• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 46 subscribers
  • Views 493 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.

AMC1306E25: Modulator clock - artefacts caused by heating up.

wolfgang gaerber
Intellectual 615 points
Part Number: AMC1306E25

Recommended Modulator Clock

Evaluating AMC1306 (first prototype series) - we see some problems concerning stability of this part.

Running with somewhat 16MHz on CLKIN - we see artefacts on heating up the device.

Isolating the problem - without thermal treatment - it looks like there are some modulator clock ranges/frequencies which introduce artefacts.

Is it possibly that this is introduced by beat frequencies with the internal spread spectrum oscillator for the isolation barrier ?

Are there recommended modulator clock frequencies ?

Whats the typical isolator internal clock ? Are they synced ? 

rgds.

Wolfgang

  • 0
    TI__Guru**** 168895 points
    Hi Wolfgang,

    Can you be a little more specific on what sort of artifacts you see? How are you applying the heat treatments?
  • 0 in reply to Tom Hendrick
    Intellectual 615 points
    Hi Tom,

    Our Setup: Delfino clocks AMC1306 with sysclock/2 , divider 6, 50% duty cycle, 16.45Mhz. On warming up the device with heatgun, detached solder iron or climate chamber - we see spikes starting from few 100LSB to FS reaching a 60-80deg celsius maximum. Further heating to higher temperatures reduces spike amplitude from max and resumes stable operation. Applying 17.8MHz clk from an external generator (to eliminate delfino running in same time domain causing problems with sampling manchester edges) shows identical behavior, other clock frequencies show no problem. Manchester signal looks consistent all the time. Data is converted using delfino builtin filter. Same behavior with insulation barrier shorted. Behavior not seen on all devices or in the same strength - having 6 x AMC1306 - on a board and 3 boards tested - we have problems on all boards.
    rgds.

    Wolfgang
  • 0 in reply to wolfgang gaerber
    TI__Guru**** 168895 points
    Hi Wolfgang,

    Thank you for the additional detail. How do you have the inputs configured during this testing? Are you actually sampling an input signal or do you have the inputs shorted? Spikes up to FS sound like you might be hitting the Vcm input limits, causing the output stream to saturate.
  • 0 in reply to Tom Hendrick
    Intellectual 615 points
    Hi Tom,

    We see this in various input configurations with inputs shorted or low load, respecting Vcm limits. The height of the spikes depends on heating and clkin frequency. There is somewhat 200KHz bandwidth. If we heat up until we see the spikes - changing input frequency by 200khz makes them disappear. This points in a different direction.

    rgds.

    Wolfgang
  • 0 in reply to wolfgang gaerber
    TI__Guru**** 168895 points
    Humm... With shorted inputs or small DC shifts, you might see an idle tone, but that should not be anywhere near full scale. When you say changing the input frequency by 200khz, are you referring to the input clock or the actual modulator input? Do you mind if I send you a note to your MyTI profile e-mail account? We can take this off line for the time being.
  • 0 in reply to Tom Hendrick
    Intellectual 615 points
    We just see slightly noise DC (shorted or no input) until we got the spikes from heating. I refer to the input clock. No problem offline. rgds.