Guitar preamp op amp tl072 input noise

[Zero999]

Why would non-inverting be less noisy?  I never heard of that before.

The only reason I can think of is it's possible to design a non-inverting amplifier, with a high input impedance, yet still have low value resistors, giving lower noise, that what you'd get with an inverting amplifier, with the same gain and input impedance.

[Ale]

I made some changes, is it possible to remove C5, R6 and R2?

[Richard Crowley]

As drawn in the latest diagram, C5 will effectively short the audio signal to ground.  It seems unlikely that is the desired effect.  Perhaps you meant C5 to be in series with the audio path instead of in parallel?  As a DC-blocking, AC-coupling capacitor.

R2/R6 form a voltage divider to set the operating point of U1D at VCC/2 (i.e. +12V)

Circuit design by committee.   

[Ale]

For semplicity I inverted the second stage. Now it should be ok. What do you think?

thanks

[Audioguru]

It looks like you do not know the simple formula for calculating the value of a capacitor in your inverting opamp circuit: 1 divided by (2 x pi x R x C).
1) C1 is so small that it cuts all frequencies below 73kHz. Therefore you will have no audio, just ultrasonics from bats.
2) C3 is so large that it cuts all high frequencies above 73Hz. Therefore if C1 would pass it you will have only deep bass and earthquake sounds.
3) R3 is so small that it shorts the volume control.

C15 can be 100nF or 220nF.

[Ale]

eheheeh you are right, I'm learning and I missed that formula 
Anyway, I made some maths to cut under 80hz and over 6khz. I hope it is correct... I have doubts on C3.
We are almost there I think 

[Audioguru]

Your preamp is still cutting low and high frequencies to sound like an old AM radio. Each RC causes its calculated frequency to be reduced -3dB which is half the power.
For the high frequencies you have two RC cuts which total -6dB which is 1/4th the power at 7.3kHz. The cuts are gradual so even 2.5kHz will be noticed to be cut a little. Most of us can notice a cut of -2dB at 20kHz. Why are you cutting the high frequencies?
For the low frequencies you have three RC cuts at 80Hz which will be at -9dB which is 1/8th the power. 230Hz will be noticed to be cut a little.

The value of R3 is still so small that it is still shorting the volume control. If you reduce the resistance of the volume control to 5k then the C15 capacitor feeding it must be huge to pass low frequencies.

[Ale]

Ok, so... since I just have on the first stage the hi-pass (70Hz) and low-pass (6kHz) I think there's no need to have another on the second stage, right? I removed the low-pass on the second stage.
If I understood well RV1 can cause problems when it goes down to gnd, that's why C1, but also need R3 to avoid shorting, right?

Is there a simple solution to this?

Thanks for your patience.

[Audioguru]

The voltage gain of the second opamp is R5/R3 which is 100k/10k= 10 times. The 100k volume control is shorted by the low value for R3 so make R3 100k and make R5 1M. Then the cutoff frequency of C1 and the new 100k for R3 is too low at 8Hz so make C1 100nF then the cutoff frequency is 16Hz and will not affect the 73Hz cutoff frequency of C11 and R16.

Why are you cutting frequencies above 6kHz? We can hear harmonics as high as 20kHz. Without hearing the harmonics then the sound is muffled like an old AM radio. But a guitar speaker probably does not produce frequencies above 6kHz anyway. 

[Ale]

Well, the maximum guitar frequencies are from 80 to 1200 Hz more or less, but as you say there are still lots of harmonics. Anyway over the 6kHz they are very low in volume and very poor, since I'm building this for a live situation and not for studio recording I think is acceptable to cut freqs to minimize the noise. Further more, I'm usually play rock music with distortions and sometimes chorus, so I will receive additional harmonics from them.
Of course I should make some field test, in the worst case I will get rid of the filter in the 2.0 version 

[Zero999]

Well, the maximum guitar frequencies are from 80 to 1200 Hz more or less, but as you say there are still lots of harmonics. Anyway over the 6kHz they are very low in volume and very poor, since I'm building this for a live situation and not for studio recording I think is acceptable to cut freqs to minimize the noise. Further more, I'm usually play rock music with distortions and sometimes chorus, so I will receive additional harmonics from them.
Of course I should make some field test, in the worst case I will get rid of the filter in the 2.0 version 

That looks much better, than the other designs. I'd say go with it.

[Audioguru]

Most amplifiers for electric guitars made by Marshall, Fender, Gibson and etc. had high input impedance of at least 1M ohms so that the low impedance electric pickup could resonate and produce a fairly high level peak at about 5kHz. They did not use an amplifier with a low input impedance that would produce a flat frequency response and they did not cut the highs. They liked the peak that helped the poor high frequency response of guitar speakers.   

[Zero999]

Most amplifiers for electric guitars made by Marshall, Fender, Gibson and etc. had high input impedance of at least 1M ohms so that the low impedance electric pickup could resonate and produce a fairly high level peak at about 5kHz. They did not use an amplifier with a low input impedance that would produce a flat frequency response and they did not cut the highs. They liked the peak that helped the poor high frequency response of guitar speakers.   

Well that's what he's got there, in the latest schematic: an input impedance of 1M. To minimise noise, it has an upper roll-off of 7.2kHz which is easy to change, if it's an issue.

[derationalize]

I have been breadboarding something similar so I have been following this thread. I have a few TL072s and wanted to make an internal preamp inside the guitar to boost the signal level. Having the opamp run in inverted mode screws up the phase relationship. This is important because I'm planning on using one channel of the TL072 to amplify the neck pickup and the other to amplify the bridge pickup separately to be able to balance the volume/gain. Since this is running off of 9v batteries I would think limiting it to a single TL072 would keep power consumption low, though there is enough space available for 2 9v batteries. If I'm using 2 batteries though, wouldn't it be smarter to just wire them up for 18v for the increased headroom? Sorry for hijacking the thread.

[Ale]

Wouldn't be easier to use 2 simple 2n3904, also for the space problem?

[Zero999]

I have been breadboarding something similar so I have been following this thread. I have a few TL072s and wanted to make an internal preamp inside the guitar to boost the signal level. Having the opamp run in inverted mode screws up the phase relationship. This is important because I'm planning on using one channel of the TL072 to amplify the neck pickup and the other to amplify the bridge pickup separately to be able to balance the volume/gain. Since this is running off of 9v batteries I would think limiting it to a single TL072 would keep power consumption low, though there is enough space available for 2 9v batteries. If I'm using 2 batteries though, wouldn't it be smarter to just wire them up for 18v for the increased headroom? Sorry for hijacking the thread.

How does an inverting op-amp screw the phase relationship? As long both channels are inverted, it won't matter.

The guitar will sound the same through an inverting amplifier, as it will through a non-inverting amplifier.

Wouldn't be easier to use 2 simple 2n3904, also for the space problem?

I agree, a discrete solution will do. I'd probably suggest J-FETs though because the input impedance can be higher, than BJTs.

[Ale]

Maybe a 2n3819 with a schema like this?

[Zero999]

Yes, you could even have the source resistor and AC coupling capacitor, at the other end of the cable, so a two wire connection can be used.

[Ale]

You mean like this? This is basicaly a DI Box, I'm working on this lately...

I also build a simple soundcard oscilloscope sunday and tested out this schematic because I also need a DI. I still have question on that, but maybe is better to open another thread, right?

[HoracioDos]

JFET PreAmp examples with calcs.



http://www.rason.org/Projects/jfetamp/jfetamp.htm
PD: I was going to move this post to https://www.eevblog.com/forum/beginners/single-mosfet-power-amplifier/ but I realized it was about mosfet not jfet.

[Zero999]

You mean like this? This is basicaly a DI Box, I'm working on this lately...

I also build a simple soundcard oscilloscope sunday and tested out this schematic because I also need a DI. I still have question on that, but maybe is better to open another thread, right?

I meant using only two wires: one for 0V and the other for both power and signal.

[Ale]

Like in the phantom power?

[Audioguru]

The neck and bridge of a guitar are some wavelengths of sounds apart so their phases are very different at different frequencies. I doubt you can hear if one signal has its phase reversed and if both signals have their phases reversed then they will have the same phase.
Instead of amplifying each pickup separately, why not use a single amplifier with a simple balance control.

A magnetic guitar pickup produces a very high output level. Then an opamp or two Jfets are not needed. The preamp input impedance must be high to allow the pickup to resonate and produce a level boost at about 5kHz so ordinary low input impedance transistors cannot be used. Use a single Jfet like this attachment:

[Zero999]

Like in the phantom power?

I meant like audioguru's circuit, but with the JFET, R1 and R2 separated from the rest of the circuit with a cable, consisting of two cores.

[Audioguru]

You can separate the Jfet from the parts on the right side if they are connected together with shielded (coaxial) audio cable.
The value of the 6.8k resistor depends on the spec's of the Jfet.