If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

# Preamplifier for Geophone (Vibration Sensor)

Other Parts Discussed in Thread: LM358, OPA376

Hi,

I have a Geophone which is giving min output 10mV and max 1V after observing on DSO.

Before giving it to an ADC I want to amplify the sensor output so that max output remains

under 5v and has enough gain and noise free output to an ADC. I'm not able to understand

how to start building the circuit. May anyone give me some starting tips on same?

Will using LM358 work?

• Hello Chetan,

It sounds like you need a system with a gain of 5V/V and a rail-to-rail output.    Without knowing more about your system, here is a very basic circuit that performs the operation you're trying to achieve.

The circuit uses the OPA376 in a standard non-inverting AC amplification circuit commonly employed for microphones, geophones, and other small signal AC sensors.  You will need to adjust the geophone bias resistor value ( R1) to match the specifications of your model.

Also depending on the required bandwidth the low-pass and high-pass filters can be modified.  The three low-pass filter R+C combinations are below, and the -3dB frequency can be calculated with f = (1/(2*pi*R*C))

R1+C3
R4+C6
R3+C1

Since the circuit is in a gain of only 5V/V the R4+C6 filter is not very effective so the dominate filtering will come from R3+C1 and R1+C3.

The high-pass filter R+C combinations are below:

(R5||R6)+C4
R2+C5

Here is the example circuit and AC results:

Have fun!

Regards,
Collin Wells
Precision Linear Applications

• Hi Collin,

However the Geophone sensor I've is closed one, it means no way I can adjust bias resistor of Geophone.

Also it's having 5 PIN output : 3 are X,Y and Z triaxial , 4th one is GND and 5th Common.

Why consider AC analysis?

As per my knowledge output of any sensor should be a DC value.

• Hi Chetan,

My recommendation for the bias resistor was only if your model required one.

I was under the impression you were interested in the changing AC waveforms associated with the seismic activity.  Do you not care about the AC content?  I don't think you're really interested in the DC value alone unless I don't fully understand your application.  Do you know how fast the input signal in your application can change?

Regards,
Collin Wells
Precision Linear

• Hi Collin,

My application target for now is very simple : To fetch Geophone's sensor output and give it to ADC on
the development board I'm having. After going through some details I understood it will be good if I use a
preamp before giving it to ADC of the development board. Also same was confirmed via DSO. As this is
the first time I'm working on Geophone (http://www.instantel.com/pdf/720B0002.pdf only consider circular shaped
geophone not the kit) so I thought observing DC value is good.

Minimum DC I was getting around 10 mV that too after keeping Geophone stable.
Maximum DC I got was 1v that when I was shacking Geophone as much possible.
Frequency was varying  from Khz to Mhz.

Please do correct me if I've gone wrong and what should be my possible step.

• Chetan;

We may have a problem with semantics-- Colin was talking about a geophone, which is used to measure ground vibration-- an AC output. However, your description of a triaxial sensor sounds more like an inertial sensor. Its output would be DC voltages in a static condition and AC + DC outputs under dynamic conditions.

To make any circuit recommendation it is necessary to also know the impedance of your transducer and the bandwidth that you need-- as well as the power supply voltage(s) that may be available.

• Hello Neil,

The Geophone show here is what I have : http://www.instantel.com/pdf/720B0002.pdf
However I don't have exact datasheet for Triaxial Geophone.

As per my understanding for Geophone I thought getting preamp to generate max 5v output
will be fine as, as said earlier I'll give that output to development board so safer voltage should
be 5 volt.

Impedance and bandwidth I can't comment.
Regarding power supply : 12 volt DC as I'm planning to have more circuitry once this phase is completed.

• **** Note: I saw Neil's post just now before I was about to post this.  Chetan please clarify what your final application is.  From the datasheet for the sensor it looks like it would be seismography but if it is not this could be why I'm having some trouble understanding some of your application requirements.  ****

Hi Chetan,

Thanks for the additional information.  After reading the datasheet this type of sensor does appear to be a seimography related geophone and you will need an AC amplifier for this type of application.  If you are only concerned with the peak or max value then follow the amplification stage with a standard Peak-Detector or Envelope Detector circuit.  The sensor is listed for a flat-response over the frequency range of 2Hz to 250Hz.  I modfied the original circuit I sent for this lower bandwidth.  Again, this is a very basic circuit and additional filtering or conditioning may be required.

http://e2e.ti.com/cfs-file.ashx/__key/CommunityServer-Discussions-Components-Files/14/1643.OPA376-G-_3D00_-5.TSC

Regards,
Collin Wells
Precision Linear

• Hi Collin,

Yes it's on seismography.

With X ,Y and Z as output for the sensor also means AC output?

I'm confused what to do.

• Chetan;

From the looks of the data sheet, it is a complete system; no preamplifier is necessary.

• Hi Neil,

I think I should clarify it again. I only have the Geophone that is shown in the datasheet.

Do you mean to say that whatever is the output of Geophone that I can interface directly

to the development board without any peamp?

• Hi Chetan,

I think you may need to do some additional research into this field to better understand what types of signals that you're trying to acquire.  In the mean time I would suggest trying the circuit I provided and connecting the output to your oscilloscope to view the results to asure they're appropriate and then connect it to your measurement board.  If your measurement board has a buffer then you may also be able to connect the sensor directly to the measurement board because the signals are quite large already.

Regards,
Collin Wells

• Hello Collin,

I think your suggestion is correct. Parallel to my more research I should get the circuit tried with the
sensor.

Though one query :
Where will the Geophone output will go in the circuit and from where amplified output comes?

Thanks.

• Hello Chetan,

The geophone output is the "VG1" source in my circuit.  You may remove the 2.2k pull-up resistor because your sensor appears to be internally biased.  The output of the circuit is marked by the "Amp Stage 1" marker.

Regards,
Collin Wells
Precision Linear

• Hi Collin,

I was not able to get OPA376 as it was out of stock so till that I thought I will implement
a circuit using LM358. So here is the circuit :

And I observed the output on DSO with Geophone attached.

Though I was not able to get perfect sine output. All I got was noise and spikes when I was shaking the geophone
with full swing and that gave me maximum output.

• Hello Chetan,

This circuit is quite different then what I've recommended and there are likely a number of reasons why it is not working for you.  What are your final goals?  Are you trying to simply detect a seismic (or other) activity and make note of it, or do you intend to try to capture the magnitude of the waveform to observe it?

To start, the gain of this circuit is 3300000/6700 = 492V/V.  We determined earlier that you only require a gain of 5V/V.  With a gain of 492V/V any input signal over 10.16mV will create an output voltage over 5V.  This is why your circuit is always displaying the maximum output, you've saturated the output with even the smallest input signals.

Also, by biasing your input to GND your are essentially cutting off the lower half of your input signal because the common-mode input range of this LM358 rail-to-rail input amplifier only includes GND, not below it.  Therefore, I would suggest either using a dual supply operational amplifier or biasing the input to mid-supply to use the optimal part of the common-mode input range of your amplifier.  This would require you to change your peak detector to be biased mid-supply as well instead of GND.

Keep in mind that you will not see a sine wave at the "Output to DSO/ADC" node on your schematic.  This output will display the peak output of the circuit because it is a peak-detector (envelope detector).  To view the amplified input signal please look at the output of the amplifier which is at pin 1.

I wish you the best in your efforts, but if you do not choose to follow the advice and/or circuits I provide then I'm not sure there is a reason to continue this thread.

Regards,
Collin Wells

• Hi Collin,

I really want to take your help. The IC you recommended was out of stock so instead I did this.
Nothing else.

My target : I want to capture the magnitude of the waveform to observer it and in later stage I want to store it in a removable media
using a micro-controller which is a very broad vision. For now I'm targeting to get at least good waveform out of the geophone I'm having.

I will be using a 12V power supply and also gain for now I'm targeting is 5 V as said.

See, I am a newbie to operational amplifiers so I was wondering if I can built an preamp from scratch on own then it will help me.
Right from the calculation till the breadboard. Any help here?

Also, there are so much confusion regarding what type of output my geophone gives.

As soon as I get hold of OPA376 I will implement it.

I really appreciate your efforts and need them.

• Hi Chetan,

The part you've selected, the LM358 is not a bad choice, I was more concerned with the component values and circuit topology you choose to use opposed to what was suggested.  Without modifying your current setup, try lowering the 3.3M resistor to a value like 26.8k to achieve a gain of 5V/V.  The gain equation for a non-inverting amplifier is:  Gain = 1+Rf/Rg where Rf is the feedback resistor and Rg is the gain setting (input) resistor.  So to achieve a gain of 5V/V, you will need to select an Rf value that is 4X the Rg value.  So if you modify the 3.3M resistor to 26.8k the equation solves for 5V/V.   Gain  = 1 + (26800 / 6700) = 5V/V.

Regards,
Collin Wells

• Hi Collin,

I did as you suggested and I think now I'm getting some valid output.
Rg = 28 Kohm (As I was having 18 and 10 kohm)

I wasn't able to save DSO output so I took snap :

1) This is the output on DSO when Geophone was in stable mode with preamp connected :

2) This is the output on DSO when I was shaking Geophone with preamp connected :

• Hi Collin,

I'm building a preamplifier like your example for my geophone, which have a natural frequency of 10 Hz.

Looking for one example, I found yours here, but I have two questions about it, could you tell me about them, please?

- Which are your criteria for choose these filters?

I mean, you have three low pass filters and two high pass filters, although one high pass filter isn't effective because of the low gain. Why are you putting on two filters for each band?

- Why are you choosing those values (R's & C's)

I know those values are for -3dB frequency, but wouldn't it better to choose any value to be more closer to the bounds of the bandwith we want?

* My application is similar to this example, a maximun gain of 5 V/V, with a frequency range from 10Hz to 240 Hz.

** My two operational amplifiers for testing are: LM358AP & OPA604AP.