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.

THS7374: How would I achieve DC output coupling if this device adds 150mV to the input?

Part Number: THS7374

Hi team,

Customer has some questions

I intend on using THS7374 in a design (unless you can recommend a different part under $2).

My product contains two analog RGB outputs, which I'd like to mux into the final output. I plan to use a CD4053 analog mux, and a THS7374 amp. The final output is a 0.7V RGB + Sync cable, with 75 ohm load. The load (monitor) will usually have DC-coupled inputs.

The datasheet mentions DC output coupling capability. How would I achieve this if this part adds 150mV to the input?

Since the device has a gain of 2, the output will at minimum be 0.3V. But the black level of 0.7V 75ohm RGB is 0V.

Perhaps I am missing something fundamental about analog video signals, or a function of this part.


Could you help?



  • Hi Jiahui,

    The primary reason for the level shift feature for this device is to ensure there is no saturation/clipping of the sync input signal which is typically hovering around 0V. This ensures the device has full sync dynamic range. As far as the output signal, is there a standard you're working with regarding the signal levels expected for your application? 

    Best Regards,


  • Hi Ignacio,

    Thanks for your reply.

    Customer also replied

    I can see the purpose of the level shift to allow both AC-coupled and DC-coupled Composite/Luma/Sync signals to work using a single-supply.
    But what about RGB signals such as VGA? Are the black levels not meant to be 0.0V?
    The purpose of the THS7374 in my circuit is to drive an analog RGB+S video cable (75R terminated, intended to drive 15kHz-capable VGA capture devices, and 15kHz-input RGBS monitors). That means 0.0V to 0.7V for RGB lines, and -0.3V to 0.7V for Csync.
    DC-coupling is preferred to remove the need for large caps. But I can do AC-coupling if it's required.
    Here's the schematic for the next prototype PCB:
    The R/G/B/CS_Pre lines are from nearby circuits, with a few k output impedance. I will dial in the correct voltages to obtain the desired amp output voltage range.
    The desired amp output voltage is 0.0V to 1.4V I believe, since it has a 75R output, and 75R termination at the monitor.
    But with the level shift, the Csync line will be fine, but the RGB lines will end up with 0.15V - 0.7V at the monitor. I fear the 0.15V minimum will cause a non-black black level.
    Part of this circuit is the reference design for a TDA3330-based NTSC decoder. This 12V IC's RGB outputs usually go to a transistor amp stage for either CRT guns, or RGB cables.
    Here's the reference design's RGB cable driver coming from pin-12 of the TDA3330:
    I only need to drive the inputs of the THS7374, so perhaps a resistor divider is all that's needed at the RGB outputs of the TDA3330.
    Could you help?
    Thanks & Best regards,
  • Hi Jiahui,

    The THS7374 allows for both ac-coupled and dc-coupled inputs. Although the input of the THS7374 allows for a 0-V input signal with no issues, the output swing of a traditional amplifier cannot yield a 0-V signal, resulting in possible clipping. This condition is true for any single-supply amplifier as a result of the output transistor limitations. Both CMOS and bipolar transistors cannot go to 0 V while sinking current. This characterization of a transistor is also the same reason why the highest output voltage is always less than the power-supply voltage when sourcing current. Therefore, if your application requires a 0V to 1.4V output (or 0V to .7V after the resistor divider) the amplifier will not be able to do this. 

    The Csync line can ideally hover around -0.1V which is concerning if your application has a Csync line that ranges beyond this value.

    As for the TDA3330 question, I could not find the circuit you were specifically referencing, however from the looks of the application it seems like the circuit is using a discrete solution to amplify the signal. The THS7374 would take its place as the amplifier stage, however the device might not require termination resistors at the output pin of the decoder (or input of the THS7374) as this will depend on the output configuration of the TDA3330.

    Best Regards,