Post by thread:[PIC:] LCD & KS0066

Exact match. Not showing close matches.
PICList Thread
'[PIC:] LCD & KS0066 - Another goofy prob?'
2003\11\05@134604 by 1zlx0fc02

Hi: That BGMicro thread was pretty timely had I aquired one, but I got a 16x2 LCD from http://www.timeline-inc.com. Yes, next time I'll pay a few more bucks for a different module, most likely. The photocopied spec sheet they sent (only the 1st page - they can't e-mail them, either and only ship once per week) is for the Hitachi LM052L module. The module itself sports the number KP-03. I've got the operating instructions for the Seiko modules off the net as well as the HD44780U spec. sheet. But as I said, I've got a KS0066 controller with the KS0065 chip ( a driver? ) After looking at the numerous C and asm code examples mostly from PIClist, I came up with the code posted below. My problem is different from that of the poster with the module from BGMicro. At this point, I've just used a pot between +5V Vdd and Vss for the contrast. I've got the first row of 16 chars blacked out since it still thinks it's a 1 row module. When I turn the contrast up towards Vdd, the dots get lighter. Very dark at 0V. I'm using 4 bit mode and the lower bits of the module are floating high. I figure that's ok because bit 1 and 0 of the upper nibble is what sets the data length regardless of the states of the lower nibble. I'll ground those other lines though, since that's really the right thing to do. It appears that this module isn't getting it's instructions. I'd have been real happy if the thing went blank like it did for the guy with the BGMicro module. At least then I'd know something was happening and could think about using a negative contrast voltage, which may not even be needed for my module. I'm using a 16F628A. My CLKOUT pin is reading 1.5MHz so the internal osc is running at 6MHz. I degubbed the code to stop after the data length setting and verified that the upper nibble of the module is getting 0x2 for 4 bit mode. I also degubbed up this point in my main code module: call LCD_init movlw 0x41 ; 'A' call LCD_putc That's all the main code module does. LCD_init sets up 4-bit xfer, 16x2, 5x7 At this point, I verified that the lower nibble of the letter "A" was present on PORTA RA3-RA0. The delay macros (from that generator cgi page at PIClist) were written for 4Mhz osc, so I threw in enough delay for module POR in the LCD_init function to account for the faster speed. I notice the LCD timing diagram calls for Data setup prior to E going high. All the code examples I've studied don't do this. The timing diagram shows E going low in the middle of a valid data frame. I think "E" should be called "/E". I'm setting up the data and strobing the E line high, then low, just like everyone else. Here's that LCD code. I hope it is readable. It came out of UltraEdit. You should see an extra \t (TAB) in a couple of comments. Maybe someone can help figure out what's wrong. I'm out of ideas. It took 3-weeks to get this LCD and I'm not enthused about paying return postage and looking for the credit to my Visa account. Thanks in advance, Mike radix dec INCLUDE "p16f628a.inc" INCLUDE "dlymacro.inc" errorlevel 0, -302 ; Pindefs #DEFINE LCD_D7 PORTA, 3 #DEFINE LCD_D6 PORTA, 2 #DEFINE LCD_D5 PORTA, 1 #DEFINE LCD_D4 PORTA, 0 #DEFINE LCD_RS PORTB, 2 ; RB2 to LCD pin 4 #DEFINE LCD_E PORTB, 1 ; RB1 to LCD pin 6 .lcdreg UDATA lcd_temp res 1 .lcd_module CODE global LCD_init global LCD_putc ; Initialize the LCD ; Hitachi POR sequence: ; LCD POR ; Display Clear ; Function set ; Disp on/off ; Entry mode LCD_init clrf PORTA bcf LCD_RS bcf LCD_E call Delay_10ms call Delay_10ms call Delay_10ms call Delay_10ms call Delay_10ms call Delay_10ms call Delay_10ms call Delay_10ms ; Function set executed first to set interface data length bsf PORTA, 1 ; 0x2 sets 4-bit xfer mode - DL=0 ; high bits xfer'd first call Strobe movlw b'00101000' ; 4 bit xfer, 2 lines, 5x7 dot matrix call LCD_CMD movlw b'00101000' ; 4 bit xfer, 2 lines, 5x7 dot matrix call LCD_CMD call LCD_Off call LCD_CLR call LCD_Mode call LCD_On return Strobe bsf LCD_E goto $+1 bcf LCD_E call Delay_10ms ; plenty of time or too much ; between nibbles? return LCD_CMD bcf LCD_RS BANKSEL lcd_temp movwf lcd_temp swapf lcd_temp, w ; high bits in low nibble BANKSEL 0 andlw 0x0f ; mask out upper nibble movwf PORTA ; put data on line call Strobe BANKSEL lcd_temp movf lcd_temp, w BANKSEL 0 andlw 0x0f ; mask out upper nibble movwf PORTA ; put data on line call Strobe return LCD_Off movlw b'00001000' call LCD_CMD return LCD_On movlw b'00001111' ; Disp on, Cur on, Blink on call LCD_CMD return LCD_CLR movlw 0x01 call LCD_CMD return LCD_Mode movlw b'00000110' ;incr cur, no disp shift call LCD_CMD return LCD_putc bsf LCD_RS BANKSEL lcd_temp movwf lcd_temp swapf lcd_temp, w ; high bits in low nibble BANKSEL 0 andlw 0x0f ; mask out upper nibble movwf PORTA ; put data on line call Strobe BANKSEL lcd_temp movf lcd_temp, w BANKSEL 0 andlw 0x0f movwf PORTA call Strobe bcf LCD_RS return Dly_10ms ; expand macro END Here's the delay code in case anyone cares. It's the function from the cgi page modified into a macro, etc. Dly_10ms macro ; Delay = 0.01 seconds ; Clock frequency = 4 MHz ; Actual delay = 0.01 seconds = 10000 cycles ; Error = 0 % FIXED_REGISTERS UDATA local d1, d2 d1 res 1 d2 res 1 .dly10ms CODE Delay_10ms ;9993 cycles movlw 0xCE BANKSEL d1 ; extra cycles movwf d1 movlw 0x08 movwf d2 Delay_10ms_0 decfsz d1, f goto $+2 decfsz d2, f goto Delay_10ms_0 ;3 cycles goto $+1 nop BANKSEL 0 ; extra cycles return ; 4 cycles (including call) endm -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

2003\11\05@164450 by John Tserkezis

1zlx0fc02@SNEAKEMAIL.COM wrote: > It appears that this module isn't getting it's instructions. I'd have been real happy if > the thing went blank like it did for the guy with the BGMicro module. At least then I'd > know something was happening and could think about using a negative contrast > voltage, which may not even be needed for my module. I had a gotcha that got me some time back. The contrast setting is different between one and two line mode, when being used on a two line display. I had existing working code for a one line display, and managed to get that working first, then changed to four bit mode, and two line mode at the same time before testing again. Spent the next two days trying to work out what went wrong around the four bit conversion, before I realised I had it right all along. All I had to do was tweak the contrast pot a bit. -- -o) /\\ Message void if penguin violated _\_V Don't mess with the penguin Linux Registered User # 302622 <http://counter.li.org> Fido: 3:712/610 BBS/FAX: +61-2-9716-8310 Internet: spam_OUTjtTakeThisOuTtechniciansyndrome.org -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

2003\11\05@233440 by piclist

On 6 Nov 2003 at 8:31, John Tserkezis jt-at-TECHNICI wrote: > .....1zlx0fc02KILLspam@spam@SNEAKEMAIL.COM wrote: > > > It appears that this module isn't getting it's instructions. I'd have been real happy if > > the thing went blank like it did for the guy with the BGMicro module. At least then I'd > > know something was happening and could think about using a negative contrast > > voltage, which may not even be needed for my module. > > I had a gotcha that got me some time back. The contrast setting is different > between one and two line mode, when being used on a two line display. > > I had existing working code for a one line display, and managed to get that > working first, then changed to four bit mode, and two line mode at the same > time before testing again. > > Spent the next two days trying to work out what went wrong around the four > bit conversion, before I realised I had it right all along. > > All I had to do was tweak the contrast pot a bit. > Not a bad idea. It took a while to get some extra pins and tweak the code, but I lashed up an 8 bit xfer circuit. I'm seeing 'A' at the LCD D7-D0, but no joy :( For a second I thought I saw 'A', so I wrote code to follow it with 'a' and 'L' - nope. Interesting trivia thing though. With the contrast set down near 0V, the 1st row is black dots and the 2nd row shows up lighter than the background. As the contrast V goes up and the black begins to match the background, the 2nd row approaches the same color of the background. At higher voltages, the 1st row becomes lighter than the background, while the 2nd row matches the background exactly. Hmm... gotta figure out what to try next. Regards, Mike -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

2003\11\06@035805 by Hulatt, Jon

> -----Original Message----- > From: 1zlx0fc02KILLspamSNEAKEMAIL.COM [.....1zlx0fc02KILLspam.....SNEAKEMAIL.COM] > Sent: 05 November 2003 18:36 > To: EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU > Subject: [PIC:] LCD & KS0066 - Another goofy prob? > > > At this point, I've just used a pot between +5V Vdd and Vss > for the contrast. I've got the first row of 16 chars blacked > out since it still thinks it's a 1 row module. When I turn > the contrast up towards Vdd, the dots get lighter. Very dark at 0V. > The "Default" internal initialisation of a two line LCD will result in exactly what you're seeing here. Set your contrast so that one line looks black, and one line looks clear, and leave it at that. > I'm using 4 bit mode and the lower bits of the module are > floating high. I figure that's ok because bit 1 and 0 of the > upper nibble is what sets the data length regardless of the > states of the lower nibble. I'll ground those other lines > though, since that's really the right thing to do. Tie the low 4 bits low. > It appears that this module isn't getting it's instructions. > I'd have been real happy if the thing went blank like it did > for the guy with the BGMicro module. At least then I'd know > something was happening and could think about using a > negative contrast voltage, which may not even be needed for my module. > I'd agree. I looked at a lot of the PIClist code for LCD's, and found the timing to be really quite optimistic. Particularly then toggling of the E line; many of the samples just set it high, NOP, and then set it low again. It seems that LCDs can miss this. Which is not all that suprising given that they run at a few hundred KHz, compared to the pic's multiple MHz. I wrote some 16F870 code with really slow timing that worked very reliably all the time. unfortunately it's on my home PC, but I can send it to you tonight if you require. The only thing that is not clear about your code... Are you *always* in bank 0 when you call LCD_init? LCD_init clrf PORTA bcf LCD_RS bcf LCD_E is making an assumption about bank0. After the first time you call LCD_CMD, you are de=finately in bank 0, but all that initalisation is risky. The next thing I'd try is putting stupid timing in everywhere. Wait 100ms between commands, and make your strobe subroutine hold the E line high for say 5ms. Give me a shout if you want my code later. Jon -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@041721 by hael Rigby-Jones

> -----Original Message----- > > From: @spam@1zlx0fc02KILLspamSNEAKEMAIL.COM [KILLspam1zlx0fc02KILLspamSNEAKEMAIL.COM] > > I'm using 4 bit mode and the lower bits of the module are > > floating high. I figure that's ok because bit 1 and 0 of the > > upper nibble is what sets the data length regardless of the > > states of the lower nibble. I'll ground those other lines > > though, since that's really the right thing to do. > > From: Hulatt, Jon [RemoveMEjhulattTakeThisOuTMONSTEREUROPE.COM] > Tie the low 4 bits low. I respectfully disagree, you do not need to tie the extra four lines low. I cannot find the document at the moment, but I am positive that the LCD handles the unconnected pins properly when in 4 bit mode. > From: Hulatt, Jon [spamBeGonejhulattspamBeGoneMONSTEREUROPE.COM] > I'd agree. I looked at a lot of the PIClist code for LCD's, > and found the > timing to be really quite optimistic. Particularly then > toggling of the E > line; many of the samples just set it high, NOP, and then set > it low again. How optimistic this is depends entirely on the speed of the PIC. The E pulse typicaly has a minimum width of 230ns. On a 40MHz PIC, one NOP between setting the E pin low and high might be getting a bit marginal. > It seems that LCDs can miss this. Which is not all that > suprising given that > they run at a few hundred KHz, compared to the pic's multiple MHz. But the parallel interface is not simply polled by the processor, if it were all the control signals would need to be an order of magnitude slower. The LCD controller has latches on the input. > From: Hulatt, Jon [TakeThisOuTjhulattEraseMEspam_OUTMONSTEREUROPE.COM] > The next thing I'd try is putting stupid timing in > everywhere. Wait 100ms > between commands, and make your strobe subroutine hold the E > line high for > say 5ms. There really should be no need to start using extreme timings. Simply follow the manufacturers datasheet to the letter, especialy with regards to initialisation and there should be no problems. Note also that the LCD modules have some requirements with regard to supply voltage rise times, over which the internal reset may not function, powering the device up in an unknown state. Also note that the busy flag is not readable until you have fully configured the module for either 4 or 8 bit mode. Regards Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. ======================================================================= Any questions about Bookham's E-Mail service should be directed to RemoveMEpostmasterTakeThisOuTbookham.com. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservEraseME.....mitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@043005 by hael Rigby-Jones

> > From: Hulatt, Jon [EraseMEjhulattMONSTEREUROPE.COM] > > Tie the low 4 bits low. > From: Michael Rigby-Jones [RemoveMEMichael.Rigby-JonesEraseMEEraseMEBOOKHAM.COM] > I respectfully disagree, you do not need to tie the extra > four lines low. I > cannot find the document at the moment, but I am positive that the LCD > handles the unconnected pins properly when in 4 bit mode. The unused data pins should be left open (floating) according to page 7 of the datasheet. www.samsung.com/Products/Semiconductor/SystemLSI/DDI/MobileDDI/BWSTN/ S6A0069X/ds_s6a0069.pdf Regards Mike ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. ======================================================================= Any questions about Bookham's E-Mail service should be directed to RemoveMEpostmasterspam_OUTKILLspambookham.com. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@045532 by piclist

On 6 Nov 2003 at 8:54, Hulatt, Jon jhulatt-at-MONSTE wrote: <snip> > The "Default" internal initialisation of a two line LCD will result in > exactly what you're seeing here. Set your contrast so that one line looks > black, and one line looks clear, and leave it at that. Hmm... it's not so black now that characters are displayed. you can still see them. My GPS and electronic organizers aren't that readable wit the contrast way down like that. I've got 8 and 4 bit code working now. I'm still writing functions, but one thing that is fouling up is the cursor home command. It hoses my first character. Set DDRAM Address, OTOH, moves the cursor and doesn't trash any characters it lands on. > > > > I'm using 4 bit mode and the lower bits of the module are > > floating high. I figure that's ok because bit 1 and 0 of the > > upper nibble is what sets the data length regardless of the > > states of the lower nibble. I'll ground those other lines > > though, since that's really the right thing to do. > > Tie the low 4 bits low. Yeah. Never leave inputs floating. Even TTL. Can you guess which line absof**kinglutely *must* be tied low? [blush - kick self in ass.] {Quote hidden}> > > > It appears that this module isn't getting it's instructions. > > I'd have been real happy if the thing went blank like it did > > for the guy with the BGMicro module. At least then I'd know > > something was happening and could think about using a > > negative contrast voltage, which may not even be needed for my module. > > > > I'd agree. I looked at a lot of the PIClist code for LCD's, and found the > timing to be really quite optimistic. Particularly then toggling of the E > line; many of the samples just set it high, NOP, and then set it low again. > It seems that LCDs can miss this. Which is not all that suprising given that > they run at a few hundred KHz, compared to the pic's multiple MHz. Min E high (PWeh) is 230ns and tcycE is 500ns. I'm ok with my bsf ; goto $+1 ; bcf My E strobes are followed by 10ms delays until I write a non-static delay function that doesn't insert a million NOP's in the code like some of the macros I've seen. That Universal LCD code on the PIClist site blows my mind. Every data word in the send function is written as bit sets if the previous bit test finds a 1 bit in the data word in a temp register. The C compilers will use bit sets to optimise code according to an algorithm, but bit testing/ setting an output port with btfsc ; bsf ; is ridiculous. > > I wrote some 16F870 code with really slow timing that worked very reliably > all the time. unfortunately it's on my home PC, but I can send it to you > tonight if you require. Thanks. Go ahead. I always like to see other code examples. Good way to pick up tricks. > > The only thing that is not clear about your code... Are you *always* in bank > 0 when you call LCD_init? The map file showed all registers in bank 0 except the lcd_temp register :-) I'll incorporarte code to eliminate unneccessary bank selects since I don't like the idea of forcing storage rules on the linker unless absolutely necessary. > > > LCD_init > clrf PORTA > bcf LCD_RS > bcf LCD_E > > is making an assumption about bank0. After the first time you call LCD_CMD, > you are de=finately in bank 0, but all that initalisation is risky. Yeah, I see what you mean. PORTA ain't in bank 1 and I could have screwed up. But I check the map file after adding UDATA sections, etc., since I learnt by shooting myself in the foot :-) > > > The next thing I'd try is putting stupid timing in everywhere. Wait 100ms > between commands, and make your strobe subroutine hold the E line high for > say 5ms. The 10ms overkill is working fine. > > > > Give me a shout if you want my code later. If I don't hear from you I'll shout again. Thanks and Best Regards, Mike -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservspamspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@051024 by piclist

On 6 Nov 2003 at 9:16, Michael Rigby-Jones Michael.R wrote: <snip> > > From: Hulatt, Jon [RemoveMEjhulattKILLspamMONSTEREUROPE.COM] > > Tie the low 4 bits low. > > I respectfully disagree, you do not need to tie the extra four lines low. I > cannot find the document at the moment, but I am positive that the LCD > handles the unconnected pins properly when in 4 bit mode. I respectfully submit that even if the module does handle it, the pins float high and waste current. I've got them connected, still, from my 8-bit test, so I can't verify if it will work and I don't feel like disconnecting them yet, but any digital designer worth his salt will tell you to tie unused inputs to something. With CMOS, not doing so is asking for trouble. This is TTL AFAIK, but the rule still applies. <snip> > But the parallel interface is not simply polled by the processor, if it were > all the control signals would need to be an order of magnitude slower. The > LCD controller has latches on the input. Yup. Thus the setup times for the gozintas in the timing diagram. {Quote hidden}> > > From: Hulatt, Jon [jhulattSTOPspamspam_OUTMONSTEREUROPE.COM] > > The next thing I'd try is putting stupid timing in > > everywhere. Wait 100ms > > between commands, and make your strobe subroutine hold the E > > line high for > > say 5ms. > > There really should be no need to start using extreme timings. Simply > follow the manufacturers datasheet to the letter, especialy with regards to > initialisation and there should be no problems. Note also that the LCD > modules have some requirements with regard to supply voltage rise times, > over which the internal reset may not function, powering the device up in an > unknown state. Also note that the busy flag is not readable until you have > fully configured the module for either 4 or 8 bit mode. I'll remember to dig for that spec. just in case I ever need to use a lame supply. I'd allow a little slack for temperature variations, especially if the PIC is running off the internal clock. Mine's running at 6MHz internally (1.5MHz on CLKOUT). It's getting a crystal since it will eventually have a timing function. It did run slower when I first got it. Regards Mke Look... an echo vvvvv {Quote hidden}> > Regards > > Mike > > > > > ======================================================================= > This e-mail is intended for the person it is addressed to only. The > information contained in it may be confidential and/or protected by > law. If you are not the intended recipient of this message, you must > not make any use of this information, or copy or show it to any > person. Please contact us immediately to tell us that you have > received this e-mail, and return the original to us. Any use, > forwarding, printing or copying of this message is strictly prohibited. > No part of this message can be considered a request for goods or > services. > ======================================================================= > Any questions about Bookham's E-Mail service should be directed to > spamBeGonepostmasterSTOPspamEraseMEbookham.com. > > -- > http://www.piclist.com#nomail Going offline? Don't AutoReply us! > email KILLspamlistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistservEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@051440 by Jinx

> > Tie the low 4 bits low. > I respectfully disagree, you do not need to tie the extra four > lines low. I cannot find the document at the moment, but I > am positive that the LCD handles the unconnected pins > properly when in 4 bit mode. Data pins have weak active PMOS pull-ups, just like those that can be enabled on PIC I/O -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listserv@spam@spam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@054215 by piclist

On 6 Nov 2003 at 9:29, Michael Rigby-Jones Michael.R wrote: <snip> > The unused data pins should be left open (floating) according to page 7 of > the datasheet. > www.samsung.com/Products/Semiconductor/SystemLSI/DDI/MobileDDI/BWSTN/ > S6A0069X/ds_s6a0069.pdf Which is for the 0069 chip and mine is a 0066, FWIW. Thanks for the extra spec. Now Jinx has indicated that they have weak pullups. My perusal of the HD44780 spec sheet yesterday ( you'd think that would be at the beginning) when I wanted to see if I was dealing with CMOS revealed nothing of the technology, but the lame spec from Timeline, Inc. led me to believe it was TTL. It's all there at the *end* of the HD44780 spec and sure enough... Pullups... max 250uA. Another mA to kill the batteries. Oh well, it could be worse. > > Regards > > Mike Ditto. :-) -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@054836 by Alan B. Pearce

>> > Tie the low 4 bits low. > >> I respectfully disagree, you do not need to tie the extra four >> lines low. I cannot find the document at the moment, but I >> am positive that the LCD handles the unconnected pins >> properly when in 4 bit mode. > >Data pins have weak active PMOS pull-ups, just like those >that can be enabled on PIC I/O That is my understanding too, and so pulling the lines low will actually increase the current draw :) Also as I remember the data sheet, the floating state of the lines is correct for the 8 bit initialisation word, as the I seem to recall the configuration words that get loaded to change between 4 bit and 8 bit modes have the appropriate bits in the "line float" state, so they get understood correctly if the chip is in 8 bit mode and is going to 4 bit mode. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@055252 by Russell McMahon

> ...................but any digital designer worth his salt will tell you to tie > unused inputs to something unless the data sheet specifically indicates otherwise........... I'm sure that's what you meant, wasn't it ? ;-) (Slow email traffic night here ....)(doing accounts/tax / .... ) Just found $5000 odd of equipment that I didn't invoice anyone for back in April this year. The night may be profitable yet :-). Government ultimately pays in this case so the money is safe, enough. Um, actually, maybe not. I've got a Genetic Engineering magazine article coming out Sunday and several people who have read it have suggested that the Prime Minister is going to be 'annoyed' with me :-(, (Not my intention but iot may be unavoidable, alas). So maybe I'm going to have trouble getting my money after all :-( RM -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email TakeThisOuTlistserv.....TakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@060914 by Alan B. Pearce

>Yeah. Never leave inputs floating. Even TTL. Can you guess which line >absof**kinglutely *must* be tied low? [blush - kick self in ass.] I did not get from the previous messages just which line this was. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email TakeThisOuTlistservKILLspamspammitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@060916 by piclist

On 6 Nov 2003 at 23:53, Russell McMahon apptech-at-PA wrote: > > ...................but any digital designer worth his salt will tell you > to tie > > unused inputs to something > > unless the data sheet specifically indicates otherwise........... > > I'm sure that's what you meant, wasn't it ? ;-) Yeah. I posted that before all that pullup info was brought up. The Slambung, er, Samsung data sheet (which I didn't have) puts it up front and Hitachi buries it toward the end of a sheet twice as long. Pretty good chips, really. Good lick collecting the Dineros (pounds?). That little aside about the article pissing off the PM was funny. Regards, Mike -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservRemoveMEmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@062209 by hael Rigby-Jones

> -----Original Message----- > From: piclist [RemoveME1zlx0fc02spamBeGoneSNEAKEMAIL.COM] > Sent: 06 November 2003 10:41 > To: spamBeGonePICLIST@spam@spam_OUTMITVMA.MIT.EDU > Subject: Re: [PIC:] LCD & KS0066 - Another goofy prob? > > Which is for the 0069 chip and mine is a 0066, FWIW. Thanks > for the extra spec. Apparently Samsung have two numbers for the same part. The link to this datasheet says "S6A0069X (KS0066U)". Oh well, glad you got it all working anyway. Cheers Mike (RJ) ======================================================================= This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. ======================================================================= Any questions about Bookham's E-Mail service should be directed to TakeThisOuTpostmasterspambookham.com. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@062831 by Russell McMahon

> Good lick collecting the Dineros (pounds?). That little aside about the article pissing > off the PM was funny. It may not be come Monday :-(. She is not the sort of person that one wants to annoy excessively. As somebody said about one of our earlier women politicians - "The softest thing about her are her teeth". It's actually quite a compliment. I thought about various ways to say things but in the end I HAD to say that she was saying things that are scientifically unsupportable (she is) and that either she is receiving bad advice (which she probably is) and doesn't know it or she knows it but is saying what she is for whatever reason. Either way someone gets unhappy. I think it is a case of corporate blindness on a large scale. Like the emperor without clothes in the fairy tale. Everyone sees things the way they want them to be and it all gets self fulfilling after a while as everyone tells everyone else what they want to hear. What I'd LIKE to happen is for politicians to realise that they haven't quite heard their advisers correctly and shift positions without anyone actually being to blame. The advisers of course haven't realised that they have been being misheard and adjust their advice to 'make it more clear'. Win win as long as we all end up working together. What's more likely is that people will want to shoot or ignore the messenger. The latter if few people pick up on the article. I'll be trying to ensure that that doesn't happen. I don't know if my contribution will cause national ructions or, more likely, sink like a stone into the pond of middle class apathy. We'll see. RM -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservEraseMEspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@063453 by Wouter van Ooijen

> > Tie the low 4 bits low. > > I respectfully disagree, you do not need to tie the extra > four lines low. I > cannot find the document at the moment, but I am positive that the LCD > handles the unconnected pins properly when in 4 bit mode. I dunno for clones, but the HD44780 has pull-ups or pull-downs (I don't remember which). This is not stated explicitly in the datasheet, but the electrical characteristics mention how strong (or weak) the pulls are. Wouter van Ooijen -- ------------------------------------------- Van Ooijen Technische Informatica: http://www.voti.nl consultancy, development, PICmicro products -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listservRemoveMEEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@064114 by Jinx

> It's all there at the *end* of the HD44780 spec and sure enough... > > Pullups... max 250uA. Another mA to kill the batteries. Oh well, > it could be worse. It's probably better. Typical spec is ~125uA and could be 50uA. From memory a 16x2 takes around 3mA, which is well within the drive capability of a PIC pin. So the display's V+ can be under s/w control to power it down if desired, for example if the PIC is put to sleep. It should also be possible to charge-pump a low voltage PIC's pin to make 5V (or at least > 4.6V I think) for the LCD's V+ if running on a couple of AAs. Alpha-numeric LCD input specs look as though they should be OK with LV PIC output voltages if > 2.2V and the only V limiting needed is on d7 when (if) reading Busy -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email EraseMElistserv@spam@mitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@084732 by Russell McMahon

> It should also be possible to charge-pump a low voltage PIC's pin > to make 5V (or at least > 4.6V I think) for the LCD's V+ if running > on a couple of AAs. Rule of thumb is a no load supply will get to within 2 diode drops of pump Vdd with silicon diodes and a single stage pump. ie about 4v8 with 3v Vdd. Your two AA's will drop below 3v fairly soon in their life. Using Schottky small signal diodes (BAT85 etc) will give you about Vdd- 0.5v boost or about 4v5 for Vdd = 2v5 or more. Synchronous switches and multiple stages get you more voltage at the cost of complexity (and diminishing returns per stage) RM -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email @spam@listservspam_OUT.....mitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@091320 by William Bross

BTW, if anybody is looking for LCD controller data sheets, Hantronix has a nice set of the ones they use, including the KS0066 at: http://www.hantronix.com/3_2.html Bill {Original Message removed}

2003\11\06@101136 by Bob Ammerman

> It's all there at the *end* of the HD44780 spec and sure enough... > > Pullups... max 250uA. Another mA to kill the batteries. Oh well, it could be worse. > > > > Regards > > > > Mike They won't kill the batteries if you leave the unused ones disconnected and let the ones that you use idle at high. Bob Ammerman RAm Systems -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spamBeGonelistservEraseMEmitvma.mit.edu with SET PICList DIGEST in the body

2003\11\06@175602 by Jinx

> BTW, if anybody is looking for LCD controller data sheets, > Hantronix has a nice set of the ones they use, including the > KS0066 at: > > http://www.hantronix.com/3_2.html > > Bill That's really helpful. I've submitted it to http://www.woe.de.vu/ and hopefully it will be added to the LCD d/s links page -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body

2003\11\08@055742 by Peter L. Peres

> From memory a 16x2 takes around 3mA, which is well within the drive > capability of a PIC pin. So the display's V+ can be under s/w control to > power it down if desired, for example if the PIC is put to sleep. No, it's not. The pic output current and the decoupling cap on the lcd will slow the swithcing down too much and the panel won't come out of reset. Use a 2N7000 to switch the low side of the panel supply if you have to. Peter -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2003\11\08@055744 by Peter L. Peres

> I respectfully submit that even if the module does handle it, the pins > float high and waste current. I've got them connected, still, from my > 8-bit test, so I can't verify if it They do not float high, they are pulled high on most controllers. Leave them unconnected and put a uA-meter between one and ground. You should see 50uA of current or so. This is very useful when connecting to a 8051 port. Peter -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

2003\11\08@065046 by Jinx

> > From memory a 16x2 takes around 3mA, which is well within the > > drive capability of a PIC pin. So the display's V+ can be under s/w > > control to power it down if desired, for example if the PIC is put to > > sleep. > > No, it's not. The pic output current and the decoupling cap on the lcd > will slow the switching down too much and the panel won't come out of > reset. Use a 2N7000 to switch the low side of the panel supply if you > have to Sorry, have to disagree to some extent. I've units with LCDs that are put to sleep that wake on an RS232 IRQ. The LCD's V+ is in my case b7 and powers down when the PIC sleeps. In fact all lines going to the LCD are low during sleep. When the PIC wakes up, all data and control lines are held at 0 for a few 10s of ms after b7 goes high, then the LCD is initialised. Works every time -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

More... (looser matching)
- Last day of these posts
- In 2003 , 2004 only
- Today
- New search...