Shop OBEX P1 Docs P2 Docs Learn Events
DS1620 High Resolution temperature reading - FYI — Parallax Forums

DS1620 High Resolution temperature reading - FYI

Beau SchwabeBeau Schwabe Posts: 6,568
edited 2011-05-04 09:18 in General Discussion
I thought I would pass this information along...

The HiRes Formula...
Temperature = Temp_read - .25 + (Count_PER_C - Count_Remain) / Count_PER_C

...would still apply, but you could simplify it a bit to...

Temperature = Temp_read·+ .75·- (Count_Remain * .0625)

or

Temperature = Temp_read·- .25 + (16 - Count_Remain) /·16

...Either way, you would eliminate the need to read the value for Count_PER_C in your program.

Email correspondence below

Hello Beau,

You are correct that the·read slope command always returns a value of 16.· The original DS1620 was based on a dual oscillator·design.· The new version uses a bandgap reference, and therefore the·count per c is fixed at 16.· ·It will remain fixed at 16.· I also agree that 1/16 = yields a resolution of .0625.

Thanks,

Thermal Applications Support
Dallas Semiconductor/MAXIM
Original Message
From: bschwabe@parallax.com [noparse][[/noparse]mailto:bschwabe@parallax.com]
Sent: Tuesday, October 03, 2006 8:04 AM
To: Thermal.Support@dalsemi.com
Subject: Tech Support: DS1620 (TS) [noparse][[/noparse]Parallax Inc.]

======================================================================MAXIM TECHNICAL SUPPORT REQUESTDate: 2006-10-03 08:03:53IP : XX.XX.XX.XX======================================================================CONTACT INFORMATION
Email Address: bschwabe@parallax.comFirst Name : BeauLast Name : SchwabeJob Title : IC layout EngineerCompany : Parallax Inc.Division : Address1 : Address2 : City : State : Postal Code : Country : UNITED STATESPhone Number : ##############Fax Number : Dist Center : Maxim USAExport Info : No RestrictionsRELATED PRODUCTS AND APPLICATIONS / CUSTOMER IMPACT
Part Numbers : DS1620Application : Demonstration application for our customersAnnual Usage : Product Line : Temperature SensorsProduct Ln Id: 11PROBLEM DESCRIPTION
Hello,I haven't been able to find this in the datasheet or I have simply overlooked it one of the two.
Can someone confirm for High resolution temperature readings from the DS1620, that the Read Slope
command or the "number of counts per degree C" always seems to read a value of 16 when I look at it?
This would indicate a resolution granularity of 0.0625 per deg C (0.1125 per Deg F)
Does the "16" value that I am reading seem correct?·

Is this value subject to change by process variation or is it a fixed constant?

Thank you
======================================================================

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Post Edited (Beau Schwabe (Parallax)) : 10/4/2006 9:03:41 PM GMT

Comments

  • manxstampmanxstamp Posts: 57
    edited 2006-10-04 20:17
    Your post includes information from Dallas Semiconductors·mentioning the 'original DS1620' and 'the new version'. Your formula applies to 'the new version'. How do you recognise this new version if you have several DS1620 ICs?

    Thanks.

    John

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔


    Manxstamp,
    Isle of Man, British Isles
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2006-10-04 20:59
    Good question John,

    I tried to look on my DS1620 for any hints i.e. RevB letter etc. but the print was so faint it was hard to read if anything was there.

    http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2735

    On the above sight, based on the "Recommended Upgrades for New Designs" header at the top, my guess is that the "new" DS1620 version has been out awhile.
    You could always call or send an E-mail to maxim to find out for sure.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • BeanBean Posts: 8,129
    edited 2006-10-04 21:35
    Beau,
    I don't mean to hijack your thread, but have you ever checked out the DS1848 from Maxim (about $5 from Newark).
    The DS1847 is the same but minus the 128 bytes of EEPROM.(about $4 from Newark).
    The DS1858 is the same as the DS1848 but also has a 3 channel 12bit ADC.
    It is I2C and has a 13bit (1/16'C) temp sensor, two digital variable resistors (single ended, not pots), AND EEPROM (128+72+72) Bytes.
    This IC is used to compensate a circuit for temperature changes. You program the EEPROM, and the chip will adjust the two resistor values for every 2'C from -40 to +95.
    It's a really cool chip that I use at work.

    Bean.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Cheap used 4-digit LED display with driver IC·www.hc4led.com

    Low power SD Data Logger www.sddatalogger.com
    SX-Video Display Modules www.sxvm.com

    Don't mistake experience for intelligence. And vis-vera.


    Post Edited (Bean (Hitt Consulting)) : 10/4/2006 9:43:07 PM GMT
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2006-10-04 21:51
    Bean,

    "...have you ever checked out the DS1848 from Maxim...", no I haven't.·

    This thread was basically started, because of something I happened across when writing a Propeller object
    for it.· Since the DS1620 is something that we sell at Parallax making an object for it made sense.· There
    are certainly other alternatives in which IC to use or which is the best fit for a particular application.
    I do like the idea of having some scratch pad EEPROM available in an application as well.· The temp comp
    pots add a nice touch also.· I might look into this and see.

    Thanks Bean!






    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • GuidoGuido Posts: 195
    edited 2006-10-05 01:19
    Bean!!

    How about showing us the code! Really interested in its sensitivity!

    Thanks

    Guido
  • jc3IIIjc3III Posts: 21
    edited 2011-05-03 19:50
    why does the Applied Sensor book use the number 2 to configure the DS1620 to continuous conversion with CPU. The book says that it is mode 2. On the data sheet 02h(2) writes low temp. value into TL register. EEh(238) initiates temp conversion, AAh(170), 0Ch(12) Writes configuration data to configuration register. I would like to know why are they using the number 2. It doesn't seem to match up with the DS1620 data sheet.
  • Tracy AllenTracy Allen Posts: 6,666
    edited 2011-05-04 09:18
    jc3lll,

    The configuration "2" is the parameter that follows the actual command, "12" 
    shiftout 15,14,lsbfirst,[12,2] ' Command to set DS1620 configuration 2.
    12 is the configure command
    2 is the particular configuration, "continuous conversion with CPU"


    I'm attaching the original change notice that came from Dallas/Maxim. The transition occured late 2002 to mid 2003. The new parts are supposed to have a date code that ends in "E2". e.g.
    0437E2 where 0437 is the 37th week of 2004. The change notice doesn't mention the change in count_per_C, but I'm pretty sure it happened at the same time. Before that, the count_per_C was a temperature dependent value greater than 100 that you had to read out in order to calc the higher resolution, but in the revision they fixed that number at 16. You can find out for sure on a given chip by reading the register and if it is 16, you have the revised chip.
    pdf
    pdf
    DS1620-change-notice.pdf
    33K
Sign In or Register to comment.