CCS C Software and Maintenance Offers
FAQFAQ   FAQForum Help   FAQOfficial CCS Support   SearchSearch  RegisterRegister 

ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

CCS does not monitor this forum on a regular basis.

Please do not post bug reports on this forum. Send them to CCS Technical Support
Need help understanding some TC77 code.....

 
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion
View previous topic :: View next topic  
Author Message
Guest
Need help understanding some TC77 code.....
PostPosted: Mon Feb 02, 2009 5:32 pm     Reply with quote
Hi All,

I'm using the following code to read temperature data from a TC77, and I'm trying to understand this code, and reconcile it with what I think the datasheet is saying:

Code:

   int8 iIndex;
   int16 data = 0;
   
   output_low(TC77_CS); //select the TC77
   for(iIndex=13; iIndex>0; iIndex--)
   {
      output_high(TC77_SCK); //clock
      delay_us(10);
      output_low(TC77_SCK);
      delay_us(10);
      if (input(TC77_SIO))
      bit_set(data,iIndex);
   }
   output_high(TC77_CS); //deselect the TC77

   data >>= 6; //to get rid of the fraction of a degree


This code reads 13 bits of data from the TC77, and stuffs them into the variable 'data'. The datasheet says that the temperature data is stored in a 13 bit 2's compliment representation, however the lower three bits are NOT temperature related (13 data + 3 non-data =16). Thus, I think this routine is missing the upper 3 bits of temperature data. I think the above routine works because these values are always zero for positive temperatures anyway. Am I reading the data sheet correctly, and interpreting it right?

So, now to extend this routine to read both positive and negative temperatures, I'd have to read all 16 bits, and then look at the highest bit to determine if the temperature was positive or negative (bit 15 is the 'sign' bit)........

I'm also not clear on the line
Code:


data >>= 6; //to get rid of the fraction of a degree


I can understand the shift to get rid of the lower three non-temperature bits, and some more shifting to get rid of the fractional part, but it seems to me the fractional part is contained in four bits, not three. Thus, I would think that a shift of 7 is correct, not 6. This has me perplexed because the code does work as written above.....

Tony
Guest
PostPosted: Mon Feb 02, 2009 5:41 pm     Reply with quote
Hi All,

Argh!!! I see that the data is shifted out MSB first, so we always get the 13 temperature bits with this routine. The question remains, however, why is the shift right operation written the way it is?

Thanks!

Tony
techdesign



Joined: 03 Feb 2009
Posts: 2
Location: Belgium

View user's profile Send private message Send e-mail Visit poster's website

tc77
PostPosted: Tue Feb 03, 2009 6:50 am     Reply with quote
It depends on which fraction you want to get rid of.

Bit3 (LSB) of the TC77 represents 0.0625°C
Bit4 is 0.1250°C
etc...

Total sample value to be multiplied with LSB value for a temperature reading.

It's quite common to skip some LSBs to have a more stable readout.
An other approach would be to add up 10 full resolution samples and divide by 10 to have a more stable readout.

Bit16 (MSB) is the plus/minus sign, so a signed int16 is perfect for a 16-bit sample. Can be converted to a float if needed.


cheers,

Michel Bavin
www.techdesign.be

Last edited by techdesign on Wed Feb 04, 2009 1:18 am; edited 1 time in total
Guest
PostPosted: Tue Feb 03, 2009 6:22 pm     Reply with quote
Hi Again,

Well, I'm still confused....... Here is the shift right line:

Code:

data >>= 6;


Does this line not say 'shift variable data 6 places to the right, and place the result in variable data'??

If so, the rightmost 6 bits are thrown out, correct? If that is true, by my read says that bits 2^1, 2^0, 2^-1, 2^-2, 2^-3, and 2^-4 are all lost. The confusion is that the upper two bits (2^1 and 2^0) contain whole data, not fractional data.

What am I missing???

Tony
techdesign



Joined: 03 Feb 2009
Posts: 2
Location: Belgium

View user's profile Send private message Send e-mail Visit poster's website

tc77
PostPosted: Wed Feb 04, 2009 2:49 am     Reply with quote
Yes, you'right to be confused.

First of all, the unsigned int16 "data" is filled from bit13 to bit0. This is a bit fishy, because of positive/negative values are best filled from bit15 to bit0.
Then the right shift >>=6 !! Even without the first problem above, this reduces the resolution to 4°C.

Best is to write your own routine i guess...

You might want to look at our TC77 code example in this project: http://www.techdesign.be/projects/012/012.htm

Look for the TC77 routine in the source code: http://www.techdesign.be/projects/012/012_v32_c.zip
Display posts from previous:   
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion All times are GMT - 6 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group