THS4513-SP interface

Hi Jamaal,

Depending upon load, you should be able to get single ended output voltages (absolute on each output) to within .5 V or so of each supply rail. With 1 V Vcm and Vs+ = 3.3 V, the Vod will be limited by the single ended V- output swing. To maximize Vod, you should set Vcm = 1.65 V (1/2 of the supply voltage range)

- I don't believe there will be significant performance impacts (noise, offset bandwidth etc.)

- The TFS35A data sheet specifies the output termination impedance required. This can be built in to a balanced input network for the THS4513. See section 11 of the attached App. Note and contact Vectron for termination network component details.

- I'll ask the device engineers about the input voltage limits.When you say "no supply voltage present" do you mean open circuit looking back into the supply?

Thank you for the response, Jeff.  1Vcm seems to be a bad idea...

 The "no supply voltage present" comment relates to my board's power sequencing.  This device would be powered from 5V, but the device driving it (through the TFS35A) is powered from 3.3V.  The 3.3V supply could come up first, driving into an unpowered THS4513.  I am wondering if this would damage the device?

Thanks again

A couple things to add...

If we assume a 100-ohm, AC-coupled network established out of the Vectron device.  Can you provide information on the proper values required for the THS4513 to establish a 20dB gain and a self-biasing input?  The datasheet lists values and topologies based on 50-ohm single ended operation, but not for differential input toplogy.  The app note you sent addresses the differential input, but I am not sure about the self-biasing aspect.

My proposed topology is a differential adaptation of Figure 68 in the THS4513 datasheet.

It may be as simple as using the values specified in Table 1 on each leg of the differential path, but those values do not agree with what would be calculated using the app note.

Thanks

Hi Jamal,

- For your question re: the input voltage allowed when the device is not powered, I was asking about the output resistance of the supply that powers the THS4513. This is important because if the device has input overvoltage protection diodes that tun on, current will flow back through the supply output to GND.Do you know what the THS4513 supply output resistance is when its off?

- I'll look into the input circuit design questions....

Hi Jamal,

- The Abs Max Vin, Vid specs apply for the "no power" (high Z looking back into the supply) and power down conditions. Vin to either input should be in the 0 to 5 V range but Vid (input differential) should not exceed 4 V. I'll double check this with the device engineers.

- For the circuit values:

* For a balanced diff amp, I would choose Rf = 348 ohms which yeilds Rg = 34.8 ohms for a diff gain of 20 db.

* Put Ct = .22 uF in series with Rt to GND on each input to self bias the THS4513 inputs.

* Calculate Rt following the procedure in the App. Note for the circuit of Fig. 21. Rs on each leg of the diff amp input is 50 ohms to get 100 ohms diff  source. I think you can assume that the .22 uF capacitor looks like a short for the calculation.

Jamaal,

With regard to your latest questions (below):

- Q: What is the behavior of the CM pin with PD_N is asserted (e.g. the device is held in power-down)? A: The input impedance is 23k ohms // 2.8 pF and the input circuit is shown in Fig. 62 of the Data Sheet.

- Q: Do the device outputs go low when PD_N is asserted? A: Yes, per Figs. 20 and 21 in the Data Sheet.