Post by thread:[PIC]: Data Bus Issue

Ok, this is half EE, half PIC. I'm designing a circuit that uses a 4 bit data bus (plus a few flow control lines) to transfer some values between a 16f877 and a Scenix SX28 chip. I'd like to put these chips close together to avoid long PCB traces. I don't have to use specific pics to transfer the data, but ideally it I could get 4 pins to line up between the two chips ie, RA0-RA0, RA1-RA1, RA2-RA2, RA3-RA3. That way all the data I transfer is correct LSB-MSB without having to do some sort of software translation on it. However, all is not going quite so easily. Both chips use a similar numbering convention, so pins on the left of the chip increase in value as they go down, and pins on the right increase as they go up. For example: RA0 |\/| RB7 RA0 |\/| RB7 RA1 | | RB6 RA1 | | RB6 (these are just for RA2 | | RB5 RA2 | | RB5 illustration only) RA3 | | RB4 RA3 | | RB4 etc So things don't line up nice. I'd like to avoid a huge copper trace spider's web trying to fix that. I'd even thought about rotating one chip 180degrees, but the problem remains. So my questions. Short of putting one chip on top of the board and one on the bottom, can anyone think of an easy solution? Secondly, does anyone have a routine for reversing the order of bits in a byte? If I transfer data and it ends up MSB-LSB order swapped, what is a good way to swap it back? I could just use a bunch of btfsc's I suppose. Thanks, Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email spam_OUTlistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

> RA0 |\/| RB7 RA0 |\/| RB7 > RA1 | | RB6 RA1 | | RB6 (these are just for > RA2 | | RB5 RA2 | | RB5 illustration only) > RA3 | | RB4 RA3 | | RB4 > etc > > So things don't line up nice. I'd like to avoid a huge copper trace > spider's web trying to fix that. That will hardly be a "huge copper spider web". This sort of thing goes on all the time during routing. Unless they are partcularly sensitive signals (and it doesn't seem that way from your description), don't worry about it. Just let the auto router do whatever it does. Most likely the signals will get routed via even more convoluted paths than what you envision. Big deal. It's all handled by computer. It might make a ratsnest, but it won't mess up the connections. Spend the extra time worrying about good grounds instead. Note that there are cases where this relaxed attitude is inappropriate, but it doesn't seem like this is one of them from your description. > I'd even thought about rotating one chip 180degrees, I'd only consider that if the reduction in complexity made the difference between a routable and unroutable board, or the signals were very sensitive. Otherwise, someone isn't going to notice one IC is reversed from the other and one of them is going to get installed backwards. ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email .....listservKILLspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body

You're right. In fact, I just figured out a decent routing in my head. Not as pretty as 4 straight lines, but then again, pretty is subjective right? :) Thanks, Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams Olin Lathrop wrote: > That will hardly be a "huge copper spider web". This sort of thing goes > on all the time during routing. Unless they are partcularly sensitive > signals (and it doesn't seem that way from your description), don't worry > about it. Just let the auto router do whatever it does. Most likely the > signals will get routed via even more convoluted paths than what you > envision. Big deal. It's all handled by computer. It might make a > ratsnest, but it won't mess up the connections. Spend the extra time > worrying about good grounds instead. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email listservKILLspammitvma.mit.edu with SET PICList DIGEST in the body

---------- From: Josh Koffman[SMTP:.....listsjoshKILLspam.....3MTMP.COM] Sent: Friday, September 12, 2003 6:53 PM To: EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU Subject: [PIC]: Data Bus Issue {Quote hidden}> Ok, this is half EE, half PIC. I'm designing a circuit that uses a 4 bit > data bus (plus a few flow control lines) to transfer some values between > a 16f877 and a Scenix SX28 chip. I'd like to put these chips close > together to avoid long PCB traces. I don't have to use specific pics to > transfer the data, but ideally it I could get 4 pins to line up between > the two chips ie, RA0-RA0, RA1-RA1, RA2-RA2, RA3-RA3. That way all the > data I transfer is correct LSB-MSB without having to do some sort of > software translation on it. > However, all is not going quite so easily. Both chips use a similar > numbering convention, so pins on the left of the chip increase in value > as they go down, and pins on the right increase as they go up. For > example: > RA0 |\/| RB7 RA0 |\/| RB7 > RA1 | | RB6 RA1 | | RB6 (these are just for > RA2 | | RB5 RA2 | | RB5 illustration only) > RA3 | | RB4 RA3 | | RB4 > etc > So things don't line up nice. I'd like to avoid a huge copper trace > spider's web trying to fix that. I'd even thought about rotating one > chip 180degrees, but the problem remains. > So my questions. Short of putting one chip on top of the board and one > on the bottom, can anyone think of an easy solution? Secondly, does > anyone have a routine for reversing the order of bits in a byte? If I > transfer data and it ends up MSB-LSB order swapped, what is a good way > to swap it back? I could just use a bunch of btfsc's I suppose. > Thanks, > Josh To reverse the bits in a byte, shift the byte to the right and shift a second byte to the left. After 1 right shift, CY has the LSB of the first byte. Shift the second byte left and CY ends up as the LSB of byte #2. After 8 shifts, byte 2 has byte #1 in bit reverse order. If you only want the lower 4 bits of #1, shift 4 times. Use swapf to exchange nybbles if needed. John Power -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Hi- Am I obtuse? Surely you can run traces between the pins, so that the like-pins can be joined. David John N. Power wrote: {Quote hidden}>---------- >From: Josh Koffman[SMTP:listsjoshspam_OUT3MTMP.COM] >Sent: Friday, September 12, 2003 6:53 PM >To: @spam@PICLISTKILLspamMITVMA.MIT.EDU >Subject: [PIC]: Data Bus Issue > > > >>Ok, this is half EE, half PIC. I'm designing a circuit that uses a 4 bit >>data bus (plus a few flow control lines) to transfer some values between >>a 16f877 and a Scenix SX28 chip. I'd like to put these chips close >>together to avoid long PCB traces. I don't have to use specific pics to >>transfer the data, but ideally it I could get 4 pins to line up between >>the two chips ie, RA0-RA0, RA1-RA1, RA2-RA2, RA3-RA3. That way all the >>data I transfer is correct LSB-MSB without having to do some sort of >>software translation on it. >> >> > > > >>However, all is not going quite so easily. Both chips use a similar >>numbering convention, so pins on the left of the chip increase in value >>as they go down, and pins on the right increase as they go up. For >>example: >>RA0 |\/| RB7 RA0 |\/| RB7 >>RA1 | | RB6 RA1 | | RB6 (these are just for >>RA2 | | RB5 RA2 | | RB5 illustration only) >>RA3 | | RB4 RA3 | | RB4 >>etc >> >> > > > >>So things don't line up nice. I'd like to avoid a huge copper trace >>spider's web trying to fix that. I'd even thought about rotating one >>chip 180degrees, but the problem remains. >> >> > > > >>So my questions. Short of putting one chip on top of the board and one >>on the bottom, can anyone think of an easy solution? Secondly, does >>anyone have a routine for reversing the order of bits in a byte? If I >>transfer data and it ends up MSB-LSB order swapped, what is a good way >>to swap it back? I could just use a bunch of btfsc's I suppose. >> >> > > > >>Thanks, >> >> > > > >>Josh >> >> > >To reverse the bits in a byte, shift the byte to the right and shift a second byte >to the left. After 1 right shift, CY has the LSB of the first byte. Shift the second >byte left and CY ends up as the LSB of byte #2. After 8 shifts, byte 2 has >byte #1 in bit reverse order. If you only want the lower 4 bits of #1, shift 4 >times. Use swapf to exchange nybbles if needed. > >John Power > >-- >http://www.piclist.com hint: PICList Posts must start with ONE topic: >[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

You aren't obtuse. That is what I ended up doing. My question mainly revolved around not having a bunch of mixed up tracks. Nice straight ones seemed to me to be less prone to future problems. In the end, I managed to route it so it worked fine. Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams David P Harris wrote: > Am I obtuse? Surely you can run traces between the pins, so that the > like-pins can be joined. -- 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

Brilliant! That's exactly what the kick in the right direction I needed. Thanks! Josh -- A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools. -Douglas Adams "John N. Power" wrote: > To reverse the bits in a byte, shift the byte to the right and shift a second byte > to the left. After 1 right shift, CY has the LSB of the first byte. Shift the second > byte left and CY ends up as the LSB of byte #2. After 8 shifts, byte 2 has > byte #1 in bit reverse order. If you only want the lower 4 bits of #1, shift 4 > times. Use swapf to exchange nybbles if needed. -- 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

> anyone have a routine for reversing the order of bits in a byte? Josh, dug these out of a far corner of the HDD. I tried them at the time to get around the same problem as you've got ============================================ From: Dwayne Reid To: <KILLspamPICLISTKILLspamMITVMA.MIT.EDU> Subject: Re: Efficient way to reverse a byte? Date: Saturday, 30 October 1999 06:51 clrf dest btfsc source,0 bsf dest,7 btfsc source,1 bsf dest,6 btfsc source,2 bsf dest,5 btfsc source,3 bsf dest,4 btfsc source,4 bsf dest,3 btfsc source,5 bsf dest,2 btfsc source,6 bsf dest,1 btfsc source,7 bsf dest,0 17 instructions, 17 cycles ========================================= From: Terry A. Steen Date: Saturday, 30 October 1999 07:22 How about: rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in rlf arg1,1 ;shift out a bit rrf arg2,1 ;shift it in Kind of long, but this should reverge arg1 and store it in arg2. At the same time, it will also reverse arg2 and store it in arg1 ============================================= From: Russell McMahon Date: Saturday, 30 October 1999 11:53 The piece of code that Tracy posted (from John Payson) does what it was originally meant to do BUT is NOT correct for the present problem. Here we require 76543210 --> 01234567 John's solution is for a 7 bit swap ignoring B7. ie x6543210 --> x0123456 As a consequence B7 & B3 is untouched and there are only 3 swaps needed making the code shorter. Rewriting it for an 8 bit swap as required here, produces - movwf source btfsc source,0 xorlw h'81' btfsc source,7 xorlw h'81' btfsc source,1 xorlw h'42' btfsc source,6 xorlw h'42' btfsc source,2 xorlw h'24' btfsc source,5 xorlw h'24' btfsc source,3 xorlw h'18' btfsc source,4 xorlw h'18' 17 instructions, time varies depending on data but fairly quick -- 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