Post Go back to editing

dc388c board not working properly

Category: Hardware
Product Number: LTC1923

I tried to use the dc388c evaluation board to stabilize my laser system. the board is not working properly as my temperature gets continuously oscillating for a long long time. can anybody suggest what to do in this case or am I making some mistake. I'm also uploading the variation of my thermistor resistance with time. I'm using %V supply and also my current is working at two extreme ends always while cooling(320mA) and heating(280mA). with time it's also not decreasing 

  • Hi,

    Thanks for contacting Analog Devices!

    Thanks for sharing the waveform. This is actually a very classic instability issue in TEC (thermoelectric controller) loops, especially with the DC388C.

    What your waveform is showing?

    From your image:

    • Thermistor value oscillates between roughly 8.6kΩ and 8.8kΩ
    • Oscillation is slow and periodic
    • It continues for a very long time (15+ minutes shown)
    • Current is hitting:
      • ~320 mA (cooling max)
      • ~280 mA (heating max)

    -> This clearly indicates:

    The control loop is saturating and oscillating (limit cycling)

    Root cause (very important insight)

    This is NOT a hardware failure.

    This is almost certainly:

    Loop compensation / tuning issue

    Your system is behaving like:

    • High loop gain
    • Very slow thermal system
    • Aggressive controller

    Result:

    • Controller overshoots temperature
    • Corrects too aggressively
    • Overshoots again
    • Repeats → oscillation

    Why this happens in your setup

    The DC388C is designed as a generic platform, but:

    -> Your laser + TEC + thermistor system is custom
    -> Thermal dynamics are very slow and nonlinear

    Mismatch occurs between:

    • Controller compensation
    • Actual thermal system response

    Key symptom confirmation

    These two observations confirm instability:

    1. Current always at extremes
    • Either max heating or max cooling
    • Never settles in mid-range

    This means:

    Controller has no fine control → behaving like bang-bang

    1. Oscillation does not decay
    • Amplitude not reducing with time

    This means:

    Loop is underdamped or unstable

    What you should do (actionable steps)

    Step 1: Reduce loop gain (most important)

    You must slow down the control loop

    Try:

    • Increase compensation capacitor (I-term cap)
    • Reduce gain resistor

    -> Effect:

    • Slower response
    • Less overshoot
    • Oscillation reduces

    Step 2: Increase integration time constant

    For thermal systems:

    • System is inherently slow
    • Controller should be even slower

    -> If controller is faster than thermal system → oscillation

     Regards,

    Tech Support

Before You Switch


Switching languages will make ADI Explorer unavailable. Resume your session by switching back to English and reopening ADI Explorer.