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.

TIDA-00325: Design files for flex PCB

Part Number: TIDA-00325
Other Parts Discussed in Thread: DLPC3435, DLP2010,

Hello,

Is there any documentation regarding the flex PCB (TIDLP-108374-01) that connects the optical module to the upper board? I've contacted Asia Optical and they told me to ask TI for the design files. A set of gerber files or ideally a schematic would help a lot.

    

  • Andy,

    Thanks for your interest in our DLP products, and welcome to the E2E forums.

    I don't believe we have the design files for these flex cables available at this time. That being said, there are resources available for you to create a flex cable. Please see this thread for more information:
    e2e.ti.com/.../2005403

    I hope this helps.

    Best Regards,
    Philippe Dollo
  • Hi Philippe,

    Thanks for the reply, it looks like we'll have to make this cable ourselves.

    The datasheet for the DLPC3435 does contain the maximum length/mismatch for the DMD signals, but it doesn't say how these signals should be filtered or buffered. The flex cable has a bunch of resistors and capacitors near the connector, I'm assuming those are to filter/buffer the signals?

    Are there pinouts for the two connectors that connect the flex cable to the optical module?

    Alternatively, is there a way for us to buy this cable directly from TI?

    - Andy
  • Andy,

    TI does not offer standalone flex cables for purchase, unfortunately. That being said, We do have a PCB/layout design guide for our TRP devices which may greatly assist you:
    www.ti.com/.../dlpa071.pdf

    I suggest you specifically refer to the DLP2010 section for your particular needs.

    Best Regards,
    Philippe Dollo
  • Hi Philippe,

    Sorry for reopening this topic, but...

    Using the design guidelines in that document you linked, as well as probing the pins on the original cable (the one that came with the EVM), I designed my own flex PCB. It doesn't work, the SPI lines behave as if nothing is connected, and the projector doesn't seem to turn on.

    Here are some of the things I've done:

    - Probed continuity on every pin in the two cables, all seemed to match. The pinout of the DLPC3435 connector matched the one in the schematic, and the pinout of the DMD connector matched the pinout diagram I got from Asia Optical (in addition to matching the pinout of the DMD connector on the original cable).

    - Checked impedance on every trace in the two cables. The highest impedance mismatch among the differential pairs is 13 mohms. The impedance of the differential pairs on my new cable are all about 300-600 mohm lower than on the original cable. The impedances of LS_WDATA and LS_CLK on the new cable are both about ~400 mohms higher than those on the original cable. I can provide a complete list of impedances and lengths if necessary.

    - The new flex PCB was designed in Eagle, using two layers. The traces were all length matched according to the design guidelines. The overall size of the new PCB is roughly the same size as the original PCB.

    - Checked the voltages on the supply rails in both cables (P1P8V, VOFS, VBIAS, VRST), they all seemed to match. All the decoupling capacitors matched, and were placed directly underneath the DMD connector on the other side.

    - Probed the SPI lines. With the new cable, the startup procedure looked identical to the one seen when no cable is connected. I've also checked continuity on the SPI lines between the flex PCB and the ZIF connector on the DMD, and the traces are in fact connected.

    - Connected the EVM to my computer using a USB cable, and opened the DLP Lightcrafter Display GUI (V6.1). With the new cable, the software was able to detect and read information from the DMD, and it didn't complain when I tried to display test patterns. The projector still didn't turn on or react.

    All of these tests were done using a TIDA-00325 evaluation module, and comparing with the original flex PCB that came with the kit. After each test I would reconnect the projector module with the original cable, and it would turn on and show the splash screen.

    Are there any more troubleshooting tests I can do? Do the DMD trace impedances need to be a certain amount, in addition to being matched? Let me know if there's any more information I can provide.

  • Can I please get an update on this?

  • Andy,

    Sorry for the delay. Thanks for being so thorough in your analysis. Let me read over this and compare it against my notes. I'll let you know in the next day or two if there are any obvious mistakes based on what you are shared thus far.

    Best Regards,
    Philippe Dollo
  • Andy,

    I'm not sure of your background and experience so I have some questions/comments:

    Are you measuring impedance or resistance? The LS_WDATA, LS_CLK and LS_RDATA signals need to be 65ohms impedance in the flex cable. Impedance is measured using a TDR or similar tool. Each pair of the differential signals need to be 100ohm differential impedance.

    There should be a 45ohm series terminator on LS_RDATA close to the DMD.

    If VRST = -14V, VBIAS = 18V and VOFS = 10V the DLPC343x is booting correctly. If these three power rails are going to their operating levels the LS bus must be working.

    Are the leds coming on?

    Regards,

    Kevin

  • Kevin,

    Thanks for the reply, I'm a new grad so I appreciate your patience.

    1. I've been measuring resistance, I thought the capacitance/inductance would be negligible. I'll look for a way to measure differential impedance in my layout software.

    2. There is a 45ohm resistor in series with LS_RDATA close to the DMD connector.

    3. Something weird is happening here. When the flash card is connected and the DMD is disconnected, I read the following voltages: VRST = 0.5V, VOFS = 4.5V, VBIAS = 4.5V. When I (successfully) program the flash card, I read VRST = -14V, VOFS = 10V, VBIAS = 18V. When I connect the DMD, these voltages drop to the previous readings of VRST = 0.5V, VOFS = 4.5V, and VBIAS = 4.5V. The correct readings only show up after I program the flash card and before I connect the DMD. My decoupling capacitors are rated for 25V on VBIAS and 16V on everything else.

    4. The DMD does not come on.

    One thing I should mention is that these flex PCBs were populated by hand and reflow soldered on a hot plate. Out of the 7 cables I made, there were 2 that did actually work when I hooked them up for the first time (DMD would turn on and show the splash screen). However they stopped working once I unplugged them. My best guess is that something went wrong when I populated/soldered these cables, but I've inspected every pin on each connector and there are no shorts/cold joints (pinouts still match the original flex cable).

    - Andy

  • Hi Andy,

    There is no activity on this thread, I hope you have resolved your issue. Please let us know, so that i can close this thread.

    Thanks,
    Navneet