• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 46 subscribers
  • Views 2712 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.

TLC3578

karthik K R
Prodigy 30 points
Other Parts Discussed in Thread: TLC3578

Hai

I am karthik.  I am using TLC3578 to read single end voltages of range +5 to -5 Volt.  I have used reference voltage of +4.096V at REFP and REFM is connected to AGND.  I wanted to know am i right with the REFM connection because the TLC3578EVM Family pdf shows negative reference voltage should be connected but the datasheet didnt mention any of the point to connect negative reference to REFM.

And i am reading the test voltages through SPI register for both REFP and REFM 0x0FFF.  For (REFP+REFM)/2 and all other channel (irrelevent of the input) is reading 0x1000.  So can anybody please tell me where the mistake is?

  • 0
    TI__Guru**** 175931 points

    Hi Karthik,

    For the TLC3578 family of parts, it is correct to tie the REFM to ground.  Take a look at Figure 33 on page 38 of the data sheet.  Notice that the REFM pin is referenced back to the ground node of the input signal source.  Also, if you look at the resistor divider network and do a little math you will notice that the input circuit is really setup for a 4.000V reference as noted on page 6 of the data sheet.  Using 4.096V won't cause the other problems you see, but that will introduce a little bit of offset error to the conversion results.

    For the results when doing the internal self tests and channel conversions, I can't really say what might be going on.  I suspect there is a problem in the configuration of the device or something wrong in the way you apply the SPI clock and MOSI.  If you can post a screen shot of your control signals (/CS, SCLK, SDI, SDO) and perhaps a schematic showing how you have the TLC3578 connected to your processor, that would help us get you up and running. 

  • 0 in reply to Tom Hendrick
    Prodigy 30 points

    Dear Mr. Tom, 

     

    First i have to say thanks for your reply which reduced my worries about schematic, because this is my first schematic design.

    I have enclosed the schematic of the TLC3578 with this post and i have mentioned the procedure below which i used to communicate between TLC3578 and microcontroller (C8051F067 from silab).

     

    1.  As soon as power on, i am sending 0xA000 to ADC.

    2. Then 0xAA00 to configure the ADC to BOB format, short sampling and internal OSC.

    3. Once i have done the above transmission, i am sending 0xD000 and i get the response of 0x0FFF.

    4. Then i am sending 0xC000 and i get the response of 0x0FFF.

    5. then 0xB000 or 0x4000, i get response of 0x0000.

    I think i am doing some mistake in this procedure but couldn't figure it out where and i am reading the data sheet once again to get clear idea.  i don't have CRO to monitor all 4 line of SPI signal.  But i hope there is no mistake in SPI coding.  Because i have used the same code for other devices and those are working good.  

    Please instruct me the procedure if i am doing wrong.

     

    .

  • 0 in reply to karthik K R
    TI__Guru**** 175931 points

    Hi Karthik,

    I don't see where your AGND and DGND nodes tie together.  These pins have to be referenced to the same GND potential.  Also, it would be very helpful to see your /CS, SDI, SDO and SCLK signals captured with a four channel oscilloscope or perhaps logic analyzer.  What does your EOC/INT signal look like?  Do you see that toggling at all?

  • 0 in reply to Tom Hendrick
    Prodigy 30 points

    Dear Tom,

    Thanks for your reply.

    I didn't find EOC/INT signal is toggling.  It is always high.  Will that signal toggle even for test signals? anyway i didnt find it is toggling for both test signal as well as input signal.

    As i have told u already, i don't have the CRO which can measure and capture 4 signals. 

    I have tied AGND and DGND nodes already which is not given in the schematic.

    Am i right with the procedure which i am transmitting?  How do i confirm whether the device is configured for hardware default or programming mode?

  • 0 in reply to karthik K R
    TI__Guru**** 175931 points

    Your setup procedure should be correct, assuming you have the proper SPI  data to clock phase relationship.  The EOC/INT should toggle with every conversion, even the internal test modes.  This is probably the easiest way to determine if the part is working as you expect it to. 

    In your case, you clear the registers from power up by sending 0xA000 and then configure for BOB, Short Sampling with EOC/INT acting as an interrupt.  If the configuration worked, when you write your 0xB000, 0xC000 or 0xD000 you should see the  EOC/INT go low ~2.7uS after the 16th SCLK (refer to Figure 11, page 28).  It will stay low until you drop the /CS input again. 

    If you change your configuration from 0xAA00 to 0xAA04, this changes the operation of EOC/INT - you should see the line go low after the 16th clock and then return high ~2.7uS later, indicating that the conversion is complete.