This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DLPC1438: 4X native resolution without FPGA

Part Number: DLPC1438
Other Parts Discussed in Thread: DLP301S, , DLPA2005

Hello

I am trying to design DLP engine based on DLP301S and I have several questions.

1. First, I want to design DLP electronics, DLPA2005 + DLPC1438 + DLP301S + actuator (4X native resolution or 2560X1440) without FPGA (partly because this FPGA seems hard to purchase). It seems feasible but I need more detailed information about the workflow that the FPGA performs. The workflow I am thinking of in my design is as follows.

Using MCU with DPI and DAC (8bit, generating waveforms of an actuator)

Read image data (2560X1440) from an external frame memory and divide into 4 subframes (1280X720). Does TI provide this algorithm or software library?

Send subframe n data to DLPC1438 => LED on => wait for t_subexposure (a multiple of 1/60 sec) sec => drive actuator => send subframe (n+1)%4 => repeat => LED off. To do this, the subframe data should be sent at every t_subexposure, not 1/60 sec. This is because, unlike display applications, minimizing actuator transition time may be desirable for 3D printing. Is this possible with DLPC1438 and DLP301S? Or should the subframe n data be sent at every 1/60? From DLP1438 datasheet (p26, 7.3.2), in External Print mode, it seems that the user can send the frame data via DPI only when necessary. So, my understanding is - DLPC1438 receives subframe data with lower pixel clock => when complete, send data to DLP301S via LVDS => DLP301S refresh the frame and hold it for a multiple of 1/60 until the next subframe data arrives.

2. If I don’t want to use built-in LED driver in DLPA2005 (LED_SEL_0 = LED_SEL_1 = low, SW4 = SW5 = SW6 = unconnected), what do you recommend for unused pins, VINL, L1, L2, VLED, RLIM, RLIM_K? Just leave them unconnected? This is to use an array of LEDs in serial connection and external LED driver.

3. If I don’t want to use measurement pins (PWM_IN, CMP_OUT, SENS1, SEN2), what’s your recommended connection? The analog comparator output (CMP_OUT) may need to be kept low.

4. From DLP301S datasheet(p9, 6.4), the max illumination at 400nm is 450mW/. If the micromirror array temperature is a main limiting factor, can I increase it (say, 2X~4X) if I apply active cooling (such as Peltier device) to maintain normal operating temperature? To some extent, lifetime decrease (say, ~20%, more or less) may be acceptable possibly due to material degradation caused by strong light intensity.

5. When uploading firmware to external flash memory for DLPC1438, what is the start address, just 0?

Best regards,

  • Hello User,

    Welcome to the E2E forum and thank you for your interest in DLP® technology!

    I will need to look into your questions and get back to you in a few days.

    Regards,

    Lori 

  • Hello User,

    I apologize for the delay because of the holidays.

    Please refer to the points below:

    1. We don't use a MCU with DPI and DAC to generate actuator waveform, we only support FPGA with actuator driver.

    Also, the actuator transition time is already done during a dark time in the illumination, so there is no advantage for print quality to hold the actuator in a fixed position. Some of the actuators on the market have a hard time staying in a fixed accurate location for seconds, so the DLPC1438 chipset cycles between subframes at a rate of 60 Hz to ensure good print quality. This means an entire frame of four subframes is displayed once every 15 Hz.

    2 and 3. From a hardware point of view, you should be able to remove the build-in LED driver. However, TI has not verified/tested the DLPA2005 without external components for the illumination block. If you want to remove the parts, you have to do the test yourself and TI does not guarantee functionality. 

    4. If the user exceeding either the illumination power density limit or the thermal limit will void the warranty and may cause early failure of the device. The limit set at 400 nm is intended to allow 405 nm center wavelength LED illumination at 3 W/cm^2, since these LEDs have some wavelength diversity down to 400 nm and a small amount below that. 

    5. If you planning to use the host commands available through the controller to program the flash, then you don't need to worry about the starting address. The ESW will take care of placing the flash data correctly as long as you select the correct flash data type. If you are doing it on your own, the start address should be 0.