Parallel output for BUF602 . It's possible?

[JhonStan]

Hi everyone, I would like to use an OP Amp BUF602 as a current buffer so I don't increase the voltage but only the current. The question is: Can I put the outputs of multiple BUF602s in parallel to increase the output current? I enclose circuit. Thank you.

[CaptDon]

In theory if all the inputs are also in parallel driven from the same source it could be possible. My concern is that each of those buffers has a low internal source resistance and each one has probably a tiny amount of offset on the output and so they would tend to fight each other to some degree. Perhaps a 1 ohm resistor in series with the output of each one? Are there any circuits shown in the datasheets for parallel operation??

[JhonStan]

Hi CaptDon, I confirm same input source. I had also thought about a small resistor in series at each output but I was not sure if it was correct. Thank you for the confirmation. Unfortunately I could not find any circuits of this component with parallel outputs.

[Terry Bites]

You'll have current sharing and thermal stability problems without a few ohms of ballast resistors in series with the outputs. Becasue of the heavy currents flowing, careful routing of power lines and placement of decoupling caps will be crucial to avoid oscillation. If you want unity gain and correct termination what about a bit of additional gain up front?

[mawyatt]

Why not use a NMOS/PMOS or NPN/PNP current buffer inside a unity gain closed loop with a high BW Op-Amp?

Best,

Edit: forgot the not!

[magic]

Given that the load is shown as 50Ω and each input is loaded with 50Ω to ground and the buffers are unity gain, I'm not sure how the current gain is supposed to happen.

Anyway, ±30mV worst case room temperature offset voltage, according to the datasheet. They will not be happy with their outputs shorted together.

[JhonStan]

To simplify the discussion, I will tell you what I would like to understand. Using a current buffer I would like to understand if it is possible to create a circuit to offer an output phase "in phase" with respect to the input signal. From 1 to 100Mhz. Current increase only because the voltage gain is managed by previous stages. In the case of using components such as the BUF, in simulation, this is possible but the output current is too low. This is why I tried to parallel multiple buffers.I could also do it with pnp and npn transistors but in the components I tested I always found an output phase shift of many degrees. The load that should be connected at the output is currently 20Ohm but I can also bring it to 50Ohm if it simplifies the project.

I am attaching an image to make me understand better.

Thank you

[magic]

The problem is up to 70mV difference between the outputs.

You could put resistors on the outputs, for example 10Ω limits idle current between the buffers to <7mA, but then output impedance is 10Ω/n (n = number of buffers) and output voltage will be less than expected due to load. Additional voltage gain could make up for it if load impedance is resistive at all frequencies.

Not sure how to make a DIY equivalent of such IC good to 100MHz. It likely uses some rather high speed transistors inside.

[JhonStan]

With 10Ω i see lowering the output but i think is acceptable. I had try another OpAmp and i reached better result in output but i nedd more test because need more input current.