DM3725 TV Out with OPA361

Hi Serge,

This E2E forum focuses on the Precision Analog products such as the OPA361. We don't have any direct association with the DM3725 digital media processor, but nonetheless, I'll assist you the best I can with your inquiry.

Originally, the OPA361 was designed to follow the video DAC output from the OMAP2420. That DAC output is a current output which was intended to flow through the OPA361's internal 500 Ohm input shunt resistor. When I review the DM3725 data sheet it appears its DAC output is also a current output DAC. However, the data sheet mentions it is intended to drive a 1.5 k Ohm load. If the DM3725 DAC output was connected directly to the OPA361 input it would drive the 500 Ohm input load instead of a 1.5 k Ohm load. The input voltage developed across the 500 Ohm resistor would be one-third of what would be developed with the 1.5 k Ohm load. You mention the OPA361 has too high of gain so that may work in your favor.

Alternately, adding a series resistance between the DM3725 output and the OPA361 input would reduce the input signal level to the amplifier. That would increase the input load resistance that the DAC output would drive. You mention a 1.6 k Ohm series resistor so now the DAC would now drive 2.1 k Ohms total. The voltage developed across the two series resistances would be higher compared to when the DAC output is loaded with 500 Ohms, or 1.5 k Ohms, but the added series 1.6 k Ohm resistor and 500 Ohm input resistor form a voltage divider. This would reduce the input voltage by 0.238x, which again might be enough to help compensate for the higher than needed OPA361 voltage gain. The OPA361 has an internal gain of 5.2 V/V so adding an attenuation of 0.238V/V in front of it results in a signal path gain of 1.24 V/V. The series resistor can be further adjusted to get the OPA361 output level where it is needed.

The OPA361 internal low-pass filter uses a Sallen-Key topology. If you add the 1.6 k Ohm resistor at the DAC output it is in series with the internal DAC current source resistance which is extremely high. That very high resistance shunts the OPA361 500 Ohm input resistor. Little deviation from the intended 9 MHz (-3 dB) filter response would be expected.

Regards, Thomas

PA - Linear Applications Engineering

Hello Sergy,

Thanks, it is my pleasure to assist you. Yes, your understanding of the OPA361 is correct.

The OPA360 is different in that it does not include the 500 Ohm input shunt resistor. Therefore, the voltage developed at the OPA360 input would be a function of the video DAC output current and whatever resistance you decide to terminate its output into. Then, there is the issue of the filter's source impedance and if it will affect the response. The OPA360 data sheet explains,

"The filter characteristics vary somewhat with signal source impedance. A source impedance greater than 500 Ohms can degrade filter performance. With current-output video DACs, a resistor to GND is often used to create a voltage output which is then applied to the OPA360 input (see Figure 1). TI’s Digital Media Processors, such as the DM270 or DM320, typically use a 200 Ohm resistor to GND to convert the current output signal. This 200  Ohm source impedance does not degrade video performance."

So as long as the resistor you select provides the necessary voltage level required at the OPA360 input and doesn't exceed 500 Ohms everything should be okay. I don't see any reason why the resistor couldn't be less than 200 Ohms, but greater than 500 Ohms will start to cause some noticeable deviation in the filter response. Below, I show an example of how source impedance affects the gain vs. frequency responses of Sallen-Key and Multiple Feedback low-pass filters(MFB). The example isn't the same filter as used in the OPA360, but it illustrates the point. The Sallen-Key filter response is pretty forgiving compared to the MFB when it comes to source impedance variation.

Let me know if I can be of any more assistance.

Regards, Thomas

PA - Linear Applications Engineering