Input Impedance Question

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi!

I need a input line level buffer stage with an input impedance of approximately 470K.

The easiest way to do this is with a non inverting unity gain stage as shown below. Note, the device that drives this is already DC decoupled with a cap.

I would like to use something like the OPA2134 dual op amp to drive a DRV134. I think this circuit will do the trick, but I just wanted to be sure there wasn't any hidden snakes laying in the grass.

An externally hosted image should be here but it was not working when we last tested it.
 
Administrator
Joined 2007
Paid Member
It's fine although you don't need R5 (and R1). Depending on what the input capacitance of the DVR134 is, then it might be good practice to add a 68 ohm or thereabouts in series with the output. And do you need a cap on the input ? That's source dependent of course but again, good practice.
 
It's fine although you don't need R5 (and R1). Depending on what the input capacitance of the DVR134 is, then it might be good practice to add a 68 ohm or thereabouts in series with the output. And do you need a cap on the input ? That's source dependent of course but again, good practice.

Thanks.

R1 was just put there for a load for the sim. Adding a series 68Ω resistor is a good idea. I'll add that one.

R5 can be deleted if it is not needed.

Yes, the source is fully decoupled with its own cap, so I did not want to add another cap in the signal path.
 
AX tech editor
Joined 2002
Paid Member
Did you check for input bias on that opamp? Input bias current through the DC input R can cause appreciable output offset. Especially when you use input capacitor: the +input bias current will flow through 470k and the -input bias current through zero.
BTW, why the need for 470k Zin? That's unusual in audio.

jan
 
Did you check for input bias on that opamp? Input bias current through the DC input R can cause appreciable output offset. Especially when you use input capacitor: the +input bias current will flow through 470k and the -input bias current through zero.
BTW, why the need for 470k Zin? That's unusual in audio.

jan

The op amp is actually a OPA2134 with a FET input. Is bias still a concern?

The input impedance is driven by an old PAS 3 tube preamp and the output impedance for that preamp is required to be 470KΩ or nasty things happen to the frequency response (ask me how I know).

This circuit is just a quick & dirty fix until I complete my own DIY preamp.
 
Administrator
Joined 2007
Paid Member
If your feeding this from a tube amp then you must be certain that not only is the tube amps output cap OK but also that it has a high value resistor to ground (even if its 10meg) on the output. If not and its floating then the cap could charge to some high voltage and this would zap the IC as soon as you connected.
 
If your feeding this from a tube amp then you must be certain that not only is the tube amps output cap OK but also that it has a high value resistor to ground (even if its 10meg) on the output. If not and its floating then the cap could charge to some high voltage and this would zap the IC as soon as you connected.

Good point. I'll simply replace the cap with a new one just to be sure.
 
Administrator
Joined 2007
Paid Member
Good point. I'll simply replace the cap with a new one just to be sure.

If it needs it then yes :)

But it must be terminated on the output side to ground via a high value resistor (which it may already have). If it hasn't and you leave it floating, then the output will assume whatever voltage is on the other side of the cap (tube amp... could be 50v to ??? who knows) and that would instantly zap any solid state unit you attempted to connect it to.
 
If it needs it then yes :)

But it must be terminated on the output side to ground via a high value resistor (which it may already have). If it hasn't and you leave it floating, then the output will assume whatever voltage is on the other side of the cap (tube amp... could be 50v to ??? who knows) and that would instantly zap any solid state unit you attempted to connect it to.

Here is the PAS 3 circuit. Not likely that there will be DC on the output since at least two caps would need to fail (double fault).

An externally hosted image should be here but it was not working when we last tested it.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.