Post by thread:[pic]: LCD mystery

Have you checked for an inadvertant Solder Connection between two of the pins, this would exhibit the qualities you have described, as the command codes may or may not work, but the character codes would probably never be correct. I'm not questioning your solder skills, I just know that it has happened to me a few dozen times. T HNX Jesse. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

>I saw something right after the label SendPSTRLoop:. You have a CALL MSG >line there, but MSG is only defined as a variable. When I ran your code in I saw this as well... You load the variable "MSG" with HelloWorld as a pointer to the start of the message table (movlw HelloWorld, movwf MSG), but when you use "call MSG," you will always call the memory address of MSG as a literal (i.e., if MSG is defined as memory address 0x23, for example, you will be jumping to the code at program memory address 0x23, always.) You can't do this :) Moreover, you actually have "addwf PCL, F" at the first line of your message table, which is correct if you load W with the incremental value for the proper character. What you want to do is just load W with a number (0 -> whatever) representing the character you want to retrieve (0=H, 1=e, 2=l, 3=l, etc.) and call the "HelloWorld" table. --Andrew _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

I'm also a beginner in PIC programming but shouldn't this be Start: bsf STATUS,RP0 clrf PORTB <= clrf TRISB bcf STATUS,RP0 clrf PORTB Drew Vassallo <spam_OUTsnurpleTakeThisOuTHOTMAIL.COM> wrote: >I saw something right after the label SendPSTRLoop:. You have a CALL MSG >line there, but MSG is only defined as a variable. When I ran your code in I saw this as well... You load the variable "MSG" with HelloWorld as a pointer to the start of the message table (movlw HelloWorld, movwf MSG), but when you use "call MSG," you will always call the memory address of MSG as a literal (i.e., if MSG is defined as memory address 0x23, for example, you will be jumping to the code at program memory address 0x23, always.) You can't do this :) Moreover, you actually have "addwf PCL, F" at the first line of your message table, which is correct if you load W with the incremental value for the proper character. What you want to do is just load W with a number (0 -> whatever) representing the character you want to retrieve (0=H, 1=e, 2=l, 3=l, etc.) and call the "HelloWorld" table. --Andrew _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads ____________________________________________________________________ Get free email and a permanent address at http://www.netaddress.com/?N=1 -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

>I'm also a beginner in PIC programming but shouldn't this be > >Start: > bsf STATUS,RP0 > clrf PORTB <= clrf TRISB > bcf STATUS,RP0 > clrf PORTB Technically, it works as he has it written. For correctness, it probably should be changed to TRISB, but since it wasn't part of his main problem, I didn't mention it in my email. There are also a number of other parts of the code that could be cleaned up, but those didn't have anything to do with the problem he mentioned, either. Once he's fixed the problem, he can go back and clean up the code if he needs to, but it's not necessary at this point. Then again, people might disagree and say that you should start out learning the right way by selecting the proper commands in the code, but he wasn't asking about that... just how to get the correct characters on the LCD. --Andrew _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

OK sir Andrew! =) Drew Vassallo <.....snurpleKILLspam@spam@HOTMAIL.COM> wrote: >I'm also a beginner in PIC programming but shouldn't this be > >Start: > bsf STATUS,RP0 > clrf PORTB <= clrf TRISB > bcf STATUS,RP0 > clrf PORTB Technically, it works as he has it written. For correctness, it probably should be changed to TRISB, but since it wasn't part of his main problem, I didn't mention it in my email. There are also a number of other parts of the code that could be cleaned up, but those didn't have anything to do with the problem he mentioned, either. Once he's fixed the problem, he can go back and clean up the code if he needs to, but it's not necessary at this point. Then again, people might disagree and say that you should start out learning the right way by selecting the proper commands in the code, but he wasn't asking about that... just how to get the correct characters on the LCD. --Andrew _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads ____________________________________________________________________ Get free email and a permanent address at http://www.netaddress.com/?N=1 -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

{Quote hidden}> -----Original Message----- > From: Olin Lathrop [SMTP:olin_piclistKILLspamEMBEDINC.COM] > Sent: Friday, April 13, 2001 1:38 PM > To: .....PICLISTKILLspam.....MITVMA.MIT.EDU > Subject: Re: [pic]: LCD mystery > > I don't see anything wrong with successive BSF/BCF assuming you want the > bit > to be 1 for one instruction time. There are two issues with using these > instruction on I/O ports, although neither effects the case you mentioned. > > 1 - The bit instructions really do a read/modify/write on the whole byte. > This means the 7 bits that are not addressed by the BSF or BCF are written > to the output latches the way they were read in. This can cause two > potential problems. The first occurs if the bit is an output. If the > external circuit held the PIC pin the opposite state it was being driven, > then it will now be driven by the PIC to that state also. Second, if the > bit is an input then the output latch essentially gets set to an unknown > value. This could cause unexpected results if the bit is switched back to > being an output and the new unknown state of the output latch is not taken > into account. > Agreed. This can be overcome by explicity setting any data latches before switching from input to output, or even better using a shadow register. > 2 - Inputs are latched at the beginning of a cycle and output are latched > at > the end of the cycle. You therefore can't write a bit value one cycle, > then > read a different bit that the external circuit was supposed to drive in > response on the next cycle. In that case you need to insert one NOP for > each instruction time you want the external circuit to have before reading > its response. > > However, a BCF/BSF reads the inputs at the beginning of the cycle. Thus successive BSFs/BCFs on the same port's data latches is a questionable activity at the best of times. Granted it usually works at low clock frequencies and with low loadings on the pin, but I have personaly been bitten with this problem running at 4MHz, the pin was driving the gate of a power MOSFET. At 20MHz, it dosen't work very often at all. Regards Mike -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

> However, a BCF/BSF reads the inputs at the beginning of the cycle. Thus > successive BSFs/BCFs on the same port's data latches is a questionable > activity at the best of times. Granted it usually works at low clock > frequencies and with low loadings on the pin, but I have personaly been > bitten with this problem running at 4MHz, the pin was driving the gate of a > power MOSFET. At 20MHz, it dosen't work very often at all. This is an issue if doing a BSF/BCF to different bits of the same port on successive instructions. It should still work fine if done on the same bit. The second instruction could possibly read the wrong value for that bit, but that will be overriddend by the operation itself. ******************************************************************** Olin Lathrop, embedded systems consultant in Littleton Massachusetts (978) 742-9014, EraseMEolinspam_OUTTakeThisOuTembedinc.com, http://www.embedinc.com -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

{Quote hidden}> -----Original Message----- > From: Olin Lathrop [SMTP:olin_piclistspam_OUTEMBEDINC.COM] > Sent: Tuesday, April 17, 2001 8:17 PM > To: @spam@PICLISTKILLspamMITVMA.MIT.EDU > Subject: Re: [pic]: LCD mystery > > > However, a BCF/BSF reads the inputs at the beginning of the cycle. Thus > > successive BSFs/BCFs on the same port's data latches is a questionable > > activity at the best of times. Granted it usually works at low clock > > frequencies and with low loadings on the pin, but I have personaly been > > bitten with this problem running at 4MHz, the pin was driving the gate > of > a > > power MOSFET. At 20MHz, it dosen't work very often at all. > > This is an issue if doing a BSF/BCF to different bits of the same port on > successive instructions. It should still work fine if done on the same > bit. > The second instruction could possibly read the wrong value for that bit, > but > that will be overriddend by the operation itself. > Agreed, consecutive operations on the same pin should at least leave the pin in a predictable state after the last instruction. Whether or not a valid pulse is generated by this would depend very much on clock speed and capacitive loading etc. And you would still have the problem of the data latches on other tri-stated pins being modifed. I guess the bottom line is that as long as the software/hardware guys are aware of the above and code/design within the limits there shouldn't be a problem with consecutive bit operations ON THE SAME PIN. I believe this issue has been resolved on the 18C series which can read back the data latches rather than the pin state? Mike -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics