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.

OPA692: Multiple output: Should no used channels need to add dummy load?

Part Number: OPA692
Other Parts Discussed in Thread: OPA3692

Recently, I design 4 channels video fanout board with OPA692, as following picture:

In the some situations, I only turn CH1, CH2~4 is not used.

In this time, CH2~4 likes antenna.

Should I add dummy load(75 ohm) in the end of termination(SMA connector) to no used channels to prevent RF noise introduced?

I noticed this dummy load recently:

  • Hi DeWei,

    Does any voltage noise introduced by these "antennas" indeed show up at the amplifier's low-impedance output? It seems to me that the amplifier output should provide a strong enough signal to ignore any received antenna noise from these disconnected loads. I see how adding terminations will ground any antenna formed by the loads, but it might burn additional power unnecessarily. These SMA loads are usually just used to simulate a transmission line load.

    Best regards,

    Sean

  • Hi Dewei,

    In the some situations, I only turn CH1, CH2~4 is not used.

    What do you mean by this? That you have all the shown components installed but make a cut in the copper tracks at the right crosses? And what about the cables? Have you connected a cable only to CH1 or also to CH2-4?

    In this time, CH2~4 likes antenna.

    What do you mean by this? That you fear that something is working like an antenna? Or that you noticed EMI, as if something is working like an antenna? And if you noticed EMI, where exactly did you noitice it? And how? By what sort of measurement? Again, are any cables connected to CH2-4 then?

    Please be more specific. No details -> no help.

    Kai

  • Thanks @kai klaas69 and @Sean Cashin's reply and I'm sorry reply too late.

    The circuit here:

    and here is portion of power

    For the 1Vpp, 1M/10MHz test, they are no problem.

    But when I reduce voltage to 0.1V, I discovered noise:

    Above Picture: OPA3692@1MHz/1Vpp

    Following Picture: OPA3692@10MHz/1Vpp

    Yellow: CH2 input of Function generator (Waveform is the same to CH1 input of function generator)

    Red: Output CH1 / 50 ohm load

    Green: Output CH8 / 1M ohm load

    I guess noise come from grounding or EMI, because when I shielding(covered) with Aluminum chassis, result is better.

    So I think maybe I added SMA dummy may:

    * Shielding unused output improving SNR


    If I expose unused sma connector to air:

    * Unused channels may be antenna if I don't connect anything.(?) EMI noise may through back with unused channels?

    I know dummy load may cause additional power consumption,

    but this loading is slight for my case, our power supply can provide enough current(1.5A for +5/-5V),

    so that's no problem.

  • Hi DeWei,

    This looks to me a little like amplifier instability. Adding those extra loads, should help. However, hopefully you do not require too >2.5V amplitude; the max linear output current from this amplifier is about 190mA, and excessive loading can increase the distortion of the amplifier.

    Best regards,

    Sean

  • Hi Dewei,

    in this frequency range the Maxwell laws of electrodynamics come into play:

    A transmission line which is properly terminated at both ends has the signal completely travelling inside the transmission line. No portion of electromagnetic field of signal ("wave packet") is leaving the transmission line. Any grounded metal coming close the transmission line will not have any influence on that. The electromagnetic field of the signal (wave packet) is entirely trapped inside the transmission line.

    If you do not properly terminate the transmission line on both ends, on the other hand, the wave packet is no longer entirely travelling inside the transmission line and portions of the electromagnetic field of signal leave the transmissione line and are travelling outside the transmission line. Metal objects coming close the transmission line will now have a relevant impact on the signal. The transmission line is no longer looking fully ohmic but becomes complex and ringing and echoing can be observed.

    So: Do not connect any cable (transmission line) to an output of OPA692 without properly terminating the cable at both ends, not even a short cable ("pig tail").

    Why not soldering SMA sockets on the printed circuit? By this you could avoid the pig tails.

    Kai