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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
04/22/05 01:28
Read: times


 
#92101 - Why not DDC ?
Responding to: ???'s previous message
Dear Prahlad,

Yes it does look odd that when you already have a MCU with 80% of its code flash still empty, you go for an analog sub system !

There were a few reasons why I did, what I did.

$ Maths. I had located some transfer functions to calculate the error terms , but it did look that they are going to take up quite a bit of the processor time. 2nd order polynomials?

$ The aspect of error sampling frequency. This is one area where the analog system shines bright. You just have to contend with three gain settings of P,I,D. With a digital approach, sampling freq. is one more variable that has to be tackled.

$ The analog front end with a digital PID. Plenty of circus is involved to get atleast 10 to 12 bit accuracy, repeatably. When I go with the analog PID, I am no longer worried about the quantization error and other non-linearties of the ADC front end.

With all this, I still have my fascination for the digital PID loop controller. It was just that in this case I was looking for a quick-fix, for the problem on hand and the analog system appealed.

All said and done I believe that a good digital PID loop with reliable Auto-Tune capabilities is really a DSP job. Not sure if they are done with 8 bitters like 8051.

Regards

Raghu

List of 11 messages in thread
TopicAuthorDate
Simple PID controller schematic            01/01/70 00:00      
   Why Analog Whynot Direct Digital Control            01/01/70 00:00      
      Why not DDC ?            01/01/70 00:00      
         RE:            01/01/70 00:00      
            Unique loads            01/01/70 00:00      
               Nothing as good as Analog.            01/01/70 00:00      
                  Copy of drawing            01/01/70 00:00      
               HOT( hydraulik OT), Happily WEOT too            01/01/70 00:00      
                  Pump develops flow and ...            01/01/70 00:00      
   Well Done            01/01/70 00:00      
      The next step..            01/01/70 00:00      

Back to Subject List