Jump to content


Photo

Secondary Size Revisited - A New Realization?

  • Please log in to reply
40 replies to this topic

#1 Jeff Morgan

Jeff Morgan

    Fly Me to the Moon

  • *****
  • Posts: 5761
  • Joined: 28 Sep 2003
  • Loc: Prescott, AZ

Posted 20 October 2013 - 03:35 PM

Last week I mailed a deposit check to Zambuto Optical so I am back in Telescope Design Mode.

Normally once I find a nice secondary mirror sizing solution just under 20% obstruction, I stop worrying about it and move on to other issues. But this time I wanted to look more closely, and a possible insight hit me. I'll throw it out to the CN wolves to find the holes in the idea, if any. I am not advocating anything here - just seeing if an idea is good or flawed.

Keep in mind (please!) that I am looking at a generalist scope - not imaging, AAVSO estimating, dedicated Milky Way Sweeper, or dedicated Planetary.

The conventional wisdom on eyepiece illumination is not go below 70% on the edge of your largest eyepiece (2" for most of us). The online calculators I have found (Newt-Web and Mel Bartels) won't even show you less than 70%. Nope. Don't look. Don't Ask. Nothing to see here. These aren't the droids you're looking for. Move along.

Ever wonder where that "consensus" figure of 70% came from? Me too. Going back through articles on the topic, one can find an interesting piece by Alan Adler in the August 2000 issue of Sky & Telescope. In personal experiments, he found edge illuminations down to 50% to be barely distinguishable from 70%, and 40% was perfectly acceptable. Better yet, he wrote a DOS program (named Sec, still available for download from Skypub.com) where you can manipulate telescope variables and allows you to peak behind the 70% Forbidden Zone, showing results down to 0% illumination (in extreme cases).

Then I realized that the field lens diameter on one the most common low power eyepieces (the 31 Nagler) is not 2", but rather 42 mm (1.65"). In fact, eyepieces with larger field stops are rather few and far between given the focal ratios of today's common telescopes. A little wiggle room!

Another thought occurred to me from an article I published on CN over the summer regarding DSO framing with eyepieces. In the article I suggested that for DSO viewing an object is well-framed when it does not exceed 75% of the eyepiece true field. Many people responded that I selected too high of a value, and that 50% object framing was more appropriate. I do not recall anyone suggesting a larger value. For discrete object viewing, it appears there may be even more wiggle room.

Thanks for indulging me this far, now let me connect all the dots - in most cases, if one targeted an illuminated field of 1" diameter (give or take a little bit) instead of 2" diameter, the majority of DSO's would still be illuminated to the 70% standard, while allowing for a step-down in diagonal size.

I suppose that this is merely a small step in the direction of what a dedicated planetary Newtonian already does.

In terms of what we lose, if Adler is right and 40% is workable, perhaps nothing. Or worst case, a handful of extremely large DSO's and the Milky Way.

#2 Alan French

Alan French

    Night Owl

  • *****
  • Posts: 4418
  • Joined: 28 Jan 2005
  • Loc: Upstate NY

Posted 20 October 2013 - 04:23 PM

I've been using a program I wrote long ago based on a "Gleanings for ATMs" article by William T. Peters and Robert Pike (Sky & Telescope, March, 1977). It nicely graphs the magnitude drop across the field, and is based on their recommendation to allow a 0.5 magnitude drop at the edge of the field (stars appear 63% as bright), and seems to work well in practice.

Reducing the illumination to 40% at the edge of the field is going to reduce a star's brightness by a full magnitude. Since we tend to center objects and concentrate our view there, perhaps not an issue. But I've found the 0.5 magnitude reduction results in adequately small diagonals.

Clear skies, Alan


#3 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10851
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 20 October 2013 - 04:49 PM

For decades I've been advocating that visual scopes can easily 'suffer' 50% illumination at the field edge, with no ill effect. And even less is no great hardship. If the fall-off is gradual, it's hard to see because the eye is primarily a contrast detector, of huge dynamic range. The outer field might be dimmed, but contrast is not altered, and this latter aspect is the most important.

For your basic 'look-see' telescope, priority one is obtaining full illumination at be field center, meaning a secondary which is just large enough to utilize the full aperture on axis can be good enough. And one can look even further on the bright side; the off-axis pupil clipping does somewhat reduce aberrations.

For example, a friend built a 10" f/5 Newt, using a low profile helical focuser and a 1.5" (!) diagonal. It just *barely* accommodates the on-axis light cone, yet 2" eyepieces reveal no visibly notable vignetting.

Too many amateurs have an unhealthy and unfounded fear of edge-of-field diminution.

#4 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1906
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 20 October 2013 - 04:57 PM

Seems like your "style of viewing" would have an effect here. People using 100 degree EPs who customarily shift their head to look around the whole field of view (and thus presumably often look directly at the edge) would be more likely to notice edge dimming.

For someone using 50-72 degree EPs, who customarily views the center of field, and only uses peripheral vision most of the time for the edge, wouldn't notice dimming at all since the peripheral vision is more sensitive to light anyway.

#5 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10851
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 20 October 2013 - 05:05 PM

If we consider a given true field, a wider apparent field (and hence higher magnification) would render vignetting less apparent due to the pattern being expanded to cover more retina. Brightness differences are more apparent when compressed to cover a smaller portion of the retina.

It's a curious notion that one would swivel one'e eye all over a 100 degree AFoV but not so for a 70--or even a 50--degree AFoV. :grin:

#6 gatorengineer

gatorengineer

    Mercury-Atlas

  • *****
  • Posts: 2553
  • Joined: 28 Feb 2005
  • Loc: Hellertown, PA

Posted 20 October 2013 - 05:40 PM

A couple of thoughts. Vignetting like coma is a personal thing, as well as edge correction. The faster the scope the shallower the drop off of illumination. Generally the faster the scope the wider the field desired, so a self re-inforcing theme. AT F8 and above it becomes pretty steep, and only at those speeds would you get to your 40 percent threshold.

I usually use Lockwoods page and shoot for a 0.5" fully illuminated for widefield rigs (almost all of my scopes). If I were going for a planetary scope I would go 0.25". I suspect my MN-86 is even less than that.

The only time I have notice illumination significantly was when I did a side by side with Petzval refactors. The newer variant with a large secondary element was MUCH brighter at the edge than an earlier variant with a smaller element.

#7 careysub

careysub

    Surveyor 1

  • *****
  • Posts: 1906
  • Joined: 18 Feb 2011
  • Loc: Rancho Cucamonga, CA

Posted 20 October 2013 - 06:07 PM


It's a curious notion that one would swivel one'e eye all over a 100 degree AFoV but not so for a 70--or even a 50--degree AFoV. :grin:


Perhaps not so curious: Don Pensack finds this approach to viewing to be a distinguishing characteristic between observers accustomed to 100 degree (or even 82 degree) EPs and those who aren't.

If you are centering an object to cover 50 or 75% of the field for viewing, then you likely aren't studying the outer edges of the field with central vision.

#8 derangedhermit

derangedhermit

    Apollo

  • -----
  • Posts: 1153
  • Joined: 07 Oct 2009
  • Loc: USA

Posted 20 October 2013 - 07:14 PM

Albert Highe includes several Excel spreadsheets with his book. One of them is named "Illumination". It includes light fall-off due to the telescope tube, the secondary, and the drawtube / drawtube baffle. The included chart goes down to 0.5 (50% drop-off) by default, and you can easily edit it to show down to 0.0 if you choose. The second sheet lets you put in any eyepiece and shows the effect on that eyepiece.

It is a very nice spreadsheet, better than anything else I have found.

I agree with your choice of selecting a secondary based on the widest field eyepiece you will regularly use, not the drawtube diameter.

You might want to also revisit the diameter of the aperture of the UTA in a truss design. (I say truss design since I think tube currents are, or should, be the limiting factor for many tube designs.)

Reducing the UTA aperture causes a very gradual light fall-off. Accepting some degree of light fall-off due to the UTA appears be a good choice in most cases - more cases, I think, than many people currently allow - since it improves illumination from a given secondary.

For example: the spreadsheet indicates (to me, anyway) that mirror+1" is a better rule of thumb than the current mirror+2", and that UTA apertures less than mirror+1" can be a very reasonable choice. This holds, at the least, for scopes up to 20" and f/4 or slower.

I suspect the optimal UTA diameter, in general, is to iteratively match it to 100% illumination from the secondary.

I think the $40 spent on the book gives a great return on investment.

#9 Jerry Hyman

Jerry Hyman

    Messenger

  • -----
  • Posts: 491
  • Joined: 29 Mar 2007
  • Loc: Mesa, Arizona

Posted 21 October 2013 - 12:44 AM

Ok Jeff, so what size mirror and f ratio did you order from CZ?

~jerry

#10 Jeff Morgan

Jeff Morgan

    Fly Me to the Moon

  • *****
  • Posts: 5761
  • Joined: 28 Sep 2003
  • Loc: Prescott, AZ

Posted 21 October 2013 - 12:48 AM

Seems like your "style of viewing" would have an effect here. People using 100 degree EPs who customarily shift their head to look around the whole field of view (and thus presumably often look directly at the edge) would be more likely to notice edge dimming.

For someone using 50-72 degree EPs, who customarily views the center of field, and only uses peripheral vision most of the time for the edge, wouldn't notice dimming at all since the peripheral vision is more sensitive to light anyway.


I would think they would be safe since the critical thing is field stop, not AFOV.

Suppose one decided that 40% at the edge of a 2" field was sufficient. The largest Ethos has a field stop of 36 mm (about 1.5"). Comparable to a 35 Panoptic or 26 Nagler. Prior to checking this out with Sec, I'm going to say such a field stop would still be in the area of high "traditional" illumination if the edge was 40%.

Of course, as has been pointed out this would vary by telescope focal ratio with faster optics have more gradual fall-offs.

#11 Jeff Morgan

Jeff Morgan

    Fly Me to the Moon

  • *****
  • Posts: 5761
  • Joined: 28 Sep 2003
  • Loc: Prescott, AZ

Posted 21 October 2013 - 12:54 AM

I'll have to order that book - it's been on my to do list for quite some time.

But if one shrinks the UTA too much, it defines the maximum true field of the scope, instead of your low power eyepiece. Perhaps that is worthwhile, I'll have to think it through a little more.

#12 Jeff Morgan

Jeff Morgan

    Fly Me to the Moon

  • *****
  • Posts: 5761
  • Joined: 28 Sep 2003
  • Loc: Prescott, AZ

Posted 21 October 2013 - 01:15 AM

8" quartz, 0.8" thickness, focal ratio TBD (primarily, I wanted to get a deposit in on the substrate before CZ ran out). Probably f/9, the maximum zenith height I can do with a CatsPerch standard chair.

#13 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 4124
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 21 October 2013 - 01:27 AM

But if one shrinks the UTA too much, it defines the maximum true field of the scope, instead of your low power eyepiece. Perhaps that is worthwhile, I'll have to think it through a little more.


Not really - most of the mirror is still capable of looking well off axis no matter what the UTA, so long as the UTA is at least the size of the primary.

I worked out a lot of the tradeoffs involved while first analyzing geometrically exactly what size the secondaries should be for large ultrafast mirrors, retaining only full illumination for a small central area. And an associated issue - how close should the barrel of a Paracorr 2 come to the edge of the primary light path, vs the larger secondary required to get it completely clear...

Best,
Mark

#14 Lamb0

Lamb0

    Vanguard

  • *****
  • Posts: 2119
  • Joined: 25 Jul 2007
  • Loc: South Eastern(ish) Nebraska

Posted 21 October 2013 - 03:23 AM

Different strokes for different scopes.

With the first iteration of my 8", (the "Test Tube"), I ended up insisting on a focuser with an extra inch of focuser travel that's often unnecessary, but comes in handy with some eyepieces and can accommodate binoviewers with a low magnification adapter that many Newtonians can't without modification. Not trusting adhesives for the secondary, I also prefer a hooded secondary holder that covers part of the mirror, so I sacrificed some of the effective diameter.

The first secondary was a 1.83" that worked well with my 1.25" 32mm Jaeger's Plossl with an effective field stop of ~28mm. Then I bought a 2" 32mm Burgess with a 41mm (measured) field stop. The field illumination near the edge plummeted to ~40%. In particular, the extended filtered views of the Veil and North American nebulae the lovely knots and whorls faded significantly. I wanted a Rich Field scope - why should repositioning the scope be necessary to see details more fully illuminated when they are already in the FoV?

Since I was replacing a defective primary mirror cell and increased the tube diameter to boost the wall thickness (.-92" was unavailable in 10" diameter at that time), and relieve the necessity of shimming the due to extra long focuser travel I upgraded to a 2.14" secondary.

Although a newly acquired eyepiece's 44mm field stop definitely still had some visible illumination drop off the edge of field the view was terrible, so I bought a Paracorr, (the original Visual Paracorr with a 38.5mm field stop). The eyepiece's astigmatism was incurable, but the 70% illumination was better, and with some tweaking I got the illumination to a very satisfactory =>75%. I later replaced the eyepiece with a 36mm Aspheric with a similarly sized field stop that delivers a very pleasing well illuminated coma corrected 2.16° TFoV.

Yes, taking into account the primary's chamfer and the secondary's hood The Mortar's obstruction is 30%, though that's less than a SCT. I figure I lose perhaps 10% to as much as 15% effective magnification due to contrast losses compared to a Newtonian optimized for "planetary" use. The seeing here rarely allows more than 200x to 240x, yet, the views of Jupiter @ 334x on a night of superb seeing are nothing less than outstanding with the Mortar's premium optics with it's superb polish, outstanding <1/20th wave p-v figure and textbook perfect Airy disk (for a 30% obstruction). Though Enterprise Optics is no more, Mr. Beck's mirror is superb! The Mortar really delivers on the Moon, doubles, and compact planetaries @ 477x. It is very competitive with a very good, well collimated and cooled NexStar 9.25". I've never seen ANY 8" SCT perform as well at any magnification, (though tracking and GoTo would be nice to have). Proper air flow with the oversized 12" tube shortens cooling time with the standard thickness (1.3125") mirror, (the Parks composite primary mirror cell doesn't hurt either), and tube currents are dramatically reduced.

As a result of my experience I believe there is no one correct answer for proper Newtonian secondary sizing. My scope does what I originally intended delivering superb Rich Field views, makes an excellent general purpose scope, and still delivers a decent view at high power, (and I still have a very good 1.83" secondary I'm willing trust for brief sessions for a ~23.5% obstruction should I choose to do so).

When I finally add a larger scope to my limited collection I needn't be worried much about illuminated field. The drop off on larger scopes is much slower, and even a 0.1° fully illuminated field will be sufficient. After all, I already have a good 8" Rich Field scope! :dob:

#15 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 44353
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 21 October 2013 - 04:07 AM

A couple of thoughts/experiences...

- Small diameter upper cages may be more sensitive to thermal issues at high magnifications since there is no buffer zone between the on axis light column and the tube wall. Tube Newts work best with larger tube diameters, partially for this reason.

- The two programs I use, Newt for the Web and Mel Bartels, both provide some information for a given field stop, Mel's shows illumination as a function of off-axis diameter. But they do quit before showing the full drop though it can be interpolated with Mel's program.

- My 130mm F/5 Newtonian had an undersized secondary, barely fully illuminated at the center, yet under dark skies, the vignetting with a 31mm Nagler was not easily noticed.

- I say get over the fixation with the size of the diagonal.. the actual effect of the size on contrast is minimal just as the effect of the size on vignetting is minimal. What I look at is the size of the fully illuminated field of view, I want a sufficiently large fully illuminated field of view and let the off-axis illumination take care of itself. I use the number 0.5 degrees though with a large scope this is too large. I just want a fully illuminated field so I know I am operating at full aperture over that center.

- There are many factors to consider, often more important than vignetting or planetary contrast.

Example:

My 16 inch F/4.42 uses a "giant" 4 inch diagonal, 25%. I could use a 3.1 inch diagonal, I have one I bought for it, but to use it, I would need to extend the tubes 2 inches to achieve a reasonably sized fully illuminated field of view. This scope is designed around eyepiece height and comfortable viewing.

Those two inches are two very critical inches, with the 4 inch diagonal, I can view flat footed at the zenith and employ the much more desirable angled focuser. If the scope were 2 inches taller, I would need a stool... or platform shoes. :)

The angled focuser allows me to view the horizon in relative comfort but has the downside that with a Paracorr, the eyepiece height is within an inch or two of the zenith height of 68.5inches all the way to 60 degree elevation.

The focuser is angled at 33 degrees from horizontal which makes viewing near the horizon possible and comfortable. This scope lives at a dark site that is open to the south, Omega Centauri culminates at 10 degrees, many other fine south deep sky objects are visible. A horizontal focuser would make viewing those very awkward and uncomfortable...

So.. the 4 inch diagonal is optimized for comfort, the scope is designed around making the viewing positions most comfortable rather than some mostly imaginary gain in planetary contrast or some fretting about the vigetting with the 42mm Nagler..

Just some stuff I think about...

Jon

#16 Lamb0

Lamb0

    Vanguard

  • *****
  • Posts: 2119
  • Joined: 25 Jul 2007
  • Loc: South Eastern(ish) Nebraska

Posted 21 October 2013 - 04:54 AM

The OP has a good perspective on what will suit the needs of many, most especially himself. That's the beauty of building/assembling/modifying our own scopes - we get to choose our own compromises rather than limiting ourselves to what a manufacturer decides is best for their balance sheets.

Are their improvements yet to be made on my scopes? Of course there are; though I won't post them in this thread. My scopes are both works in progress, even the little ETX60.

As handy as Newt is I often prefer Mel Bartels' program, and I'm glad the old SEC.exe is still available from S&T for download as I like the graphical output, simple though it is - even if it does require DOSBox or something similar to run nowadays. Additional/updated tools, especially those not requiring a particular OS or walled garden would be welcome; but so long as there are forums like CN we needn't feel like we're out on a limb sawing the tree behind us to design, optimize, and otherwise improve our scopes to meet better satisfy our desires. ;)

#17 don clement

don clement

    Vendor (Clement Focuser)

  • -----
  • Vendors
  • Posts: 812
  • Joined: 02 Feb 2011
  • Loc: Running Springs, California

Posted 21 October 2013 - 07:04 AM

70% is about 1 F-stop or ~1/1.414. Seems to me like the MTF would be the best tool to quantify the change in image contrast due to changes in secondary size. The rest appears to be about personal perceptions akin to personal perceptions in the audiophile world. IMO vacuum tubes sound best.

Don

#18 kfrederick

kfrederick

    Gemini

  • *****
  • Posts: 3017
  • Joined: 01 Feb 2008

Posted 21 October 2013 - 07:13 AM

Maybe try a chief in 8 inches should be a nice size. Use a curved secondary and a Chief can make fast work at least f5 . Done correct should give APO views . Optics in a box/ no dew spider/ If you wont long it folds back . Much going for a chief . Carl has made at least 3 CHief primarys . I am sure as a newt be a very nice telescope . Just fun to take some risk for the reward .Be a shame to put a spider over Carls 1/40 wave surface . I like your 17 bet it is nice views

#19 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 44353
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 21 October 2013 - 07:14 AM

70% is about 1 F-stop or ~1/1.414. Seems to me like the MTF would be the best tool to quantify the change in image contrast due to changes in secondary size. The rest appears to be about personal perceptions akin to personal perceptions in the audiophile world. IMO vacuum tubes sound best.

Don


My thinking is like this:

MTFs show planetary fine scale contrast. Such contrast is only of interest relatively close on-axis because for the eye to begin to see what exists at the focal plane requires relatively high magnifications and therefore quite narrow fields of view.

Jon

#20 Mark Harry

Mark Harry

    Vendor

  • *****
  • Vendors
  • Posts: 6189
  • Joined: 05 Sep 2005
  • Loc: Northeast USA

Posted 21 October 2013 - 07:18 AM

I tackle it somewhat differently. Most eyepieces used are 1.25" (in my case!) I size the secondary so that 100% fully illuminates .3-.4" at the focal plane. It has worked well over the years, 1.25" eyepieces have a max width at the field stops of about 1.100~". In various focal ratios of Newts that are typical, this criteria works just fine. When I use Plossls, Orthos or Naglers of med/hi powers, just about 100% illumination with no vignetting is the norm; where I feel it's most important.
Low powers, where there may/may not be noticeable light falloff, centering whatever detail I want to examine allows slight peripheral vision, which is vastly more sensitive to light levels encountered at the field edges, is scarcely noticeable. IE, what I'm looking at DIRECTLY with the fovea, is framed nicely within the field of view smack in the center.
The only time I ever -notice- say a 32mm Plossl having falloff, is with something like a 6" F/10 or similar long focus instrument. But never with F/4-5 where the falloff at lowish powers is more gradual.
****
I never much catered to the bigger 2" eyepiece format with smaller scopes. I know I'm likely missing out on some really mindblowing wide field views. But the detail and contrast of F/8-10 instruments with lower abberations is a pretty good place to be for a compromise for quality, vs extreme dedicated DSO/lowest power viewing.
fwiw,
M.

#21 gatorengineer

gatorengineer

    Mercury-Atlas

  • *****
  • Posts: 2553
  • Joined: 28 Feb 2005
  • Loc: Hellertown, PA

Posted 21 October 2013 - 07:53 AM

just a comment. In this day and age we also have to work with what is commercially available. You wont find a high end secondary less than 1.30". Also mounting and having adjustment on anything smaller becomes extremely difficult.

Based on the 1.30 commercially available size, i wouldnt go any slower than F7, which gives a 0.50" fully illuminated at a focal distance 6" above the focuser. This size gives you a co of 16.25%, which is refractoresq.... The only advantage of going slower would be a reduction in coma which is proportional to the square of the focal length.

#22 MitchAlsup

MitchAlsup

    Surveyor 1

  • -----
  • Posts: 1577
  • Joined: 31 Aug 2009

Posted 21 October 2013 - 11:12 AM

Then I realized that the field lens diameter on one the most common low power eyepieces (the 31 Nagler) is not 2", but rather 42 mm (1.65").


The focuser baffle on my large DOB has a 1.58" opening, for similar reasons.

#23 MitchAlsup

MitchAlsup

    Surveyor 1

  • -----
  • Posts: 1577
  • Joined: 31 Aug 2009

Posted 21 October 2013 - 11:12 AM

Then I realized that the field lens diameter on one the most common low power eyepieces (the 31 Nagler) is not 2", but rather 42 mm (1.65").


To focuser baffle on my large DOB has a 1.58" opening, for similar reasons.

#24 hottr6

hottr6

    Vanguard

  • *****
  • Posts: 2442
  • Joined: 28 Jun 2009
  • Loc: 7,500', Magdalena Mtns, NM

Posted 21 October 2013 - 01:55 PM

You wont find a high end secondary less than 1.30".

I have recently bought both 0.75" and 1" quartz secondaries from Protostar. I doubt that there are better commercial secondaries.

#25 gatorengineer

gatorengineer

    Mercury-Atlas

  • *****
  • Posts: 2553
  • Joined: 28 Feb 2005
  • Loc: Hellertown, PA

Posted 21 October 2013 - 02:29 PM

I stand corrected. Glad to hear Protostar is alive and well.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics