Quantitative and qualitative test of etalons with an Ha tube

[christian viladrich]

Hello,
The weather beeing so bad these last two months, this was the opportunity to take a plunge in the quantitative assessement of air-spaced etalons.
This work builds on discussions with Bob Yoeles, Oliver Smie (Beloptik), Serge Koutchmy (IAP) and Cyril Bazin (IAP).
But most of all, François Rouviere (co-author of Solar Astronomy) deserves the main credit and a big thank for establishing the formulae to calculate the FWHM, FSR and delta CWL of etalons from the mesurements.

The general line of the procedure is presented here. Sorry, the page still under work ...
http://astrosurf.com/viladrich/astro/in ... sting.html

The main interesting thing is there, with the measurement of a SM III-60 mm and a SM III-60 mm RichView etalon:
http://astrosurf.com/viladrich/astro/in ... ing-2.html
Both are excellent with a FWHM < 0.6 A, the SM III 60 mm behing slightly better and a thad more uniform.

This is preliminary, but still very encouraging. I plan to test two SM 90 mm in the coming month.

In a nutshell, the idea is to use the etalon as a Fabry-Perot interferometer, the etalon receiving diffuse light from an Ha light :


The system of interference fringes is observed visually (qualitative evaluation) or with a camera (for quantitative estimation).


After some least-square fitting with Lorenztian functions and some math, the FWM, FSR and delta CWL can be derived :-)

[christian viladrich]

Uniformity of the etalon over its aperture can be easily checked. Here a SM-40 mm etalon. Note the dip developping on the central fringe as we move away from the central spacer. This is due to a drift in CWL (moving away from Ha) :

[christian viladrich]

Another test (this one only qualitative for the time being), is a test using collimated light from an Ha lamp. The small refractor is used as a collimator :


Here is how two mica-spaced etalons look like :


The test is very sensitive to non uniformities (mica cleavage, striae, inclusions or refractive index inhomogeneitie). In fact, this looks more terrible that it actually is. Both these etalons are very uniform at the eyepiece !

[EGRAY_OBSERVATORY]

Hi Christian, can you attach those images as JPEG's/jpg's please as others standards can not be viewed via SolarChat.

Many thanks

Terry

[Bob Yoesle]

Hi Christian,

A possible caveat - despite the high quality etalon fringes - the SMII and SMIII etalons are clearly tuned to have high CWL's. Compare them to your Tucson era Coronado 40 mm SM etalon, which as indicated by the small central ring is very nearly on-band, and therefore requires very little tilt when actually being used.

Here's one of my Tucson era SM90's, and the central ring is also relatively small and shows minimal change when moving off axis. This results in a uniform etalon gap and excellent contrast performance, and due to minimal tilt to come on-band, no banding.

SM90 Tucson fringes.jpg (392.08 KiB) Viewed 237 times

With the SMII and SMIII etalons, when actually used for solar observation, their high CWL requires a greater amount of tilt - which can produce banding. In actual use the high CWL also requires a significant amount of rich view tuning mechanical pressure, which appears to be applied only to the central spacer. From what I have observed this deforms the etalon gap to be significantly non-uniform, reflected in a greatly varying central ring diameter on axis versus off axis:

SM90II fringes.jpg (586.36 KiB) Viewed 237 times

Left - SM90II uncompressed with large central ring = CWL very off-band; Center - compression applied to center > central ring compresses to a dot = center on-band; Right - same compression as Center image - off axis, widened central ring and significantly off-band.

This non-uniform etalon gap results in a widened FWHM, and significantly affected overall contrast performance. Therefore in my experience, despite their excellent un-tilted and un-compressed etalon fringes, the introduction of excessive tilt and/or rich view compression renders their performance compromised when compared to the earlier minimally tilt-tuned versions:

Closu compare SM.jpg (470.76 KiB) Viewed 237 times

SM90II RichView on Left, 2003 era SM90 on right.

[DeepSolar64]

Bob,
It seems that it's pretty much agreed upon that the Coronado Tuscon etalons are better than the later Meade Coronado ones. Your tests seem to confirm this. For me I am way to late in the game so I have Meade ones. I would like to look through an old Tuscon one. I just wonder how much the difference really is.

James

[christian viladrich]

Sorry for the images ...
Here is the set up for the test in diffuse light:
http://astrosurf.com/viladrich/astro/in ... ASI290.jpg

An example of interference fringes with the SM III 60 mm :
http://astrosurf.com/viladrich/astro/in ... .8-dng.png

[christian viladrich]

Sequence of images showing slight non uniformity of the CWL accross the aperture of a SM 40 etalon :
http://astrosurf.com/viladrich/astro/in ... ale4mm.png

[christian viladrich]

Optical setup for testing etalons in collimated beam :
http://astrosurf.com/viladrich/astro/in ... -setup.jpg

Results for two mica-spaced etalons :
http://astrosurf.com/viladrich/astro/in ... 94-410.jpg

[christian viladrich]

Hi Christian,

A possible caveat - despite the high quality etalon fringes - the SMII and SMIII etalons are clearly tuned to have high CWL's. Compare them to your Tucson era Coronado 40 mm SM etalon, which as indicated by the small central ring is very nearly on-band, and therefore requires very little tilt when actually being used.

Hi Bob,
Yes indeed, the SM III 60 is 0.6 A from Ha, and requires a tilt of 0.77° to be set on Ha.
The SM III 60 Rich View is 1.25 A away from Ha (at minimum correction), and requires a tilt of 1.11° to be set on Ha.

Both etalons present a slight (red) shift in CWL accross their aperture from the central spacer to the outer rim. This shift is very symetrical on the SM III and a little bit less on the SM III 60 mm RichView.
I haven't made any measure with the RichView fully compressed since it would mean going further away from Ha (to the red), which is unecessary in real life.

It could be that the smaller SM III 60 RichView is more symetrical the the larger SM III 90 mm RichView (better balance of the pressure). I hope to have a classic SM 900 in test soon, but no SM III 90 RichView.

In any case, as you've pointed out, larger delta CWL means larger tilt to tune the etalon on Ha, which means more banding.
Maybe, it is easier to acheive a good accuracy in FWHM compared to CWL. This is the case for hard coated filters. So, it might be a manufacturing choice. There is no free lunch ...

[EGRAY_OBSERVATORY]

Thanks Christian.

Terry

[Bob Yoesle]

I would like to look through an old Tuscon one. I just wonder how much the difference really is.

Hi James,

I don’t know if there will be an in-person NEAF this year, but there are usually a few original Tucson SM filters in attendance. If you’re ever out on the west coast, please feel free to get ahold of me for a demo.

Maybe, it is easier to acheive a good accuracy in FWHM compared to CWL. This is the case for hard coated filters. So, it might be a manufacturing choice. There is no free lunch ...

Hi Christian,

I think you are right, and getting a precise thickness of spacers is apparently a bit more difficult than getting uniform thickness. This probably required much more time and effort devoted to spacer production, and may have been a contributor to a Tucson SM90 costing more than double what a MEADE SM90II or III etalon costs.

I think the more stringent thickness requirement is one of the reasons that Lunt moved away from tilt-tuned internal etalons. Being in a collimated lens system with magnified field angels, they needed to be almost on band and have minimal tilt to perform well. Air pressure tuning also has the advantage of being less sensitive to spacer thickness variation etalon-to-etalon, and compared to mechanical Rich View tuning, does not affect the etalon gap because the tuning is accomplished by uniform air density changes, not varying mechanical compression which can change the etalon gap spacing.

There is also a significant difference in the mechanical support to the etalon edges, as the Tucson SM90 had silicone support “buttons” and a ‘cleaner’ base support. The SM90II etalon plates have essentially no radial support, as the threaded sleeve which changes the Rich View tuning lies along the side of the top etalon plate. This may make the etalon more susceptible to de-contacting, and shows that Rich View pressure is only applied to the etalon center, thereby resulting in an unequal compression and change of the etalon gap thickness in the center versus the edge.

SM90 v SM90II.jpg (238.1 KiB) Viewed 199 times

I know for a fact that this type of mechanical pressure tuning was used for some internal etalon "modules" implemented by Tucson Coronado, but this was with etalons that probably had a CWL very near to being on-band and required very little additional mechanical pressure to be tuned to be exactly on-band.

[christian viladrich]

Hi Bob,
Have you seen anything special between the two plates of the SM III 90 RichView etalon ? Is there a central spacer ? Do you see the lateral spacers ? Is there any pressure applied to the lateral spacers though the RichView mechanism ? There was a Lunt patent similar to this.
I don't think the etalon plate is bent. My guess is that all the spacers are compressed. This is just a guess.
This beeing said, it might that the compression forces on the central and lateral spacers are not well balanced.

[marktownley]

A gold standard thread gents!

[Bob Yoesle]

Hi Bob,
Have you seen anything special between the two plates of the SM III 90 RichView etalon ? Is there a central spacer ? Do you see the lateral spacers ? Is there any pressure applied to the lateral spacers though the RichView mechanism ? There was a Lunt patent similar to this.
I don't think the etalon plate is bent. My guess is that all the spacers are compressed.

Hi Christian,

I don't have a SMIII etalon. From your photos it appears the newer SM90III telescopes use a non-RichView tilt-tuned etalon as the primary etalon, and a tilt and RichView III etalon (which I assume is functionally/mechanically identical the the RVII) as the double stacking etalon.

There is nothing special between the SM90II RichView etalon plates besides the typical-looking spacer chips.

There is a single central spacer within the uncoated central region, and the typical rear occulting foil disc opposite of the area of the front etalon where the central pressure bolt applies mechanical pressure to the center of the front etalon plate.

There are 9 peripheral (lateral) spacers which appear identical to every other etalon I have examined.

None of the peripheral spacers have any pressure applied to them throughout the range of compression applied via the central pressure bolt/pin:

SM90II RV disassembled.jpg (485.46 KiB) Viewed 164 times

Note the central pin (with plastic insert tip) extends about 10 mm above the "shelf" of the internal portion of the ERF cell / RichView tuning mechanism.

The SM90II tilt and compression RichView tuning therefore may have a requirement to be both tilted and have the central spacer compressed due to the high CWL, which may potentially produce both uneven contrast due to banding if excessive tilt is required, and a widened band-pass if excess compression on the central spacer is applied - D below (greatly exaggerated for clarity):

Etalon Gaps & tuning SM.jpg (48.62 KiB) Viewed 140 times

A - A large unobstructed etalon with a thermally unstable uneven etalon gap.

B - A large etalon with a central spacer to stabilize the etalon gap for greater thermal stability.

C - A large high CWL etalon with with central pressure tuning resulting in an uneven etalon gap.

D - As in C with the additional need for tilt tuning.

The other RV etalon I have is a small internal RichView etalon, and identical with regard to the pressure being applied only by the center pin attached to the collimator lens, but it is not tilted.

Rich view etalon.jpg (131.5 KiB) Viewed 87 times

It is my understanding that the PST etalon is either tilt-tuned only (early models?), or has peripheral compression applied and no tilting (later models?).

[DeepSolar64]

Hi James,

I don’t know if there will be an in-person NEAF this year, but there are usually a few original Tucson SM filters in attendance. If you’re ever out on the west coast, please feel free to get ahold of me for a demo._Bob Yoesle

Thanks Bob for the offer. I f I ever make it out there I will look you up.

[Bob Yoesle]

Yes indeed, the SM III 60 is 0.6 A from Ha, and requires a tilt of 0.77° to be set on Ha.
The SM III 60 Rich View is 1.25 A away from Ha (at minimum correction), and requires a tilt of 1.11° to be set on Ha.

Hi Christian - Just some additional thoughts, presumptions, and observations...

For a 0.7 Å air-spaced etalon the Jacquinot spot is 0.99° in diameter. This results in a field angle exceeding ~ 0.5° will begin to be shifted noticeably off-band. This would seem to imply that a tilt-tuned etalon should not have a CWL tuned so high that the tilt needed to bring it on-band would exceed about 1.0° or it may give rise to noticeable banding.

Therefore the SM60III looks like it will perform OK as primary etalon, but the SM60III RV would not due to the excess tilt needed to bring it on-band. It would need to use only enough tilt to remove the ghost reflections, and use the RV compression tuning to get it on-band. However, it appears that just as with my SM90II RV, the compression needed to come on-band would likely (if it is indeed the same as the SMII RV) only be applied to the central spacer. This results in an uneven etalon gap and a poor overall FWHM of about 1Å +/- .

However, in my experience - used for double stacking alone - this etalon will achieve some desired narrowing of the band-pass of the double stacked system - it's just not nearly as good as it otherwise could have been. The RV etalon used alone, however, is clearly deficient compared to a non-RV tuned etalon using a lesser amount of required tilt-tuning to be brought on-band.

[p_zetner]

Thanks for this informative thread!
Very enjoyable reading.
Cheers.
Peter