Simple SO2R Audio ( notes) 
I finally implement
SO2R controller simple audio outlined in previous entries. Design adjusted it slightly during implementation to better take into account the characteristics of my season. For example, instead of taking audio directly from
radios (rig1 and rig2) the volume of sound cards after being processed in the filters. For its part the two sound cards already have a harness simple "Y" leading a right channel and left another two separate headsets, so the driver just to have a separate entry for each channel (other settings may be easily adapted if necessary). Moreover LEDs eliminated because they represent an unnecessary consumption of energy, initially thought it might be important to see where he was the key (in the design of previously
EA5FV Dani described using a mechanical key which can visually know the button that channel is pressed). This design is simpler, if in doubt press the center button (audio and rig2 rig1 to left and right ear respectively). Finally (and I always forget that
GP3 !!!!) in
12F675 not be used for input without special consideration, fortunately there was still free the GP5
so I made the change and now. The microcode is trivially simple, are available in assembler code (
. asm) if it wants to modify and binary code (
. hex) if you want just an integrated record now. The code is so simple that it only takes a few minutes to write. The code is an initialization routine that really is repetitive in all the programs I write, probably initializes many variables that will not be used but I learned an enormous amount of hours doing odd behaviors debugging is extremely important to establish an operating environment driver that is absolutely defined, I can use the 4 things that this program really needs to invest dozens of hours later looking for a flaw that ends happening because there is some aspect of generating default ending conflict. Things like the status of interrupts, timer records and WDT (Watchdog) of the processor must be "tied" very firm to conduct occurring after not rare. To give some sign of life (heartbeat) because it has no visual indication when starting exercise both relays a couple of times. The routine operation in sequence simply check
switch "A + B" the
"A" and "B" detect when going down (entry into low) were each routines where appropriate relay is activated and makes a small "debouncing" before continuing to review the switches. The code could do much more "elegant", for example the routines in each switch are identical and could use the same code for all three but since I did not seem right time to apply the method of "brute force" and get three separate chunks of code (the capacity actually used tiny 12F675 still a tiny fraction), the code is also easier to understand.
As in every design there is always some little detail or online bad soldier does not boot up at first, but gave very little work actually make it work. As is usual in simple designs of my station I use a small metal box of Altoids pills to implement (after eat the pills, of course), the finished device can as being used during the operation. The result of the practical use during the
CQ DX MM (Manchester Mineira
) confirmed the usefulness of design, actually a very useful addition and very inexpensive to implement. During the contest I had to share the connection of switches because they were invested over the headphones, and oddly enough I generated considerable confusion, once aligned with the headset buttons operation is even natural.
I finally implement SO2R controller simple audio outlined in previous entries. Design adjusted it slightly during implementation to better take into account the characteristics of my season. For example, instead of taking audio directly from radios (rig1 and rig2) the volume of sound cards after being processed in the filters. For its part the two sound cards already have a harness simple "Y" leading a right channel and left another two separate headsets, so the driver just to have a separate entry for each channel (other settings may be easily adapted if necessary). Moreover LEDs eliminated because they represent an unnecessary consumption of energy, initially thought it might be important to see where he was the key (in the design of previously EA5FV Dani described using a mechanical key which can visually know the button that channel is pressed). This design is simpler, if in doubt press the center button (audio and rig2 rig1 to left and right ear respectively). Finally (and I always forget that GP3 !!!!) in 12F675 not be used for input without special consideration, fortunately there was still free the GP5 so I made the change and now. The microcode is trivially simple, are available in assembler code (. asm) if it wants to modify and binary code (. hex) if you want just an integrated record now. The code is so simple that it only takes a few minutes to write. The code is an initialization routine that really is repetitive in all the programs I write, probably initializes many variables that will not be used but I learned an enormous amount of hours doing odd behaviors debugging is extremely important to establish an operating environment driver that is absolutely defined, I can use the 4 things that this program really needs to invest dozens of hours later looking for a flaw that ends happening because there is some aspect of generating default ending conflict. Things like the status of interrupts, timer records and WDT (Watchdog) of the processor must be "tied" very firm to conduct occurring after not rare. To give some sign of life (heartbeat) because it has no visual indication when starting exercise both relays a couple of times. The routine operation in sequence simply check switch "A + B" the "A" and "B" detect when going down (entry into low) were each routines where appropriate relay is activated and makes a small "debouncing" before continuing to review the switches. The code could do much more "elegant", for example the routines in each switch are identical and could use the same code for all three but since I did not seem right time to apply the method of "brute force" and get three separate chunks of code (the capacity actually used tiny 12F675 still a tiny fraction), the code is also easier to understand. 
As in every design there is always some little detail or online bad soldier does not boot up at first, but gave very little work actually make it work. As is usual in simple designs of my station I use a small metal box of Altoids pills to implement (after eat the pills, of course), the finished device can as being used during the operation. The result of the practical use during the CQ DX MM (Manchester Mineira ) confirmed the usefulness of design, actually a very useful addition and very inexpensive to implement. During the contest I had to share the connection of switches because they were invested over the headphones, and oddly enough I generated considerable confusion, once aligned with the headset buttons operation is even natural. 
0 comments:
Post a Comment