• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 66 subscribers
  • Views 1037 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

cable length between TLC 5940 chips

Mirko Nuncium
Prodigy 30 points
Other Parts Discussed in Thread: TLC5940, TLC59711

Hi, Is there a limitation of cable length between daisy chained TLC5940's? I am trying to connect 14 chips using ethernet cable of 6 meters distance and encountering problems.

  • 0
    TI__Intellectual 2540 points

    Mirko,

    What is the spacing along the cable?  In other words, are the drivers placed every 6m / 14 = 0.42m or is each driver 6m from the previous?  What are they mounted to.  In general, the TLC5940 is not optimized for really wide spacing between drivers in a cascaded configuration.  We have parts intended for this purpose, so you might look at TLC5970, TLC59711.

    John

  • 0 in reply to John Perry
    Prodigy 30 points

    Hi John, thank you for the quick answer,

    They are daisy chained in between eachother with cables of various lengths, from 0.5 meters to 6 meters (the longest) So chain is in total about 14 meters long. I use Arduino mega as the main driver. ethernet cable for comunication between chips and additional cable for 12v power with shared ground,  Chips are driving RGB 12v power leds.

    I tested the chain of 5 chips with shorter cables ( about 05metrs) and they work very well, but once i connected 6 meters cable, problem appeared.

    Do you know what is the maximum length i could use? and is there a way to expand it? I should stick to 5940's since i already purchased them and printed the circuit boards for them .

  • 0 in reply to Mirko Nuncium
    TI__Guru 61245 points

    Hello Mirko,

    When you say "problem occurred", what exactly happened? Are the parts in the daisy change work properly up to a certain point in the chain and the ones after the first failing unit are failing as well? If this is the case, I expect that there is a timing issue between SCLK and SIN. Please measure these signals on each part with reference to the parts GND and check if there is a timing issue.

    There are 2 possible issues you could have in this case:
    - The slopes of the SCLK signal are getting flatter and flatter, the longer the SCLK line. You might need a buffer somewhere in the row to get the slopes steep again.
    - There is a short delay between the signal available on SIN and the MSB available on SOUT. This could as well lead to timing issues between SIN changing and SCLK rising for long distances. If this is the case, most probable you need to delay the SCLK for the later parts somehow.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 30 points
    Hi Brigitte, I have a flicker problem with long cable even when only one module is used, meening that SOUT-SIN connection is not in use.  In other case, with 3daisychained tlc's, when i connect 3rd tlc in the chain with cable that is 1meter long, my leds driven through 3rd chip are getting artifacts, they flicker when they should be off, and have low frequency when they are on, while first two chips are doing fine. Third chip also burns after some time i have a 1mf capacitor between gnd and vcc. You mentioned that i could make a buffer, how could i do that?
  • 0 in reply to Mirko Nuncium
    TI__Guru 61245 points

    Hello Mirko,

    Please check first if you have a GND problem on the module. Is the supply voltage changing through the long cable that could explain the flickering? So connect just one module through a long cable and measure the voltages on the module with respect to the GND pin of the IC. If all voltages are fine, go on with the next check below.

    When you observe the flicker issue and the supply voltage of the LEDs and the IC are stable, please measure the communication interface while writing to TLC5940. Through a long cable, the timings of the communication could be wrong writing strange content to the memory.

    Before making sure that all signals are correct and stable, it does not make sense to check if a buffer helps.

    Best regards,
    Brigitte

  • 0 in reply to Mirko Nuncium
    Prodigy 10 points

    Sounds like the frequencies of the ics are causing that problem. Set a 100uF cap between the power supply ground and 5V (yes, "shorten" the circuit) and on each ic an 0.1uF cap as close as possible to the ics power supply (gnd and vcc). I solved the same problem by doing so.

     Read this article for more information: http://www.eetimes.com/design/analog-design/4009954/Choosing-and-Using-Bypass-Capacitors-Part-1-of-3- 

  • 0 in reply to John Perry
    Prodigy 10 points

    @John - Any way the LTC59711 can be daisy chain in lengths upto ~300-500ft for controlling multiple analog PWM servos?