Chinese semiconductor thread II

tokenanalyst

Brigadier
Registered Member

Huali Chuang Science has developed the world's smallest optical six-dimensional MEMs force sensor and completed a multi-million yuan A+ round of financing​


Recently, Huali Chuang Science announced that it has completed a multi-million-yuan A+ round of financing, which was exclusively invested by Polylite, a leading listed company in the field of metal 3D printing. It is disclosed that the funds raised in this round of financing will be used to deeply explore high-performance optical multimodal sensing technology, combined with the advantages of metal 3D printing technology in product lightweighting, special material application, mass production cost, etc., to provide a powerful and cost-effective six-dimensional force sensing solution for the embodied intelligence field represented by humanoid robots, and promote the development of the embodied intelligence industry.
Huali Chuang Science has developed a "head-to-toe" six-dimensional force sensing solution. In response to the gap in humanoid robot knuckle force perception, in June 2024, Huali Chuang Science released the world's smallest known optical six-dimensional force sensor Photon Finger , which for the first time miniaturized the size of multi-dimensional force sensors to the millimeter level, overcoming the miniaturization problem of six-dimensional force sensors. At the same time, with the support of metal 3D printing, the cost has been greatly reduced, and the products have been used in companies such as Tencent, Qiangnao Technology, and RoboSense. In view of the high performance, lightweight, and easy integration required for the six-dimensional force sensing of the humanoid robot wrist, in December 24, the six-dimensional force sensor PhotonR40 with a central through-hole structure was launched, which greatly reduced the difficulty of integrating the force perception of the humanoid robot wrist joint. This product used metal 3D printing technology for the first time to complete the mass production of the core structure, and has been used by many customers.

1738595835013.png

Please, Log in or Register to view URLs content!
 

tokenanalyst

Brigadier
Registered Member

Impact of even aberrations on alignment accuracy in image-processing-based systems​


Abstract​


The alignment system is a critical sensor for ensuring the overlay performance of lithographic tools, with lens aberrations playing a significant role in determining alignment accuracy. This study investigates the influence of aberrations on alignment accuracy through both simulations and experiments. The findings reveal that even aberrations lead to variations in the sensitivity of odd aberration-induced misalignment. These variations are explained by analyzing the asymmetry of the optical system’s impulse response function. Sensitivity curves were generated by numerical simulation using Hyperlith and validated through experimental measurements. The results refine the understanding of the impact of lens aberrations on alignment accuracy. The proposed sensitivity curve offers a foundation for optimizing error-correction models, ultimately improving alignment accuracy in lithographic processes.

Please, Log in or Register to view URLs content!
 

tokenanalyst

Brigadier
Registered Member

55 billion yuan Hefei 8.6-generation OLED project first ordered equipment​

Hefei Guoxian, a subsidiary of Visionox, has started construction of its 8.6th generation organic light-emitting diode (OLED) production line.​


1738596854115.png

Ion doping equipment is a device that injects ions in the form of gas to improve the characteristics of low-temperature polycrystalline silicon (LTPS) and low-temperature polycrystalline silicon oxide (LTPO) thin-film transistor (TFT) devices. This is one of the core equipment of OLED process.

This is the first time that Hefei Guoxian, a subsidiary of Visionox, has ordered 8.6-generation OLED-related equipment.

It is understood that the total investment of the 8.6-generation AMOLED production line project of Hefei Guoxian Technology Co., Ltd. is 55 billion yuan, with a designed production capacity of 32,000 glass substrates per month (size 2290mm×2620mm), which can produce larger-sized AMOLED panels, covering multi-size application fields such as tablets, laptops, cars, and displays.

This high-generation AMOLED production line has been upgraded in terms of technology, size, efficiency, and application. It is the world's first application of precision metal mask-free technology solutions, which can greatly improve the cutting efficiency and economy of medium and large-sized OLED products. It will drive the update and iteration of downstream terminals through technological upgrades, further tap the growth potential of AMOLED, and help Hefei move towards the "World Display Capital", reshaping the global AMOLED panel supply landscape.

It is understood that Hefei Guoxian Technology Co., Ltd. ("Hefei Guoxian") is responsible for the investment, construction and operation of the production line. The company will be funded by Visionox and Hefei Investment Platform. Visionox and Hefei Guoxian have worked together in the transformation of innovative technologies to create a high-generation AMOLED production line with leading technology, strong compatibility and wide coverage. The production line brings ViP, LTPO+, COE, Tandem and other technical upgrades "black technology" to meet the needs of medium and large-sized OLED panel products for low power consumption, high brightness and long life.

Please, Log in or Register to view URLs content!
 

Wahid145

Junior Member
Registered Member
I was wondering, the 294L was two 150L and 140L Bonded together. So I asked the AI (kimi.ai and DeepSeek), is it due to equipment limitation (No access to LAM Etchers) or is it industry norm to join two Decks (150L and 140L decks) to achieve a tall 294L.

Apparently, most of industry, SK Hynix, Samsung, just grows the Layer without Joining two Decks. But due to limitation YMTC chose this path to bond two Decks, and it has an advantage of achieving higher Density of Memory (20GB/mm2) but is much harder to do so and more prone to defects during Hybrid bonding.

If anyone has more insight on this topic do share. My main concern here is how far AMEC/Naura etching has come compared to leading players like LAM
 

sunnymaxi

Major
Registered Member
I was wondering, the 294L was two 150L and 140L Bonded together. So I asked the AI (kimi.ai and DeepSeek), is it due to equipment limitation (No access to LAM Etchers) or is it industry norm to join two Decks (150L and 140L decks) to achieve a tall 294L.

Apparently, most of industry, SK Hynix, Samsung, just grows the Layer without Joining two Decks. But due to limitation YMTC chose this path to bond two Decks, and it has an advantage of achieving higher Density of Memory (20GB/mm2) but is much harder to do so and more prone to defects during Hybrid bonding.

If anyone has more insight on this topic do share. My main concern here is how far AMEC/Naura etching has come compared to leading players like LAM
Yin Zhiyao, Chairman of AMEC: Our etching machine has achieved the goal of processing one atom accurately every time (Core Things) Mr. Yin said that AMEC's laboratory technology can achieve an accuracy of 0.02nm, while a silicon atom is 0.25nm, which means it can process at the atomic scale. It is really amazing, and may surpass American companies!

i don't think so, you can find better information that this.. AMEC chairman itself said this.. Etching Empire has completed.

AMEC expanding more into semi tools .jpg
 

tokenanalyst

Brigadier
Registered Member
Apparently, most of industry, SK Hynix, Samsung, just grows the Layer without Joining two Decks. But due to limitation YMTC chose this path to bond two Decks, and it has an advantage of achieving higher Density of Memory (20GB/mm2) but is much harder to do so and more prone to defects during Hybrid bonding.
No necessary, Xtacking showed that hybrid bonding technology is maturing fast and this proves that is ready for prime time.
If anyone has more insight on this topic do share. My main concern here is how far AMEC/Naura etching has come compared to leading players like LAM
Cryogenic Etching.

1738602105508.png
 
Last edited:

tokenanalyst

Brigadier
Registered Member
more details about Imecas Hard-Rad SiC device.
Silicon Carbide Semiconductors can work at high temperatures making SiC electronics in Venus a theoretical possibility.


China's first high-voltage radiation-resistant silicon carbide power device was successfully developed and passed space verification.​


Power devices, known as the "heart" of power electronics systems, are the core for realizing electric energy conversion and control. They are also one of the most basic and widely used components in the field of national economy and people's livelihood. Their core technology research and development and innovative development have attracted much attention.

The latest news from the Institute of Microelectronics of the Chinese Academy of Sciences said that the teams of Liu Xinyu and Tang Yidan of the institute and the team of Liu Yanmin from the Space Application Engineering and Technology Center of the Chinese Academy of Sciences have cooperated to jointly develop the first domestically produced high-voltage radiation-resistant silicon carbide (SiC) power device and its power system, which has successfully passed the first phase of space verification and realized its in-orbit application in the power system.

Liu Xinyu said that the silicon carbide payload jointly developed by the cooperative team flew into space on the Tianzhou-8 cargo spacecraft in November 2024, starting a scientific experiment journey in the orbit of the Chinese space station.

After more than one month of on-orbit power-on testing, the silicon carbide load test data was normal, the on-orbit test and application verification of high-voltage 400-volt silicon carbide power devices were successfully completed, and the static and dynamic parameters in the power system met expectations.

In Tang Yidan's view, the first phase of the mission of the silicon carbide payload has been successfully completed, successfully realizing the verification of the adaptability of the first domestically-produced high-voltage 400-volt radiation-resistant silicon carbide power device to the space environment and its on-orbit application in the power system. This indicates that under the demand for space payloads measured in "grams", silicon carbide power devices will become the preferred solution for significantly improving the efficiency of space power supplies, and are expected to drive the upgrading of space power systems.

Liu Xinyu pointed out that China has successfully developed the first domestically produced high-voltage 400-volt radiation-resistant silicon carbide power device, passed space verification and realized its in-orbit application in the power supply system. This will provide a new generation of power devices to choose from for China's future lunar exploration, manned lunar landing, deep space exploration and other fields, and provide strong support for China to build a world space science power.

He believes that as the performance of silicon-based power devices approaches its limit, third-generation semiconductor materials represented by silicon carbide, with their unique advantages such as large bandgap width, high breakdown field strength, and fast saturated electron speed, can greatly improve the transmission power and energy conversion efficiency of space power supplies, simplify heat dissipation equipment, reduce launch costs or increase loading capacity, and increase the power-to-volume ratio by nearly 5 times, meeting the needs of high energy efficiency, miniaturization, and lightweight of space power systems, which is of great strategic significance to the development of the new generation of aerospace technology.

Therefore, countries around the world regard silicon carbide as the commanding height of competition for the next generation of power devices, formulate corresponding plans, and vigorously develop it.

Please, Log in or Register to view URLs content!
 

gelgoog

Lieutenant General
Registered Member
I was wondering, the 294L was two 150L and 140L Bonded together. So I asked the AI (kimi.ai and DeepSeek), is it due to equipment limitation (No access to LAM Etchers) or is it industry norm to join two Decks (150L and 140L decks) to achieve a tall 294L.

Apparently, most of industry, SK Hynix, Samsung, just grows the Layer without Joining two Decks. But due to limitation YMTC chose this path to bond two Decks, and it has an advantage of achieving higher Density of Memory (20GB/mm2) but is much harder to do so and more prone to defects during Hybrid bonding.

If anyone has more insight on this topic do share. My main concern here is how far AMEC/Naura etching has come compared to leading players like LAM
The US government restricted sales of tools which can do more than 128 layers.
Please, Log in or Register to view URLs content!

The two decks have more layers than that. So YMTC figured out some way out of this tools restriction.
The VNAND industry in general is also long term moving to two decks.
 
Last edited:

Wahid145

Junior Member
Registered Member
I am well aware of (almost) all the answers from above. I just want to know that, YMTC will be able to achieve 300+L in the near future. I remember vividly, when YMTC released 232L, the Leading players were still at 100 something Layers and were thinking of introducing 200+ layers after some quarters of YMTC 232L release! How cool is that from YMTC, coming out of nowhere and being at the bleeding edge.

But then the american sanction hit it and almost decimated it (layoffs and stopping factory expansion etc, early 2023 and I think even the CEO had to step down due to having american Green Card). That was really upsetting to see how a leading edge company could be hurt like that with sanctions.

Seeing YMTC introduce 294L (272 Active Layers actually) gives me immense joy but want them to be in their prime of 2022. Where they are better then the Industry "leaders". I hear SKHynix is about to introduce 320+Layer NAND in this Quarter. So, YMTC has some catchup to do, never rest until you leave these unsanctioned Companies far behind. Just like DeepSeek is doing to OpenAI.

People from the future are gonna look back at this current time frame and be spell bounded as to how all these Chinese companies are out innovating the so called "Industry Leaders" with immense restrictions placed on them

No necessary, Xtacking showed that hybrid bonding technology is maturing fast and this proves that is ready for prime time.

Cryogenic Etching.

View attachment 144923
I hear AMEC is working on this technology but forgot if they mastered it or not.
 
Top