Because of the Thanksgiving holiday in the U.S., TI E2E™ design support forum responses may be delayed from November 25 through December 2. Thank you for your patience.

  • State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 358 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

SM72445: I2C Values for the ADC settings A0-A6

SM72445: I2C issues

Glenn Weile
Prodigy 40 points
Part Number: SM72445

It appears TI has a misprint on page 14 of the spec sheet regarding the address selection input lines... The I2C2 and I2C0 labels in the address selection table need to be reversed (ie pulling up I2C2 and leaving I2C1 and I2C0 tied to ground gives the device an address of 0x04 not 0x01 as documented)....

According to page 15 of the spec sheet, the device should give me a NACK after the 7th byte has been written to the E3, E4, or E5 'registers' --- it does not !

When I do a read operation on 'register' E1 and vary the voltage on the AVOUT and align the 10 bit values to integers... I see that the VIN value changes while the VOUT value does not ! (had scoped out voltage levels on pins 15 (AVIN) and 17 (AVOUT) just to make sure - and tied AVIN to ground... same problem...

The reserved bits in 'registers', E0, E1, E3, and E5 --- consistently have their first bytes set to 0x56, 0x7F, 0xE0 and, 0x18 (not the default 0x00 value outlined in the spec sheet).... only register E4 register has the reserved bits set to 0 (as documented)...

Writing values to "register' E3 are read back correctly with the exception of byte 7 which always comes back as 0x00 --- byte 7 is not NACKED by the SM72445 as I explained above... and the write operation on bits 7 to 0 of the E3 Register do not alter the SM72445's operation (for example outputting the PLL clock on pin 5 by setting bit 1 does not output the clock on pin 5 !

Now... the board is a custom board, there were other devices on the I2C bus (all worked fine)... that I had removed just to make sure there were no conflicting issues... the only device communicating with the SM72445 is the micro (running its I2C subsystem at about 50 KHz which is well within the 100 Khz I2C specification by Philips... both the clock and data lines are pulled up via 2.2 K ohm resistors... the digital and analog power contains less than 0.005 volts of noise --- ringing and ground bounce are less than 0.025 volts... of course the board has a solid ground plane...

Any detailed timing specs or additional information regarding the I2C operations of this chip would be much appreciated...

  • 0
    TI__Guru* 75370 points

    Hi Glenn,

    Thank you for posting.

    Our group owns the application assistance. This this is an old device, and unfortunately we do no have much resources to support with its digital interface.  On the other hand, there are some customers using the I2C interfaces, but we haven't heard similar issues being reported.

    Actually the SM72445 can operate well in MPPT without activating its I2C interface, which were how all the certified SM3220 products being configured back in 2009-2010. If you still want to use micro to control the MPPT, you may consider TI's C2000 solution and you can access to plenty of support resources.

    Thanks,

    Youhao Xi, Applications Engineering

     

  • 0 in reply to Youhao Xi
    Prodigy 40 points

    Are you saying TI is unable to supply the timing specifications for this IC ?

  • 0 in reply to Glenn Weile
    TI__Guru* 75370 points

    Hi Glenn,

    I just want to help.  What I meant is that the resource to support the I2C interface of this device is rather limited. If you can stay with the analog solution provided by this device, we can support you right away. We would have to find help from the digital experts in other groups but it would take longer time.  In stead, using C2000 solution seems to be a quicker path.

    Thanks,

    Youhao

  • 0 in reply to Youhao Xi
    Prodigy 40 points

    My apologies.... Mr Youhao... I know you're trying to help... but plugging that mixed signal processor (C2000) on this board is something I'd expect from some 'frat boy' salesman barely out of school... It was my mistake to think that this forum was supposed to be helpful as to the support ticket TI sent me... this has not been the case. Ask yourself Mr Youhao... the question... Do you want to be in tech support or sales ?

    I'm not the only engineer dude in the company I work for seeing bugs on this IC my friend... and you should be aware that I'm a senior engineer that more or less has to tell management that 'a mistake was made in selecting the IC and there is zero support behind it' --- it is also my job to come up with a solution when 'tech support fails'... 

    I think this may be the 5 th time in 35 years being a senior engineer that I've had to ask for tech assistance from a manufacturer (and usually it has come from the design group of the IC)... It's sad how it is Mr. Youhao I suppose because the SM72445 seemed a perfect fit.... that we now have to spin the boards and go with the Linear Tech / ADI LT8491 IC... 

    No big deal Mr Youhao... we were only going to purchase about 3000 of these SM72445 ICs this year... you may as well mark this thread as closed, unsolved, whatever..... 

    And PS... Mr. Youhao.... IMHO... you're better off in sales than tech support...

  • 0 in reply to Glenn Weile
    Prodigy 40 points

    Oh... I forgot to mention in my I2C communications issues with this rotten IC Mr. Youhao... that the IC ACKED every 'command register address from 0x00 to 0xFF - should not have the IC NACK invalid command register values ?  And I'd always get an ACK back during a write operation on the length value no matter what it was... 

    Forgive me Mr Youhao... I guess I'm getting a bit old and frustrated these days... but you know back in the old days we would always hang around the coffee machine as we waited for our scopes to warm up... these days we hang around the coffee machine waiting for our scopes to boot up...

    I love the old Lecroy 9360 scope I've got, just as reliable as the day it was built.... depicts exactly what I read in from this useless chunk of silicon TI for some odd reason TI is still selling. 

    I hope you don't mind Mr Youhao, that I've also brought this to the attention of a few suppliers unfortunate enough to have this rotten silicon sitting on their shelves... 

    The IC is trash Mr Youhao... as the scopes, logic analyzers, protocol analyzers... and other engineers all tell me.... 

  • 0 in reply to Glenn Weile
    TI__Guru* 75370 points

    Hi Glenn,

    Sorry for the frustration that you have had wit this device. If you wish I may give you some background info through PMS.  

    Thanks,

    Youhao