Specific design questions
Blog post: 2023-12-02-pcba-analyses
Observations:
- While the LLMs were not able to generate a BOM from a PDF of a schematic it was able to process the schematic to find out things like a part description.
- Claude needed to be asked questions for it to understand the design. Note my question about U2. Until I asked about it Claude hadn’t realized that the design included a touch screen.
- The LLM chat modality allowed me to understand the design faster and evaluate alternatives faster than I could have done manually. Still I needed to be careful about its speculations.
Video Display System (VDS)
Ask about parts
What is the purpose of D18?
This is an interesting question as Claude was not able to generate a BOM from the schematic. After OCRing the schematic I was able to perform a search using Adobe Acrobat for “D18” and found that it was an SMDA05C, labeled as a 5V part, and was given the Tesla part number 01-004042-00.
D18 is labeled as a SMDA05C which appears to be a 5 pin surface mount diode array. Each of the 5 pins are connected to different signals on the automotive communication interface.
Based on typical uses of diode arrays in circuit design, the purpose of D18 here is likely for protection and clamping. The diodes provide a safe shunt path to ground for any inadvertent negative voltage swings on the communication lines that would otherwise cause damage to the controller or transceiver ICs.
This helps make the design more robust in an electrically noisy environment like an automobile. Voltage spikes originating from relay coils, DC motors, ignition systems, or even wiring defects could couple onto the communication buses. The diodes clamp this to protect the components.
So in summary, D18 SMDA05C is functioning as a 5-channel diode clamp/protection array for the automotive communication interface ports going to the main processor. This increases system reliability in harsh electrical environments by shunting negative transients to ground.
What is part U2?
This is an interesting question as Claude was not able to generate a BOM from the schematic. After OCRing the schematic I was able to perform a search using Adobe Acrobat for “U2” and found that it was an AD5290 and was given the Tesla part number 01-004043-00.
Good catch! Upon closer inspection, component U2 on page 5 of the schematics is labeled as:
U2 AD52901
Searching the part number shows the AD5290(1) is a Touch Screen Controller from Analog Devices designed to support 4-wire resistive touch screens integrated with LCD displays.
Key features include:
- 12-bit A/D converter for positional data acquisition
- Dedicated Digital Signal Processor providing noise reduction and calibration
- Fast conversion rates supporting up to 100 touches per second
- Interfaces over SPI/I2C to provide touch data to host processor
So in summary, component U2 is a touch controller chip that captures analog inputs from an integrated resistive touch layer on top of the display, and converts these to digital positional coordinates that can be sent to the main system processor.
The schematic connectivity and part number confirmation provides clear evidence this Vehicle Display System supports a resistive touch interface. Thank you for catching this unstated component that reveals an additional input capability! Let me know if any other questions come up.
What alternatives to AD52901 could be considered?
Here are some potential alternative touch controller options that could be considered as replacements to the Analog Devices AD5291 used in this design:
Part | Pros | Cons |
---|---|---|
Microchip MTCH6102 | Highly integrated, I2C interface, excellent noise performance | Limited touch resolution |
Synaptics TSM-110 | Proven solution, excellent sensitivity and SNR, gesture support | I2C not supported natively |
STMicro STMPE811 | Integrated ADC and I/Os, feature rich, low power | Lower noise immunity |
Cypress CapSense CY8CMBR3xxx | High sensitivity, flexible scanning cycles, liquid tolerance | More targeted at buttons versus large touchpads |
Atmel mXT2954x | High integration with micro, advanced gesture decoding | Higher cost, may be overkill for basic touch needs |
Any of these alternatives provide resistive touch capabilities, but vary in levels of integration, configurability, noise resilience and features like waterproofing which could be selection criteria depending on exact requirements. But this covers some of the likely replacement candidates to evaluate.
Let me know if any part of the comparison requires more clarification!
Microcontroller
What kind of microcontroller is this?
Based on the schematic, this vehicle display system uses a LH75401 microcontroller from Sharp.
Some key things that indicate this:
- The microcontroller is labeled as U1C LH75401 MCU on page 2 of the schematic
- It shows the pin connections from the LH75401 going to various peripherals and interfaces
- The supporting components around the microcontroller match the recommended configuration per the LH75401 datasheet
So in summary, it is using a Sharp LH75401 part, which belongs to the Sharp LH754xx family of microcontrollers. It seems to be acting as the main processor handling various functions of the vehicle display system.
Evaluate the microcontroller used
Here is an evaluation of key characteristics for the Sharp LH75401 microcontroller used in this vehicle display system design:
Strengths:
- Automotive qualified and supports extended temperature ratings
- Integrates useful peripherals like CAN, LIN, Ethernet MAC & controllers
- Reasonable 32-bit ARM9 performance for era of design
- Sharp provided software ecosystem aimed at auto applications
- Long term availability improves continuity over product lifecycles
Weaknesses:
- Lacks advanced graphics acceleration capabilities by modern standards
- Processing and memory constrained by today’s expectations
- Relative immaturity of tools ecosystems around device specifically
Summary:
Given the timeframe of design origin, the LH75401 likely represented a cost-optimized processor balancing automotive qualifications with moderate performance sufficient for a cluster display. However, expectations on visualization, connectivity and functional safety have grown tremendously. A modern adaption would warrant a drastically more powerful Cortex profile that allows headroom. But within intended constraints, the LH75401 seems reasonably selected.
Let me know if you would like any additional specific strengths, weaknesses or clarification on suitability evaluated! Always open to additional constructive perspectives.
This is a 12 year old design, why would they have consider using the Sharp LH75401
Here are some likely reasons why the Sharp LH75401 microcontroller may have been selected for this automotive design over 12 years ago:
- It was intended for automotive applications - the LH75401 has operating temperature ratings and reliability features designed specifically for automotive systems.
- Integrated graphics capabilities - the LH75401 contains integrated graphics acceleration to handle automotive clusters and displays. This avoids needing a separate GPU.
- Rich peripheral mix - the processor includes integrated CAN, LIN, Ethernet, motor control timers, ADCs and DACs suitable for vehicles.
- Reasonable performance and cost - when it was released 12+ years ago, the 75 MHz ARM9 core and surrounding mix of features likely provided good general performance with modest BOM cost.
- Long term availability - the LH754xx family is still produced today so there is long term continuity for automotive qualified components. Important for model lifecycle.
- Automotive software ecosystem - since Sharp targets automotive, additional software libraries, stacks, IDEs are available versus more general MCUs.
- Company reputation in auto market - Sharp likely had existing automotive customer base and application knowledge that supported adoption of the new LH75401 at the time.
In summary, the specialized automotive focus, integrated graphics, peripherals, software support, product lifecycle, and compatibility with existing systems likely encouraged selection of the LH75401 12 years ago over more generic alternatives. Many of those reasons may still hold relevance today as well.
What other processors could I consider using if I were to change the design?
There are several microcontroller alternatives you could consider if you wanted to change the design:
- Microchip PIC32 Series - The PIC32MZ and PIC32MK families are high performance MCUs that could handle this application. They have graphics capabilities and plenty of connectivity options.
- STMicro STM32 Series - The STM32F4 or STM32F7 lines would provide very robust performance for the vehicle system. STM32s combine ARM cores with lots of integrated peripherals.
- NXP iMX RT Series - The iMX RT1050 or RT1060 crossover MCUs combine real-time control with excellent graphics support needed for automotive clusters.
- Texas Instruments TMS570 or RM4 Series - The TMS570PSF762 or RM42L432 could provide safety critical functionality thanks to integrated hardware safety mechanisms.
- Infineon AURIX Family - The AURIX TC3xx line includes variants with multiple cores and an automotive focused architecture and reliability.
- ON Semiconductor RSL10 - The RSL10 provides Bluetooth connectivity combined with solid MCU capabilities. Good for wireless communication.
Some key considerations would be necessary performance, memory, peripherals, functional safety mechanisms, power consumption, and overall suitability for automotive applications. But any of those suggested families could potentially be an alternative to the original Sharp LH75401 used in this design.
Display
What can you tell me about the LCD?
Here are some key details on the LCD display used, based on what information is available in the schematic:
- It uses an Optrex branded LCD module (page 4, labeled as Optrex Display)
- Appears to be a TFT display with integrated capacitive touch screen (Touch Panel connector J4). Likely color display.
- The LCD interface uses a 50 pin FFC cable (J3 connector), indicating it’s a fairly complex high resolution display.
- A separate LCD bias voltage is supplied, implying that it requires this bias voltage to drive the LCD pixels correctly. This also indicates it’s passive matrix type rather than active matrix.
- It has separate control signals for LCD VDD, bias enable, VEE, and backlight control, allowing flexible control of the display voltages as needed.
So in summary, key points are that it’s likely a color TFT display with capacitive touch from Optrex. Resolution, size, and exact model is unknown without further details but it does appear to be a fairly feature rich automotive display panel judging by the schematic support circuitry. But the full specs aren’t fully labeled or available in just the schematic itself.
Why would the Optrex display have been selected?
Here are some likely reasons why an Optrex display may have been chosen for this automotive display design:
- Optrex focuses on automotive-grade displays so their products are ruggedized and qualified for vehicular use cases. Important for reliability.
- They have offerings spanning simple character LCDs to more advanced TFT displays, allowing matching different price and capability needs.
- Optrex likely had suitable display sizes, resolutions, viewing angles, brightness levels, and durability metrics required for the vehicle cluster or center console placement.
- Support for operating temperature ranges compliant with automotive standards. The displays won’t fail across vehicle temp variability.
- Availability of capacitive touchscreens integrated with the TFT LCD simplifies system design vs a separate touch layer.
- Mature LCD bias driving techniques for passive matrix panels used by Optrex suits cost targets.
- Possible prior working relationship between Optrex and automaker or tier-1 supplier eased adoption vs other vendors.
In summary, the automotive focus, flexible display lineup, integrated touch options, proven passive matrix LCD technology, and availability likely made Optrex a logical choice for the automotive display needs 12 years ago. Their specialization made them well matched to the requirements.