H'alpha - double pass module

Catalin Fus · Post by **Catalin Fus** » Fri Mar 09, 2012 5:18 pm

Hi everyone!!

I'm trying to put together a h'alpha system based on this :

http://tinyurl.com/7r67kc3

and I'll have some questions related to the ERF filter or pair of filters that do the same job. First of all, this is the configuration I've thought of :

ERF - Ø102mm maximum - RG610/RG630 Schott glass
102mm f6.3 apo
1.25'' or 2'' KG3 filter
2'' 2x / 4x Powermate
35mm 50/50 beamsplitter (AR coated)
H'alpha filter in a mount that permits 3degr tilt
30mm flat mirror ~ 92% reflectivity.

Do you feel that the ERF + KG3 filter combination is safe enough?
Can the ERF be replaced with a simple red filter?

Thanks in advance to all of you who will reply :-)!!

Any other suggestions regarding the system, it's function or safety are gladly welcomed!!

Cata

Post by **Merlin66** » Fri Mar 09, 2012 7:25 pm

Welcome on board!
Certaintly an interesting concept. A double pass of a 1.5A Ha filter...
I haven't seen this configuration actually working. In the very early Lunt pre-production designs there was talk of a similar double pass arrangement being proposed, but this obviously didn't make it to production.
(There's a guy in Germany who sells 2A filters which can be tilt stacked to give a much narrower bandwidth.....)
The RG610/ KG3 will make a good ERF, the problem will probably lie in the tuning arrangement.
Keep us posted on developments.
(EDIT - corrected the mis-read! should have been 1.5A)

Catalin Fus · Post by **Catalin Fus** » Fri Mar 09, 2012 7:55 pm

thank you!!!

the filter is 0.15nm => 1.5A. Am I wrong? I guess is a double pass through a 1.5A filter, not 15A.

I've talked to the optician at Omega (Robert Johnson) and he said that if I keep everything aligned, there should be no problem.
The tilting should be easy to make

.
He has also the double filter system for sale, the one used by Dave Groski...but I think it poses greater challenge for tuning than this double pass unit, where only one piece is moving/has to be adjusted, not two.

My bet is that if I can make everything aligned correctly, should be no problem. I will see :-)!!

Right now I'm waiting for the flat mirror and the beamsplitter to be made.

If anyone would be interested, after I finish the project (hopefully it will be successful), I can give plans for the parts.

Cata

Post by **Merlin66** » Fri Mar 09, 2012 8:01 pm

Whoooops - you are correct! Mis read the spec.
Normally to get an on-band double stack one filter has to be on the blue side and the other on the red side, the "interference"/ overlap gives the narrowband outcome.
Tilting filters in an optical beam normally shifts the bandwidth towards the blue.

Catalin Fus · Post by **Catalin Fus** » Fri Mar 09, 2012 8:08 pm

What I can understand from your reply is that a double filter system should provide a narrower band than a double pass unit?

Cata

Post by **Merlin66** » Fri Mar 09, 2012 8:21 pm

Cata,
No. They are basically using the same interference method.
I use a couple of 0.7A SM60 front etalons in a tuned double stack arrangement and can get down to 0.4A - using tilt filters (if the centre bandwidth is initially slightly greater than the Ha wavelength) would give similar results.

See message #5849 in my other thread...
http://solarchat.natca.net/index.php/en ... 60-on-ed80

Catalin Fus · Post by **Catalin Fus** » Fri Mar 09, 2012 8:29 pm

OK!! Understood! Thanks for the info

!!

Post by **Merlin66** » Fri Mar 09, 2012 8:41 pm

http://en.beloptik.de/index.html
This is Oliver Smie's website.
Looking at the optiacl arrangement with the telecentric barlow, the light through the beamsplitter and onto the filter must be non-collimated - to allow the image to come to a focus....
I'd think this beam may have to be f30 or greater to avoid issues with the filter coating transmission?????

(EDIT: I'm very surprised to see he has raised a patent on the double filter stack!!! We've been using very similar systems for the last few years!! - See his download section...)

Catalin Fus · Post by **Catalin Fus** » Fri Mar 09, 2012 9:03 pm

I guess so. In the dialogue I had with Bob from Omega Optical, he didn't mentioned about such a f-number to be reached....but my logic says you are right and I should design / use adequate telescope and amplification.

My options are a bigger aperture OTA, stopped to 100mm by ERF... or only 4x / 5x Powermates / Meade TeleXtender.

He mentioned that in order to keep the designated spectral function , the filter MUST work in parallel rays.

Well, I've seen the patent and I can not comment on that as we live in a free world. Dave Groski made it years before him, but not in that way. He might be afraid somebody will stole his idea and start selling same design units for profit.

Cata

Post by **Merlin66** » Fri Mar 09, 2012 9:10 pm

Cata,
If a proper collimating system is used prior to the filters then a refocusing lens could be used after the filters to give a better outcome

Bob Yoesle · Post by **Bob Yoesle** » Fri Mar 09, 2012 9:11 pm

“Do you feel that the ERF + KG3 filter combination is safe enough?
Can the ERF be replaced with a simple red filter?”

The Schott RG filters are a simple red glass filters - and needed to block UV. However, they are useless for IR, and that is where the Schott KG filter comes in for safety. However, if you’re not using a Baader DERF or Lunt ERF, the small KG filter could have a considerable heat load, which may have to be dissipated, and the KG glass can only go up to a certain temp ~ 300 F or so I believe without possible failure. So aperture may be an issue with using simple RG glass - but 102 mm may be fine. Think possibly of ventilation.

Using a TeleVue powermate will be better than a simple standard barlow, but it will not be a true telecentric optical system, and therefore you likely will not have uniform contrast performance over a full solar disc view. A true telecentric system would have to be custom designed specifically for the objective’s focal length and focal ratio. Having a double pass system will likely compound this issue.

swisswalter · Post by **swisswalter** » Fri Mar 09, 2012 9:19 pm

Hi Ken

interesting point. Is that also true for a PST mod? Will it be better to put such a lens after the PST etalon in the optical train ?

Post by **Merlin66** » Fri Mar 09, 2012 9:28 pm

No.
If you use the standard PST etalon assembly it already has the barlow, for collimating, up front and a 20mm f10 re-focusing lens at the rear.
just "plug and play"

Catalin Fus · Post by **Catalin Fus** » Fri Mar 09, 2012 9:49 pm

@ Bob Yoesle - Thanks for the input! I was thinking seriously of buying an RG filter and apply a coating to reflect/cut IR above 700 - 750nm.
I can do it (have access to an optical company) but in this case, do I need another filter in the system (like KG3) ?
Also, I can have designed a special Telecentric unit for the telescope I'll use but I would strongly prefer some other solution in case I switch refractors....I would like the unit to be usable on multiple aperture/focal ratios.

@Merlin66 - do you feel there is 'an issue' in the design of the 'patent pending' unit and it needs another lens for re-focusing?

Cata

Post by **Merlin66** » Fri Mar 09, 2012 9:56 pm

IMHO yes, I'd definately have a fully collimated beam through the filters and use a re-focusing lens to image.

Catalin Fus · Post by **Catalin Fus** » Fri Mar 09, 2012 10:05 pm

One quick question

I've found while browsing the internet, this Heliopan

http://tinyurl.com/7yewly6

is this what is needed as ERF? looks like Schott RG645 with IR treatment.

I see there are many sizes available and surface quality should be good as it shouldn't affect the picture quality

swisswalter · Post by **swisswalter** » Fri Mar 09, 2012 10:05 pm

No.
If you use the standard PST etalon assembly it already has the barlow, for collimating, up front and a 20mm f10 re-focusing lens at the rear.
just "plug and play"

Hi Ken

thank you very much

Post by **Merlin66** » Fri Mar 09, 2012 10:40 pm

I just remembered Marcello Lugli and his Cromix Ha filters!
http://digilander.libero.it/malug/Inglese/Cromixsun.htm
There's a certain similarity!!

Post by **marktownley** » Sat Mar 10, 2012 12:05 am

Yes, seen these 'other' filters from Omega. I have been doing some prototyping with Omega Ha filters for some time now. I look forward to seeing how your project develops.

Parallel or collimated beam is a must for this to work, barlows will not, they are not telecentric, telecentrics will be better (than barlows) but you will still find sweet spotting issues.

I hope you don't think me blunt in saying it, but these omega ha systems are easy to build in theory, but are quite a bit harder to get to a working system that come close optically to commercially available ready mades...

Catalin Fus · Post by **Catalin Fus** » Sat Mar 10, 2012 8:57 am

Mark, I like tinkering

!!

I'm pretty sure it will not be a 'walk in the park' but I have to see myself how everything is working. If others have done it and used it with results, they might work.

Can you detail a little about the area where a system like this (double pass with single filter or double filter) can lose vs a commercial unit , with same 0.8A band?

Cata

Post by **marktownley** » Sat Mar 10, 2012 9:45 am

Mark, I like tinkering

!!

I'm pretty sure it will not be a 'walk in the park' but I have to see myself how everything is working. If others have done it and used it with results, they might work.

Can you detail a little about the area where a system like this (double pass with single filter or double filter) can lose vs a commercial unit , with same 0.8A band?

Cata

Hi Cata,

Tinkering is one of my past times too

The big difference I have found are lower transmission values at 656nm, contrast is not as good and reflections as a result of laminate layering particularly in the blocker element. These can be dealt with by means of circualar polarisers but this negatively effects throughput at 656nm...

I've tried a few optical configurations with the Omega Ha's and the best result i've had so far is mounting the etalon externally on a (stopped down) 6x30 finder, with the blocker tilted accordingly in the ep end. Whilst never going to set the world alight in terms of being a revelationary new scope it certainly has produced the best quality views i've got from the filters- I think principally because the etalon is receiving parallel (well f108) light as opposed to say f25 / f30 beams from either a stopped aperture or powermate induced telecentric beam. Much more even illumination and lack of sweetspotting in my 'finder scope' configuration... But, i graciouslly accept more of a toy than a serious solar telescope... :whistle:

Mark

Catalin Fus · Post by **Catalin Fus** » Sat Mar 10, 2012 1:47 pm

Hehehe...well, that is quite encouraging, right?

You made the double filter one?

What is the 'solar add-on' that you would recommend to someone who has refractors from 102mm to 160mm ? Should I think of one based on PST with BF10/BF15?

Thanks!
Cata

Bob Yoesle · Post by **Bob Yoesle** » Sat Mar 10, 2012 2:23 pm

“I was thinking seriously of buying an RG filter and apply a coating to reflect/cut IR above 700 - 750nm... I can do it (have access to an optical company) but in this case, do I need another filter in the system (like KG3) ?”

If you want IR blocking all the way to ~ 2500+ nm the answer is probably yes. Most standard dielectric AR coatings will not go that far, so an ITF (which uses metallic silver and dielectric layers) or KG type of glass filter will be needed for absolute safety..

“Also, I can have designed a special Telecentric unit for the telescope I'll use but I would strongly prefer some other solution in case I switch refractors....I would like the unit to be usable on multiple aperture/focal ratios.”

All depends on what you mean by “usable.” All the designs I have seen discussed here are usable. What I haven’t seen is an analysis of - or imaging performance over - a large field of view. I suspect they all suffer from some degree of sweet spot contrast non-uniformity and /or vignetting, as the physics of light is inviolate. If you really want to be able to use a H alpha filter on multiple apertures and focal ratios, you will need to use front mounted etalons.

Contrast performance depends greatly on instrument and field angles through the etalon, and with internal etalons these must be optimized via quite specific optical parameters. Unfortunately, there is no such thing as plug and play in this particular regard - collimators and telecentrics will perform optimally only with very tight adherence to specific parameters.

For example: A system involving using a collimator lens there are two items are necessary to achieve good contrast performance over a non-vignetted full disk image - the focal ratio of both the objective and the collimator should be close to the same (e.g. the collimator doesn’t act as a field stop for vignetting), and the ratio of the focal lengths of the objective and collimator is not greater than two or three to one to keep field angles acceptable through the etalon (< 0.75 degree).

This boils down to needing to use relatively large collimators of long focal length and large internal etalons. A PST mod (with a relatively small - 200 mm FL collimator) used with a 100/1000 objective will not have optimum full-disk contrast performance. The full disk field angle (half a solar diameter = 0.25 degree) will be increased to 1000 ÷ 200 = 5 x (0.25) = 1.25 degrees.

However, it will perform well for narrow field hi-res work, as we have seen many examples of superb hi-res imaging with these designs. Switching an H alpha “module” using a given collimator or telecentric between telescopes with different apertures and focal ratios (focal lengths) will therefore be problematic except for the case of narrow field observation and imaging.

Catalin Fus · Post by **Catalin Fus** » Sat Mar 10, 2012 2:55 pm

Bob, many thanks for your reply!

What I can understand from your answer, regarding the optics part is the following :

1 - I should use a collimator lens (the one placed internal, before the filter), that has a focal ratio FR ~ 6.3

2 - I should use the above with a focal length of ~ 210 to 320mm (635mm / 2 or 3), negative FL I guess.

This lens should be placed internal at its focal length into the light path, just like the PST one....

Hope it makes sense....I'm just beginning to understand what this involves and I hope I'm not bugging anyone around here with so many questions.

Thanks!
Cata

Bob Yoesle · Post by **Bob Yoesle** » Sat Mar 10, 2012 3:24 pm

Hi Cata,

Yes, that is correct. However, since you will now have a collimator that is significantly larger than your H alpha filter (~ 50 mm vs 25 mm) significant vignetting will have entered the picture.

DIY designs all depend on what compromises and trade off’s you are willing to have. There are no simple implementations to get all the factors to work out. Most of these designs therefore seem to end up giving you increased field angles and poorer full-disk contrast performance in exchange for larger aperture and better resolution at narrower fields of view.

Catalin Fus · Post by **Catalin Fus** » Sat Mar 10, 2012 3:58 pm

Bob,

but Sun's image is not smaller than 25mm? I mean, there should be vignetting, but the idea is if I can see it or not...

Will have to think , think and think again

!!

Thanks
Cata

Bob Yoesle · Post by **Bob Yoesle** » Sat Mar 10, 2012 6:39 pm

Hi Cata,

While I’m not an expert by any means with optics, I have learned a little from others in this regard...

We are not discussing the sun’s image diameter, as it has not yet been focused by a re-converging lens (again, this is not a telecentric based system).

Perhaps the illustration below will help. To make it easier to see the geometry involved, I have used a positive collimating lens instead of a negative lens, but the principles are identical.

First you can see the importance of the f ratios leading to identical light cones (i.e. similar triangles for full illumination).

Second, you can see the importance of using a larger and longer focal length collimating lens to better control (reduce) the field angles presented to the etalon.

Third, you can see that the etalon size acts as a field stop, so that using a 20 mm diameter etalon with a 50 mm collimating lens significantly stops down the collimated beam. This could significantly impact image brightness and possibly result in vignetting.

After the etalon we place a suitable re-converging lens to form an image of the sun, and the focal length of this lens determines the final image size.

This is to the best of my understanding anyway. I hope others will contribute if I have gone astray.

Post by **marktownley** » Sat Mar 10, 2012 7:26 pm

Good explanation there Bob, thanks!

Post by **Merlin66** » Sat Mar 10, 2012 8:45 pm

Bob,
In a fully collimated system the angles you show should not have an impact on the resulting parallel beam through the etalon - on axis.
However, when we consider linear plate scale and image size of the sun's image at the focal points, and trace the top and bottom extremes of the light path through the collimator and then through the etalon...that's where the impact of the aperture as an aperture stop comes in. This angular deviation does imapct on the size and positioning of the "sweet spot/ band". Likewise edge vignetting becomes an issue - loss of contrast etc.
(I have similar problems with my spectroscope designs...the collimator lens must match the telescope focal ratio, and the height of the slit (similar to the size of the solar image)causes off axis issues on the grating and it's performance)
Hope this helps.

Catalin Fus · Post by **Catalin Fus** » Sun Mar 11, 2012 12:50 pm

Well, I'm a little bit confused now

!!

I will sort out all infos I got from you and what I've talked with one optician....and will be back soon with a ray trace or something alike.

Many thanks for all your opinions!!

Cata

Bob Yoesle · Post by **Bob Yoesle** » Sun Mar 11, 2012 3:53 pm

You're welcome Mark and Hi Ken,

Thanks for your input!

Hi Cata, to help you and others understand the field angles (from the edge of the sun's disk) referred to, they are depicted by the red rays in this diagram adapted from Ken Kobayashi:

See this CN thread for additional discussion: http://www.cloudynights.com/ubbarchive/ ... t/all/vc/1

These are the field angle rays which must be optimized via the scope / collimator design to keep the sweet spot under control. As you note Ken, the on-axis rays are collimated, but the off-axis field rays are not. For an extended object such as the sun, this becomes a significant issue with etalon performance. The best contrast performance comes from front mounted etalons where the 0.25 degree field angel is maintained unchanged. Andy Lunt has a good discussion of internal etalon perfromance issues in this Lunt Blog post:

http://luntsolarsystems.com/blog/intern ... erformance

The 0.25 degree field angle will increase in the collimator based system by the ratio of the objective/collimator focal lengths. As noted in earlier, it appears the “rule of thumb” for the collimating lens' focal length is that it should be ~ ½ the objective’s focal length to keep the native 0.25 edge of the sun’s disk field angle from becoming so enlarged that it falls outside of the etalon’s bandpass for optimum contrast performance. The larger the focal length of the collimating lens, the larger in diameter it must become, and hence the etalon must become larger as well by the same ratio (~ ½ the objective diameter).

From what I have seen, this is borne out with commercial implementations. Solar scopes with small internal etalons have more pronounced sweet spots, and those that have larger collimator/etalons have less prominent sweet spots. The Lunt LS80 with an internal 50 mm clear aperture etalon appears to have no issues with sweet spots in the single stacked mode.

If full disk contrast uniformity is of lower priority, this requirement can be relaxed, and larger apertures for higher resolution narrow field use will be very functional with smaller collimators & etalons.

Post by **Merlin66** » Sun Mar 11, 2012 5:57 pm

Bob,
Thanks for that!
Yes, that's a better diagram for explaining the off-axis effects.
The choice of collimator, as you say, is critical. (BTW in the PST it is 50% of the objective focal length.)

colinsk · Post by **colinsk** » Sun Mar 11, 2012 6:04 pm

Designing a telecentric for you focal length is not an insurmountable task. You need to enderstand the part of an optics text that explains pupils. The entrance puple needs to be as far in front of the scope as possible. A few meters works but a 100 is better. Infinity is ideal.

Post by **Merlin66** » Sun Mar 11, 2012 8:49 pm

Hello Colin! What a small world!
A couple of observations/ comments..
1. Looking at Bob's diagram - if the collimating lens is a telecentric design and properly positioned then the red rays would be parallel to the optical axis, not inclined. The exit pupil of a telecentric, as Colin said is at infinity.
2. If the collimator is not not telecentric, the the red rays would deviate to cross the optical axis at the position of the "exit pupil". The "eye relief" ie the distance from the lens to the exit pupil can vary depending on the lens design (Think in terms of eyepieces..) a rough first guess is the focal length of the lens.
3. Assume the light beam is as shown, and the etalon plates are mounted between the collimator and the re-imaging lens. The path distance between the plates on axis will be the shortest, and the edge inclined red path the longest.
If the etalon is designed say for Ha +2A then the rays on axis will "pass" Ha +2A, the larger optical spacing (red beam) will cause the CWL to shift towards the blue; let's say Ha -2A. This would mean that the tuning would vary from Ha+2A at the centre and Ha-2A at the edge of the FOV - somewhere in the middle would be a "zone" at Ha - The sweet band/ spot.
4. This also says that the doppler shift will show the red wings closer to the centre than the edge of the FOV.
5. The size of the etalon will/ can cause vignetting and loss of contrast.

Re Blocking filters - ideally this should match the finesse of the etalon, typically 7-8A bandwidth. This would also allow tilting to show doppler shifts. The narrow band filters CATA is trialing would work as a BF, but ONLY on Ha - ie it would limit the tuning range.

Catalin Fus · Post by **Catalin Fus** » Sun Mar 11, 2012 8:54 pm

Thanks Bob for the material provided!! It was worth lecturing!!

Now, what I wanted to do (still want!!) is to place somekind of Telecentric Assembly, into the light path of the objective lens (my 102mm f6.3 in this case) to have the rays parallel. Is this impossible? This is the part where ' I'm confused ' because I never intended to use Dave's design and put the filters/the double pass filter after the focal point of the objective lens, in a collimated beam, using another two objective lenses for this purpose.
This would mean for me to start building everything from scratch and not worth the hassle to build another tube for this objective (102mm f6.3)...I could simply use another design, with an 80 - 100mm short FL lens and go from there with the math, just as discussed here

.

I asked a friend of mine, who designs and knows optics, to take a look over the data and see if he can work out a Telecentric Assembly, specific to the refractor I have.
He uses also a Thousand Oaks H'alpha filter with a refractor so he knows the whole problem with tilting, sweet spots, amplification to obtain f/30 etc etc.

As soon as I'll have the data, I'll gladly post it here.

Regarding the dimension of the filter, all I can do is think about buying a bigger one, from here :

http://www.lot-oriel.com/de/en/home/filters/

but first I have to see how it might work with the 25mm one from Omega, in double pass.

Cata

Post by **Merlin66** » Sun Mar 11, 2012 8:58 pm

Cata,
I'm sure you'd get good performance using a negative telecentric lens ie TV Powermate or the TS/ Baader versions.

Catalin Fus · Post by **Catalin Fus** » Sun Mar 11, 2012 9:04 pm

Merlin66,

what I do not understand now after all these messages, in the design proposed on Omega's store, is if I will need another lens, to bring back in focus the parallel rays that exit above the beamsplitter.

What's your idea?

Thanks
Cata

Post by **Merlin66** » Sun Mar 11, 2012 9:08 pm

IMHO, as I mentioned before, the filters should be in a parallel collimated beam and a re-imaging lens would be required......

Catalin Fus · Post by **Catalin Fus** » Sun Mar 11, 2012 9:14 pm

I asked again just to be sure. We have talked about both designs and I was afraid I haven't understood correctly!

Thanks again for your input!! Greatly appreciated!!

Cata

Post by **Merlin66** » Sun Mar 11, 2012 9:41 pm

Vanter,
No, I was just considering the Omega narrowband filter as proposed by Cata with the beamsplitter; similar to the two narrowband filter designs of Smie and Lugli.
The KG3 is a completely different problem/ issue....

Post by **marktownley** » Sun Mar 11, 2012 9:51 pm

The KG3 is a completely different problem/ issue....

Ha ha ha!

colinsk · Post by **colinsk** » Mon Mar 12, 2012 2:23 am

Hello Colin! What a small world!
A couple of observations/ comments..
1. Looking at Bob's diagram - if the collimating lens is a telecentric design and properly positioned then the red rays would be parallel to the optical axis, not inclined. The exit pupil of a telecentric, as Colin said is at infinity.
2. If the collimator is not not telecentric, the the red rays would deviate to cross the optical axis at the position of the "exit pupil". The "eye relief" ie the distance from the lens to the exit pupil can vary depending on the lens design (Think in terms of eyepieces..) a rough first guess is the focal length of the lens.
3. Assume the light beam is as shown, and the etalon plates are mounted between the collimator and the re-imaging lens. The path distance between the plates on axis will be the shortest, and the edge inclined red path the longest.
If the etalon is designed say for Ha +2A then the rays on axis will "pass" Ha +2A, the larger optical spacing (red beam) will cause the CWL to shift towards the blue; let's say Ha -2A. This would mean that the tuning would vary from Ha+2A at the centre and Ha-2A at the edge of the FOV - somewhere in the middle would be a "zone" at Ha - The sweet band/ spot.
4. This also says that the doppler shift will show the red wings closer to the centre than the edge of the FOV.
5. The size of the etalon will/ can cause vignetting and loss of contrast.

Re Blocking filters - ideally this should match the finesse of the etalon, typically 7-8A bandwidth. This would also allow tilting to show doppler shifts. The narrow band filters CATA is trialing would work as a BF, but ONLY on Ha - ie it would limit the tuning range.

You knew I would end up here eventually!

A note about studying telecentrics. There are two lens systems that are called telecentric. One has the exit pupil at infinity. They are use in robotic vision systems. They show a group of cylinders as having circular tops. A normal lens system would show the sides of the off axis cylinders. The telecentric only shows the tops. This is what you will find in most literature about telecentrics. It is important to know that the telecentric we need has the entrance pupil at infinity. This will keep the off axis rays with the same angle through the etalon as the center ray.

Also, the discussion above assumes an air spaced etalon with a refractive index of 1.000. If you have a gap material with a larger refractive index it is not as sensitive to off axis effects. This comes straight from the etalon equations. Mica is easy to fabricate and has a much higher index of refraction. >1.5

Bob Yoesle · Post by **Bob Yoesle** » Mon Mar 12, 2012 4:14 pm

I think that for a double pass system, the telecentric system might be a better route than a collimator based system, as the etalon positioning is very tolerant of variation near the focal plane in a telecentric system, and the field rays would appear to be the same leaving the etalon as entering.

See this link for a good discussion of telecentric optical design:

http://home.comcast.net/~g2baraff/Solar ... g/Tele.pdf

But again, designing custom optics for such a system would seem almost mandatory.

For the LS200 double pass system, Lunt appeared to use (at least to my untrained eye) a more complex system with both telecentric and collimator properties:

Catalin Fus · Post by **Catalin Fus** » Mon Mar 12, 2012 8:04 pm

Hi Bob,

thanks again

!!

This is the layout that I was thinking to do, but without the last collimator lens. Merlin66 added this idea later as the scheme seemed incomplete.

After the discussion here, I asked an optician friend of mine to try and design a Telecentric unit specific to my 635mm focal length refractor. We will see what is the result and what is doable.

After looking in the PDF link you gave, it seems that the SUN in the system I've been talking about is 18mm in diameter...so the 25mm filter would be enough.

Cata

colinsk · Post by **colinsk** » Tue Mar 13, 2012 1:02 am

The diameter the sun at the focal point of any system is found by the equation fl/107.

colinsk · Post by **colinsk** » Tue Mar 13, 2012 1:07 am

I think that for a double pass system, the telecentric system might be a better route than a collimator based system, as the etalon positioning is very tolerant of variation near the focal plane in a telecentric system, and the field rays would appear to be the same leaving the etalon as entering.

See this link for a good discussion of telecentric optical design:

http://home.comcast.net/~g2baraff/Solar ... g/Tele.pdf

But again, designing custom optics for such a system would seem almost mandatory.

For the LS200 double pass system, Lunt appeared to use (at least to my untrained eye) a more complex system with both telecentric and collimator properties:

You would also need 1/4 wave plates to avoid reflections as the flat surfaces could not be tilted. To see how to use 1/4 waveplates see my etalon article part II.

colinsk · Post by **colinsk** » Tue Mar 13, 2012 2:21 am

[quote]Hello Colin! What a small world!
A couple of observations/ comments..
1. Looking at Bob's diagram - if the collimating lens is a telecentric design and properly positioned then the red rays would be parallel to the optical axis, not inclined. The exit pupil of a telecentric, as Colin said is at infinity.
2. If the collimator is not not telecentric, the the red rays would deviate to cross the optical axis at the position of the "exit pupil". The "eye relief" ie the distance from the lens to the exit pupil can vary depending on the lens design (Think in terms of eyepieces..) a rough first guess is the focal length of the lens.
3. Assume the light beam is as shown, and the etalon plates are mounted between the collimator and the re-imaging lens. The path distance between the plates on axis will be the shortest, and the edge inclined red path the longest.
If the etalon is designed say for Ha +2A then the rays on axis will "pass" Ha +2A, the larger optical spacing (red beam) will cause the CWL to shift towards the blue; let's say Ha -2A. This would mean that the tuning would vary from Ha+2A at the centre and Ha-2A at the edge of the FOV - somewhere in the middle would be a "zone" at Ha - The sweet band/ spot.
4. This also says that the doppler shift will show the red wings closer to the centre than the edge of the FOV.
5. The size of the etalon will/ can cause vignetting and loss of contrast.

Re Blocking filters - ideally this should match the finesse of the etalon, typically 7-8A bandwidth. This would also allow tilting to show doppler shifts. The narrow band filters CATA is trialing would work as a BF, but ONLY on Ha - ie it would limit the tuning range.

You knew I would end up here eventually!

A note about studying telecentrics. There are two lens systems that are called telecentric. One has the exit pupil at infinity. They are use in robotic vision systems. They show a group of cylinders as having circular tops. A normal lens system would show the sides of the off axis cylinders. The telecentric only shows the tops. This is what you will find in most literature about telecentrics. It is important to know that the telecentric we need has the entrance pupil at infinity. This will keep the off axis rays with the same angle through the etalon as the center ray.

Also, the discussion above assumes an air spaced etalon with a refractive index of 1.000. If you have a gap material with a larger refractive index it is not as sensitive to off axis effects. This comes straight from the etalon equations. Mica is easy to fabricate and has a much higher index of refraction. >1.5

Ok, reading the posts from Mark on Cloudy nights I seem to be wrong so I have to go back to the textbooks. He knows more about this stuff than me and says that the exit pupil needs to be at negative infinity for an etalon.

See this thread:

http://www.cloudynights.com/ubbthreads/ ... /page/0/vi

It is too bad they deleted his images. They used to do that to me as well. When the thread was off the first page they would delete my drawings.

Bob Yoesle · Post by **Bob Yoesle** » Wed Mar 14, 2012 6:07 am

Ken Kobayashi had a very good description on CN of the exit pupil issues in collimator-based internal etalon systems:

http://www.cloudynights.com/ubbthreads/ ... in/2632129

“An optical system can take light from a point source (e.g. a star) and turn it into an arbitrarily small parallel beam. That's what a visual telescope does - the objective focuses the light into a point, then a collimating lens (aka "eyepiece") turns it back into a very narrow parallel beam. The diameter of the exit beam is called the exit pupil.

If you choose a collimating lens with a short focal length, you can make the exit pupil as small as you want. But at the same time the magnification of the system goes up. If you point it at the sun, the light from the limb will be tilted many degrees relative to the light from disc center. You may be on the H-alpha line at disc center, but many angstroms off at the limb.

So, if you want to use an etalon in a collimated beam and achieve full-sun coverage (or as large a FOV as possible), you want to use as low a magnification as possible. But that means the "exit pupil" (diameter of the collimated beam) becomes large. Which means you need a larger etalon.

The etalon places a limit on the aperture too. If you double the aperture and keep the same magnification, the exit pupil doubles in size, and you need to double the size of the etalon.”

Emphasis added.

Thus the use of a larger collimator/focal length will result in the best possible contrast uniformity over the largest possible or practical field of view. A relatively small short focal length collimator will "magnify" the field angles more, and not give a very large area of uniform contrast - resulting in a "sweet spot." Again, the etalon size should be matched to that of the collimator to avoid creating a field stop and vignetting - see the previous post regarding collimator geometry.

If you use a telecentric based system, you can use a smaller etalon - but the telecentric will likely have to be custom designed and fabricated. The cost of this would have to be weighed against using a larger etalon and more straightforward off-the-shelf collimating optics... although it might be possible to use off-the-shelf optics to make the telecentric as well - but this is way beyond my limited knowledge of telecentric design.

However, from what I am experiencing, building a simple collimator-based system from readily available optical components will require using a positive vs. negative collimator lens, as larger long-focus negative lenses are all but impossible to find. This type of system may require the use of folding mirrors, and requires taking great care to ensure proper optical alignment is maintained.

colinsk · Post by **colinsk** » Wed Mar 14, 2012 1:33 pm

That is why I am working on understanding the telecentric design. My goal is a 6" scope with a 40mm mica etalon and full disk views at .5A. Narrower views at .3A. I can simply request a custom telecentric and get on made but at a very large cost. So I am working on learning the design parameters so I can start running OSLO and get closer to a final design.

Negative Concave lenses are easy to 25mm but I will need them larger.

Post by **Merlin66** » Wed Mar 14, 2012 7:50 pm

Colin,
Check out Surplus Shed...they had a few "interesting" negative lenses available.