Clean Buffer with 3.3V Supply for Guitar high Impedance Signals for Teensy Audio

There's not such thing as "totally clean" - can you be more precise? What actual performance are you wanting?

That link uses the MCP602 opamp which is a surprizing noisy opamp (i.e. not designed for audio) at 29nV/√Hz.
It is a micro-power opamp (supply draw is 0.23mA per amp, rather than the expected 4mA or so). It can hardly
handle any load at all, the output dropping substantially if loaded with less than 25k.

An AD8656 is 2.7nV/√Hz and thus a full 21dB quieter. It can drive 1k ohm loads with ease.

CMOS opamps like this benefit from input protection circuitry to protect from static discharge, BTW. The normal Schottky
diodes to the rails should do nicely.
 
Hi, Markt,
Obviously you are not a Guitar man? Guitar electronics are a own world often....
In this world clean means, that distortion is not obviously hearable. Low even harmonics will be seen welcome making the tone richer. Still very nice sounding clean tube amps from Fender ~1960 do not necessarily have negative feedback and therefore use the dynamics of speaker and cabinet to give back the resonance body which was stolen from the Guitar.
In this circumstances totally clean will be perhaps <1pc harmonic distortion. Frequency range for electric Guitar application is 80Hz...7kHz.

If you have a look, at that schematic, then you can see, that the maximum gain of the amplifier is set to about 4. Actually I use it with Humbucker Pickups at gain=1 together with standard level of the audio shield of 1.3Vpp. I think, this is the reason, why the relatively high input noise of MCP602 is not a problem here. At 1.3Vpp it's output range with the input impedance of the audio shield is sufficient to Overdrive these 1.3Vpp. The low current consumption helps enable the RC filter for supply voltage.

Thank you for pointing to AD8656! As explained I do not see the necessity to go for it in this application, but it is always good to learn.
Christof

Edit: Perhaps you can help me with this, as I am not feeling secure at this calculation:
Calculation for gain=1:
21nV*sqrt(20.000Hz (bandwidth))= 3E-6V.
We have a range 1.3Vpp at 16bits: 1.98e-5V per bit.
So this source of noise would be less than 1 bit for gain=1.
 
Last edited:
Connected and supplied via USB from my PC. Using MCP602 at gain=1 and in my unshielded plastic enclosure with a shorted input plug with standard input range of 1.33Vpp, the audio peak function detects peak<0.0004. (-71dB) noise level. With full gain it is peak<0.0006.

Edit: Guitar plugged into input of the buffer at gain=1 brings peak<0.0006 noise level. If I touch the strings, which are connected to Gnd of cable: peak<0.0013 (-60dB). A different Humbucker Guitar brings this noise level, whether I touch the strings or not.

So noise from other sources seems to be well more relevant than the input noise of this opamp in this setup.
 
Last edited:
This is what I meant about totally clean - if you are sampling at 16 bit for a 3.3V peak-to-peak signal you have quantization noise of 20µV rms,
and given guitar signals are often about 60mV, this is an SNR ratio of around 70dB. Not what would be generally called "totally clean" for audio.

This is why numbers are needed to describe performance, not adjectives!

My point about the puny drive capability of the MCP602 is thus perhaps the important point. Don't use micropower opamps for general use...
 
Hi,
As said you will need filtering of the supply voltage taken from 3.3V of Teensy, which is done here with an R330. If you have 4.5mA per opamp you will have some voltage drop. The existing pcb, that I used, uses a dual opamp. So it is nice to have low power consumption.
Of course things could be done differently for example with separate power supply. Feel free to make a complete suggestion, that is similarly simple and significantly better in reality for this application!
 
Inspired by MarkT's insisting for numbers, I did some additional evaluations.

Teensy 4.1 with audio shield supplied from and connected to my PC via USB.
Setup open without shielding. MCP602 in buffer.

Switched over from peak.read() to rms.read() which leads to nicer numbers for same physics. :) Better to compare to known numbers.
Comparison of a line input, which is directly connected to analogue GND to one, which is connected to the buffer. The buffer input is connected to GND.
For my Humbucker Guitar a line sensivity of 1.33Vpp without gain is good.
As the noise RMS value shall be compared to a sine signal with maximum amplitude, the RMS number is reduced by 1/sqrt(2) = 3.1dB to form a RMS Signal-to-Noise-Ratio.
SNR.jpg
So all in all the high impedance input buffer with gain=1 lead to a degradation of unweighted Signal-to-Noise-Ratio from 77,8 to 76,7dB for sensivity of 1.33Vpp.
It is astonishing, that lower line sensivity does only show better SNR, if the buffer with higher gain=4 is used.
These numbers are quiet nice in comparison to top analogue UKW stereo receivers, which could achieve about 70dB for stereo. http://www.hifi-forum.de/viewthread-84-783.html
Wrapping the whole setup with aluminium foil for shielding did not improve SNR.
Datasheet of SGTL5000 https://www.nxp.com/docs/en/data-sheet/SGTL5000.pdf says, that SNR of 85dB can be achieved.
As far as I understand SNR=76,7dB corresponds to "Effective Number of Bits" ENOB =13.2 bits


Has somebody done similar investigations about SNR of the audio shield?
 
Last edited:
You must log in or register to reply here.
Back
Top