Spin Semiconductor

Hi,
I designed this circuit to use the FV-1 in a guitar stompbox. It's pretty close to the datasheet's circuit, with added buffers, dry level and wet level.
http://www.wuala.com/ricothetroll/publi ... atic2.png/
http://www.wuala.com/ricothetroll/publi ... oard2.png/
To have a better quality image, please clic on "Télécharger 'FV-1_Schematic2.png' sur votre ordinateur" (might be in english "download...")
My problem is : the output of the FV-1 is very noisy ! The noise only appears on the "wet" signal ("wet" pot, before IC3A) and not on the "dry" one, so I'm quite sure it comes from the FV-1.
Is it a problem with my layout ? What can I do to make the FV-1 silent ?
Best regards.
Eric

It is difficult to say what the problem may be. A few things:

Can you post a better quality schematic? It is difficult to read, it seems to have lost a lot of detail.

What is the program you are running? Code could be an issue.

The right input does not seem to be driven, if it is not used then tie it to the left input.

Hi Frank,
Thanx a lot for answering me.
I have put the schematic + board on imageshack, maybe you'll be more lucky with it then :

http://img847.imageshack.us/img847/8987 ... matic2.png
http://img535.imageshack.us/img535/4051/fv1board2.png

I'm using the factory settings. You're right about the R channel, I left it floating to eventually be able to upgrade the circuit to stereo. I'll try to connect both inputs together, see if the noise decreases then. Should I connect both outputs too ? Wouldn't there be phase issues ?

The remaining unrouted wire is because I have two wires coming from the DC In connector, as I couldn't route it on the PCB.

Best regards.

Eric

Hi Eric,

ricothetroll wrote:Hi Frank,
Thanx a lot for answering me.
I have put the schematic + board on imageshack, maybe you'll be more lucky with it then :

http://img847.imageshack.us/img847/8987 ... matic2.png
http://img535.imageshack.us/img535/4051/fv1board2.png

Much better, the board was OK on the first set but the schematic was a problem, these are both clean.

ricothetroll wrote: I'm using the factory settings. You're right about the R channel, I left it floating to eventually be able to upgrade the circuit to stereo. I'll try to connect both inputs together, see if the noise decreases then. Should I connect both outputs too ? Wouldn't there be phase issues ?

That may be part of the problem, the factory programs sometimes add the left and right input so if right is not being driven it may be picking up noise and adding it in. Tie the right input to left if not being used.

Do not connect both outputs together, that is not necessary.

Hi,
I linked both inputs. I feel like the S/N is much better now ! By the way the FV1 is still noisier than my TL072 in/out.
Thanx a lot for your support !
Best regards.
Eric

Drive the inputs of the FV-1 as hot as possible (3.0Vp-p max ) and see if that helps. I've had a instances where people were only hitting the FV-1 with 1Vp-p and ADCs like to be run closer to their full range.

Hi all! I compiled this chart, and noticed that the sound after the FV-1 is very noisy. http://www.tonepad.com/getFile.asp?id=122 Input of the circuit to ground, and I took it for oscilloscope, at the entrance of FV-1 no noise at the output noise is. What ideas and advice? Thanks in advance!

What does the spectrum of the noise look like? The DAC is a delta-sigma type so noise is modulated out of band, look at the output after the low pass filter and see what the noise looks like.

White noise. After the filter is about 10 mV. I tried using a 15pF capacitor to harness quartz, tried without it. Noise is. Greetings from the sunny Crimea!

Are you using a battery for power or a power supply?

Is there any noise on the power rails?

Are you using a PCB or a breadboard or a vero board?

Picture?

frank wrote:Hi Eric,

That may be part of the problem, the factory programs sometimes add the left and right input so if right is not being driven it may be picking up noise and adding it in. Tie the right input to left if not being used.

Do not connect both outputs together, that is not necessary.

Frank, to minimize noise, is it really necessary to tie the right input to the left input even if ADCR register is never used inside the program?
What about tying it to the ground?

If the ADCR is not used then you can leave it floating, just make sure the code never reads it. Lots of the demo code reads both inputs. Do not tie it to ground, if you want to tie it off tie it to a capacitor to ground, this will let the ADC bias the pin to mid.