• State TI Thinks Resolved
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 119 views
  • Users 0 members are here
Support feedback
Options
Options
Related

SN74LVC1G125: Amplifier ICs and Alternative Approaches for 24 MHz Clock Transmission to Image Sensor

Conor
Guru 12155 points
Part Number: SN74LVC1G125
Other Parts Discussed in Thread: DS90LV017A, DS90LV012A, DS90LV011A, DS90LV018A

Tool/software:

Hi,

We are currently evaluating a design where a 24 MHz clock needs to be transmitted via cable to an image sensor.
In general, we understand that differential signaling (LVDS) is more advantageous than single-ended CMOS in terms of EMI reduction and signal integrity.
However, the image sensor we are using only supports single-ended input and does not accept differential signals.

At present, we are using the switching buffer SN74LVC1G125DCK, but even with EMI filters, radiated noise remains a significant issue.
Our customer has also suggested that replacing the buffer with an amplifier IC similar to a video amplifier, which has slower edges, might help reduce EMI.

Given these constraints, we would like to request your advice on the following:

  1. Does TI have amplifier ICs (e.g., video amplifiers) that are suitable for clock transmission in the tens of MHz range?

  2. For sensors that only accept single-ended input, would it be practical to use LVDS for the cable section and then convert back to CMOS right before the sensor?

  3. Could you recommend suitable TI driver/receiver devices for such an implementation, or alternative methods that TI would recommend for EMI reduction in this use case?

Best regards,

Conor

  • 0
    Guru 316920 points

    1. Any digital buffer can be made to output slower edges by adding a resistor. (This should be used with a Schmitt-trigger buffer at the receiver, and is still sensitive to noise.)

    2. Yes, LVDS is commonly used this way.

    3. DS90LV011A/DS90LV017A and DS90LV012A/DS90LV018A.

  • 0 in reply to Clemens Ladisch
    Guru 12155 points

    Hi Clemens,

    We would like to confirm whether our understanding regarding the use of video amplifiers for clock transmission is correct. Specifically:

    1. Video amplifiers are designed for accurate amplification of analog signals and are not suitable for transmitting clock signals (square waves).

      • Potential concerns include distortion of rise/fall edges and increased jitter.

      • CMOS logic-level compatibility is not guaranteed, which may cause unstable logic recognition at the receiver.

      • Their wideband design may also lead to higher power consumption and additional noise components.

    2. Therefore, using a video amplifier would not provide significant EMI reduction and may even degrade clock quality, so this approach is not recommended.

    3. For clock transmission, instead of video amplifiers, TI recommends using differential drivers/receivers (e.g., DS90LV011A/017A with DS90LV012A/018A) or logic buffers combined with series resistors.

    Could you please confirm if this understanding aligns with TI’s recommendations?

    Thanks

    Conor

  • 0 in reply to Conor
    Guru 316920 points

    This is mostly correct.

    It would be possible to slow a video amplifier down by adding a low-pass filter at the input, but this is no better than a digital buffer for the reasons you listed.

    A single-ended signal with slow edges might work. But if your EMI filters did not help, then a differential signal is likely to work better.