[AudioI] ground loop on octocard input

Fri Feb 4 21:53:41 UTC 2022

Please send through the examples, hardware, software or otherwise, it is very interesting. Feel free to discuss anything hearing, acoustics and audio here. It is possible Omar may want to source an off the shelf solution.

What sort of amplification is required for the seismological instrument. What sort of bandwidths ?

Omar, that tone is heavily saturated, I imagine that the seismic digitizers which Jonathan is talking about have input filters on them if they are that low frequency. Is it possible that this low frequency tone is entering the system before your preamplifier and getting amplified before being captured ?

If you have the time and inclination, and you want to test further on the ground loop concept, the Pro soundcard has an isolated power supply built in. It would be interesting to see if it changes things at all. It also has balanced +-15 to 20V rails for powering pre-amplifiers on it - so it would be interesting to see if having the isolated power supply changes anything. You can see the connectors of the pro soundcard here :https://github.com/Audio-Injector/AudioInjector-Analogue-Breakout-Standard/blob/master/AudioInjectorAnaBreakStdv1.pdf

You could build your own preamplifier suited to the task. If you were inclined to learn more about that side of things, you could also modify an existing RCA preamplifier to change their gain and bandwidth to suit your needs.

Matt

On 5/2/22 07:41, JONATHAN RIZZO wrote:

Hi Omar,

Using a soundcard for a geophone isn’t really the right application, assuming you are interested in seismic signals. 

A seismic signal digitizer is a better option. I can give you some examples if you need them. They usually sample at a much lower rate so 60Hz noise isn’t much of an issue. 

Jonathan

 On Feb 4, 2022, at 3:19 PM, Omar Marcillo via People <people at lists.audioinjector.net> wrote:

 Matt, 

 Quick additional info for the ground loop issue:

 this is a screenshot  of a geophone connected to the amplification board. The almost square signal is 60 Hz. 

 IN this case I am using the 5V from the pi to power the board, I get the same noise if I used an external 9V battery to power the amplification board.

 Omar

Sent from Outlook <http://aka.ms/weboutlook> 

--------------------------------

From: Matt Flax <matt at audioinjector.net <mailto:matt at audioinjector.net> >
 Sent: Sunday, January 30, 2022 9:09 PM
 To: Omar Marcillo <omar_marcillo1 at hotmail.com>; people at lists.audioinjector.net <mailto:people at lists.audioinjector.net> <people at lists.audioinjector.net <mailto:people at lists.audioinjector.net> >
 Subject: Re: [AudioI] ground loop on octocard input 

If that is possible, it would be interesting to see if removing the switching supply kills the noise.

Another idea is to look a little more into the nature of the noise. If for example you use a tool like jaaa to visualise the frequency spectrum of the input lines, you may see the noise there. You could confirm the nature and frequency shape of the noise - this may give further information on diagnosing.

Other ideas are to put an isolated supply either in your preamplifier circuit or as the wall adapter. That would certainly break any ground loops.

For the preamplifier circuit, a nice low power 5v isolator is this chip :

https://www.analog.com/en/products/adum5000.html

You can put two on a circuit board and connect their outputs to create isolated +-5v rails. I would be happy to create you a suitable schematic if you would do the layout.

Matt

On 31/1/22 12:14, Omar Marcillo wrote:

 HI Matt,

 No, the 2V is not necessary I actually added that so the output is only between 0 and 4V, I tried also with no bias but I still see the same effect. 

 The 2v is generated by the board, with a voltage follower.

 The board uses a 7660 to invert the input power, which is a switched capacitor voltage converter.

 Maybe I could try with a single supply amplification board.

 Thank you.

 Omar 

Sent from Outlook <http://aka.ms/weboutlook> 

--------------------------------

From: Matt Flax <matt at audioinjector.net>
 Sent: Sunday, January 30, 2022 4:03 PM
 To: Omar Marcillo <omar_marcillo1 at hotmail.com>; people at lists.audioinjector.net <people at lists.audioinjector.net>
 Subject: Re: [AudioI] ground loop on octocard input 

Hi there,

OK, lets check the situation;

You have an octo running on a USB supply. You have a preamplifier running from a 9v external battery. The amplification board generates a negative rail line using a switching chip.

When everything is using batteries, there is no low frequency noise. When the Pi is plugged into the wall you get a low frequency noise, probably ground loop hum.

About the DC bias of 2v, is this necessary ? It is used to bias the Geophone ? how are you generating the 2v supply ?

Can you see the IC used on the switching supply to generate the negative rail ? Can you read it's model number ? Can you tell whether it is a charge pump device or a different type of switching supply ?

An initial gut feeling says that if it is a ground loop, then it is most likely an EM generated ground loop through the Geophone in some way. This is because you have only one connection to the mains supply (the USB charger).

It is possible that this noise is not due to a ground loop. It is possible that the noise is due to the method used to generate the negative rail.

Matt

On 29/1/22 07:23, Omar Marcillo via People wrote:

 Hi, 

 I am trying to connect a geophone (a seismic passive sensor ) to the octocard input. My geophone is a passive sensor with two terminals with variable voltage, with very low voltage signal (+/- 10mV) so I added an instrumentation amplifier  (AD620 board) that is powered with an 9V external battery. If the pi is connected to AC power I get a 60 Hz signal that masks everythign (I think it is ground loop somewhere). If I use a power source that is not connected to AC (a small portable charger) the signals looks fine.

 I tried adding a small DC to the output of the instrumentation amplifier board but I still have the same ground loop problem.

 about the amplification board, it takes 5V voltage and creates -5V to power the circuit, the board has -/+ inputs and varaible gain,  I connect one terminal of the geophone to + input  and the other to GND. I add 2V offset and  feed the ground and biased output to the octocard.

 Any suggestion or ideas to fix the ground loop will be appreciated.     

 Omar

-- 
Checkout the community email list :
https://lists.audioinjector.net/mailman/listinfo/people

-- 
Checkout the community email list :
https://lists.audioinjector.net/mailman/listinfo/people

<test_AUDACITY1.png>

 -- 
 People mailing list
 People at lists.audioinjector.net <mailto:People at lists.audioinjector.net> 
 https://lists.audioinjector.net/mailman/listinfo/people

-- 
Checkout the community email list :
https://lists.audioinjector.net/mailman/listinfo/people

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.audioinjector.net/pipermail/people/attachments/20220204/fa94a7a3/attachment-0001.htm>