LCD mini click (PID: MIKROE-2453)

[solveering]

My review of the LCD mini click (PID: MIKROE-2453):

As part of an ongoing project, I required a small, minimalist LCD display and the MIKROE-2518 fit the requirements quite well (it is not included in the click, as expected, even though it is shown). Unfortunately, getting any form of documentation for the display itself was close to impossible. The ME support team gave me the 'we don't have the technical docs' reply, which is strange and somewhat disturbing. Nevertheless, I purchased two... for anyone else that may be interested to know, DigiKey carries these too (so shipping is less time & $$s) as well as others that are likely made by the same OEM and have datasheets available for them.
Anyway, this is about the Click board...
The ME site does not provide a lot of information other than to list the two components that are used on the board (MCP4161 and MCP23S17) but it is possible to find the schematic. Unfortunately, it is wrong. After working through setup, configuration and a number of other items, I concluded that the one item I was relying on to make the display turn on and show some text (by following the libstock example), was not working as intended. To be fair, anyone using this with a development system may not be experiencing these issues, but when the datasheet says that the unit has an "Input Voltage : 3.3V or 5V", I expect the unit to work either with 3.3V or 5V. The reality is that the interface works either with 3.3 or 5V signals, but the board needs 5V. If you look at the datasheet, the 3.3V OR 5V is supported by the fact that all voltage levels (Vcc) are tied to either 3.3 or 5. But what should be stated is that the LCD display requires 5V and is tolerant to a wide range of signal voltages. If you don't have 5V, it will not turn on.
Once I figured that out (three days into what was meant to be a simple add-on), I could not for the life of me figure out the data stream. The library setup is simple enough: enable the SPI-LCD library and use the example. The only problem is that the information generated by the library is not what the HD44780 on the display is expecting. There is a good chance that I did something wrong, but the setup is so simple that I have not figured out what it could be.
The solution to this issue came by replacing the Click with a direct connection from the processor and then simply using the LCD library. And now the display worked like a charm (I had to use a potentiometer for contrast adjustment, but that is secondary).
Lastly, the information about this board states that the interface uses two CS lines, an RST,SCK MISO, MOSI, 3.3V, 5V, Gnd, PWM and INT lines. The MISO and INT lines don't do anything so you can leave them out. If you compare that to the 4bit mode used for the LCD, you will see that, except for the contrast adjustment, you can achieve the same using 6 digital lines + PWM so only one extra pin if you don't count the MISO/INT issue. Considering that you can pick any GPIO pins with the LCD library and you must use a hardware SPI port for the click board, I just don't see the point.

But the bottom line is: if you don't have a development board or 5V available and can't get this to work right out of the box, you are probably best off using something else. I would give this a 1 out of 5. I am very disappointed and question whether this board was ever tested before being sold (I bought two, so this is not a Mfg defect).

For anyone interested in looking closer into this, I have attached logic probe data of the LCD library output and SPI-LCD output as received by the HD44780 controller on the LCD. The output to the display should have been the same ("TEST") but clearly something is different... I would be interested in hearing from anyone that has actually used this board successfully.

[lana.arsic]

Hi,

I apologize for the inconvenience, old version of the schematic was published
this is one of older click boards which should be revised.
But now our developers have corrected it, thank you for notification.

I agree that information that it requires 5V should be added
and it is now corrected on website.

All our click boards are made so they would be placed into mikroBUS sockets,
and purpose of LCD mini click is to place LCD with mikroBUS socket.

INT pin is used as interrupt pin from MCP23017, you can read more about its function in datasheet:

https://download.mikroe.com/documents/d ... 21952b.pdf

and SDO is used for SPI.

All our click boards are tested before sending, but if you doubt that they don't work,
you can open a ticket.

Kind regards,
Lana

[solveering]

The schematic being incorrect is just one aspect of this. The main problem I encountered was that the SPI-LCD library generates an output to the LCD (so measured at the FFC connector) that differs from what the LCD library generates (SPI-LCD output does not display anything, LCD library does). I was under the impression that the purpose is to provide an easy way to sent data to an LCD by toggling the pins of the port expander on the click board to replicate the data that would be sent if it was a direct connection to the LCD + having the convenience of a software controlled contrast setting.
The two images showing the logic probe shows that the data getting to the HD44780 controller is different. I have also attached the raw Ideofy file (http://www.ideofy.com/la-08_en) in case you want to take a closer look (software to view these is free).
As to the MISO and INT pins, I understand what they are, but they don't serve a specific purpose when used with the SPI-LCD library (I saw no data on them). If the Port Expander library is used with some script to access more advanced functions of the HD44780 (such as perhaps custom characters or the like) then they may have some limited purpose though I don't see what would cause an interrupt as the data is all in one direction. Reading data from the display... not sure that makes much sense either. But I won't argue with them being there as there might be some form of purpose. What I would argue with though is why the LCD uses 4bit mode when the Port Expander on the click board has 12 unused pins. It would have been more logical to connect it in a 8bit mode and then use software to determine which mode is to be used... but anyway, not trying to be picky, just providing my review of and comments on this board.

[lana.arsic]

Hi,

Libraries are different, but you should get proper output
using provided example with SPI-LCD library also,
I have tested on EasyPIC PRO v7, which exactly MCU did you use?

I have forwarded your complain to our developer to consider to make revision,
but at this moment it is in 4bit mode, so it can be used with compiler's library
which supports 4bit mode only.

Thank you for your comments and for your time, we really appreciate it.

Kind regards,
Lana

[solveering]

I used a PIC18F25K50 on a solder-less breadboard. The connections were fine, I tested and retested them several times (captured the serial data, etc.).
My recommendation is that you capture the signals at the LCD and compare the output for the two libraries (like I did and attached previously). if you see a difference between the two, then there is your issue. If not, then I made a mistake somewhere (please let me know if that should be the case).