Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
10/15/05 12:09
Modified:
  10/15/05 12:10

Read: times


 
Msg Score: +1
 +1 Good Answer/Helpful
#102450 - Some hints...
Responding to: ???'s previous message
Hallo Donald,

if you are willing to give us many details of your project, then you can expect a lot of help from us!

Here are some hints:

1. Resolution of built-in ADC/DAC is only achievable in quiet systems, means, when the micro is "sleeping" during conversion. So, during conversions the UART should be off: No serial transmission should be invoked or being recieved. So, using a micro containing the ADC and DAC need not necessarily be advantageous. You can also succeed with a AT89S52 and using external ADC/DAC.
By the way, do you also recieve serial data, or do you only send it?

2. Resolution of ADC/DAC is also limited by the layout of your board. There should be a clear separation between analog ground and digital ground. Both grounds should be connected at only one point, near the converters.

3. In any case you will need solid ground planes for the analog and digital section. So, the use of a 4-layer board seems to be essential.

4. Concretely spoken, there are at least three individual grounds: The digital one of the micro, the analog one of the analog section and the ground which is in connection to chassis (chassis ground), means where all the connectors are referenced to. Sometimes there can also be a fourth ground, the power ground of high current consuming drivers. All these grounds must be routed to a central star ground point, so that the voltage drops of the many and possibly noisy ground return currents will not interfere with each other.
Right at the beginning of development phase you should make a plan how to route the individual grounds. This ground management will be your guiding light later, when designing the layout of PCB.

5. Before thinking about concrete layout and software issues, design all the analog circuits, which drive the pressure generating stuff and which read the transducers. Find out how these circuits are to be controlled and which tolerances are allowed (timing, resolution, hysteresis, etc.), so that you clearly know what the micro has to do.
After this phasis you can think about the code and where to locate these parts on the PCB.

6. Try to program the micro with a simple test program, so that you are sure, that your programming setup will reliably work. Don't use your main program for this testing.

7. Make always small steps and test it and check it immediately, so that you are absoluely sure, that what you are testing right now will definitely work.

Good luck,
Kai

List of 18 messages in thread
TopicAuthorDate
ADUC848 project getting started            01/01/70 00:00      
   Some hints...            01/01/70 00:00      
      More!            01/01/70 00:00      
         Bible?            01/01/70 00:00      
            The "book" is 3 general chapters that re            01/01/70 00:00      
      ADUC848 Project: details for hints            01/01/70 00:00      
         ground loops, oversampling            01/01/70 00:00      
         You have a head start, most have proble            01/01/70 00:00      
            Is 10 readings per second sufficient?            01/01/70 00:00      
               required sampling rate            01/01/70 00:00      
   Thanks, but...            01/01/70 00:00      
      books?            01/01/70 00:00      
         books?            01/01/70 00:00      
            Avoid That            01/01/70 00:00      
               C on a PC?            01/01/70 00:00      
                  OK            01/01/70 00:00      
               Starting C on the '51 with no prior C kn            01/01/70 00:00      
                  Point taken            01/01/70 00:00      

Back to Subject List