Simple Hi-Z to Line-In level buffer opamp based circuit - design review appreciated

[fab672000]

Hi, I found this opensource stereo input buffer circuit that I would like to try.

In theory this circuit could handle two high impedance mono inputs and provide a proper impedance line level output.

I had a look at the low pass RC filter (at the right end of the circuit) and calculated the frequency cut and I get approx. 159Khz which is way too high, especially for guitar hi-z buffers I think.
So I plan to replace the 1nF capacitor by a 6.8nF one (probably ceramic is enough for that low pass filter even though original impl use an MKT expensive cap), now I get 1/(2PI.RC) = approx. 23Khz with 6.8nF

I like the opamp design and also the use of a low power and fairly transparent TLC2262 opamp there (buffer will be battery power supplied), but would be interested to learn about other jfet alternatives (like the JRC chips we see in many designs)

Thoughts?

 

[JohnRoberts]

If you have an active stage why make the LPF passive where it can be influenced by termination, cable capacitance , whatever?

That said 160kHz is not crazy high for an audio path that is likely to have multiple LPF in series. They all add together.

JR

 

[fab672000]

>That said 160kHz is not crazy high for an audio path that is likely to have multiple LPF in series. They all add together.
Interesting, did not think about what's filtering behind the buffer, so I'll try remember that for the next projects too, thanks!

Now I just had a look to the line-in interface where I plan to mainly use that buffer: the audio T4 rev D. shield and I don't see any rc filter at the input of their (SGTL5000) chip, did I miss something? datasheet is here


Anything else in that design that could be improved?

 

[Newmarket]

Hi, I found this opensource stereo input buffer circuit that I would like to try.

In theory this circuit could handle two high impedance mono inputs and provide a proper impedance line level output.

I had a look at the low pass RC filter (at the right end of the circuit) and calculated the frequency cut and I get approx. 159Khz which is way too high, especially for guitar hi-z buffers I think.
So I plan to replace the 1nF capacitor by a 6.8nF one (probably ceramic is enough for that low pass filter even though original impl use an MKT expensive cap), now I get 1/(2PI.RC) = approx. 23Khz with 6.8nF

I like the opamp design and also the use of a low power and fairly transparent TLC2262 opamp there (buffer will be battery power supplied), but would be interested to learn about other jfet alternatives (like the JRC chips we see in many designs)

Thoughts?

I'd say the idea of the lpf on the output is to knock out ultrasonics from hitting following stage. I would probably omit it myself and leave that to any following input stage. Reference to (electric) guitar signal seems irrelevant here. You can lpf e guitar far lower than that without issue. And likely with an active filter stage.

 

[fab672000]

I'd say the idea of the lpf on the output is to knock out ultrasonics from hitting following stage. I would probably omit it myself and leave that to any following input stage. Reference to (electric) guitar signal seems irrelevant here. You can lpf e guitar far lower than that without issue. And likely with an active filter stage.

Indeed, I think most cabinets cut around 6-7kHz if I remember correctly. But the reason why I don't want to cut that much is that I will have digital embedded dsp effects running in series before/after the amp + cab IR. So, I don't want to assume/limit myself to 7kHz for now. Also, thanks to JR; I have learned that I should be mindful of potential additional RC filters in my signal chain if I use this box with another line level device in series, so I'll either recalculate the low pass filter to a higher value or keep the current value unchanged.