At the moment I am building a controller for my 640x400 controllerless lcd screen. Now, I have done the math and know I have very little time before the next set of data has to be shoved out, but it is possible. Now, knowing that is all good I would have been finished by now but, I am having trouble with the data lines. They seem to be... not logical. Like, if they are floating, the screen is blank, but if one is either pulled high or pulled low, that bit is lit up. Also, it seems that one wire 'bleeds' into the other. I am not sure exactly why this is going on. I got all the negative voltages how others said they did it. I hope someone on here can help, as I tried the microchip forum and they didn't help at all. Thanks! LCD Display: EDM-LG64AA-44D Datasheet: Here - Craig
You're going to have to post more information. Your circuit, type of device you're using as a controller, firmware, etx...
Alright my mistake. The circuit right now is extremely simple. THe negative voltage is generated, well, simply by some old wall transformer things so, not much to it there. I am using a Pic Microprocessor (18F4550 @ 48Mhz) to do the Frame, Load, and CPX clocks. The source code is really simple. Now, I have also an asm version but I didn't really put much effort to it, but it will have the timings down more to the exact specs. Simply each of the control wires goes to one of the PORTA i/o's on the microprocessor. The data lines are just sitting randomly on the breadboard and I just stick it in the 5v or 0v rails attempting to see the lines down the display. This afternoon i'll get a pic, but its really simple setup. I hope that covered the general important parts I forgot. Code: #include <p18f4550.h> #pragma config PLLDIV=5, CPUDIV=OSC1_PLL2, USBDIV=1, FOSC=HSPLL_HS, FCMEM=OFF #pragma config MCLRE=OFF, LVP=OFF, PWRT=ON, BOR=OFF, WDT=OFF, PBADEN=OFF #define LCD_CPX LATA0 #define LCD_FRAME LATA1 #define LCD_LOAD LATA2 void main() { unsigned char seg, line; ADCON1 = 0x0F; CMCON = 0x07; TRISA = 0; TRISB = 0; while(1) { LATAbits.LCD_FRAME = 1; for(line=0; line<200; line++) { for(seg=0; seg<160; seg++) { LATAbits.LCD_CPX = 1; Nop(); LATAbits.LCD_CPX = 0; Nop(); } LATAbits.LCD_FRAME = 0; LATAbits.LCD_LOAD = 1; Nop(); LATAbits.LCD_LOAD = 0; Nop(); } } }
I was affraid of this, getting no reply. Does any one have atleast any guesses to what is going on? If not, I probably will only give this lcd one more try in like a month then call it a lost cause. =/ Thanks for the attempt anyway. =)
Well looking at the data sheet it seems that the display generates vee itself and you just need a pot to divide that voltage down to a contrast you are happy with. In digital circuits having floating lines is a killer, for every wire determine if it should be high or low and then connect it as such. Can you find any app notes or better data sheets with more information? Rod
I know floating is bad... but I don't understand why the digital inputs arn't acting so... digital. I am really lost on that one. Also, there is no sign of a on-board negative voltage generator, but I of course put my meter to it, and nope, just 5v when floating (guessing it has some resistance to Vdd or something). Saddly to say, this is the only datasheet. I searched and searched. Also I tried getting information on the actual chips on the back, with no results. Edit: Finally got some pics. Not sure if it will help much but atleast you can get a view of the nice display.
It sounds more like a contrast adjustment problem. Its possible the reason why you dont see different results on the screen for the digital inputs is that the contrast is not adjusted correctly. Remember, there is a very small 'sweet spot' over the entire range of the contrast adjustment to see the display properly. Try each digital input HI/LO again, but adjust the contrast. PS: fully adjusting contrast too far one way or the other will give you a full screen, etc
High impedance digital inputs can float. Sometimes they have protection diodes in to prevent them going too high or too low and this is why you can see them floating either to vdd or vss, Remember that your meter can affect the reading you get. I would suggest that you try slowing everything down, putting millisecond delays rather than microsecond delays and trying again. My experience with working with LCD's is that they are a pain in the arse to work on without a scope because I never really trust PIC's to work as programmed and trying to debug two things at the same time is a bit of a headache. Rod
Rod10K sorry you misunderstood me. The inputs are working correctly, and it is showing what it should. The problem is that the only way to get it to read 0 is to leave the wire in extremely high-impedence. Even a simple pull-down transister's internal resistance will cause it to read as a 1. I'll screw around with the contrast more but I don't think it is the situation this time. Also, some times one wire will 'bleed' into the other. Like, if I pull 1 wire low making that column go dark, the one next to it also will. On a final note, the code I posted gives a 20Hz refresh rate, so it isn't too timed yet (will rewrite into asm later after this situation is solved). Maybe you can have some luck on finding information. Here are the chips on it: The first 2 there are multiple of them. I figure those are the main lcd drivers, but I wasn't able to get any information. The last 2, there is only one of each, probably some sort of special-purpose ICs. I'll report if I find anything with the contrast and a few of my own experiments. Thanks for your help so far. Much better than the other forum I tried. =) *** Again Thank you very much. Edit: Ok well, I still haven't figured this thing out. If no one has any more ideas... I guess i'll just push it to the side for now. Would have been nice to have a good sized lcd working though.
an electronics shop by me has surplus lcds like this with no controller, and they have a tutorial about how someone got it to work. It's pretty detailed, so it might give you some ideas. One thing it mentions is to alternate the polarity every frame, otherwise you can damage the lcd, maybe this has something to do with the bleeding? I havn't played around with an lcd like this though, hope this helps! http://www.vetco.net/LCD_DOCS/AN001-Driving_INMP001_LCD.pdf
Well the graphics lcd I am using doesn't use anything like that, so I am a bit confused on that. Well, i'll check out some other lcd's, Maybe another good deal, because it seems this lcd is just being way to much of a pain in the butt!
In this senario it was not needed. The data lines are being directly driving high or low... so pull-up resisters are not needed. Honestly I think I tried just about everything to get this thing to work, and without a scope to see the actual signals, I am just about out of luck. I'll be posting another thread about a scope.. =P
