares offers secondaries down to 1/30 wave. You would need to have a really good primary to benefit significantly from that kind of accuracy.

#9 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1150
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 06 February 2013 - 01:29 AM

why do the graphs of field illimination show a central flat drop of 0.06 mag?

#10 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 06 February 2013 - 05:24 AM

why do the graphs of field illimination show a central flat drop of 0.06 mag?


Area lost because of the secondary obstruction.

Jon

#11 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 06 February 2013 - 06:10 AM

Assuming then that the wave front errors combine by RMS summation, a secondary that matched the error of the primary would make the the overall error only 22.5% worse.


This what Robert Royce says:

"To calculate the impact of errors on the final wavefront one method is to take the error of the primary mirror plus the error of the secondary mirror (if its surface error its times 2) times the cosine of the angle of incidence of the diagonal, in this case 45 degrees or .7071. So, assuming a 1/10 wave primary mirror and a 1/10 wave secondary mirror (wave front accuracy, not surface) we have the following: .1 + .1 * .7071) = .171 or Lambda/5.8. But some people think this is much too severe an application of arithmetic. I like the product of the complement method (which I developed independently) but it may only be foolishness. It goes like this: assume you have a .1 wave mirror and .1 wave diagonal. .9 is the complement of .1 or 1 - .1 = .9 so the entire formula is 1- (.9 * .9) * .7071 = .134 or Lambda/7.44 It looks nicer. Then there is the root sum squared method. This is valid for systems having many elements. This method is be as follows: ((.1^2 + .1^2)^.5) * .7071 = .141 or Lambda/7.07. Confusing?"

Sizing your diagonal Robert Royce

In any event, he includes the 45 degree cosine factor and concludes that the wavefront error of two 1/10 wave front mirrors will be somewhere around 1/7 wave. He does not seem to use the RMS summing.

My thinking is pretty simple:

You go to the trouble of making a large 1/10 mirror. That is a major investment. A high quality secondary is far less expensive and yet has a similar effect on the image so it's cost effective to invest in the best possible secondary.

As similar situation is recoating a mirror set. The reflectivity of the primary and the secondary are of equal importance in the brightness of the image but they do not cost the same. If you recoat the secondary with high reflectivity and the primary with standard coatings it will be much less expensive than recoating both with high reflectively coats and you gain about 50% of the reflectivity advantage...

Jon

#12 dpwoos

dpwoos

    Apollo

  • *****
  • Posts: 1390
  • Joined: 18 Oct 2006
  • Loc: United States

Posted 06 February 2013 - 07:52 AM

Antares offers secondaries down to 1/30 wave. You would need to have a really good primary to benefit significantly from that kind of accuracy.


In fact, it has been suggested to me (by someone who has reason to know the inside story) that you would also need to have a good imagination to believe the 1/30 wave number to begin with. I think that Protostar guarantees 1/10 wave not because the numbers coming out of the interferometer aren't ever better, but rather that these better numbers don't contain any information.

#13 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 07 February 2013 - 08:00 AM

Antares offers secondaries down to 1/30 wave. You would need to have a really good primary to benefit significantly from that kind of accuracy.


In fact, it has been suggested to me (by someone who has reason to know the inside story) that you would also need to have a good imagination to believe the 1/30 wave number to begin with. I think that Protostar guarantees 1/10 wave not because the numbers coming out of the interferometer aren't ever better, but rather that these better numbers don't contain any information.


Optical flats for Fabry–Pérot interferometers need to be much flatter than lambda/30... it is possible.

Jon

#14 dpwoos

dpwoos

    Apollo

  • *****
  • Posts: 1390
  • Joined: 18 Oct 2006
  • Loc: United States

Posted 07 February 2013 - 09:55 AM

Antares offers secondaries down to 1/30 wave. You would need to have a really good primary to benefit significantly from that kind of accuracy.


In fact, it has been suggested to me (by someone who has reason to know the inside story) that you would also need to have a good imagination to believe the 1/30 wave number to begin with. I think that Protostar guarantees 1/10 wave not because the numbers coming out of the interferometer aren't ever better, but rather that these better numbers don't contain any information.


Optical flats for Fabry–Pérot interferometers need to be much flatter than lambda/30... it is possible.

Jon


Never said it wasn't possible, but rather that (I was told) that it isn't possible in this case.

#15 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 07 February 2013 - 10:56 PM

Never said it wasn't possible, but rather that (I was told) that it isn't possible in this case.



I suspect that there are folks here like Glenn, Mike Jones, Ed Jones, David and others who probably know and if it is possible, some have probably made lambda/30 flats.

Jon

#16 dpwoos

dpwoos

    Apollo

  • *****
  • Posts: 1390
  • Joined: 18 Oct 2006
  • Loc: United States

Posted 08 February 2013 - 07:58 AM

I don't think that the problem is production - some number out of every batch might very well be 1/30 wave. Rather, the problem is creating and maintaining an interferometric testing setup that can find them with certainty. If you accept, based on provided interferometry, that a flat advertised as 1/30 wave is so, then that is up to you. I don't, and so have (and will continue to) purchase secondaries with what I think are more realistic and trustworthy numbers.

#17 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 08 February 2013 - 11:30 AM

I don't think that the problem is production - some number out of every batch might very well be 1/30 wave. Rather, the problem is creating and maintaining an interferometric testing setup that can find them with certainty. If you accept, based on provided interferometry, that a flat advertised as 1/30 wave is so, then that is up to you. I don't, and so have (and will continue to) purchase secondaries with what I think are more realistic and trustworthy numbers.


I am not buying any flats. But I would be interested in an explanation of why you think it is it not possible to manufacture and interferometrically test flats to 1/20 or 1/30th wave.

Jon

#18 dpwoos

dpwoos

    Apollo

  • *****
  • Posts: 1390
  • Joined: 18 Oct 2006
  • Loc: United States

Posted 08 February 2013 - 12:08 PM

As I have written several times, I don't think it is impossible, but rather that in practice claims of 1/20 or 1/30 wave secondaries as sold to the amateur astro community are not (to me) credible. The reason I think this is because someone I trust in the optical fabrication business with knowledge of the testing methodology and equipment told me so. I also know that it is very easy to confuse information and noise - even bona fide scientists/researchers struggle with this. I also know that businesses can easily find themselves in a spiraling race, where facts are stretched in an effort to distinguish themselves from the competition. All of this makes me very skeptical of these claims, and I am happy to stick with more modest numbers (e.g. 1/10 wave).

#19 GlennLeDrew  Happy Birthday!

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10643
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 08 February 2013 - 01:16 PM

I don't make flats, except an 18" long octagonal flat for a 1m f/2.5 LIDAR telescope. But it had a very much looser requirement on flatness than 1/30 wave (thank goodness.)

If sizable batches of flats are made, lambda/30 (P-V, of course) specimens are certainly possible to find, being more likely from the center of a bunch. But I wouldn't go out of my way to find one. A 1/10 wave unit is more than good enough, and the difference between it and a 1/30 wave job is probably impossible to discern in use.

#20 dpwoos

dpwoos

    Apollo

  • *****
  • Posts: 1390
  • Joined: 18 Oct 2006
  • Loc: United States

Posted 08 February 2013 - 01:52 PM

I certainly think that it is possible that some samples from a batch of secondaries would be 1/20 or even 1/30 wave. What I am questioning is the ability of the vendors of secondaries to the amateur community to pick them out via interferometric testing. If a 1/10 wave flat is plenty good enough (as you say, and I agree) then why the 1/20 and 1/30 wave claims? To me, this sounds like all marketing hype which I guess also fuels my suspicions that the testing is not legit. Of course, I am not claiming that the secondaries in question are not good 1/10 wave - they probably are.

#21 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 08 February 2013 - 04:55 PM

If a 1/10 wave flat is plenty good enough (as you say, and I agree) then why the 1/20 and 1/30 wave claims?



It's not a question of what you and I think is good enough, some people want the best and are willing to pay for it. You are claiming that manufacturers who claim 1/30 wave and are providing interferometric backup to those claims are not creditable.

Such a claim require substantiation, what actual evidence do you have that Antares secondaries are not what they claim? Glenn seemed to think it was possible to sort out a batch. Newport optics will sell you a 6 inch diameter 1/20th wave flat.

Jon

#22 dpwoos

dpwoos

    Apollo

  • *****
  • Posts: 1390
  • Joined: 18 Oct 2006
  • Loc: United States

Posted 08 February 2013 - 05:11 PM

Jon,

I already told you (more than once) why I think this, and so why rehash it? I never claimed it is "impossible" - why do you belabor the "it is possible" point? I don't believe the claims because I don't believe in the testing, and (apparently) you do. Go for it.

#23 ed_turco

ed_turco

    Apollo

  • *****
  • Posts: 1388
  • Joined: 29 Aug 2009
  • Loc: Lincoln, RI

Posted 08 February 2013 - 05:18 PM

I would be very interested in what wavelength light this 1/30 wave claim comes from. Seems like in the '80s, we went through accuracy inflation by means of the light used to express wavelength accuracy.

I am not saying anything about what light is used in testing, but in expressing the accuracy.

Think about it; 1/30 wave of red light is nowhere as good as 1/30 wave in green. Green wavelengths are around 1/50000 of an inch. In red light, 1/30 wave is a totally different proposition.

I love these colors! LOL!

#24 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1900
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 08 February 2013 - 05:33 PM

I believe what is being asserted here is that the interferometric readings used to select 1/30 wave (and others below some threshold) simply don't have meaningful precision at that level in the equipment being used.

For example I have two scales that both read out to tenths of a gram, but if I weigh something on the two scales, the "tenths" readings aren't the same. One or both scales are not actually measuring at that level of precision, despite what the readouts say.

#25 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43856
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 08 February 2013 - 06:21 PM

I don't make flats, except an 18" long octagonal flat for a 1m f/2.5 LIDAR telescope. But it had a very much looser requirement on flatness than 1/30 wave (thank goodness.)

If sizable batches of flats are made, lambda/30 (P-V, of course) specimens are certainly possible to find, being more likely from the center of a bunch. But I wouldn't go out of my way to find one. A 1/10 wave unit is more than good enough, and the difference between it and a 1/30 wave job is probably impossible to discern in use.


Glenn:

With your knowledge and skills at interferometry as well as fabrication, do you think that Antares testing methods are likely robust enough to actually determine with reasonable accuracy the flatness of a diagonal in the 1/10th lambda to 1/30 lambda range?

Antares Optics

:question:

Jon






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics