INITIAL PHASES AND CONTROL OF VOLTAGES INTERNAL TO THE BLANK
In these chapters I will describe the phases that led me to the creation of a parabolic mirror 410 mm f/4.5, focusing first on the phases that led me to create a good spherical surface, necessary for the subsequent parabolization phase. Let's start with the preliminary checks that have been carried out.
The blank I started with was already roughed out, with the basic curvature defined and a first polishing almost complete. Before continuing, But, I wanted to check for the presence of any internal tensions in the glass, an aspect often overlooked but fundamental for the stability of the mirror in the long term.
The test I used is based on the principle of photoelasticity: when a transparent material has internal tensions, it modifies the polarization of the light passing through it. By illuminating the glass with polarized light and observing it through a second polarizer, any tensions manifest themselves as colored fringes or bright patterns; in the absence of tension, instead, the image appears uniformly dark.
To create the set-up I used a simple but effective solution: I used the white screen of an LCD TV as a polarized light source, while as an analyzer filter I recycled a pair of 3D cinema glasses.
The result was immediately evident: while the screen appeared almost black, a clear pattern appeared on the surface of the blank, with a true rainbow of colors near the edge. In the center the classic black cross could be seen, sign of marked internal tensions. To confirm that it wasn't an artifact, I rotated the mirror 90°: the pattern rotated with it, definitive proof that the tensions were in the glass and not in the set-up.
Figure 1. Internal tensions in the blank for two 90° orientations.
At that point the choice was forced: before proceeding with the fine machining, I decided to subject the blank to an annealing cycle to relax the internal tensions. It is a delicate operation, which requires time and an adequate oven, but it is the only way to relax internal tensions and ensure that the mirror remains stable in the long term.
ANNEALING OF THE BLANK
I decided to proceed with annealing the blank after detecting internal tensions with the photoelasticity test. The choice of cycle was made based on the physical characteristics of the piece.
The blank is made of borosilicate and has a thickness of 30 mm, values that guide the choice of temperatures and times. For low expansion glasses such as borosilicate, higher relaxation temperatures are adopted than for soda-lime, but with very controlled ramping and cooling. I calculated the plateau times and thermal transitions to ensure that the entire glassy mass reached the necessary thermal homogeneity.
Detail of the annealing cycle performed
- Heating ramp: temperature increase with trend of +70 °C/h up to 565 °C, i.e. the relaxation temperature chosen for this blank.
- Maintenance plateau: 5 ore a 565 °C, corresponding to approximately 1h for every 6mm of thickness, to allow the entire glass mass to reach a uniform temperature.
- First controlled cooling: descent of -10 °C/h up to 510 °C, followed by a maintenance of 3 hours to uniform the temperature with the center of the blank.
- Intermediate cooling: descent of -18 °C/h up to 370 °C, value below which internal tensions are considered frozen.
- Controlled final cooling: continuation of cooling to approx -30 °C/h up to room temperature.
The entire cycle had a total duration of approximately 42 hour.
The annealing was carried out at a company that owns ceramic kilns and which helped me set up the cycle. Given the overall duration of the treatment, we agreed to carry out the annealing over the weekend so as not to interfere with normal work activities.
To minimize local thermal gradients, the blank was placed on a refractory sheet and also surrounded with refractory blocks.. This arrangement creates a hot mass around the edge of the mirror as it cools, preventing it from cooling too quickly compared to the core, reducing the risk of introducing new tensions or deformations.
|
tempo |
Temp. |
Delta t |
Delta T |
|
[h] |
[°C] |
[h] |
[°C/h] |
|
0 |
25 |
– |
– |
|
8 |
565 |
8 |
68 |
|
13 |
565 |
5 |
0 |
|
19 |
510 |
6 |
-9 |
|
22 |
510 |
3 |
0 |
|
30 |
370 |
8 |
-18 |
|
42 |
25 |
12 |
-29 |
Post annealing verification with birefringence test:
After annealing I repeated the photoelasticity test with the same set-up used previously. The improvement is evident: the surface of the blank was predominantly dark, indication of absence of significant internal tensions. Four lighter areas remained visible, arranged at 45° to the vertical; rotating the blank though, the zones did not rotate with the glass, confirming that these were artefacts of the set-up and not residual tensions in the material.
Prima
Later, after
Figure 3. Internal stresses before and after blank annealing
A significant reduction in internal tensions was achieved and a good blank was obtained to proceed with the polishing and figuring phase.
FIRST OPTICAL TESTS
After oven annealing, the first optical measurements were performed.
The mirror was tested vertically, supporting it with a strap that rests 180° on the edge, to minimize punctual constraints.
The measurement was performed using a home-built Bath interferometer.
First of all, the radius of curvature of the mirror was precisely measured (3684 mm) and the frequency of the interferometer laser (633 nm). After which several interferograms were acquired, processed and averaged to obtain the measurement.
Below is an example of one of the interferograms:
Figure 4. Initial interferogram
And below is the result of the average of about ten measurements:
Figure 5. Map of post annealing aberrations of the blank
As can be seen from the image, the result of the measurement indicates that the optical surface is dominated by astigmatism aberrations, spherical, clover and so on and so forth, For a total weight of 879 nm rms WFE. Evidently the relaxation of internal tensions caused the distortion of the optical surface. Even the conic is far from the sphere K=-1.135 against K=0 which we generally try to reach before starting for the parabolization phase. It is not essential to start from the sphere itself, but it is the path that usually allows you to obtain the best results.
I decide not to go back to trouble, but to try to fix the situation with pitch and cerium oxide.
Definitely the longest road, but it will allow me to experiment with various techniques, necessary to understand how to remove and then manage astigmatism, which is essential given the very thin thickness of the blank (diameter/thickness ratio 15:1).