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

What's wrong with FCD100 and FDL55?

  • Please log in to reply
94 replies to this topic

#51 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 13 September 2018 - 02:27 AM

I can add so much that CDGM's H-ZK1 is near identical to Ohara's S-BAL22, and H-K9L to good old BK7, Hoya's BSC7 and Ohara's S-BSL7.

Vla -

 

Thank you for yet another informative reply.  I did not find S-BSL7 on Ohara's downloadable Excel file but did find it in data sheets on the Ohara site.  S-BAL22 doesn't seem to be in either the Excel file or the data sheets; I had to poke around on the Web to find data for it.  Maybe it has been discontinued.  Hoya BSC7 was in the current Hoya Excel file, as were FCD100 and  LACL60.  Web searches also turned up data for the CDGM glasses you mentioned.

 

The similarities you noted are, indeed, striking.  They suggest efforts among competing glass manufacturers to duplicate successful/popular types, but I wonder how the quality of batches compares across brands.

 

My thanks again for taking the time to add significantly to this thread -

 

     Larry



#52 Alan French

Alan French

    Night Owl

  • *****
  • Posts: 6,080
  • Joined: 28 Jan 2005
  • Loc: Upstate NY

Posted 13 September 2018 - 12:51 PM

Vla -

 

Thank you for yet another informative reply.  I did not find S-BSL7 on Ohara's downloadable Excel file but did find it in data sheets on the Ohara site.  S-BAL22 doesn't seem to be in either the Excel file or the data sheets; I had to poke around on the Web to find data for it.  Maybe it has been discontinued.  Hoya BSC7 was in the current Hoya Excel file, as were FCD100 and  LACL60.  Web searches also turned up data for the CDGM glasses you mentioned.

 

The similarities you noted are, indeed, striking.  They suggest efforts among competing glass manufacturers to duplicate successful/popular types, but I wonder how the quality of batches compares across brands.

 

My thanks again for taking the time to add significantly to this thread -

 

     Larry

Vla,

 

You made me curious as I was looking at S-BSL7 in the Ohara Excel file just the other day. Downloaded it again and S-BSL7 is the seventh entry. 

 

I do not find S-BAL22 either, however.

 

Clear skies, Alan



#53 Vla

Vla

    Viking 1

  • *****
  • Posts: 611
  • Joined: 08 Mar 2014

Posted 13 September 2018 - 04:55 PM

Larry, Alan (it's been many moons smile.gif,

 

The two glasses are in the catalog in the latest OSLO version, so I wouldn't write them off just yet. You don't know unless you ask directly, and probably not even then. On the thread subject, I (finally) made a fairly complete, although somewhat dated (at least a few years, except for the ED glasses) relative partial dispersion chart. It saves time when looking for matching glasses.

Looking at the ED's, Yuri's remark could be taken literally, since FPL55 is a little lower than FPL53, which means little less well positioned vs. bulk of potential matching glasses for either doublets or triplets. Of course, matching P(F.e) is only the first and most important requirement, but the violet P(g,F) also matters, as well as the raw power determined by the refractive index. Relatively weak ED glass will require steep curves, likely with a significant difference between two inner radii (doublet), generating a lot of higher order SA residual even at medium to slow focal ratios (for instance, BSM81, which on the chart seems to be a perfect match for FPL53).

 

I entered the two newcomers (FPL55 and FCD100) into the top chart, and another small negative for FPL55 is visualized, a bit lower Abbe number. All else equal, smaller Abbe differential means more spherochromatism.

Attached Thumbnails

  • rpd0c.png

Edited by Vla, 14 September 2018 - 12:19 PM.

  • Bomber Bob and danielguo like this

#54 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 13 September 2018 - 09:11 PM

 

You made me curious as I was looking at S-BSL7 in the Ohara Excel file just the other day. Downloaded it again and S-BSL7 is the seventh entry. 

 

I do not find S-BAL22 either, however.

 

Alan - You are correct, I did get S-BL7 data from the Ohara Excel file, as well as from an Ohara data sheet, and also from both for Ohara S-FPL53, S-FPL55, S-LAL61, and S-BSM81.  For what interest it might be, I found a data sheet for Ohara S-BAL22 at
<https://refractivein...&page=S-BAL22>.

 

Data for the Hoya glasses came from the Excel file at the Hoya site.  Data sheets for the CDGM glasses, were found via Google searches.

 

Below is a summary table adding key parameters of the mating-type glasses Vladimir mentioned in his post #50.  It is meant to be appended to my post #36 with an updated version of the original table.

 

 

 On the thread subject, I (finally) made a fairly complete, although somewhat dated (at least a few years, except for the ED glasses) relative partial dispersion chart. It saves time when looking for matching glasses.

 

....

 

I entered the two newcomers (FPL55 and FCD100) into the top chart, and another small negative for FPL55 is visualized, a bit lower Abbe number. All else equal, smaller Abbe differential means more spherochromatism.

 

Vladimir - Wow, many thanks for posting the new chart of Vd vs. P(F,e) (what I've called 'PF,e').  You are certainly right that those charts, including the one of Vd vs. P(g,F) in your telescope-optics.net work, do make identification of potential mating types easier.

 

For whatever reasons, the values of P(F,e) I arrived at for the tables put several glasses at somewhat different relative positions than suggested in the plots you provide.  I did not find explicit values for P(F,e) in the Ohara or Schott data sheets, so calculated those from refractive-index and/or partial-dispersion values that were in the data sheets:
     PF,e =  (nF - ne)  / (nF - nC)

 

Rounding errors in those calculations could account for some of the discrepancies.

 

      Larry

 

addendum to Selected Optical Parameters of some Refractor Glass Types.r6.jpg



#55 Vla

Vla

    Viking 1

  • *****
  • Posts: 611
  • Joined: 08 Mar 2014

Posted 14 September 2018 - 08:55 AM

Larry, you're right, those ED P(F.e) values on the pic are not agreeing. I read directly of screen in ATMOS, but could be that the displayed value does not follow cursor position accurately. I went on to calculate by hand, and I still got different values for some (maybe you'll remember, I did get lower value for FPL55 before, and now I got the same). For Ohara glasses I used their online PDFs, for CaF2 and FCD100 OSLO catalog (I checked Schott and Nikon CaF2, they are nearly identical, giving somewhat lower 0.45614 value), OK4 and H-FK61 their respective mfg data. Specific values are:

 

FPL51 - 0.45649

FPL53 - 0.45455

FPL55 - 0.45356

FCD100 - 045420

H-FK61 - 0.45574

OK4 - 0.45453

CaF2 - 0.45654

 

I put these new values on the pic (bottom, the top is still as ATMOS shows it, except for FCD100 and FPL55 which I inserted according to your table).



#56 Riccardo_italy

Riccardo_italy

    Viking 1

  • -----
  • Posts: 857
  • Joined: 29 Sep 2014
  • Loc: Italy

Posted 14 September 2018 - 09:19 AM

Now, looking at the Table it's clear why CaF2 remains superior to any fluorite glass.

 

Not because of the Abbe number, but because of the P(F.e): it's easier, with fluorite, to find a mating element with an as close as possible P(F.e) number but as farther as possible Abbe number.

 

That's the reason why when you really need a FAST doublet (proportional to diameter), you go with fluorite.

 

Almost same diameter, 100-110mm (where you can find really fast scopes). Among ED scopes you have:

- FPL53 chinese made air-spaced triplet at f6.5,

- CFF 105mm f6 triplet with FPL55 and all aspheric lenses.

- Skywatcher Esprit, FPL53 triplet at f5.5

 

But if you look at fluorite doublets, you have the Borg 107mm at f5.6. I think any ED doublet can be made so fast and with acceptable spherical and chromatic aberrations.

 

So, Fluorite is important when you need to build a fast scope.



#57 MooEy

MooEy

    Messenger

  • *****
  • Posts: 442
  • Joined: 02 Oct 2007

Posted 14 September 2018 - 10:40 AM

Acrylic looks like a very nice material...



#58 Jeff B

Jeff B

    Fly Me to the Moon

  • *****
  • Posts: 7,466
  • Joined: 30 Dec 2006

Posted 14 September 2018 - 10:48 AM

This is so cool guys and thanks.

 

The way I look at it is that the good stuff (CaF2, FPL-53, FPL-55, ...) are all on an almost vertical Vd line with a bit of vertical spread in P(F,e), but the "main sequence" line of the glass slopes away from the that vertical Vd line at a robust angle.  This gives CaF2 with its slightly higher P(F,e) value greater distance from the main sequence line, hence a bit better performance with a mating glass that's close to the main sequence and similar P(F,e).

 

Good stuff!

 

Jeff



#59 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 14 September 2018 - 05:45 PM

...I went on to calculate by hand, and I still got different values for some (maybe you'll remember, I did get lower value for FPL55 before, and now I got the same). For Ohara glasses I used their online PDFs, for CaF2 and FCD100 OSLO catalog (I checked Schott and Nikon CaF2, they are nearly identical, giving somewhat lower 0.45614 value), OK4 and H-FK61 their respective mfg data. Specific values are:

 

FPL51 - 0.45649

FPL53 - 0.45455

FPL55 - 0.45356

FCD100 - 045420

H-FK61 - 0.45574

OK4 - 0.45453

CaF2 - 0.45654

 

I put these new values on the pic (bottom, the top is still as ATMOS shows it, except for FCD100 and FPL55 which I inserted according to your table).

Vla -

 

Thank you very much for checking the calculations.  They are certainly at the mercy of variations in posted data values as well as calculation errors.  Below, I've outlined the calculations that gave rise to the value of P(F,e) I put in the table for FPL55.  I'm doing this as part of the thread, rather than as a personal message, in case the causes of differences might be of interest to others tempted to undertake similar calculations (though I suppose that group to be rather small).

 

Values present in Ohara S-FPL55 data sheet downloaded from
<https://www.oharacor...f/S-FPL55.pdf>:
   refractive-index values
     nC  1.43733
     ne   1.43986
     nF   1.44196
   partial-dispersion values
     (nF - nC)  0.004635
     (ne - nC)  0.002529
     (nF - ne) not directly given, derived as (nF - nC) - (ne - nC)  0.002106

Calculations:
     either directly from refractive-index values or from partial-dispersion values:
     PF,e = (nF - ne) / (nF - nC)  0.45437,  0.4544 rounded to four places

 

For CaF2, the P(F,e) value I used for the table, 0.4573, was calculated directly from refractive-index values on a Lasertec page:  <http://www.lcoptical...-material.html>
Very similar values, usually differing slightly only in the 5th decimal place  were found on Schott data sheets -
Schott LITHOTEC-CAF2: <http://www.matweb.co...07c94a349abdf1>
Schott ML: <http://www.yuanch.com/pdf/fhg.pdf>
As I mentioned in notes with the original table, different sources gave quite different values for some CaF2 parameters, e.g. values of Vd, ranging from 94.96 to 95.31.

 

Corrections to my calculations would be very welcome.

 

An added note relevant to the topic:  Difficulties of identifying ideal mating types for the ED glasses and CaF2 might have been increased by the dramatic reduction of available glass types that followed mandated removal of those containing lead or arsenic.  An interesting article about this is at
<http://www.ora-blogs...ization.html>. 

The ban on lead- and arsenic-contining glasses was reportedly of significant environmental/health value, but one wonders whether optical designs before it might have benefitted from glasses now absent from manufacturers' catalogs owing to the ban .

 

    Larry

 

Edited to correct refractive-index values for FPL55; original entries were wavelengths owing to copy/paste error from data sheet.  Correct refractive-index values were used for calculations of values in original tables.  Thanks to Vla for pointing out the error in this post.


Edited by LMO, 14 September 2018 - 10:00 PM.


#60 Vla

Vla

    Viking 1

  • *****
  • Posts: 611
  • Joined: 08 Mar 2014

Posted 14 September 2018 - 08:31 PM

Larry, what you used as indexes are the wavelengths. Seems that FPL55 does have the lowest P(F,e) value of all EDs, so that probably answers what Yuri had in mind saying it's a bit worse than FPL53 (if I remember correctly). We could have saved a lot of time if we just asked him. But I think it was worth it going through this. We know more about it than before.

 

Going into the numbers, that error is irrelevant, but I don't see Ohara giving nF-nC partial dispersion here. I used the same PDF and calculated partial dispersions directly from the indexes given. It is limited to 5 digits and gives nF-nC=0.00463 and nF-ne=0.00210, for P(F,e)=0.45356. If I'd use nF-ne=0.002106, P(F,e)=0.45486. So the relative dispersion value is very sensitive to the index differential value, up to the 6th digit, while it itself is sufficiently accurate if rounded to 4 digits.

 

Edit: Yes, I see nF-nC, right up in the corner. So P(F,e)=0.4544 according to Ohara. Since OSLO uses 6-decimal refractive index, dispesions based on its data should be accurate (CaF2 and FCD100). For the rest (other than Ohara) will see what I can get from OSLO.

 

All right, after little confusion, OSLO also confirms your numbers for FPL 55 and 53, gives a bit lower value for H-FK61 (0.4551) and OK4 remains where it was. FPL51 drops to 0.4556, and here's the corrected graph, since I can't edit the old post anymore. After all this, we can say that we know that we don't know what Yuri really meant ')

Attached Thumbnails

  • rpd0c.png

Edited by Vla, 14 September 2018 - 11:22 PM.


#61 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 14 September 2018 - 10:18 PM

Larry, what you used as indexes are the wavelengths.

...

 

Oops, thanks for pointing out that error in the post.  I've edited the post to enter the refractive-index values actually present in that Ohara FPL55 data sheet, and the correct values were used in calculations for the tables in my earlier posts.  In fact, I find that my original Excel file used 6-decimal versions for the refractive-index values, but I can't presently re-locate the source for those.

 

In any case, you've pointed out a central problem -- results for the relative partial dispersion values are very sensitive to precision of the refractive-index values, which greatly complicates (naive) efforts such as mine to get some insight into glass pairings that producers of excellent refractors might actually be using.

 

    Larry



#62 Vla

Vla

    Viking 1

  • *****
  • Posts: 611
  • Joined: 08 Mar 2014

Posted 14 September 2018 - 10:45 PM

No, why, it only takes a 6-decimal index value. But, as already mentioned, there is more to it than just matching closely enough relative partial dispersions, firstly blue/red and then, if possible, violet too. The raw power of matching glass given by its refractive index vs. power of ED glass, will - in addition to its Abbe # differential - determine how strongly curved need to be the inner (the most strongly curved) lens surfaces. The stronger matching element for given ED glass, the more strongly curved surfaces required, and the more of higher order spherical. So for more complete information on which glasses are near optimal matches, one needs to know the refracting index as well. Then comes production (quality) consistency, price, ease of shaping i.e. working with, stability etc. We can't tell just looking at a graph like this which glass is actually viable for production, but at least we can tell which ones are not.


Edited by Vla, 15 September 2018 - 07:41 AM.

  • Jeff B likes this

#63 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 15 September 2018 - 06:36 PM

No, why, it only takes a 6-decimal index value. But, as already mentioned, there is more to it than just matching closely enough relative partial dispersions, firstly blue/red and then, if possible, violet too. The raw power of matching glass given by its refractive index vs. power of ED glass, will - in addition to its Abbe # differential - determine how strongly curved need to be the inner (the most strongly curved) lens surfaces. The stronger matching element for given ED glass, the more strongly curved surfaces required, and the more of higher order spherical. So for more complete information on which glasses are near optimal matches, one needs to know the refracting index as well. Then comes production (quality) consistency, price, ease of shaping i.e. working with, stability etc. We can't tell just looking at a graph like this which glass is actually viable for production, but at least we can tell which ones are not.

Vla -

 

Some notes:

 

1. I found the source of the 6-decimal-place values I had for the Ohara glasses.  Curiously, it was an earlier version of the Ohara Excel file that I had downloaded -- 'OHARA_20171130_6.xlsx' vs the one now provided by Ohara, 'OHARA_20180518_5.'  The file names suggest both dates and precision, Nov. 11, 2017 6-place for the earlier one, and May 18, 2018 5-place for the second, current one.  Below is a set of the 6-place values from the earlier file; they can be compared with the 5-place values i entered in post #59 here (as corrected after you'd pointed out my initial, careless entries from the wrong column in the Ohara PDF data file).

 

FPL55 values in old vs. new Excel files are unchanged for nd (6th place = 0), Ve, P(g,F), and (calculated) P(F,e).  The old Excel file gives Vd = 94.88 vs 94.66 in the new files and by my calculations.

 

The 5-place values in the current Excel file are the same as those in the Ohara PDF file.  The significance of Ohara's drop back to only 5 places in the current files it provides is unclear.

 

2.  Your point about the importance of the actual refractive index of the mating glass is certainly well taken and another well-covered one in your telescope-optics pages, e.g. page/topic 9.2.  I included nd values for the glasses in my tables to provide at least one reference for consideration of that point.

 

3.  The practical issues associated with production using particular glass types are well beyond my grasp at present, though there have been some discussions of them in various places, and they are outside the focus I'd intended for this thread.  It was a focus you clearly grasped and one to which you've added very helpful input.  My thanks yet again for that.

 

    Larry

 

6-decimal-place values for FPL55 in an older, 2017 Ohara Excel file.
(The 5-decimal-place values from the newer, 2018 Ohara Excel file and current Ohara data sheets are given in parentheses for comparison.)
   refractive-index values
     nC  1.437328 (1.43733)
     ne   1.439857 (1.43986)
     nF   1.441963 (1.44196)
   partial-dispersion values
     (nF - nC)  0.004635 (unchanged)
     (ne - nC)  0.002529 (unchanged)
     (nF - ne) not directly given, derived as (nF - nC) - (ne - nC)  0.002106
Calculations:
     either directly from refractive-index values or from partial-dispersion values:
     PF,e = (nF - ne) / (nF - nC)  0.45437,  0.4544 rounded to four places



#64 25585

25585

    Voyager 1

  • *****
  • Posts: 11,468
  • Joined: 29 Aug 2017
  • Loc: In a valley, in the UK. Dark end of the street.

Posted 16 September 2018 - 03:01 PM

The actual reason is this:

 

It is well known that GSO Dobsonians use mirrors made from BK7 and since they are perfectly corrected chromatically, using BK7 in a refractor has to be a good thing.  lol.gif

 

 

Sorry for the silliness but I just couldn't help myself.

 

Humor break over, now back to your regularly scheduled programming.

 

Jon

Strange that BK7 is looked down on for binoculars, as opposed to BAK4.

 

I have several pairs of Swift Osprey binoculars, that are BK7 Japanese made. Very bright and clear.



#65 Vla

Vla

    Viking 1

  • *****
  • Posts: 611
  • Joined: 08 Mar 2014

Posted 16 September 2018 - 03:33 PM

Since picture is worth a thousand words, probably the best thing to do to see the difference between FPL53 and FPL55 is to put them together with the same matching glass and see the outcome. I've noticed on the chart a new (for me) fast apo matching glass, Schott N-KZFS2, which has more of Abbe # differential than any other with a close P(F.e) value. News is that OSLO says it costs as much - or little - as BK7, while the old KZFS2 was ten times as much. Anyway, P(F,e) value only slightly favors FPL53, and when we put them into a 152mm f/8 system, it looks like this. For all practical purposes identical. For comparison, there's the system with FPL51, and it is actually real close. None of them would pass the "true apo" wavefront criterion only because of the violet g-line (the F-line with FPL51 is 0.26 wave p-v, but since significant part of it comes from the midfront deformation, the RMS is smaller, 0.061 wave, well below the "diffraction limited" 0.0745; similar for the red with FPL55/53).

 

 

Attached Thumbnails

  • fpl.png

  • peleuba, Jeff B, danielguo and 1 other like this

#66 Jeff B

Jeff B

    Fly Me to the Moon

  • *****
  • Posts: 7,466
  • Joined: 30 Dec 2006

Posted 16 September 2018 - 05:51 PM

Nice Vla and I like to see those differences in the plots.  Thanks!

 

I noticed in the graph above that the marker on the right for good old FPL-51 was basically pointed right at N-KZFS2.  How are the internal radii?   Do they look "reasonable?

 

Any chance you could run up the graphs for 152mm F10 doublet?  Just a thought.

 

Jeff



#67 Vla

Vla

    Viking 1

  • *****
  • Posts: 611
  • Joined: 08 Mar 2014

Posted 16 September 2018 - 08:25 PM

Jeff, I actually had it at f/10, only scaled it down to f/8 to make it comparable to the other two glasses. It is a true apo according to both, wavefront error and polychromatic (photopic Strehl). By letting error in the violet grow little more (to 0.64 wave p-v), the poly Strehl reaches 0.97.

 

As for internal radii, at f/8 they are with FPL51 290/294mm vs. 355/363mm with FPL53 and FPL55. That is all due to the smaller Abbe# differential of FPL51 vs. N-KZFS2, and brings in noticeable higher order SA residual.

Attached Thumbnails

  • f10.png

  • peleuba, fredsf and Jeff B like this

#68 Jeff B

Jeff B

    Fly Me to the Moon

  • *****
  • Posts: 7,466
  • Joined: 30 Dec 2006

Posted 16 September 2018 - 11:18 PM

Thanks Vla.  Actually the FPL-51 F10 version looks like a very nice design.   The off-axis performance is pretty good.  It shows to me that a modest increase in focal ratio can make a very nice improvement to the optical performance relative to F8.  But it would be a foot longer and optically slower, which for me visually, is just fine but not so much with the imaging crowd.   Then again, almost all fast ED scopes have some form of reducer/flattener in the chain for imaging.  I imagine F10 versions for the FPL-55/53 glasses would be very, very nice but I would have to wonder if I could readily see the differences between the F8 and F10 versions.  It would be nice experiment though.  

 

I wonder how much cheaper it would be to make the FPL-51 F10 versus the FPL-55/53 F8.

 

Cool, and again, thanks.

 

Jeff



#69 dscarpa

dscarpa

    Skylab

  • *****
  • Posts: 4,164
  • Joined: 15 Mar 2008
  • Loc: San Diego Ca.

Posted 17 September 2018 - 09:58 AM

 ES charges $400 for a 127 7.5 with FCD100 vs one with glass similar to FPL51. David 



#70 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 18 September 2018 - 02:58 AM

I'm attempting here a summary of what I've learned so far from what has turned out to be a very instructive exercise.

 

1. Mating type for FPL53: 

 

Although 'ZKN7' (Schott N-ZK7) is often cited as the 'ideal' mating type for FPL53, if I've gotten the various values correct in the posted table, N-BK7 looks to be even better in key respects.  Compared to N-ZK7 as a match for FPL53, N-BK7 appears to have a more favorable PF,e value, a more favorable Pg,F value, and a somewhat higher refractive index, although a somewhat higher Vd value. 

 

Further, contrary to the suggestion by Thomas Back in his 'Brief History of Astro-Physics Lenses' that Astro-Physics used ZNK7 with FPL53, Roland has frequently noted practical problems with use of ZNK7 causing him to abandon its use, and Alan French has put it simply in various places of one thread "AP has never made any telescopes using FPL53 with ZKN7." "The AP lenses with FPL53 did not use ZKN7. Because of issues with glass quality, AP had stopped using ZKN7 even before the adoption of FPL53."  and " If Tom Back claimed AP used ZKN7 with FPL53, he was mistaken."
<https://www.cloudyni...l-53-and-caf2/>

 

BK7 is the only mating type I've seen specified for an A-P scope -- used with FPL53 in the relatively recent 175 f8 EDF triplet, with the best specified color correction among A-P scopes up to its time.  Sky-Watcher also uses the same FPL53/BK7 combination for its admired ProED doublets (80mm, 100mm, 120mm).

 

2. FPL55 vs FPL53:

 

Again, from the values in my table, if BK7 is chosen as the mating type, FPL55 looks to be at least as good, maybe even better, a match than FPL53; FPL55 has a value of PF,e even closer to BK7's, the same value of Pg,F as FPL53, a somewhat higher refractive index than FPL53, and even a slightly higher Vd.

 

Vla's comparisons in this thread of FPL53/KZFS2 and FPL55/KZFS2 doublets showed those two FPL glasses to provide very close results, very slightly better ones with FPL53 (p.3 of this thread, post #65).  Although it's not clear that this is the explanation for the slightly better performance of FPL53 in these particular comparisons, the table shows a PF,e value for FPL53  very slightly closer to that of KZFS2 than is the PF,e of FPL55.  The reverse is the case for BK7, favoring FPL55 as just noted.

 

It would be interesting to see models of doublets and triplets comparing results with the FPL55/BK7 combination to results with FPL53/BK7.

 

In any case, the model comparisons Vla has already contributed to this thread do suggest that, in theoretical terms, there's 'not much' wrong with FPL55 compared to FPL53.

 

3. FPL53 vs CaF2:

Alan French (bravely) initiated a recent CN thread discussing this comparison, with reference to the examples Vla provided in his telescope-optics.net pages:
<https://www.telescop...o_examples.htm>
<https://www.cloudyni...l-53-and-caf2/>

 

The telescope-optics.net page examples cited by Alan compared doublets (100 f10 FPL53 mated with ZNK7 vs CaF2 mated with K5, example #15 vs #16) and triplets (140 f7 FPL53 mated with ZNK7 vs. CaF2 mated with K7, examples #18 vs. #19).  To these can be added the doublet comparisons, in both Fraunhofer and Steinheil configurations, (100 f7 FPL53 mated with ZNK7 vs. CaF2 mated with K5, examples #12 vs. #14). 

 

In all cases, The FPL53/ZKN7 versions yielded polychromatic Strehl values as good as, or slightly better than, those from the CaF2 combinations.

 

4. FCD100:

 

Vla's models in this thread of the FCD100/ZNK7 combination in doublet and triplet designs and of FCD100-based triplets using combinations of differing mating types chosen to 'offset each other's deviations' from ideal matches -- H-K5/FCD100/N-BK7 and H-ZK1/FCD100/H-K9L -- point to the latter path as the most likely preferable one to be taken with FCD100.  Vla noted that something like the second version comes close to what Roland has cited as the degree of correction in the new Stowaway 92 (p2 of this thread, post #30 and post #48).

 

It is interesting that Hoya's introduction of FCD100, April 20, 2011, included the comment  "Combining it with optical glass of high refractive index allows us to achieve unprecedented advanced chromatic aberration correction."
<http://www.hoya-opti...ews/index.html>
The comment uses the same strong wording of high promise that Roland employed in announcing the new FCD100-based Stowaway 92  -- "unprecedented color correction."

 

Could FCD100, used with some appropriate mating type, or combination of mating types in a triplet, in fact, be better than either FPL53 or FPL55, with 'unprecedented' correction at least over the main visible range?

 

Comments pointing out errors in these conclusions and, particularly, ones noting any misunderstandings underlying errors, would be welcome and appreciated.

 

       Larry


  • jay.i likes this

#71 Vla

Vla

    Viking 1

  • *****
  • Posts: 611
  • Joined: 08 Mar 2014

Posted 19 September 2018 - 03:31 PM

My take:

1. I don't think Thomas Back made it up that AP started using ZKN7 in the EDT series with FPL52 and for the last runs used it with FPL53. It is also too specific to be an error. So if anyone is, directly or indirectly, saying that didn't happen, he should show something tangible to back it up. A few years ago I had a discussion here with some apo makers and retailers, and they did not convince me ZKN7 had insufficient quality. My impression (because they were not talking openly and specifically about all aspects involved) was that they tried to use lower quality ZKN7, perhaps even ZKN7-like (other than Schott) glasses, because the real thing, in needed quality, was expensive. On the other side, Schott was committed to ZKN7 as one of those glasses they plan on keeping in the future, which just wouldn't make sense if it is an inherently poor optical glass. So I certainly wouldn't wright it off on that basis. BK7 does have advantage of better price, and perhaps some other properties, but those are not strictly optical criteria.

 

Optically, BK7 is inferior match for FPL55 to ZKN7 primarily because of its significantly smaller Abbe# differential. Blue P(F,e) partial dispersion is not quite as close as with ZKN7, but the amount of secondary spectrum added because of that is negligible (its a bit closer violet P(g,e) dispersion is also near-negligible). But smaller Abbe differential means more of higher-order SA residua, and that is something affecting all wavelengths, including the optimized one (in the same direction goes its little higher refractive index, but that is also near-negligible). The two f/10 6-inch doublets below are illustrative of the difference in higher-order SA level between ZKN7 and BK7.

 

match.png

 

2. I can only repeat that the two systems sporting FPL53 and FPL55 with N-KZFS2 are for all practical purposes identical in performance level. No human eye would be ever able to tell the difference. A slight tweak could make either little worse than the other, and that would still be unnoticeable.

 

4. Hoya's line (I had to check it out) saying that FCD100 works gives exceptional performance when combined with glasses of high refractive index is a bad error. Unless they've come up with some new, revolutionary configuration. Otherwise, higher index glasses will do the same thing as low Abbe# differential, i.e. force on strong inner radii and higher-order spherical. Here's one illustration of what the higher refractive index does.

 

indx.png


Edited by Vla, 19 September 2018 - 03:36 PM.


#72 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 19 September 2018 - 04:49 PM

My take:

 

[full details in post above]

Great public thanks to Vladimir (Vla) once again -- I couldn't have hoped for a more complete, thorough, and informative response!

 

Maybe Alan will provide the basis for his statement that Roland never paired ZKN7 with FPL53. 

 

    Larry



#73 Mitrovarr

Mitrovarr

    Gemini

  • *****
  • Posts: 3,362
  • Joined: 12 Sep 2004
  • Loc: Boise, Idaho

Posted 17 September 2019 - 11:49 AM

Just a random note - I have a SW 150ED. There is some chromatic aberration. However, I've noticed that it all seems to be in the deep violet range (the main effect is changing the perceived colors of some very bright white stars, like Vega). I think this supports the idea that it might be using FPL-55.

Despite the CA it's a very nice, very sharp telescope and I do like it.

#74 LMO

LMO

    Vostok 1

  • *****
  • topic starter
  • Posts: 184
  • Joined: 15 Sep 2013
  • Loc: Salt Lake City, UT

Posted 03 July 2020 - 03:02 PM

With Roland's permission, I am quoting below, in its entirety, a July 1 post from him to the A-P Users' Group.  It responds unequivocally to the question posed  for this thread by my OP, and also covers some other interesting topics raised in the course of the thread.

 

    Larry

----------------------
From: "Roland Christen
Subject: Re: [ap-ug] AP 6" f12 Super Planetary NASA Glass?
Date: July 1, 2020 at 9:18:42 PM MDT

 

The original triplets used glass made in the US by Kodak and Bausch&Lomb for the NASA space program. Later models used Ohara equivalent to KzFSN4, but I never used K7 glass in any of our designs. We used a number of different outer elements over the years. The design achieved better and better color correction in later models which allowed good correction at faster focal ratios. Outer elements included BK7, FK5, BaFN10 and BaFN13 glasses.

When Ohara FPL52 became available we switched to ED Triplet designs. Eventually FPL53 supplanted FPL52 with slightly better color correction potential. In the last couple of years Ohara switched to FPL55 which has much better workability and hardness while maintaining the excellent color correction of FPL52 and FPL53. FPL53 is a very soft and fragile glass and is not generally available any more. Could any knowledgeable optician determine if a lens has FPL53 or FPL55? The answer is no.

 

Hoya has an equivalent glass, FCD100, which has slightly better color correction when used with specific crown glasses, and is now available in larger sizes also. I suspect the Chinese glass companies will make an equivalent soon in larger sizes (above 85mm dia), but I'm not sure if they can match the internal quality of Ohara and Hoya.

If anyone wants to try their hand at designing with those original KzFSN glasses, you can use ATMOS lens design, a very easy to use optical design program, which we sell for amateur and professional use. It has a very powerful optimization routine that can take any 3 glasses and come up with a triplet design in just minutes. Better than watching TV while being stuck at home during this pandemic ;^))

 

Rolando
----------------------


  • JKAstro, Jeff B, eros312 and 3 others like this

#75 Alan French

Alan French

    Night Owl

  • *****
  • Posts: 6,080
  • Joined: 28 Jan 2005
  • Loc: Upstate NY

Posted 03 July 2020 - 04:29 PM

Great public thanks to Vladimir (Vla) once again -- I couldn't have hoped for a more complete, thorough, and informative response!

 

Maybe Alan will provide the basis for his statement that Roland never paired ZKN7 with FPL53. 

 

    Larry

I'm glad Larry answered this, because I missed this long ago. :)

 

Clear skies, Alan




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