Version: **

A reference design kit based on Infineon’s EZ-USB™ CX3 USB 3.0 camera controller and EN801/EN805 image signal processor (ISP) from Eyenix website. This camera kit uses an IMX307 image sensor from Sony with a manual focus lens.

  Figure 1. Kit block diagram

Eyenix ISP has a 4-lane MIPI-CSI2 RX interface for connecting image sensors and a 4-lane MIPI CSI-2 TX interface for connecting to EZ-USB™ CX3.

Infineon EZ-USB™ CX3 enables USB 3.0 connectivity to EN801/EN805 using a MIPI-CSI2 interface with data speed up to 1 Gbps per lane.

 Figure 2. Kit photo

Key features of the kit:

• Bandwidth: 4x CSI-2 lanes, 1 Gbps per lane
• Color formats supported: YUV422 8-bit
• USB camera protocol: UVC
• Supported resolutions:
  
  • 720p @30/50fps
  • 1080p @25/30fps
  • 1080p @50/60fps
  • 1440p @25/30fps (supported by EN805 + IMX335)

Eyenix EN801/EN805 ISP has the following features:

• RGB interpolation
• WDR
• Digital noise reduction
• Adaptive contrast enhance (ACE)
• Histogram equalizer for defog
• Lens shading compensation
• Down and up scaler
• Defect pixel correction
• Hue controller
• High light masking
• Auto exposure, Auto white balance, Auto focus
• Edge enhancement
• Gamma correction
• On-screen display (OSD)
• Adaptive lighting control

The kit also supports control through an on-screen display (OSD) menu. This menu can be navigated using the Python-based camera control tool as shown in Figure 3. The tool controls the ISP menu via USB CDC class through a PC.

  Figure 3. Menu control tool

Figure 4. On-screen display (OSD) menu

See the Eyenix website for more information on this camera control software and applications including surveillance cameras, webcams, document scanners, and machine vision systems.

Version: **

1. Is the XENSIV^TM PAS CO2 sensor affected by acoustic and vibration noise?
   XENSIV^TM PAS CO₂ sensor has a robust packaging, and is acoustically isolated from the surrounding noise. The sensor has been designed in such a way that only CO₂ molecules can diffuse within the measurement chamber, while significantly attenuating the surrounding noise. As shown in Figure 1, the sensor performance remains within the specifications up to 101 dB SPL for pink noise, which is one of the most common signals in biological systems and is widely present in modern-day music.
   Figure 1. Acoustic robustness of XENSIV^TM PAS CO₂
   
   

   Infineon recommends that you isolate the sensor from vibration sources. The operating frequency of the sensor is 40 Hz. If vibration is expected at exactly 40 Hz with very high acceleration (≤ 0.3 g), you should do the following to minimize the vibration impact:

   • Mount the sensor in the X-direction of vibration.
   • Enable the denoiser filter.
   • Operate the sensor in continuous mode.

However, for a typical application, the impact of vibration is minimal. As shown in Figure 2, for a test setup with three sensors mounted on a commercially available air purifier, the test result shows that with the denoiser filter enabled, the sensor is robust against the vibration generated by a different fan of the air purifier.

 Figure 2. Vibration robustness of XENSIV^TM PAS CO₂

For more details, visit the CO₂ sensor website and see the application note, General design in guidelines for XENSIV™ PAS CO2 sensor.

2. How much indoor area can be covered by one sensor for a typical application?
   Even though CO₂ is generated locally, after a certain period, the CO₂ concentration is distributed throughout the room. Therefore, irrespective of the room size and number of people, the sensor is capable of accurately estimating the indoor CO₂ concentration within an indoor environment. For more details, see the application note, General design in guidelines for XENSIV™ PAS CO2 sensor.
   
   
3. Are scientific literature available indicating the harmful effects of CO₂ for human body?
   Several studies have been done on the effect of CO₂ on the human body at various concentration levels. The summary is provided in the chart. See Source [1] for the details of the study.
   
   

   Figure 3. Impact of CO₂ on the human body at various concentration levels

   Source
   

   1. https://www.sciencedirect.com/science/article/pii/S0160412018312807
   2. https://books.google.de/books?hl=en&lr=&id=TEnO6lILSj4C&oi=fnd&pg=PR8&dq=Co2+sensors+and+health&ots=Xy1umDNsmq&sig=ZjljEpPMeRBse75NgUiwbh01KuM&redir_esc=y#v=onepage&q=Co2%20sensors%20and%20health&f=false
   3. https://ieeexplore.ieee.org/abstract/document/8767280

4. Is it possible to obtain MEMS separately from the module along with the compensation software to integrate them into a custom DSP solution?
   No, currently it is not possible to purchase MEMS separately.
   
   
5. How is the sensor calibrated? Does it need recalibration after delivery?
   Each sensor is calibrated in the factory when it is shipped from Infineon. For the final product, you  will not have to perform calibration. However, you should use the automatic baseline offset correction (ABOC) feature during operation to minimize errors due to aging. See the application note, FCS_ABOC_XENSIV™ PASCO2.
   
   
6. How can I generate a known level of CO₂ in ppm?
   CO₂ sources are numerous such as bike tire CO₂ inflators, soda water, and human breath. To arrive at a known level, capture the CO₂ gas in a Tedlar bag and pump it into a gas chamber using a 6-V diaphragm pump. CO₂ concentration can easily reach more than 10000 ppm in such a setup. A reference CO₂ sensor can be used to measure the known level. See the application note, Recommended performance evaluation methodology for XENSIV™ PAS CO2 sensor.
   
   

   One such setup is as follows:
   

7. Can the XENSIV^TM PAS CO₂ sensors be battery-powered?
   XENSIV^TM PAS CO₂ can be used for battery-powered applications. Infineon has a reference design which can help operate the sensor at 3.3 V (a low-cost boost converter circuitry along with the sensor). For details, see the application note,

   XENSIV™ PAS CO2 for low-power application.

8. When is the CO₂ sensor available? Where can I get eval kits?
   

    

   Evaluation kits are already available with major distributors.

9. What are the power requirements of the XENSIV^TM PAS CO₂ sensor?
   The average power consumption to power up the sensor is 29.7 mW. During the measurement phase, the power consumption is 26.4 mW. Infineon provides a power calculator based on Microsoft Excel that helps in determining the average power consumption based on the end application.
   

   Figure 4. Average power consumption

   For more information, see Section 3 of the application note, XENSIV™ PAS CO2 for low-power applications.

10. What is the long-term stability of these sensors?
    

     With the automatic baseline offset correction (ABOC) feature enabled, the sensor accuracy drifts by 1% every year.

    

    Figure 5. Values of pressure and acoustic stability

    See Section 4.1.5 - Transfer function table in the datasheet.

11. How can I reduce a high CO₂ level?
    According to ASHRAE, an ideal living room should not have more than 1100 ppm of CO₂. Therefore, as soon as the CO₂ level increases beyond 1100 ppm, you should ventilate the room.
    
    
12. How is the sensor calibration done?
    
    • Does the sensor calibrate single gas or more gases?
    • How long does the calibration process take?
    • What is the required equipment?
    • How often calibration must be done?
    • Can the calibration be done anywhere or only in the laboratory?
    • Does the sensor have automatic calibration or auto leveling?

Each sensor is calibrated during our production for the complete operating range. Production calibration is done with highly accurate CO₂ gas bottle and verified with an ideal reference sensor.

The sensor might show small amount of offset after assembly due to stress generated from the assembly process on the light source. Therefore, to get the best performance, this offset should be corrected using either FCS or ABOC.

For the FCS calibration, you need to have a reference sensor. In the ABOC mechanism, the device keeps track of the minimum value recorded over a week. The offset to the reference baseline is computed and used to calculate the correction factor to be applied for the week after.

Figure 6. Operation of the ABOC feature

See the application note, After-assembly calibration scheme for XENSIV™ PAS CO2 for more details.

13. Can this device be installed on a microcontroller/system-on-chip such as Arduino/Raspberry Pi?
    Yes. Infineon provides C/C++ libraries for interfacing the sensor.
    See the following GitHub repos:
    
    • For devices compatible with Arduino.
    • For PSoC™ 6 MCU library.

14. Does the response time (T63 = 75s) meet the certification requirements and accuracy for medical applications?
    The sensor is not specifically designed to cover medical applications. However, the response time is defined based on the diffusion process.
15. What are the design rules for integrating the CO₂ sensor into a package for airflow speed and direction?The device must be positioned in such a way that the diffusion of CO₂ molecules can happen easily, with a large enough opening. The recommended opening is at least 14 mm x 14 mm.
    

     

    Figure 7. Airflow direction

 See the application note, General design in guidelines for XENSIV™ PAS CO2 sensor for details.

16. What is the time constant (CO₂) of the sensor?
    Time constant/response time of the sensor is the time it takes to measure the CO₂ concentration. This is due to the diffusion time of CO₂ molecules.
    For VDD3.3 = 3.3 V, VDD12 = 12 V, Tamb = 25°C, % RH = 30 %, p = 1013 hPa, the response time, T63 = 75 seconds.
    See Section 4.1.5 of the datasheet for more information.
    
    
17. What are the response times in different airflows?
    The response time is independent of the airflow. You should not place the sensor directly within the airflow. For more details, see the application note, General design in guidelines for XENSIV™ PAS CO2 sensor.
    
    
18. How do I waterproof the sensor board for wet environments?
    The sensor is not designed for applications where it must be waterproofed. The sensor may be damaged if submerged in water.
    
    
19. Is the CO₂ sensor tested in an environmental chamber?
    A test chamber is used, which  ensures that there is no leakage, and it can maintain laminar flow while exposing the sensor to the target gas concentration. Inside the test chamber, you should accommodate a reference CO2 sensor. Additionally, to get an overview of the complete test conditions, the pressure sensor, humidity sensor, and temperature sensor should also be considered.
    For more information, see the application note, Recommended performance evaluation methodology for XENSIV™ PAS CO2 sensor.
    
    
20. Can the sensor be used within battery-driven consumer goods, with a supply voltage lower than 12V, such as 3.3 V?
    Infineon has addressed the requirement; a reference design is available to ease the design process. You can use this design to realize 12V with relatively low effort. For more information, see the application note, Reference design Sensor2Go kit, and the detailed knowledge base article (KBA).
    
    
21. What are the major technologies used in the sensor?
    The sensor uses a microphone, emitter (MEMS + filter), and the embedded software that implements the sensing algorithm

Note: For more information, see the Community webpage for CO2 sensor where you can find answers to your questions and ask your own.

Version: **

Question:
Why does the CYU3P_GPIF_ERR_INVALID_STATE GPIF error occur in FX3 SDK?

Answer:
The GPIF error CYU3P_GPIF_ERR_INVALID_STATE occurs when the state machine reaches an invalid state. An invalid state is a state that is not defined the .h file generated by the GPIF II designer tool. Mostly, this happens when mirror states are used in the state machine. To know more about mirror states, see section 4.5 – GPIF II constraints of the GPIF Designer User Guide (gpif2_designer_userguide.pdf). This document is available with the FX3 SDK and can be found in the following location after installing the SDK: C:\Program Files (x86)\Cypress\EZ-USB FX3 SDK\1.3\doc\GPIFII_Designer.

When mirror states are used in the state machine, the state numbers of all mirror states associated with a state will not be defined in the .h file generated by the GPIF II designer tool. So, in a state machine that makes use of mirror states, if at any point in time the state machine reaches any mirror state, there are chances for the GPIF error CYU3P_GPIF_ERR_INVALID_STATE to be triggered. This will not affect the functionality of the state machine.

Another possible reason for the GPIF error CYU3P_GPIF_ERR_INVALID_STATE is also associated with the use of mirror states. Figure 1 illustrates the  reason with the  example of Slave FIFO state machine provided with AN65974.

Figure 1. Slave FIFO State Machine

The slave FIFO state machine in Figure 1 uses SLWR and PKEND for identifying mirror states. While building the state machine, the GPIF II designer warns that SLWR and PKEND are removed from the outgoing transition equations from the state IDLE and the state  DSS_STATE. If SLWR and PKEND are removed from the outgoing transition equations from the IDLE state, then the following are obtained:

1. IDLE -> READ : Equation: !SLCS & !SLRD & !SLOE when (WR,PKEND) = (1,1)
2. IDLE -> WRITE : Equation: !SLCS & SLRD when (WR,PKEND) = (0,1)
3. IDLE -> SHORT : Equation: !SLCS & SLRD when (WR,PKEND) = (0,0)
4. IDLE -> ZLP : Equation: !SLCS &SLRD when (WR,PKEND) = (1,0)

This means that the transition equation from the IDLE state and its mirror states can be either (!SLCS & !SLRD & !SLOE) or (!SLCS and SLRD). 

Now, assume that the DSS_STATE is not used in the state machine. If you initiate a read, by first asserting SLCS and then asserting SLRD in the next clock cycle, it might cause a problem. The transition equation (!SLCS & SLRD) will be matched in the first clock cycle and the state machine will transition to an INVALID state. This is because for a read operation, (WR,PKEND) will be (1,1). When (WR,PKEND) = (1,1), the state machine can jump to a valid state only if the transition equation is !SLCS & !SLRD & !SLOE. Since this is not satisfied, the state machine will jump to an INVALID state.

The only way to avoid this problem is to place a valid state in the location matching the "!SLCS & SLRD" equation when (WR,PKEND) = (1,1). This is the DSS_STATE state in the state machine. If this DSS_STATE state is not present in the state machine, it will trigger the GPIF error CYU3P_GPIF_ERR_INVALID_STATE.

So, while developing a custom state machine that makes use of mirror states make sure that the state machine can transition to a valid state for any combination of inputs to avoid CYU3P_GPIF_ERR_INVALID_STATE. If required, a valid state (like DSS_State in AN65974) may be used.

Original KBA:USB-Serial: Programmed Configuration not Updated in Self-powered Mode - KBA231991

Translated by: Kenshow

タイトル： USBシリアル：プログラムされた構成がセルフパワーモードで更新されない-KBA231991

バージョン：**

質問：

USBポートの電源を入れ直しても、USBシリアルデバイスにプログラムされた新しい設定が更新されないのはなぜですか？

回答：

図1は、デバイスのVBUS / VCCピンがUSBコネクタのVBUSから給電され、USBシリアルデバイスのVDDD / VCCIOピンがUSBコネクタのVBUSから直接または間接的に給電されないセルフパワーシナリオを示しています。 

上記のシナリオでは、プログラミング後にUSB-Serialデバイスがホストから切断されても、VDDD / VCCIOを介して電力が供給されるUSB-Serialデバイスの内部コアは以前の構成を保持します。

注：VCCおよびVCCIOは、部品番号CY7C65213およびCY7C65213Aに対応します。

            VBUSおよびVDDDは、部品番号CY7C65211、CY7C65211A、CY7C65215、およびCY7C65215Aに対応します。

図1.セルフパワー構成シナリオ

プログラムされた新しい構成を更新するには、次のいずれかの条件が満たされている必要があります。

1. USBシリアルデバイスのVDDD / VCCIOピンの電源を入れ直す必要があります。
2. USBデバイスは、リセットピンを最低1usローにアサートしてリセットする必要があります。
3. USB-シリアルデバイスは、USB-シリアル構成ユーティリティのReset Deviceオプション（図2を参照）を使用するか、デバイスの元のホストアプリケーションからCyResetDevice（）呼び出しを発行することにより、接続先のホストからリセットする必要があります（USBシリアルSDK の Cypress USB-Serial API ドキュメントを参照してください）。

図2.USBシリアル設定ユーティリティのリセットデバイスオプション

したがって、図1に示すようにUSBシリアルデバイスがセルフパワー設定の場合、つまり、VDDD / VCCIOが内部ソースを使用して電力を供給され、ボードがホストに接続されていないときに内部ソースがアクティブのままである場合、新しくプログラムされた設定は、USBが再接続された後でも、デバイスがリセットされるまで更新されません。

English: How FX3/CX3 Bootloader detects the state of PMODE ... - Cypress Developer Community

Translated by: Kenshow

タイトル： FX3/CX3ブートローダがPMODEピンの状態を検出する方法 - KBA231617

バージョン：**

PMODEピンの状態を検出するために、FX3ブートローダは、以下の順序でPMODEラインの内部プルアップおよびプルダウン抵抗を有効にします。

1. ブートローダは、最初にすべての PMODE ピンを内部的に 50kΩの抵抗で HIGH 電圧にプルアップします（FX3のデータシートのデジタルI/Oの項を参照）。5 µs待機した後、ブートローダはPMODEピンの状態をサンプリングします。
2. ブートローダは PMODE ラインのプルアップを解除し、10kΩ抵抗で PMODE ラインをLOW電圧までプルダウンします(FX3 データシートの 13 ページを参照)。ここでも5µs待機した後、PMODEピンの状態をサンプリングします。この後、内部プルダウンが除去されます。

このプロセスは、デバイスの接続/リセットされるたびに繰り返されます。

ブートローダーは、PMODEピンがフローティングであるかどうかをチェックします。これには、プルアップ抵抗とプルダウン抵抗が有効になっているときにサンプリングされた PMODE ラインの状態が使用されます。

注：特定のI / Oのパワードメイン電圧に対応するHIGH / LOWとして解釈される電圧の範囲を理解するには、FX3データシートの表8 - DC仕様を参照してください。

PMODEピンがフローティングの場合：

内部プルアップ時には、その端子の電圧は HIGH になります。内部プルダウンを行うと、そのピンの電圧は LOW になります。内部プルアップ抵抗とプルダウン抵抗を変更したときの電圧レベルの変化は、ブートローダがピンがフローティングしているかどうか を検出するために使用します。

PMODE ピンが外部から HIGH にプルアップされているか、外部から HIGH の電圧が供給されている場合：

ブートローダが内部プルアップを有効にすると、そのピンの電圧は HIGH になります。次に、内部プルダウンを有効にすると、そのピンに分圧（外部抵抗と内部10kpull-down）が形成されます。ここで、外部抵抗を選択する際には、その端子の電圧がLOWレベルにならないように注意してください。もしLOWレベルになると、ブートローダはそのピンをフローティングと解釈します。

PMODEピンが外部からプルダウンされている場合：

ブートローダによって内部プルアップ抵抗がそのピンに配置されると、（50kの内部プルアップ抵抗と外部プルダウン抵抗で）分圧されます。ここでも、外部抵抗を選択する際に、そのピンの電圧がHIGHレベルにならないようにしてください。次に、ブートローダがそのピンに内部プルダウン抵抗を配置すると、そのピンの電圧自体がLOWになります。したがって、ブートローダが内部プルアップ抵抗を配置したときのピンの電圧がHIGHレベルの場合は、ブートローダーはピンがフローティングであることを検出します。

PMODEラインの外部プルアップおよびプルダウン抵抗の推奨値は10kです。

FX3 / CX3デバイスがホストまたはリセットに接続されている場合の、PMODE電圧レベルのディップとスパイクの影響。

開発者コミュニティの次のディスカッションスレッドを参照してください。

https://community.cypress.com/thread/50671

前述のように、ブートローダは、PMODEピンの状態を検出するための内部プルアップおよびプルダウン抵抗を有効にします。ブートローダによって有効にされるこれらのプルアップおよびプルダウン抵抗は、PMODEラインで見られるディップとスパイクの原因になります。これらのピンの機能には影響しません。