[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
[ATM] Re: Foucault test limitations
Greetings All,
This post is somewhat premature as I still need to gather more data. However,
since the topic is active I thought I'd throw my two cents into the pile.
First, let me list my series of suppositions on the limitations of the Foucault
test.
A) Its my belief that the LARGEST limitation of any type of Foucault Knife
Edge test is its complete blindness of Primary Astigmatism in the numeric
results.
B) Operator error is probably the second largest limitation.
C) Finally, the Couder mask (or lack there of) can change the results of a
test. I think this is the point that Vladimir Sacek has been trying to make.
The number of zones and the size, shape, and position of the zone openings, is
of substantial importance.
Going out on a limb, I'll make the following statement:
"It is my personal opinion and experience that Couder mask Foucault testing
on large fast optics is perhaps critically flawed." My specific reference to
this flaw will be limited to an example optic of 316mm diameter (12.5") at
F/4.5, using a 7 zone Couder mask, as imaged here:
http://lerch.no-ip.com/atm/Comp/Mask.jpg (35KB)
Since this is obviously a controversial statement to make, I don't expect anyone
to just believe what I write. Consequently I offer the following chronological
series of events that I hope support my position.
Before listing the events, It is important to note that NO experiments were done
to see how different prescriptions of Couder masks effected the results. This
will be left as an exercise for the reader, if so inclined.
Below is a list of chronological events that lead to my above statement, which
all started approximately one year ago.
A) An optic of 316mm diameter F/4.5 was tested by a reputable test operator, and
independently confirmed by members of my local ATM mirror lab, using a Couder
mask with the following prescription:
i) Each zone opening was 2 inches in height, above / below diameter
tested
ii) Each zone had the following radiuses (in inches)
Zone Inner Outer Mean radius
1 1 2.52 1.76
2 1.76 3.41 2.97
3 2.97 4.12 3.77
4 4.12 4.72 4.42
5 4.72 5.26 4.99
6 5.26 5.74 5.5
7 5.74 6.19 5.97
The results for this Couder mask Foucault testing were as follows:
Strehl = 0.991
Surface RMS = 4nm
P-V Wavefront @ 550nm = 1/14.11
Complete results seen here:
http://lerch.no-ip.com/atm/Comp/Couder.gif (21KB)
B) The same optic was then tested with my Robotic Digital Knife edge test (a
maskless test based on the work of Mike Peck). The test used the same mean zone
radius from the Couder Mask above. These are the results:
Strehl = 0.796
Surface RMS = 21nm
P-V Wavefront @ 550nm = 1/3.9
Complete results seen here:
http://lerch.no-ip.com/atm/Comp/FigXp_SOG_Ave.gif (22kb)
C) The same optic was then tested using Jim Burrows 1D Hartmann test. These are
the results:
Inside ROC:
Strehl = 0.679
Surface RMS = 27nm
P-V Wavefront @550nm = 1/2.5
Complete results seen here:
http://lerch.no-ip.com/atm/Comp/Hart_Inside.gif (106kb)
Outside ROC
Strehl = 0.824
Surface RMS = 19.2nm
P-V Wavefront @550nm = 1/3.9
Complete results seen here:
http://lerch.no-ip.com/atm/Comp/Hart_Outside.gif (106KB)
D) At this point in time we have conflicting results. The Classic Couder mask
Foucault result indicate a near perfect optic. While the newer and
unsubstantiated test methods show the optic as undercorrected. Unfortunately
this optic was made unavailable for further testing.
E) In an effort to continue investigating the disparity in the test results, a
new optic was obtained. The new optic is the now infamous Charlie optic, owned
by a member of our local ATM mirror grinding club. Charlie's optic is of the
same diameter and F-ratio as the previous optic, so that continued investigation
could be resumed.
F) We tested Charlie's mirror using my Robotic Foucault test across 3 diameters
in 60 degree increments. We then sent the optic to R.F. Royce for
interferometric testing, expecting to settle the issue once and for all.
Unfortunately, Charlie's optic suffered from large amounts of primary
astigmatism, which made it appear that my Robotic Foucault test failed
miserably. Here are the results from Interferometric testing of Charlie's optic
from R.F. Royce:
Strehl = 0.44
Surface RMS = 45.9nm
P-V Wavefront @ 550nm = 1/1.3
Complete results seen here:
http://lerch.no-ip.com/atm/2ndTry/results/testdata.JPG (86KB)
G) A little stymied at this point, I almost conceded that my Robotic Foucault
test was flawed. Fortunately, Dave Rowe released his FringeXP software which
gave us amateurs a method to reduce Fringe images from interferometric testing.
Consequently I built an interferometer based on the design of Peter Ceravolo,
using a reference element from Peter as well. I then put Charlie's optic back
on the test stand and re-tested it using my home built interferometer, here are
the results:
Strehl = 0.43
Surface RMS = 40.1nm
P-V Wavefront @ 550nm = 1/1.4
Complete results seen here:
http://lerch.no-ip.com/atm/Inter/Fringes_and_Foucault/316x2829_Charlie's_Mirror/Inter_Axis_B_ave.gif
(45KB)
F) At this point, I now know my interferometer gives very similar results as Mr.
Royce's interferometer for an optic of this diameter and F-ratio. So now, what
to do about the discrepancy in the Robotic Foucault results? After some
consideration, I came to believe that perhaps the Primary Astigmatism in
Charlie's optic might be the culprit.
G) With Charlie's permission, I put his optic on the polishing machine and
polished out the astigmatism. After One Hundred and Five figuring sessions (Oh
My!) I finally found a surface that was somewhat acceptable and passed Jim
Burrows criteria while testing with the interferometer. I also re-tested the
optic with my Robotic Foucault test, and here are the results:
Interferometry:
Strehl = 0.837
Surface RMS = 18.5nm
P-V Wavefront @ 550nm = 1/1.82
Complete results seen here:
http://lerch.no-ip.com/atm/Inter/Fringes_and_Foucault/316x2832_Charlie_Re-Figure/Refig_2/Inter.gif
Robotic Foucault test:
Strehl = 0.841
Surface RMS = 18.1
P-V Wavefront @ 550nm = 1/4.35
Complete results seen here:
http://lerch.no-ip.com/atm/Inter/Fringes_and_Foucault/316x2832_Charlie_Re-Figure/Refig_2/Robo.gif
F) For a final piece of the puzzle, I also did an impromptu crude star test,
that is lacking in many ways. This star test I'm about to show is full of
faults caused from the following list of encountered problems
i) The atmosphere was exceptionally turbulent
ii) We didn't have a tracking platform, and imaged as the star drifted thru
iii) We didn't have a micrometer attached to the focuser, so we guessed
(poorly!)
iv) We only had a Zero power red dot finder to help aim the scope
v) We had to use a 2x barlow to get the bare CCD to focus
Given the difficulties encountered, here is a comparison of a simulated star
test (derived from the interferometry data) and the actual star test:
(apologizing in advance for the excessive defocus!)
http://lerch.no-ip.com/atm/Inter/Fringes_and_Foucault/316x2832_Charlie_Re-Figure/Refig_2/Charlie_Star_Test.jpg
So, In conclusion I'm pretty confident of the following factual observations
#1) My interferometer gives similar results as Mr. Royce's interferometer
#2) The interferometer appears to agree with my crude star test images
#3) Robo agrees with the interferometer so long as there is LIMITED astigmatism
#3) Robo agrees with Jim Burrow's 1D Hartmann Test
#4) Robo Disagrees with Couder mask Foucault testing, thus the reason for my
statement that Couder Mask Foucault testing is flawed for optics of this
Diameter and F-Ratio, when tested using the previously described Couder Mask.
PS, I have several more instances of data on different sized optics show a
strong agreement between Robo and Interferometry, but I'll save those for a
later date.
BTW, if your curious what 105 figuring sessions look like when animated, have a
look at this 1Meg video :)
http://lerch.no-ip.com/atm/Inter/Fringes_and_Foucault/316x2832_Charlie_Re-Figure/Z5.avi
Take Care,
James Lerch
http://lerch.no-ip.com/atm (My telescope construction,testing, and coating site)
"Anything that can happen, will happen" -Stephen Pollock from:
"Particle Physics for Non-Physicists: A Tour of the Microcosmos"
" Press on: nothing in the world can take the place of perseverance.
Talent will not; nothing is more common than unsuccessful men with talent.
Genius will not; unrewarded genius is almost a proverb.
Education will not; the world is full of educated derelicts.
Persistence and determination alone are omnipotent. "
Calvin Coolidge
_______________________________________________
ATM mailing list http://www.atmlist.net/