Matrix user forums

Ηello to all!
I would like to use an GLCD to display some messages . So I tried to send some characters to GLCD to test the component. As GLCD I use a VARITRONIX MGLS240128TA-03 . The GLCD is new but not work. I was checked the connections with PIC but I don't find something wrong. The test code works with simulation mode, the PIC works ( I was checked with oscilloscope), so what is the reason for that? The component maybe? I post below the test code and an icon of GLCD , for any help. Thanks in advance.

There are many different GLCD controller available. Flowcode GLCD components only support a few of them. I don't think your display uses a supported controller.

Thanks Kersing for info. As I saw in GLCD pcb , the controller is T6963C , that is supported by FC7 components. Below is data sheet of GLCD that shows the controller. Is that supported from FC7 components?

Hello,

The T6963C display driver IC is like a lot of these display drivers, they can work in a lot of different ways. They are designed to be hugely flexible and support a lot of different options in terms of the layout of the pixels and make up of the display.

Therefore just because the display works with the driver chip in one configuration doesn't mean it will work out the box with the fixed component library.

Do you have any example code that you can point to that does work with the display, we can compare this to the component code. If the code is different then we can maybe add a new component or a property to alter the generated code.

Thanks Ben for info. You can see at http://en.radzio.dxp.pl/t6963/, that has some code samples for this controller. These are for ATMEL MCU, I hope it helps to make a component for PIC. Below are some info for interfacing of T6963C. Thanks.

Hello,
I've tested last week a display with the same controller but in my case it was a Winstar WG12864F. So far I got only some random lines and nothing else. Before to ask for support I will test some more display and also make some oscope check.

In your configuration I noticed that you enabled the Watchdog but not checked the Auto clear.
I believe you should disable it or check the Auto clear.
Also in the Loop I would add a delay or a blinking led just to avoid a continuous call of the component macro.

For Benj:
If you make a new component please leave some adjustable pin to play with, thanks!

Regards
Giuseppe

Hello,

Here is a new version of the component you can try. I've made a couple of minor changes that should hopefully allow things to work correctly. This display driver seems to have two layers, one for text and one for graphics or maybe two for double buffering tricks and so maybe one of the layers is not being cleared correctly.

Simply copy to your "Flowcode 7/components" directory and restart Flowcode.

For Benj:
If you make a new component please leave some adjustable pin to play with, thanks!

I haven't touched the pin configuration, is this ok as is or are there other pin configurations that need to be catered for?

Let us know how you get on.

Hello Benj,
I've tested the new component but no luck in my case!

Giuseppe

Ok thanks, I'll have another run through the code tomorrow.

Not working for me!Youtube Video

Many thanks to you both for testing. I was trying to use the bit set and bit clear routines to avoid having to read modify write for each pixel but it looks like this may be causing the problems.

I've now done things exactly as they are happening in the library so hopefully this will now work better for you. I've also disabled the text area of RAM just in case this was not being cleared and displaying uninitialised pixels on screen, everything the Flowcode component currently does is in the graphical area of RAM.
Again let me know how you get on.

Hello Benj,
I'm sorry but I get only garbage on the display.
I'm attaching flowchart, schematics and some snaps just to compare my settings. As this display has more control pins it could even be my wrong configuration.
In the flowchart if I disable all gLCD macros and I use only the LED blinking at 1 sec it works.
If I enable all macros and try to initialize and print then the LED blinks about once every 12 sec!

Is there any test that you suggest? I can use o'scope with serial analyzer if it helps.
Thanks again for checking.

Regards
Giuseppe

Here also my display.

Thanks
Giuseppe

Sorry I've got a bit side tracked by other projects. Please feel free to keep shouting for a solution.

Many thanks for the schematic and project code they look fine to me. My only suggestion would be to add a flashing LED to triple check that the MCU is actually running and at the correct speed.

https://www.matrixtsl.com/wiki/index.ph ... ED_flasher

I'll have another pass through the code for you today, surely we must be close now

Right here is another version of the component to try,

I've improved the clear function to hopefully clear all the variations of RAM memory on the display.

Hello Benj,
don't worry for your delay. I understand you have so many things to do and no problem for that. I don't pretend you answer all our request of course!

Anyway I'm confused. I don't understand your request regarding the blinking led! It's there!
If you read my post I said:

"In the flowchart if I disable all gLCD macros and I use only the LED blinking at 1 sec it works.
If I enable all macros and try to initialize and print then the LED blinks about once every 12 sec!"

It looks like the PIC is hanging when communicating with the display. I also tried some delay but unsuccessfully so far.
Please let me know.

Regards
Giuseppe

Woh! You just sent the new component while I was writing my answer!
OK, I'll try and let you now.

Thanks again.
Giuseppe

"In the flowchart if I disable all gLCD macros and I use only the LED blinking at 1 sec it works.
If I enable all macros and try to initialize and print then the LED blinks about once every 12 sec!"

Sorry I missed that.

Hello Benj,
I'm sorry but still the latest component didn't work. Almost same random pixels.
I would anyway invite the other two users SILVESTROS9 and GUTI to test as well, because I could have done mistakes of course.

I've captures 1 sec digital data with a Picoscope MSO. If you have Picoscope 6 installed, you can easily zoom back and forth to see details. The traces on the scope are from D0 to D13 and you can see the connections below.

The data are captured at power on and triggered with D0 (pin C/D on the display) falling edge.

Thanks!
Giuseppe

Hello eveyone!

Yesterday, i tried the last project Benj submitted and my result is different. The screen just clears out, nothing seems to be responding. When i tried to compile a little bit of code to try the arduino to flick the led, the led responds properly. Disconnecting some digital pins make some difference though, when removing bit 1, some lines were drawn on the screen. Rebooting the arduino clears the lines previously drawn. The result is the same with the 3 screens I've tried, all of different sizes. I'm willing to try anything you guys want.

Thank you all, hope i helped!

Hello
I recommend that you review the position of the jumper on the product you are using. LCD product; Can you share your brand's datasheet?

I experienced a similar problem while using Winstar RAIO RA8835 LCD. For the 6800 and 8080 there was a selection jumper.

Another issue seems to be that the LCD pins are connected directly to the processor. When the PIC18F2550 is directly connected, it may be that the LCD draws no current.

The other subject; Try feeding the LCD's supply voltage, PIC18F2550. Do you try GND with two different sources in common?

Hello,
thanks for your suggestions! My datasheet is attached above but there aren't jumpers on the pcb.
Also I don't see why it should not work with the configuration above and one single power supply. It's not the PIC to supply current to the LCD but the battery. I can try to separate the sources but I'm afraid that's not an issue.
Thanks
Giuseppe

Could the power supply or LDO not supply enough current?

Software should also add delay after gLCDintial operation (ex: 100ms)

You should check PortC1 connection and internal oscillator settings.

I can help you with the entire system, if you can brace the circuit diagram.


and What is LCD brand? (is not T6963,hantronix or toshiba ) ex: Winstar, Raystar etc
Usually (jumper) is on LCD nin as Dummy Resistor.

If you read the entire post from beginning you'll see front/back of my gLCD.

celebriums wrote:Could the power supply or LDO not supply enough current?

I can exclude that because it's supplied with 2 18650 Li-Ion battery with a regulator to 5V. Vcc is always stable at 5V (checked by oscope).

celebriums wrote:Software should also add delay after gLCDintial operation (ex: 100ms)

I already tried to put delays after each gLCD call but no change.

celebriums wrote:You should check PortC1 connection and internal oscillator settings.

See my data above. I got data from PortC so I can't see anything wrong in the Port. Also I measure 2 MHz at RA6 which is fosc/4 as expected because I set the internal osc at 8MHz.

celebriums wrote:and What is LCD brand? (is not T6963,hantronix or toshiba ) ex: Winstar, Raystar etc
Usually (jumper) is on LCD nin as Dummy Resistor.

This is a good point. The controller for sure is a T6963 by Toshiba (it's the most upper right chip in my picture above) however I couldn't find a specific schematic. It could be some smd resistor to set but I don't have useful info. Of course if you have a similar (old!) schematic I would be thankful!

Regards
Giuseppe

Hi
Thanks for information.
I think the difference between your t6963C and the communication speed of the microcontroller.

I watched the video, I saw you work on a breadboard instead of a PCB.

In addition, can you test the functions in the following software? If not, will you try using an external oscillator?
Note:
Changed the codes.
Please pay attention to the transparent option in the codes.
please pay attention to the LED flashing speed and forget the external oscillator option.