Newhaven Products > Character LCDs

Issues upgrading to NHD-0220D3Z-V3

(1/2) > >>

I have a device that currently uses the NHD-0220D3Z display (Rev 1.0).  As that product is no longer available, I am upgrading to the NHD-0220D3Z-V3.  As far as I can tell, this should be a more-or-less drop-in replacement, with little to no work on my end.  Unfortunately it has not turned out that way.

Comm setup:  RS232 @ 9600 Baud.  I have both R1 and R2 as open connections.

When I connect a -V3 display to my device, I get what seems to be a stream of characters, usually non-English, written to the display.  I strip down my code to do nothing (I do set-up my UART for the comm, but I never write anything to the display) and I get the stream of characters.  I change it to display just a single letter (e.g. "A") or send a single command (I've tried clear screen, display baud rate, set contrast and set brightness, together and individually) and I still get a stream of characters.  For each test, I was able to plug in my original display and always got the expected results (the character was displayed or the command was followed).

I found an engineering notice from 05/22/2012 that says something about a firmware update that may be incompatible with customer code, but no details on what is incompatible.  Also, the datasheets for the two displays are nearly identical with only a few minor differences (e.g. -V3 typically draws 35 mA vs 34 mA for the original).

Any thoughts on what to check would be appreciated.


It sounds like it is something to do with the timing.  Not the baud rate of the rs232 itself, but rather the timing between writes/commands.  Can you try adding a delay between writes?

I tried delays of 1, 2, 5, 10, 20, 50 & 100 ms between each write with no luck on the -V3 display.  The Rev 1.0 display was successful regardless of the delay length.

I used my DMM and measured a few pins on the connector.  Because I am using RS232, I don't need to use pins 5, 7 or 8 on the display's connector.  These pins are connected to my micro, which is set-up to drive these pins as outputs with a low signal (so they are non-floating).  When the display is not connected, I measure 0V at each of these pins on my board.  When I connect the display, pin 5 goes up to ~0.3V and pins 7 & 8 drive up to 4.9V.  If I take pin 8 (the data line) and directly connect it to ground (rather than using the low output of my micro), the -V3 display functions as expected with no need for delays between my commands.

The best I can tell, is that even though R1 and R2 are open (thus indicating RS232 in use), the -V3 display will still listen to the SPI/I2C data line if it does not have a hard short to ground.  The Rev 1.0 display shows similar voltages on pins 5, 7 & 8, but seems to ignore pin 8 data when in RS232 mode.

It isn't going to be easy for me to modify my PCBs to hard short pin 8 to ground, so any other ideas to get RS232 working would be appreciated.


I have just tested that display to confirm operation with only the use of 3 pins, without any other pins needing to be grounded.  The 3 pins I used were pins 1-3, RX, VSS, VDD.  Using only these 3 pins, the display was fully functional.  We have never seen this issue before where that pin needs to be grounded in order for the RS232 to work.  Do you have some connection to these pins?  Or are they completely disconnected from any other devices/circuitry as they should be?

The other pins are connected to my micro, but I am not doing anything with them - I am not attempting to send or read any data from those pins.  I don't think that I've said before that my micro runs at 3.3V, so I have a bidirectional voltage level translator between my micro and the display.  It is a TI chip, part number is TXB0108.  The Rev 1.0 display works just fine with this setup, but the -V3 display does not.

--- Quote ---... are they completely disconnected from any other devices/circuitry as they should be?
--- End quote ---
I am little surprised by this.  No where in your datasheet does it indicate that these pins need to be disconnected, let alone fully disconnected from everything.  Also, it is generally not good electrical engineering practice to have floating inputs, even if they are unusued, so connecting them to ground or +VDD is often considered a best practice.

I made a test cable that only had pins 1, 2 & 3 connected and no luck, still getting the character stream.  Exact same cable on the Rev 1.0 display works just as you describe.  I'm using a 14-way ribbon cable with IDC connectors and it is about 6" long.  To make my test cable, I just clipped wires 4-14 in the middle and left them floating.

Looking at the display board, the silk screen on the board itself says 2002D3 VER 1.1, but the sticker is for the NHD-0220D3Z-NSW-BBW-V3.  What I believe is a lot number above the model number is 112680-11-03-30.



[0] Message Index

[#] Next page

Go to full version