Two Serial LCD's wired in parallel?

Comments

• rbehm Posts: 102

  2019-04-14 11:52

  You did not say what kind of serial interface these displays have. It it is real RS232 25ft is in the spec of RS232 and should nt be a problem. With pure logic level I would not even try this distance without decent buffers and possibly galvanic isolation.
  
• Jtravers555 Posts: 7

  2019-04-14 12:57

  My first idea was to try logic level, trying to keep it simple.
  
• kwinn Posts: 8,697

  2019-04-14 14:08

  Jtravers555 wrote: »

  My first idea was to try logic level, trying to keep it simple.

  
  Logic level should work over 25 feet, although you may need a pulldown resistor on the lcd's. My preference for this kind of setup would be to use an optoisolator on the remote display and drive it's led with current from the prop pin or a transistor if required.
• iseries Posts: 1,452

  2019-04-14 16:34

  I would to wireless and either connect it to my existing network or use a send receive par. The freedom to move it is well worth it.
  
  Mike
  
• JonnyMac Posts: 8,923

  2019-04-14 16:56

  You might buffer the signal with a TC4427; it's dual-channel so you can feed both inputs with your data and get two stiff outputs from the chip.
• jmg Posts: 15,144

  2019-04-14 21:03 edited 2019-04-14 21:08

  Jtravers555 wrote: »

  For my project, I'd like to have two displays, one in the basement local to the project, and a remote one, about 25ft away.
  
  I'm using Sparkfun Serial Enabled displays, 5v white on blue, and already have the local one working fine. I was wondering if it was possible to wire two of these in parallel? Anyone tried that before?
  
  I'm using 9600 baud for the LCD's, and am hoping that a 25ft run of CAT5 won't pose any problems for that. From my digging, it seems like it could work. Searching for the two serial lcd's pulled up too many false answers to have any value.

  
  That sounds easy enough to set up and try.
  Is there a schematic for the LCD and interface ?
  Things to watch would be Power supply droop over the 25ft, especially with backlight currents, and and RX noise effects will be determined by the LCD hardware.
  You could always parallel conductors in your cable for the 5V power ?
  Sparkfun images suggest they use PIC16F88, but no 5V cap values given & 5V decoupling looks 'minimal', adding a 100uF ~ 470uF power cap could reduce spikes on backlight ON/OFF ?
  
  It may even need a local 5V regulator, if the cable drop is too large ?
  Google suggests 0.188 ohms/m loop, so that gives 0.188*25*60m = 0.282V - probably tolerable. Try slowly lowering Vcc on one, and see when it fails.
  
  I would try 100~220 ohms series driver resistances, into each cable RX leg. That isolates the near display, from the remote one, and you can compare displays to see any issues.
  
  
• localroger Posts: 3,451

  2019-04-14 22:34

  If you are running twisted pair, make sure each signal wire (RX or TX) is twisted around a constant voltage carrier (GND or power). If you twist TX around RX you will get massive capacitive coupling and crossover, but otherwise it should work fine.
• Peter Jakacki Posts: 10,193

  2019-04-14 23:30 edited 2019-04-14 23:39

  It is rather amusing to see 3/5V 1-wire passive pull-up bidirectional "networks" that operate with microsecond timing being run over long distances yet it seems a simple "RS232" labeled 9600 baud into 2 receivers needs buffers. Really?
  
  Just to set the facts straight about RS232 receivers, not as they were back in the 60's but ever since the MC1489 receivers were used. Proper RS232 receivers would have a hysteresis that would trip on a positive signal and require a negative signal to reset it so the noise margin was at least 5V. However if you examine the specs of RS232 receivers you will see that they are lazy and simple turn on with sufficient positive voltage at around +1.3. Well, that's not so bad except that they reset at about +1V which means they only have a noise margin of 300mV and also means that the negative voltage is redundant. So a 0 to 3V (that's 10 times the noise margin) works just as well as a +/-6V (typical) signal but probably be better because its source impedance is typically ~40 ohms vs the 300 ohm source impedance that all RS232 drivers had to have since the 60's when circuits were a lot easier to break.
  
  So I would not even hesitate to connect these displays up in parallel and drive them directly from a Prop I/O over that distance but maybe with a 100 ohm in series at the Prop pin as some kind of protection.
  
  Anyway, I notice that many manufacturers use the label "RS232" when they are only talking about asynchronous serial, whether it be inverted or non-inverted levels and only logic level. RS232 only ever referred to the electrical specification, never the data format.