Guitar Buffer / D.I.Box

>>1170443
Maybe you're exceeding the input voltage range of your ADC. Try adding a pot between the amplifier and your ADC.

>>1170467
A 500k pot would be enough?

Also, could I put a pot right after the 9V power supply?

File: pic_26_2.jpg (282KB, 800x480px) Image search: [Google]

282KB, 800x480px

I have tested it with a oscilloscope. This is the result:

In the last image I see the amplitude is just the same in the input and in the output. So why my circuit works on an amplifier but it adds distortion when I connect it to the FPGA line-in input?

>>1170493
A better solution is to replace R1 and R2 with a single potentiometer. This will adjust the amplitude going in to (and out of) your buffer. 10kohm to 10kohm should be enough.

>>1170623
But the amplitude is the same. Or at least it is what the oscilloscope shows.
Which is the explanation then to have a clean sound on an amplifier and a dirty one in the fpga?

File: 1405269130821.jpg (28KB, 429x399px) Image search: [Google]

28KB, 429x399px

>>1170711
If the amplitude really is ok, then let's look at the fpga.

What's the input impedance of the fpga? How many bits is the ADC? Does the ADC have enough resolution to accurately sample the peak to peak input voltage? Do you have another component, like a cap for example, in front of the input to the fpga that could be limiting the sound? Do you have sufficient bypass capacitors on the power pins of the fpga? Are you sure that you have the audio output circuitry correct on the fpga? Do you have additional circuitry after the fpga and before the headphones that may be distorting the audio?

>>1170758
In home I will make some more test and try to fill the rest of information, but...
>What's the input impedance of the fpga?
I guess it is 600Ohm, it use to be the line-in default value, isn't it?
The FPGA model is Terasic DE1-SoC.
> How many bits is the ADC?
It is 24-bits@96kHz, but I'm using 16-bits@48kHz.
>Does the ADC have enough resolution to accurately sample the peak to peak input voltage?
Yes, it has.
> Do you have another component...
No, it is just the guitar+buffer+FPGA+headphones.
>Do you have sufficient bypass capacitors on the power pins of the fpga?
I don't know what you mean, I am using the FPGA Codec and its audio input. Anyway, I will put the same circuit on the computer and check it to comprobate the FPGA is not damaged.
>Do you have additional circuitry after the fpga and before the headphones that may be distorting the audio?
No, there is nothing in between

This is how it sounds on my PC. It's the same as in the FPGA.

Single notes:
http://vocaroo.com/i/s0KywfRA7xqO
Chords:
http://vocaroo.com/i/s0ZVWKeSQ9EP

>>1171521
Is the delay effect something you process with fpga or is it before?

>>1171573
There is not any delay.
I know in the first audio it seems there are multiple notes, but there are three notes played one by one. The distortion it creates also modifies the frequency, but I don't know why.

The audio is recorded on the PC with the same setup as the FPGA. I connected the guitar to the buffer and the buffer to the PC. I am also using the motherboard line-in input. The sound is just the same as in the FPGA.

When I connect the buffer to the guitar amplifier (power amplifier), it works perfect, totally clean. I don't know what it's happening.

Do you think I should give a try to the mic input? I guess I will try this later.

>>1171619
Ok.
Let's try to enhance a bit your buffer. First add two capacitor between +vcc and gnd. First one is a big electrochemical one about 220microF and second is a small plastic one about 220nF.
This allow you to filter you power supply because ac coming from C1 go through R1 and end up un +vcc.

Then you can add a resistor before Vin and ground. Try a res about 1M. It should adapt impedance and capture maximum of the signal coming from the guitar

>>1172062
Also could you provide a test with sin 1khz? If you don't have GBF try "signal generator" on fdroid repo on your phone.

In my opinion, there is not just one problem. I hear delay and distortion.

>>1171619
This sound sounds familiar to me. I couldn't quite place it at first, but eventually I recalled a situation where there was a "noise reduction" chip that I was using. The chip would cut off the audio signal if it was below some noise threshold. It doesn't explain the overdrive sound, but it explains the sharp cutoff when the signal gets to a low amplitude. There may be a similar noise reduction built in to both your computer sound card and the firmware on the fpga.

In all honesty, I've done roughly what you're trying to do, and I found it easier just to use the op amp as an inverting amplifier instead of a buffer. Most modern op amps have a plenty low enough output impedance, even in an inverting or non-inverting configuration. Plus, using it as an amplifier gives you the opportunity to adjust the gain to match whatever you are wiring it to.

Tl;dr there may be a noise reduction algorithm in the fpga that is contributing, but it may also have to do with op's choice to use an op amp as a buffer instead of an inverting amp. Check the data sheet.

>>1172064
Also, I totally agree with this anon. >>1172062

Put some decoupling caps on your power pins and an impedance matching resistor from your guitar input down to ground. Or just remove R1.

>>1172062
I have made this circuit on a simulator, but the output signal has an amplitude of picoVolts, I guess I have done something wrong. Can you check if this was what you mean?

>>1172063
I tested it before at 5 kHz in the oscilloscope and the sinusoidal wave had a little peak, but later it was corrected. It had a peak on the rising and falling edges.

>>1172064
If I don't get you wrong, are you suggesting building a linear booster and adjust it for an output amplitude of gain=1? Some years ago I built a MXR Microamp clone using a TL072. Would it fix my impedances problem? I need an impedance output below 600 Ohm.


If there is something I can build different to the circuit I showed here, I'll be glad to hear about it.

Thanks for all your replies, guys. You are helping me a lot.

File: impedanceMatchingMod.png (499KB, 556x306px) Image search: [Google]

499KB, 556x306px

>>1172078
Sorry, I forgot the picture.

>>1172070
R1 is necessary to get vcc/2 at the aop input

File: ma_man.png (11KB, 711x328px) Image search: [Google]

11KB, 711x328px

>>1172078
>>1172079

File: BLOCk_Pass.png (2KB, 252x200px) Image search: [Google]

2KB, 252x200px

>>1172078
>>1172079
This gives you nothing because AC can pass through a capa but DC can't.
So you blocked your power supply (DC) trough aop.

Simple way to remember is pic related.

Also, you link your wire between C1 and R3 to gnd so all current went to gnd not your aop input.
( U=RI and R is 0 so I is max)

File: NanananananananananAAAAAAAna.png (23KB, 1028x612px) Image search: [Google]

23KB, 1028x612px

>>1172100
I've also check the sound you posted earlier. I used audacity to get the waveform and I notice the difference between high and low amplitude. I think you should try another circuit that gives you amplification <1.

Do I guess right if I say that you are using a active guitare?

>>1172103
No, I am using a guitar with passive pickups.

>>1172093
Thank you very much for this diagram. I have built it and the sound is better, but still a little bit distorted.
Here is an audio sample:
http://vocaroo.com/i/s02CccMN9Kg5

File: Trump_fuck_with_Lepen.png (33KB, 1610x788px) Image search: [Google]

33KB, 1610x788px

>>1172206
>http://vocaroo.com/i/s02CccMN9Kg5

It sound alright I think. Maybe you can reduce a bit the input resistor to decrease the signal level if you are just doing it for fun or you can try this diagram. It 's a linear amplifier.

Relation between input and output is
>Vout= -Vin*Rvar/R2

So set R2 and change amplification with potentiometer.
If Rvar maximum value is the same that fixed R2 value you get amplification max =-1

Let me know if it worked

Hey! OP again.

I have changed the TL071 for a TL072 I had. Now it works far better. IMO it is getting a clean and rich tone. I will post some audios tomorrow.

Without a buffer the sound was poor in harmonics, it was a dull tone.

If you have some suggestion to improve the sound quality I am open to them.

Seriously, thank you very much for your support. I will also post the final diagram.

>>1172308
>>1172310
>Improve sound quality
I think this diagram will do the trick because the problems come from a too high level in the output of the buffer (amplification =1) so if you reduce this level it will be ok.

IDK the english term for it in english but in french it's "écrêtage". That mean cut the top of the peak of a signal. This gives you unwanted harmonic which are distortion.
>>1172103
You can see it on the left side of pic: maximum and minimum level of sinus squared in the bad way.

File: angorarabbit525255B325255D1.jpg (78KB, 790x617px) Image search: [Google]

78KB, 790x617px

>>1172318
Sorry only minimums.

>>1172318
En anglais, on dit "clipping"

>>1172392
Merci. Tu es d'où? J'habite le Haut de France.

>>1172310
OP again. I have another question.

Now I am happy with the sound I am getting with the TL072, but I realized about something.
The TL072 pin-out it's not the same as the TL071. In fact I don't know how it is working at all and why it is producing a better sound.

Anyone can explain this to me?

Also I realized that with the TL072 the circuit must be working in another way, it doesnt matter if there is a battery or not, so I think the OP-Amp is not working and are passive components which modifies the sound.

>>1172646

Did you follow the pinout on the datasheet?

Are you sure of your circuit? I mean everything go to the good place?

>>1172614
Eastern Canada. I don't speak much French, but I'm surrounded by people who do so I can make short sentences and read a bit.

>>1172650
I mean, I had the circuit with the TL071, as it distorted I introduced a TL072 I had from other projects. But I didn't checked the pin-out datasheet.

Both have differents pinouts, but it sounds better, I don't know how nor why.

File: Marine_casse_toi_de_mon_pays.jpg (52KB, 1093x342px) Image search: [Google]

52KB, 1093x342px

>>1172653
not the same pinout!

youre very retarded, OP. build it again and take care of pinout, jesus christ.

>>1173437
I know perfectly it is wrong, what I couldn't understand is why it was producing any sound at all.

Anyway, the project is finished and working. I changed the OPA for a 741, which has the same pinout. It works cleaner.

>>1173463
I think it worked because of "diaphonie", don't know the english for it.

>>1173463
im pretty fucking sure you have everything bridged and you are literally simply passing the signal through.