27 replies to this topic

[Matthew Paul]

Hello, 

I've never been able to get a Ronchi grating to function if the light source did not also traverse through the grating.

When testing at ROC or with a DPAC system, this is no problem as the light source can always be placed behind the grating and everything works great. But, in other tests such as those where the light source and grating are separated by feet of distance, I have never been able to get the grating to show lines under test. I can see the illumination, but when looking through the tester, all I get is a mess of diffraction. I've tried gratins from 133 to 200 lpi

For a little background, I've made several sets of optics, a Newtonian, Dall Kirkham, Hyperbolic Corrected Newtonian, and now I am making an 8" RC. 

 

With this test where the tester and light source are separated, at different conjugate foci, the standard knife edge test functions perfectly well -- and this is adequate. Though, I would like to get the Ronchi grating to work for me if possible.

I have tried to place multiple different slits of differing sizes over the light source without success. The only thing that I have not tried is to cut a small piece of the grating and place that indecently over the light source, and view through another grating at the tester. 

 

Could I call on the experience of those here and ask if you have any tips or tricks for testing through a Ronchi grating in this application? Am I missing something? 

Thank you, 
Matt

[MKV]

You need a coherent source to see the Ronchi bands (as in a star at infinity test). Ordinary LED  collimated beam with a narrow band filter should work.  A weak laser (less or equal to 1 mW in power) operating in non-lasing mode only. If you don't; know how to do that, stay away from lasers. 

[BGRE]

Never had a problem with Ronchi test on an actual star.

If the object/source and image distances are unequal the magnification will not be 1 and having identical object/source and image gratings will not work well. Ideally the image of the source grating should match the grating used to analyse the image beam.

 

Without a source grating there is an upper limit to the source size:

https://www.research...e_Ronchi_ruling

When the magnification differs from 1 the size of the source image with respect to the analysis grating should be used.

The effective source size should be a fraction of the grating period for good fringe visibility,

Edited by BGRE, 13 June 2025 - 08:10 AM.

[MKV]

Never had a problem with Ronchi test on an actual star.

The OP used artificial stars in autocollimation and at ROC. 

 

Starlight is coherent, as is a laser beam, hence the beam doesn't have to go through the grating in order for the bands to appear at the at the focus. 

 

When an incoherent source is used i the workshop, the source has to go through the grating, and the best way to make sure the grating is  oriented the same for the course and the grating to move together as a unit), even though this reduced resolution in half.. 

[davidc135]

I have a multi colour + white LED torch/flashlight behind a diffuser and approx 50 micron pinhole which works well in all colours where I place the grating at focus. In this arrangement the grating can easily be swapped for a knife edge or even an ep. The problem can be the poor quality of cheap gratings.

 

David

[MKV]

I have a multi colour + white LED torch/flashlight behind a diffuser and approx 50 micron pinhole which works well in all colours where I place the grating at focus. In this arrangement the grating can easily be swapped for a knife edge or even an ep. The problem can be the poor quality of cheap gratings.

Ronchi test is normally used with an extended and diffused light source.

[dan chaffee]

I have tried to place multiple different slits of differing sizes over the light source without success. The only thing that I have not tried is to cut a small piece of the grating and place that indecently over the light source, and view through another grating at the tester. 

 

When I made the slit narrow enough, I have had no problem seeing the 

lines at ROC on a knife edge tester. I was using approx. 100 lpn grating, which 

is probably on the coarse side, but it got the job done.   

[BGRE]

The OP used artificial stars in autocollimation and at ROC. 

 

Starlight is coherent, as is a laser beam, hence the beam doesn't have to go through the grating in order for the bands to appear at the at the focus. 

 

When an incoherent source is used i the workshop, the source has to go through the grating, and the best way to make sure the grating is  oriented the same for the course and the grating to move together as a unit), even though this reduced resolution in half.. 

Not true, read the paper I linked for example.

Taking advantage of the van Cittert-Zernike theorem and using a grating over the source was tried about 7 years after Ronchi's original tests.

Early versions used either a pinole or slit source.

A source with a long coherence length isn't required as the spacing between fringes is achromatic.

A relatively small source is required if a matched (source image to analysis grating) grating over the source isn't used.

[davidc135]

Ronchi test is normally used with an extended and diffused light source.

Here's a DPAC Ronchigram of an apo doublet using a pinhole source. Any lumpiness probably due to the mediocre 80 lpi grating.

 

Outside focus showing some sphero-chromatism and slight tde.

 

David

 

Edited by davidc135, 13 June 2025 - 05:29 PM.

[MKV]

Not true, read the paper I linked for example.

Taking advantage of the van Cittert-Zernike theorem and using a grating over the source was tried about 7 years after Ronchi's original tests.

Early versions used either a pinole or slit source.

A source with a long coherence length isn't required as the spacing between fringes is achromatic.

A relatively small source is required if a matched (source image to analysis grating) grating over the source isn't used.

The example you posted back in 2023 shows an extended source. If a pinhole or a slit were okay, why would Malacara not mention them?

 

https://www.cloudyni...ons/?p=12789267

Edited by MKV, 13 June 2025 - 08:31 PM.

[BGRE]

Covering the source with a suitable grating makes the fringe system much brighter as a larger source can be used than when a slit or a pinhole is used with a conventional incoherent source making it more convenient for most applications. IIRC Malacara does mention this advantage.

However, in some circumstances (when using either a real or an artificial star for example) using a grating over the source may not be convenient or useful particularly if the magnification from the source to its image isn't close to 1.

The fringes can also be more useful for analysis if the grating is such that only 2 beams are of interest. Using a cosinusoidal aperture rather than a rectangular one for each grating "ruling" is one method proposed by Joseph Braat et al.

[Dale Eason]

The example you posted back in 2023 shows an extended source. If a pinhole or a slit were okay, why would Malacara not mention them?

 

In the pdf version I have they do mention them and discuss that they are usually replaced by the screen itself.

[MKV]

In the pdf version I have they do mention them and discuss that they are usually replaced by the screen itself.

Thank you, Dale. I'll give it a try.

Mladen

[Gleb1964]

Starlight is coherent, as is a laser beam..

Not exactly.
There are spatial and temporal coherence. Star light has spatial coherence, but not temporal.
To make a separate source work with Ronchi grating one need match temporal coherence or source size with grating, taking into account magnification factor between source and image.

No spectral filtering is needed, as Ronchi test is achromatic. Using very narrow spectral band source would complicate test with odd diffraction effects.

A point source, a slit source or source covered with Ronchi grating would work. Source grating grating should be matched by magnification factor to the image grating.

Edited by Gleb1964, 14 June 2025 - 10:45 AM.

[MKV]

Not exactly. There are spatial and temporal coherence. Star light has spatial coherence, but not temporal.

I know that, Gleb. I said they are coherent because they are -- in their own way. When it comes to seeing Ronchi bands using the screen only at the eye/camera end (the source doesn't go through the screen, as illustrated), one can see the Ronchi bands. When I used a pinhole and an incoherent source (e.g. an LED), I didn't see any bands. Maybe I missed something. I will repeat the test. 

 

 

edit: relabeled for more clarity

Edited by MKV, 15 June 2025 - 11:32 AM.

[TG]

I used a beam splitter and a red laser with the lens removed, in DPAC mode, with the Ronchi grating placed just before my eye. Saw bands, np

Edited by TG, 14 June 2025 - 02:45 PM.

[RichA]

I know that, Gleb. I said they are coherent because they are -- in their own way. When it comes to seeing Ronchi bands using the screen only at the eye/camera end (the source doesn't go through the screen, as illustrated), one can see the Ronchi bands. When I used a pinhole and an incoherent source (e.g. an LED), I didn't see any bands. Maybe I missed something. I will repeat the test. 

 

ronchi no screen source.jpg

What if you put the ronchi screen on the front of the beam splitter.

[BGRE]

Back reflections from the Ronchi grating into the viewing arm may then be an issue.

 

Its also optically equivalent to placing matching gratings over the source and in the viewing arm which doesn't test the proposition that fringes can be seen when only a grating is used in the viewing arm provided the source is small enough (and brightt enough).

Edited by BGRE, 15 June 2025 - 02:18 AM.

[davidc135]

Back reflections from the Ronchi grating into the viewing arm may then be an issue.

 

Its also optically equivalent to placing matching gratings over the source and in the viewing arm which doesn't test the proposition that fringes can be seen when only a grating is used in the viewing arm provided the source is small enough (and brightt enough).

Doesn't post 9 above prove that proposition?

 

I used both a Geoptic 50 micron white pinhole and a crude, home made, similar diameter pinhole over the rgb and white LED torch with the same results. A beam-splitter directed the beam to the 50mm f8 objective and flat. A coarse 80 lpi grating was used.

 

But when I've tried the same thing using a 125 lpi grating, the result was blotchy and unusable. I put it down to poor grating quality but maybe the pinhole was too big?

 

David

Edited by davidc135, 15 June 2025 - 03:13 AM.

[BGRE]

Yes, as did Ronchi's original work circa 1922.

It just that Mladen hasn't managed to replicate that result.

[Gleb1964]

..

When I used a pinhole and an incoherent source (e.g. an LED), I didn't see any bands. Maybe I missed something. I will repeat the test. 

 

If you have "really" point source , which is below diffraction for given beam speed, you should be able to observe Ronchi bands without problem. 

Now imagine adding another incoherent point source with some offset. You would see two Ronchi patterns overlapping each other with reduced visibility. If your source big enough, it like comprising many incoherent point sources making many shifted Ronchi patterns, reducing patterns visibility or varnishing it entirely. 

 

If your extended source is coherent (or partially coherent), like when using laser illumination or narrow band filtered source, you may get Ronchi pattern, but it would be complicated with diffraction artefacts, like on picture below. 

 

Edited by Gleb1964, 15 June 2025 - 03:22 PM.

[Matthew Paul]

I tried a bunch of different things without much results. Grating in front of source, etc. I do have a small slit from my spectrometer that I'll try before just using the knife edge. 

 

Thanks for all the feedback, tips, and tricks. 

[MKV]

Doesn't post 9 above prove that proposition?

I don't see how. You covered your source with the screen (apparently without any diffusion). That's not what the test setup shows in #15. The test sketched in #15 is akin to testing on a star -- the Ronchi grating is only at the eye or camera position. The star light doesn;t pass through the grating before it reaching the test mirror. The grating is only positioned at the eye end (Ronchi eyepiece)

 

To test in an analogous way to a star, an autocollimation setup could be used as illustrated below

 

 

The obvious advantage in this case would be double sensitivity (double-pass null) compared to the actual  Ronchi star test (single-pass null).

 

In this case the outgoing diverging signal would never encounter the Ronchi screen, at it gets converted to a collimated (parallel) light beam, and becoming an extended source, which is then returned to the mirror focused as it passes through the Ronchi screen. The question remains: will an incoherent source (i.e. LED) work or not.

 

Another way would be to simply expand a point source (pinhole or a slit) into a collimated beam using a beam expander (such as a collimating telescope). This would actually represent a single-pass null arrangement we have in an actual star test.

 

BTW, if I don't use a diffuser with an LED (in a "standard" Ronchi test with the source and the eye end being coplanar and moving together) I usually get uneven illumination in my ronchigrams.

 

Mladen

[MKV]

If you have "really" point source , which I below diffraction for given beam speed, you should be able to observe Ronchi bands without problem. 

Well, I used a laser with the lens removed. The light cavity is pretty much Airy disc size, even if slightly elliptical, and I saw Ronchi bands without a problem.

 

[MKV]

I used a beam splitter and a red laser with the lens removed, in DPAC mode, with the Ronchi grating placed just before my eye. Saw bands, np

Ditto, but did you try with an LED light source?