The Caustic Test, From Amateur Telescope Making Book Three


Page 429 Page 430 Page 431 Page 432
caus01.gif (549179 bytes) caus02.gif (565079 bytes) caus03.gif (637014 bytes) caus04.gif (536087 bytes)
Page 433 Page 434 Page 435 Page 436
caus05.gif (595160 bytes) caus06.gif (581130 bytes) caus07.gif (607679 bytes) caus08.gif (570059 bytes)
Page 437 Page 438 Page 439 Page 440
caus09.gif (583894 bytes) caus10.gif (621421 bytes) caus11.gif (600512 bytes) caus12.gif (543009 bytes)
Page 441 Page 442 Page 443 Page 444
caus13.gif (599001 bytes) caus14.gif (481691 bytes) caus15.gif (569165 bytes) caus16.gif (614038 bytes)
Page 445 Page 446 Page 447 Page 448
caus17.gif (642767 bytes) caus18.gif (549221 bytes) caus19.gif (610654 bytes) caus20.gif (599340 bytes)
Page 449 Page 450 Page 451 Page 452
caus21.gif (620270 bytes) caus22.gif (614096 bytes) caus23.gif (577801 bytes) caus24.gif (608112 bytes)
Page 453 Page 454 Page 455 Page 456
caus25.gif (562327 bytes) caus26.gif (557293 bytes) caus27.gif (507946 bytes) caus28.gif (215436 bytes)

There is an Excel spread sheet that seems to be an accessory program for the above chapter of 'ATM Book Three'.  
The Excel spread sheet is by Ken Hunter.  The link for it is at


Bill Thomas uses a CCD to capture images of the caustic.  His test bypasses some of the difficulties of the standard wire test used for caustic measurements.
Find the original at:




I tried to use my home built tester in place of the 'ding bat' shown on page 445 in 'ATM Book Three'.  I can tell you that the 'ding bat' will need to rest on something better than the average camera tripod to get repeatable measurements to less than .001 inch.

I tried a possible method using the grating lines in place of the left/right drive on the 'ding bat'. Then I would need to write a formula to work the caustic formula in reverse to find a radius on the mirror that satisfies the fixed grating line locations. Then the mirror mask instead of having regularly spaced holes would have holes as prescribed by the reversed calculation.  Although I never followed through with this I did discover a method to clear up multi imaging while using the Ronchi grating.

causline.jpg (75216 bytes) I masked off all but one line in the 100 line per inch grating on its lower half using vinyl electrical tape and a toy microscope.  Now the light source is a single slit instead of a multi slit. Doing this improves the resolution of the Ronchi test, particularly when working in the sensitive range between minimum and maximum radius of curvature of the mirror under test.  



Another Tester called the DAFT tester uses an eyepiece to view the image of a pin hole ( not a slit ).  A string is placed in front of the mirror mask openings to produce a diffraction pattern.  If I understand the description, the test finds the same locations longitudinally ( to and from the mirror ) as the Foucault test.  So if you have a sturdy Foucault tester already it can be adapted to the DAFT test.  The home built tester design given in this program is not sturdy enough to carry the extra lenses required. See: daft.htm.   The link for this article is at:

The Dall Null Test, From Amateur Telescope Making Book Three

page 149 page 150 page 151 page 152 page 153
dall01.gif (288163 bytes) dall02.gif (299055 bytes) dall03.gif (276031 bytes) dall04.gif (262493 bytes) dall05.gif (277088 bytes)

Another way to test precisely is to put a lens between the tester and the mirror.  This can cancel the spherical aberration caused by testing at the center of curvature.  The test is then a 'null' test as if the mirror were a sphere. 


There is no point in setting up the above tests unless the mirror is already corrected to better than half wave since correcting to a half wave can be done more easily using a Ronchi program printout.  If you have more time than money then best to just wait for good testing nights and use an LED for an artificial star and the knife edge, and eyepiece tests to finish .  Likely windless, overcast nights will be best.  Be advised that different combinations of eyepiece and Barlow can produce spherical aberration.  People that use the eyepiece test all the time usually have a favorite high power eyepiece for testing.


Back to mirror testing page


Advanced Amateur and Professional Mirror Testing

A large high quality optical flat can be used to test parabolic mirrors. Often the flat is called a Cassigrain flat since it has a small hole in the center. Light from a small laser goes through the hole in the Cassigrain flat and shines on the mirror under test.  The laser light source must be defocused to illuminate the entire mirror under test. If the mirror under test is parabolic and the light source is at the focus then the light from the parabolic mirror strikes the flat and returns to the mirror under test as a collimated beam and is focused back through the hole in the Cassigrain flat to the knife edge or ronchi grating.  Using this method errors in the figure of the parabola under test are doubled since the light strikes the same point on the parabola twice.  It is also possible to generate interference patters of the parabolic mirror surface deviation.

See diagmak.htm for more information.


Back to mirror testing page




In the past interferometry required a null type test but now with advanced data capabilities it is possible
to remove spherical aberration from the display of a parabolic mirror display.
( A parabolic mirror being tested at its center of curvature. )

Following a link at Wikipedia we found this page helpful in understanding interferometry.
The local copy of this page.   Note, the local copy is not updated.


Also refer to the talk given at the Riverside Telescope Makers Conference at:

Zero One Two Three Four Five Six Seven Eight Nine
Ten Eleven Twelve Thirteen Fourteen Fifteen Sixteen Seventeen Eighteen Nineteen
Twenty Twenty one Twenty two Twenty three Twenty four Twenty five Twenty six Twenty seven Twenty eight Twenty nine
Thirty Thirty one Thirty two Thirty three            

Local copy not updated. 


How the James Webb Telescope primary mirrors were made:
Titled subject begins about 5 minutes into video.
Testing methods for aspherical mirrors:

Back to mirror testing page