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.

DAC8562-Q1: gain

Part Number: DAC8562-Q1
Other Parts Discussed in Thread: DAC8562,

Hello

I use the same program to run dac8562 and dac8562-q1.I found dac8562 work well,but dac8562-q1 can't output the same. The init program is below.

When run DAC_Write_3Bytes(0x19, 26214, DAC1_CS);,dac8562 Boutput 1V,dac8562-q1 Boutput 1.9V.I had test several dac8562 and dac8562-q1 and the result is same.//26214/65536*2.5 = 1

void DAC1_init()
{
    DAC_LDAC1_Set();

    DAC_Write_3Bytes(0x28, 0x0001, DAC1_CS);//Reset all registers and update all DACs (Power-on-reset update)

    DAC_Write_3Bytes(0x20, 0x0003, DAC1_CS);//Power up DAC-A and DAC-B

    DAC_Write_3Bytes(0x38, 0x0001, DAC1_CS);//Enable internal reference and reset DACs to gain = 2

    DAC_Write_3Bytes(0x02, 0x0003, DAC1_CS);//Gain: DAC-B gain = 1, DAC-A gain = 1 (power-on default)

    DAC_Write_3Bytes(0x30, 0x0000, DAC1_CS);//LDAC pin active for DAC-B and DAC-A

    DAC_Write_3Bytes(0x18, 0, DAC1_CS);

    DAC_Write_3Bytes(0x19, 0, DAC1_CS);

    DAC_LDAC1_Clr();

    __asm("nop");
    __asm("nop");

    DAC_LDAC1_Set();
}

Thanks a lot.

  • Hi,

    There should be no difference between the two devices from a digital interface perspective.  I suspect that you have an issue with your hardware setup.  How do you have the boards wired? Are you testing this on the evaluation board or some hardware you created?  You should start by verifying all of the digital input, supply, and ground are correctly connected.  You might also want to use an oscilloscope to review an SPI command to verify the edge timing is correct.  Sometimes even the wrong SCLK polarity will still work with a device if there is significant parasitic on the digital lines.  It might be possible that your DAC8562 setup was only marginally working.

    Thanks,

    Paul

  • Hello

    Here is the waveform when I set the gain.

    CH1 is DIN,CH2 is SCLK,CH3 is SYNC.

    I can't find problems from it.I  run both dac8562 and dac8562-q1 in the same board,and the dac8562-q1's gain remain x2.

    And I find dac8562-q1's reference pin outputs 2.400~2.490V,less than 2.495V.And dac8562 output is 2.503V.

    I also test several CPOL,CPHA,bitrate ,and results have no change.

    Any ideas about it?

    Thanks a lot.

  • Could the data for the Q1 device be left-shifted by one bit on accident? If you measure the DIN pin on both devices simultaneously, are they identical? Are the clocks identical between the two?

    If you write full-scale value to both devices (0xFFFF), is the DAC8562 = 2.5V and the DAC8562-Q1 = 5V?

    Can you share a schematic as well?  

  • Hello

    CH1 is DIN,CH2 is SCLK.

    The first 3 bytes is to set dac8562-q1's gain,and the second is to dac8562.

    I translate the first 3 byte to the second,and you can see the DIN and clocks are identical between the two.

    When I set gain x1 and write full-scale value to both devices (0xFFFF), the DAC8562 = 2.5V and the DAC8562-Q1 = 3.4V(Vcc).

    Why same waveform have different result?I suspect dac8562-q1 cant set gain to 1.

    Thanks a lot.

  • How does the CS1 and CS2 lines look relative to these two lines?  Are they gated correctly?  If you connect CS1 to CS2, do they both update correctly (the same in that case)?

  • The first waveform has CS1 in CH3 and CS2 in CH4.They are correct.

    Then I connect CS1 to CS2 and change program to control CS1 and CS2 as the same.In the second waveform the dac8562 and dac8562-q1 use the same CS,DIN.CLK,and their outputs are different.DAC8562-q1 still can't set gain to 1.

  • Hi,

    When do you enable the internal reference? I noticed that when the reference is enabled, the gain is reset to 2x.  

    Moreover, the command to set the internal reference might accidentally be getting issues if you hold the DIN line high while still clocking the device with chip select low.  If you look at table 17 in the DAC8562-Q1 datasheet, you can see that all the bits are 1 or "don't-cares".

  • Please check the waveform then suspect the command issue.

    Now the issue is dac8562-q1's internal reference not 2.495~2.505V,and dac8562-q1's gain can't set to 1.

    I have show the waveform and the program before,and if they don't have problem and dac8562 can set gain to 1,the only problem is dac8562-q1's function.

  • It is concerning that the reference is not 2.5V.  If you look at the reference voltage with an oscilloscope, do you see any oscillation?

    Is the device temperature high? I wonder if we have a soldering or assembly issue that is causing the device to reset randomly, and the cap on the reference pin is providing an external voltage while it discharges.  Was this a hand soldered board? If so, you might want to check all the solder joints with a scope.

    As a basic check, try power cycling the device and only issue the reference enable command to only the one DAC.  Do not issue any other commands - does the reference turn on to the correct value? If it does not, and you are confident that that the board is correctly assembled, then I suspect the device might be damaged or there is some power supply glitch issue.

  • I test 2 samples from ti and they work well.They can set gain to 1 and reference is 2.495V.

    The abnormal dac8562-q1's device marking is ZAAQ and they have 84 in the left of first line.

    Most of them are mounted by machine,and I mounted two by hand,and they all can't set gain to 1.

    Can ti find the reasons or they are fakes?

  • Hi,

    If these came from an authorized TI distributor, then they should be authentic devices.  If that is the case, you can reach out to the distributor and tell them that you would like to quality return.

    https://www.ti.com/info-store/distributors.html

    If these are not from an authorized distributor, then TI cannot warranty the part.  You could provide all the markings on the device, and I can see if it is a valid lot trace code and date, but that does not necessarily mean the device is or is not authentic.  I cannot verify authenticity without a quality return event. 

    ZAAQ is just the device name markings.  I will need the two other digits from the top of the device for the date, and markings on the bottom of the device for the lot trace code.

    Thanks,

    Paul