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.

ADS1247 SYNC

Other Parts Discussed in Thread: ADS1247

Hi, I have a ADS1247 that I want to synchronize with other like devices every second.

The datasheet is not very clear with regards to what the latency of the device is after initiating an SPI SYNC command.  Here is a cut and paste from page 34 of the datasheet:

The filter is reset two system clocks after the last bit
of the SYNC command is sent. The reset pulse
created internally lasts for two multiplier clock cycles.
If any write operation takes place in the MUX0
register, the filter is reset regardless of whether the
value changed or not. Internally, the filter pulse lasts
for two system clock periods. If any write activity
takes place in the VBIAS, MUX1, or SYS0 registers,
the filter is reset as well, regardless of whether the
value changed or not. The reset pulse lasts for 32
modulator clocks after the write operation. If there are
multiple write operations, the resulting reset pulse
may be viewed as the ANDed result of the different
active low pulses created individually by each action.

Here's how I understand it: 1) SYNC command -->2) wait 2 clks-->3) filter reset (2 Clks)-->4) filter pulse (2Clks)-->5) Write Activity (if any)-->6) filter reset (32 modulator clocks)-->7) filter reset wait time (approx 1 sample time)

Is my flow correct?  Basically I want to understand how many clock cycles we get delayed when using the SYNC command (when in constant conversion mode) 

What is a multiplier clock cycle?  I can't find anything in the datasheet that defines this.

Can you clarify the last sentence?  It is unclear to me.

  • Hi, the datasheet gives two pieces of information with regards to sample rates, and I would like to understand the impact individually, and together

    1) Internal OSC frequency tolerance is roughly +/- 5%.  Is this between parts?  I.e. one part will always be the same frequency no matter the age or temperature? Or does this value vary by temperature? Does this tolerance also impact the sample rate?  Does the sample rate vary over temperature by +/-5% (following the internal OSC tolerance)?  Is there any way to normalize / calibrate the parts' sample times so that they are reasonably close

    2) Table 13 pg 26 shows the nominal and actual data rates.  How does this correlate with OSC frequency?  Is the actual data rate correspond to the nominal frequency tolerance?  If not, how does it correlate?

    I have multiple units that will be communicating with a computer over USB.  I am trying to develop a good method to reasonably synchronize the AtoD readings of the units.  I am struggling with such issues as the non deterministic nature of USB, and these SPS variances between parts.

    Thanks for your help!

  • Hi Scott,

    Welcome to our forum!  I've merged your two questions regarding the ADS1247 since they are both 'similar' and we're working on a comprehensive answer.  When you say you have 'multiple units', is this a single board with multiple ADS1247 devices on it or multiple boards strung together via USB?

  • Hi Tom-

    There are up to 12 individual boards, all with a micro and ADS1247, each attached to a computer's USB port (or external hub).  Each unit uses a Stream Pipe, and the ADS1247 will be running in continuous mode.

  • Hello Scott,

    The ADS1247 is a delta-sigma converter and the data rate is proportional to the master clock frequency.  The internal oscillator frequency will vary from device to device and is guaranteed to +/-5% accuracy with a typical frequency of 4.096MHz. 

    Figure 37 shows the typical data rate variation when using the internal oscillator against temperature and voltage supply.

    The device can be used with the internal oscillator or fed with an external clock source through the CLK pin.  In applications where users need to synchronize multiple devices, the user will need to drive the devices with a common external master clock and use the sync command and the START pin to synchronize the devices.

    Since the devices are in different boards, I am not sure if it will be possible to accurately synchronize the separate devices in this application.  You could try using the START pin but there will be variation between the internal oscillators of the devices.

    Best Regards,

    Luis