Jump to content


Photo

DIY 127mm Petzval

  • Please log in to reply
11 replies to this topic

#1 Chemisttree

Chemisttree

    Explorer 1

  • *****
  • Posts: 82
  • Joined: 05 Feb 2008
  • Loc: San Antonio, TX

Posted 16 August 2013 - 03:20 PM

I'm playing around with a Petzval design consisting of a SS 127mm f9.4 objective and an 80mm ED f6.8 doublet. Using the formula for effective focal length of two doublets as,

EFL = (f1*f2)/(f1+f2-d) (plossl efl formula)

Where:
f1 = focal length of 127 mm objective (1200 mm)
f2 = focal length of 80 mm ED doublet (545 mm)
d = distance between elements (450 mm nominally)

I can use these two doublets to construct a 127 mm f4.8 widefield... maybe. I could change the spacing (d) and build an f3.8 without vignetting the rearward ED element but I'm worried about aberrations if I go with too fast a scope.

The ED doublet came out of my AstroTech 80 mm ED with FPL-51 glass. I'm not sure of the roc on any of the surfaces or of the spacing between the elements.

What do you guys think? How do I calculate SA and CA for this setup?

#2 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10654
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 16 August 2013 - 04:01 PM

At first blush, this would seem to be merely a focally reduced telescope. A rear doublet designed for an infinite conjugate (as telescope objectives are) must introduce aberrations in this configuration. And perhaps worse field curvature.

#3 Chemisttree

Chemisttree

    Explorer 1

  • *****
  • Posts: 82
  • Joined: 05 Feb 2008
  • Loc: San Antonio, TX

Posted 16 August 2013 - 06:41 PM

Yes, I'm sure you're right. I'm hoping the f9.4 light cone isn't too far from ideal. I'll put it on a stick and see what happens.

#4 gatorengineer

gatorengineer

    Mercury-Atlas

  • *****
  • Posts: 2532
  • Joined: 28 Feb 2005
  • Loc: Hellertown, PA

Posted 17 August 2013 - 01:01 PM

Please keep us posted on this experiment.... I have had alot of similar wonderings over the years.... One of the 6" F8's working at around F5 would be incredible.....

#5 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43894
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 18 August 2013 - 03:23 AM

Yes, I'm sure you're right. I'm hoping the f9.4 light cone isn't too far from ideal. I'll put it on a stick and see what happens.


It's not the F/9.4 light cone you are concerned about, it's the effect of the doublet on that light cone and how the two work together. In general it seems that focal reducers are also field flatteners, I am not sure how the math works for a flattener.

Jon

#6 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10654
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 18 August 2013 - 08:47 AM

Jon,
A focal reducer is a flattener too only when designed to do that. Simply adapting an objective lens, even if by itself produces a flat image, will likely only worsen the existing field curvature of the front objective. Or if the front objective is also flat, there may result some field curvature with the combination.

#7 Jon Isaacs

Jon Isaacs

    ISS

  • *****
  • Posts: 43894
  • Joined: 16 Jun 2004
  • Loc: San Diego and Boulevard, CA

Posted 18 August 2013 - 10:44 AM

Jon,
A focal reducer is a flattener too only when designed to do that. Simply adapting an objective lens, even if by itself produces a flat image, will likely only worsen the existing field curvature of the front objective. Or if the front objective is also flat, there may result some field curvature with the combination.


Glenn:

I fooled around with binocular objectives in an attempt to make a focal reducer for my 80mm F/7, my goal was F/5. The reducer part was relatively easy but the field was messy and my ST-80 with the 2 inch focuser did a better job.

So, any idea of what sort of optics reducers/flatteners use?

Jon

#8 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10654
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 18 August 2013 - 11:54 AM

A first stab at flattening a field is sometimes undertaken by 'bending' a lens in the design phase. That is, the basic power of the elements is retained, but one or more is overall bent into a meniscus form. But the characteristics of the objective drive the specifics, And of course there's more than one way to skin a cat; different designs can achieve much the same result.

#9 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1613
  • Joined: 02 Mar 2007

Posted 19 August 2013 - 08:49 AM

CA will be determined by the achro lens, since ED lens won't correct CA induced by some other element. It would be waste of ED doublet to use it in such arrangement.

The smaller doublet, placed in the converging cone, will also generate spherical aberration, coma and astigmatism that it doesn't in a collimated cone, for which it is designed. To calculate the aberrations, it is needed to know lens radii. If they are both Fraunhofer type, which is likely, they can be closely approximated based on the f.l.

Vla

#10 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1613
  • Joined: 02 Mar 2007

Posted 19 August 2013 - 09:36 AM

Field can be flattened with a small positive doublet placed in the converging cone of the main lens. While both doublets will contribute Petzval curvature of the same sign, the rear doublet can be shaped up and positioned so that it induces astigmaticm which flattens out their combined Petzval (in general by inducing astigmatism opposite in sign and somewhat stronger than that of the front doublet). Here's an example: a 100mm f/10 front doublet with a 400mm f.l. positive doublet placed at 744mm from its rear surface.

The front doublet is very close to the regular Fraunhofer, only with slightly modified 1st and 4th radius, to induce some coma to offset slight residual at the second doublet (less than 0.1 wave RMS 1/2 degree off axis). The rear doublet, however, wouldn't work well in collimated light (3/4 wave p-v spherical and quite a bit of coma).

This arrangement reduces secondary spectrum by about 30% vs. standard achromat.

la

Attached Files



#11 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10654
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 19 August 2013 - 10:49 AM

Vla,
I'm not clear; is that 'recipe' for a front Fraunhofer modified or not?

How is the off-axis performance compared to the front doublet only?

How about for semi-angles larger than 0.5 degree (which is rather restrictive, given that one of the desired attributes is a larger field)?

The savings in tube length are not notable, given the placement of the rear doublet near to the final focus.

#12 wh48gs

wh48gs

    Surveyor 1

  • *****
  • Posts: 1613
  • Joined: 02 Mar 2007

Posted 19 August 2013 - 07:31 PM

Glenn,

Like I said, front doublet is very close to the Fraunhofer, the only difference is that the 1st and 4th radius are deliberately slightly out of balance needed to fully correct for coma. This means if this particular rear doublet would've been used with the Fraunhofer, which is fully corrected for coma, the combo would have some residual coma (about f/7 paraboloid level). I guess I induced coma to the front doublet because it is the easiest way to correct for residual coma in designing, but I expect it could be done with adjusting the rear doublet as well.

The purpose was to illustrate what kind of positive doublet (reducer) will flatten field of an achromat, and why a doublet corrected for object at infinity won't do a good job. This simple doublet reduces f.l., flattens the field, reduces chromatism and astigmatism about 30% vs. comparable achromat. 1-degree field well within diffraction limited minimum means that there will be no noticeable abberations in the eyepiece. The astigmatic blur would reach 0.025mm at 16mm off axis, which is fairly good for imaging as well. For comparison, standard achromat's blur that far off on flat field is about 0.05x0.15mm.

Disadvantage is more complicated alignment and greater sensitivity to miscolimation. I believe Vixen had this kind of achromat.

Vla






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics