DAC38RF89: Single-ended, DC-coupled Outputs

Hi,

The DAC has a single ended clock input at PIN A12 that you can use.

To DC couple to this clock input, you need to make sure the clock driver has a common mode of 500mV and a peak to peak swing of 800mV or higher. Also note the clock input is internally terminated with 50ohms to ground.

Thanks,

Eben.

Hi,

I was not talking about the clock input, but the DAC *outputs*.

We want to use the current-sink outputs of the DAC in a DC-coupled way.

Thx

Hi,

My mistake.

You can use the schematic below to model the DAC output.

To successfully DC couple, you need to bias the DAC output to 1.8V. Also you must ensure the peak to peak voltage swing + common mode voltage never exceeds the compliance voltage range of 1.3V to 2.3V on each complementary output pin.

The current sources (IG1 and IG2) in the model can be changed based on table 45 in DAC38RF89 datasheet. The model below is shown for full scale current of 30mA (or DACFS = 10)

Thanks,

Eben.

Thanks,

Eben.

Hi,

thanks for your answer. Now it is clearer to me how to simulate, however, I don't know what.

I was looking for some example circuitry to DC-couple the DAC directly into a 50Ohm load to ground.

Or is this not an option at all for current-sink outputs? I don't want to use a transformer.

Thanks again

Hi,

What is the min and max frequency you will generate from DAC?

Thanks,

Eben.

Hi,

what we are trying to do is to mimic signals that come out of particle detectors. There is quite a variety of different technologies for that purpose,

and so there is also a variety of pulse shapes and lengths.

Common to all is that there might be nothing for an extended period of time (0V), but then there is one pulse of more or less known shape.

The rise time can be less than 100ps. The pulse length can be up into the us-range. So we need a frequency range from DC to more than 3.5GHz.

Thx

Hi,

no suggestions any more?

Thx

Hi,

Is it possible to terminate the 50ohms load to a bias voltage instead of ground?

Thanks,

Eben.

Hi,

unfortunately, no. One can think of the load as an oscilloscope with 50Ohms to

ground.

Thx, regards