It is currently 10 Dec 2019, 00:34

All times are UTC [ DST ]


Search found 4 matches

Author Message

 Jump to forum   Jump to topic

Posted: 18 Mar 2019, 18:22 

Hi all, I'm a translator living in Japan, I mainly do programming stuff so I contacted happycube a little while ago to see if there's any translations I could do for free to help out with LD-decode, and he suggested this article. I finished the part he was most interested in (section 4.2), and will probably do the rest of the article sometime soon. Hope you also find it interesting!


------------
http://sts.kahaku.go.jp/diversity/document/system/pdf/085.pdf

"Historical Development of Laser Disc (LD) Technology with Respect to Efforts to Hasten Its Technological Development and Practical Application" by Sumitaka Matsumura

4.2 Element technology in disc production
Section 4.1 noted the basic workflow for the LD disc manufacturing process, but there were a number of element technologies that caused the disc performance or quality to vary in actual production. The following talks about the optimum pit shape, reduction in micro-defects, and reliability (life expectancy) as examples of typical element technologies.

4.2.1 Optimum pit shapes
The most important parameter that caused LD disc performance to vary was pit shapes, particularly in audio/video quality performance. Optimum pit shapes were investigated from a variety of angles for disc production.

(1) Theory and reality
In theory, the ideal pit shape is a "rectangular" type as shown in Figure 4.17 (a). In such cases, when the playback laser wavelength is λ (lambda) and the disc surface refractive index is n, the optimum pit length becomesλ/4n. Whenλ = 632.8nm and n = 1.49, the optimum pit length becomes 106nm.
However, it is almost impossible to make pit shapes where the edges are perfectly perpendicular; in reality they become "mesa" types as shown in Figure 4.17 (b).
Whether pit shapes are good or bad has an effect on playback signal characteristics and manufacturability, and there is a significant amount of tradeoffs for these qualities. For example, it would be better for the playback signal characteristics if the edge angle was as close to perpendicular as possible, but such a pit shape causes a problem where the stampers cannot accurately replicate the pits during the manufacturing process.

(2) Pit shapes and playback signal characteristics
For the playback signal characteristics, the audio/video quality will improve when the signal has more of the AC component (RF signal) and less of the noise component. The size of the RF signal is mainly determined by the pit depth, pit width, and pit length. In principle, the pit depth is determined by the thickness of the photoresist coating, but configurations are made so that the playback RF signal is maximized for discs in consideration of the amount of the coating removed during development and the pit depth reproduction rate during manufacturing. Because the pit shapes are mesa types in reality, the optimum depth for disc pits was deeper than λ/4n, therefore a thickness of 140-150nm was optimal for the photoresistant coating. Similarly, optimization settings are made to the laser power during cutting for the pit width and to the optical modulator duty (duration of the laser being on) for the pit length.

[image omitted]
Edge

(a) Rectangular type pit

[image omitted]
Edge

(b) Mesa type pit

Figure 4.17 Pit shapes (schematics)

In addition, the dimensional accuracy of pits was improved in order to reduce the noise component, requiring a prevention of minor deviations in shapes. To do this, it was important to narrow down the focus point of the laser beam recorder's objective lens as much as possible. Specifically, a laser with a short wavelength (460nm or less) laser was used with an objective lens that had a high numerical aperture (0.9 or greater). Such special lenses were originally for high-power microscopes, and Olympus lenses were used exclusively. Furthermore, a high resolution type photoresist was used to prevent microscopic deviations in shapes.

(3) Pit shapes and manufacturability
Pit shapes needed to be replicated by the stamper as faithfully as possible during the manufacturing process, but there were pit shapes that were easy to replicate as well as pit shapes that were difficult to replicate. Generally, pits with low depths, narrow widths, and rounded edges were easier to replicate.
A stamper that was molded with easy-to-replicate pit shapes had wide margins for the molding conditions, could have short cycle times, and rarely had pit defects upon mold release. The pit defects that occurred upon mold release were called names such as "plowing" and "pit mekure" ("pit curling"), and they occurred when the edges of the stamper's pits scratched the edges of the disc material's pits immediately after molding. With such scratched edges, abnormal waveforms would occur in the playback signal, resulting in a defective disc. To improve mold releasing for such cases, solutions included smoothing the pit edges by controlling the baking temperatures and times when making the disc masters, or putting lubricants into the molding compounds. In addition, if the cycle time was made too short, the pit replication may become softer and the playback signal characteristics may degrade.

(4) Applying Optical Simulations
As previously mentioned, during development trial and error was repeatedly used in a variety of techniques to determine the optimum pit shapes, but computer simulations were also being done at the same time, and these simulations were used in detailed cause analyses and to determine optimal parameters. The paper "Diffraction theory of laser read-out systems for optical video disc", submitted by the British applied physicist H.H. Hopkins in 1972, was the theory that formed the basis for the theoretical analyses related to playback signals from reflective optical discs. Based on this theory and in accordance with the analysis purposes and computing ability of computers, under conditions such as geometrically limited pit shapes or repeating fixed patterns, various analysis methods were developed such as analysis methods for finding solutions in one or two dimensions, using finite element methods or boundary element methods, and methods for finding solutions in three dimensions without limitations on the pit shapes. Various simulations were also done for mesa type pits, and these simulations were used for investigating the optimum pit shapes and cause analysis for problems. Such optical simulation technology was only used by optical device manufacturers in the past, but with the development of optical discs, this technology also became established in electronics manufacturers. This was achieved by electronics manufacturers employing technicians from optical device manufacturers and training their new technicians, in addition to electronics manufacturers and optical device manufacturers working together.

Note: Summary of H.H. Hopkins' simulation technique
By applying a Gaussian distribution function to the intensity of the laser light output from the objective lens as thus:

[image omitted]

The light will reflect off the pit surface and pass through the objective lens, and the intensity of the light that enters the light sensor is represented by the following Fourier series:

[image omitted]

Here, t is time, I0 is the DC component of the detected light, the second and following terms are the AC component (the high-order harmonic RF component), and we get the following:
In addition, R(m,n) is an expression that reflects the pit shape using the reflection coefficient parameter of the pit surface.

Figure 4.18 (a) and (b) show examples of simulation used for cause analyses of color flashes and dropouts.
(a) shows a normal playback waveform, while (b) shows how the waveform changes when the shape of just one pit in the center is molded smaller. (b) is a typical example of symptoms such as "dropouts" and "color flashes", which are discussed in the next section.

[image omitted]
(a) Playback waveform when pits are normal

[image omitted]
(b) Playback waveform when one pit is abnormal
Figure 4.18 Example of a playback waveform optical simulation

In this manner, simulation methods were used to investigate the playback signal wave in relation to various pit shapes (height, width, length, pitch, etc.), and this was very helpful in investigating the optimum pit shapes for obtaining the best playback signal. In addition, by investigating the relationship with pit abnormalities, simulations were used for analyzing quality problems when there were playback signal waveform abnormalities.

4.2.2 Reducing micro-defects
One side of a LD disc has around 14 to 28 billion pits, but it can be said that it is impossible to create a disc that has no pit defects whatsoever. Therefore, cases where there is a defect in a pit and the playback signal becomes abnormal were referred to as "dropouts", and a correction method where the dropout was replaced with the previous line was used (see figure 4.19 (a)). However, if the size of a defect is the same size as a pit or less, or if part of a pit has changed shape, the defect will be reflected as-is in the video signal without dropout correction being used. In such cases, the phase of the playback signal's RF waveform will be out of phase, therefore one pixel and the surrounding area will appear as a bright, differently colored point when looking at the video. This defect is called "color flash noise" and has an appearance as shown in Figure 4.19 (b).
After investigating the causes of color flash noise, the following three factors were almost always reasons:
① Imperfect pit shape: An RF signal was output but the amplitude was insufficient.
② Foreign material inside the plastic layer: When noise occurs due to colored foreign material, the focus is lost due to the different refractive index of the foreign material in the plastic, causing insufficient RF amplitude.
③ Degradation of the reflective aluminum layer: Sub-micron sized corrosion or exfoliation occurs due to the passage of time, resulting in noise.
① and ② do not change over time, and defective products do not normally make it to retail markets normal quality inspections. The cause of ① was determined to be mainly due to micron sized or smaller defects on the glass master surface during the mastering process, ultra-fine impurities (organic materials or particles) in the ultrapure water or chemical solutions used during various processes, etc. Therefore, by using a higher grade for the relevant materials and being thorough in product quality control, it was possible to prevent color flash noise from occurring. Similarly, ② was addressed by mostly eliminating foreign/residual materials with the help of plastic manufacturers, which in turn mostly eliminated color flash noise. ③ is related to reliability (life expectancy), which is discussed in the next section.

[image omitted]
(a) Dropout (after correction)
[image omitted]
(b) Color flash
Figure 4.19 (a) and (b): A dropout and a color flash

4.2.3 Reliability (life expectancy)
LD discs are a "non-contact format", therefore it was often said that one of their qualities is that they are an "almost permanent" storage medium, with hardly any degradation over time. Unlike "contact formats" such as video tape and VHD which had unavoidable degradation due to friction, LD discs were a "non-contact format", qualitatively implying that friction was not a problem for this "almost permanent" medium. Starting from the R&D stages of LD, there were technological development and evaluation testing done in order to preserve the reliability of LD discs, but in actuality, verifiable results could not be obtained until several years had passed since LD products' marketplace introduction.
One factor that had a large effect on the reliability of LD discs was something that was called "snow noise". Snow noise was a type of color flash noise that occurred due to the sub-micron sized corrosion or exfoliation of the reflective aluminum layer over the passage of time. One difference from the color flash noise mentioned in the previous section is that snow noise appears as flickering snow across the entire screen, and another difference is that the symptoms do not appear immediately after production -- rather they increase gradually over a period of months/years. In addition, this defect was bothersome because instead of appearing on all of the discs manufactured in the same line, it only appeared on some discs. Consumer complaints of snow noise occurred through the first half of the 1980's, but the cause was narrowed down and solutions were implemented in the latter half of the 1980's. Meanwhile, testing methods were established and gradually refined. As stated in the previous section, the cause of snow noise was degradation of the reflective aluminum layer over the passage of time. Investigations revealed that the coating of the reflective aluminum layer was improper, and miniscule changes might occur over long periods of time. These miniscule changes refer to compressive stress due to the growth of the oxide coating of the reflective aluminum layer, causing sub-micron sized exfoliation or aluminum oxide crystal growth, resulting in cases where noise would occur during playback. Because these are microscopic changes, a disc would appear unchanged to the naked eye, therefore analyses were performed with optical and electron microscopes. In addition to suppressing the oxide coating growth and relieving the compressive stress that occurs, it was necessary to establish coating guidelines that would increase the adhesion to the plastic base plate in order to stop these microscopic changes from occurring over time.
The coating guidelines for the reflective aluminum layer were optimized by using different guidelines for trial manufacture then repeatedly performing testing by evaluating the life expectancy after exposure to heat/humidity/acceleration. Some of the process guidelines that were most effective were ① the storage environment of the plastic base plate immediately before vapor deposition, ② the degree of vacuum during vapor deposition, and ③ the vapor deposition rate. In addition, ④ using an alloy doped with approximately 1% of other metals was determined to be particularly effective in regards to the vapor deposition materials (aluminum chips).
For the plastic base plate in ①, moisture absorption was performed on the clear base plate after molding for 15 minutes or more in an environment with a relative humidity of 60%, the aluminum oxide coating adjacent to the base plate was sufficiently grown immediately after vapor deposition, preventing successive growth of the oxide coating.
For the degree of vacuum and vapor deposition rate in ② and ③, a high degree of vacuum and a high speed vapor deposition rate resulted in the aluminum oxide coating becoming a single layer of columnar crystals, allowing for compressive stress from successive growth of the oxide coating to easily occur, but with a low degree of vacuum and a low vapor deposition rate, a laminate of fine granular crystals occurred, preventing compressive stress from occurring.
For the alloy used in ④, the materials themselves prevented growth of columnar crystals and were used for creating a laminate of fine granular crystals.
Figure 4.20 shows cross-section images of the reflective aluminum layer before and after the improvements.
Through these improvements, the snow noise problem was solved, allowing for LD disc life expectancy to reach a level where no practical issues should occur.

[images omitted]
酸化被膜
Oxide coating

柱状結晶
Columnar crystals

ディスク基盤
Disc base plate

(a) Structure of the reflective aluminum layer immediately after deposition

酸化被膜の成長
Growth of the oxide coating

圧縮応力
Compressive stress

(b) Growth of the oxide coating after deposition

酸化アルミニウム結晶の成長
Growth of aluminum oxide crystal

剥離
Exfoliation

圧縮応力
Compressive stress

(c) Exfoliation and growth of aluminum oxide crystal

酸化被膜
Oxide coating

粒状結晶
Granular crystals

基板の水分で形成される酸化被膜
Oxide coating formed with moisture on the base plate

ディスク基盤
Disc base plate

(d) Structure of the reflective aluminum layer after improvements
Figure 4.20 Cross-section images of the reflective aluminum layer improvements

 Jump to forum   Jump to topic

Posted: 17 May 2019, 18:41 

I talk about the comb filters in Blu-ray recorders from Japan a bit in this thread: using Blu-ray recorders from Japan for LDs

 Jump to forum   Jump to topic

Posted: 12 Jun 2019, 02:11 

With a Bluray recorder can you choose between recording on mpeg-2 or mpeg-4 or whatever?

You can set the bitrate (using nebulous terms such as SR and XR) , but I don't think any recorders will let you choose the compression format, and Sony ones are the only ones that record external analog signals in AVC, all other companies use MPEG2, AFAIK.

Toshiba ones will at least record the audio in PCM, but there's no digital audio input, so the LD PCM has to go through a conversion from digital to analog and back to digital again, so directly capturing the PCM with a PC sound card would be better anyway.

 Jump to forum   Jump to topic

 Post subject: Re: Capturing PCM from LD
Posted: 14 Aug 2019, 07:25 

Hmm, was just going to mention bit rates, or rather bit depth. I haven't used Audacity much but I seem to remember it showing the bit depth wrong (or at least weirdly), since a 16-bit wav captured in Sound Forge would be shown as 24-bit, and a 24-bit wav file would be shown as 32-bit. I guess you tried different bit depths already but if not, maybe try setting it to 24-bit (since LD PCM is 16-bit).

Another software that I've seen recommended from a number of people is Cockos Reaper, it's not freeware but it has a 60 day demo, so maybe try it before getting a new capture device: https://www.reaper.fm/index.php
Page 1 of 1 [ Search found 4 matches ]


All times are UTC [ DST ]


Jump to:  

cron