• State TI Thinks Resolved
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 28 views
  • Users 0 members are here
Support feedback
Options
Options
Related

OPA237: OPA237 output waveform abnormality

jeno.zhang
Genius 3515 points

Part Number: OPA237

Hello,

The customer's product uses OPA237NA for controlling LED current. Not a new design, it has been used in the product for many years. The latest batch of components has encountered abnormal output waveform issues. During the production process, there were 40 drivers, of which 20 had issues. Please help analyze the problem.
The schematic diagram is shown below, where Q2 is 78L05, XA1 and XA2 are input signals and varying voltage signals.

image_1787834043760.jpg
The normal waveform is shown below,

image_1787834022169.jpg
The abnormal waveform is shown below,

image_1787834017346.jpg

The chip packaging label is shown below,

image_1787834245242.jpg

 

 

 

 

 

  • 0
    TI__Mastermind 29718 points

    Jeno,

    1. This looks like a stability problem.  
    2. I am looking into this to see if any changes were made to the die to degrade stability.
    3. I need to know the model number of the transistor connected to the output pin (Q2).  Also, most of input components values are not legible.  Knowing the values will help me understand the origin of the stability issue. 

    Best regards, Art

  • 0 in reply to Art Kay
    TI__Mastermind 29718 points

    Jeno,

    My colleague helped me find a little additional information on this device (Thanks Carrie!).

    1. This device did undergo a die redesign.  The device needed to switch to a modern process, and this necessitated a die redesign.  
    2. Here is the part-change-notice PCN for this device:  PCN_OPA237
    3. The new design was characterized to compare old vs new.  For most parameters the performance is very similar.  For some parameters the new die performance is better, and for some parameters the new die is worse.
    4. I think the key parameter here is stability of the device.  Overshoot vs capacitive load is a good measurement of stability.  More overshoot and ringing means less stable.  After reviewing the overshoot vs. capacitive load plots for old die vs new I see that under some conditions the old die was better and for other conditions it is worse.  Specifically, the stability is worse in the inverting configuration.  Also, the stability for non-inverting looks worse for falling edge signals.  
    5. I cannot post our internal validation documents.  However, based on your images, the new device is unstable.  
    6. There are two causes of instability.  Capacitive load on the output of the op amp, and capacitance on the inverting node to ground.  Note that a capacitive feedback path (as in the customers circuit) is not necessary a problem for stability.
    7. Instability can generally be resolved with circuit changes (compensation).  I think it is likely that the differences between the old version of the device and the new device are impacting the stability.  However, it is also possible that we can make a circuit change to compensate (correct) the instability.  To simulate the circuit and potentially correct the instability issue I will need to know the circuit details.  I will need to know all the component values.
    8. Other than adding compensation components to solve the stability problem, the only other option is to change the op amp for a device that is stable. To recommend a substitute op amp I will need to know the component values in your circuit.
    9.  Operational Amplifier Stability Theory and Compensation covers the theory of amplifier stability in detail.  

    Best regards, Art