This thread has been locked.

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.

to lower the frequency range of a Geophone

Other Parts Discussed in Thread: OPA188, LM7705, TINA-TI, INA128, INA188

Hi, I have a 2Hz. natural frequency Geophone HS, which will give a signal of the vibration velocity of 1 volt/inch per second, and wish to measure the vibration velocity down1 Hz. I saw there is a German company has a similar solution but not complete the same. please  see the web page https://e2e.ti.com/support/amplifiers/precision_amplifiers/f/14/t/161153 .  Can some one show me how it can be done?

Darren

  • Hi, I forgot one important features of this application. the Geophone (HS-1 from GEospace) has a similar output vs. frquency curve to the one in example. The output drop drastically as the froquency get below the natural frequency, what I need is to have a flat output vs. frquency curve below 1 Hz. There is a high end model that has 1 hz. natural frequency but I wish to go below 0.5 Hz. even with the high end one.

    Darren
  • Hi Darren,

    We're not Geophone experts here, just amplifier experts, but I'll give it a try. What about using an amplifier with increasing gain at low frequencies such at the resulting transfer function of both the amplifier and the Geophone is flat:

    In that example I show a geophone with a corner frequency at 10Hz (-3dB) and a single-pole role off implemented by R1, C1, VG1, and VCVS1. The gain of the amplifier increases at low frequencies to counteract the decline in signal output from the geophone. I chose a low frequency corner of 0.1Hz (implemented by R5 and C2) and a low frequency gain of 40dB.

    This circuit only restores the original geophone output at frequencies below 10Hz, so additional gain may be required to raise the geophone output to the necessary signal levels at all frequencies.

    Implementing this in reality may be somewhat difficult. Capactior C2 is very large and so it may exhibit unexpected behavior due to dielectric absorption. The OPA188 was selected because of its excellent low frequency noise and thermal drift, which will be very important in this application. 

  • John:

    Thank you for the answer. Now further questions,

    1.  Can I use 1.7M to replace 11.5 k for R5 so to use a 1u for C2?  Where is the 40 db gain coming from? I know I may need to change other resistance value to maintain the gain value.

    2. the frequency range of the 40db increase, is that fixed to 100 times of the cutoff frequency, 0.1 Hz. in this case, or it can be changed.

    3 what is the best choice for Voltage-Controlled Voltage Sourced IC,  VCVS1, will LM7705 work?

    4. what voltage shows on VF1, the original from Geophone?

    Thanks

    Darren

  • Darren,

    1. The 40dB of gain is the DC gain of the amplifier, set by the equation (1+ R5/R4) . If you increase R5 to 1.7M you will need to increase R4 to about 17000kOhms. This will greatly increase the noise of the circuit which is why I recommended the values you saw. If you would like to learn more about noise in op amp circuits I highly recommend you view the TI Precision Labs series on noise, starting here: https://training.ti.com/ti-precision-labs-op-amps-noise-1?cu=14685 

    2. To design the amplifier, I simply saw what the attenuation of the Geophone would be at 0.1Hz (in the example 40dB) and set that as the DC gain of my amplifier. Then set the corner frequency of the amplifier to be 0.1Hz. Likewise you could chose to set the DC gain of the amplifier to 60dB with a corner frequency of .01Hz for a similar effect. 

    3. VCVS1 is just part of the geophone model in simulation. You do not need to buy an IC for this.

    4. Yes, VF1 just displays the geophone output.

    If you are new to designing circuits with op amps I really recommend watching the precision labs series of videos: http://www.ti.com/lsds/ti/amplifiers-linear/precision-amplifier-precision-labs.page?DCMP=tipl&HQS=hpa-pa-opamp-tipl-pr-lp-en 

  • Also,

    After examining Geospaces curves for the HS-1, I remembered that a Geophone should have a 2nd-order response, not a first order as I showed above. Consequently that circuit will not give a flat frequency response. The DC gain of the circuit must decrease at 40dB per decade to maintain a flat frequency response. I've attached a file with a new circuit that should give a flat frequency response for a 40dB/decade roll-off. Transfer Function Correction.TSC

  • John:

    I can not open the STC extension file, can you put the diagram in pdf or publish on the forum?

    Thanks

    Darren

  • Sorry I meant TSC extension.
  • Darren,

    .TSC is the extension for a TINA-TI simulation file. TINA-TI is a free circuit simulation program provided by TI, that you should consider downloading as you work on this project. It will help us share circuit ideas back and forth and can possibly reveal problems with the circuit before it is built. Here is an image of the circuit in my previous post:

  • Thank you, John.

    Darren

  • What is the preamplifier for?

  • John:

    Now I think I understand what was you intended to do. the right part is a Inverted shelving Low-Pass Amplifier that has a corner at 0.15 Hz. and 0 db frequency at 15Hz. However, the -60 db will continued from 15Hz and goes on. In order to correct that you have the none- inverted shelving high-pass filter that will has a 0db gain, because non-inverted; to the right of corner (19Hz.) there is a compensate gain of 20 db thru 195hz.

    If so, then I may need to point out that the high-pass shelving filter needs to be a differentiator instead of an integrator as show, sothe C4 should be placed between R7 and ground.

    Darren
  • Darren,

    That's correct, the slow of the shelving low-pass will continue on above 15 Hz. When I did this design I hoped that would be acceptable since I thought higher frequency seismic events are local, rather than teleseismic. The pre-amplifier is there only the reduce the gain requirements in the shelving low-pass filter and provide the proper termination impedance for the geophone. It is NOT intended to correct the transfer function, the feedback capacitor value is only selected to attenuate high frequency noise from this stage without affecting the overall transfer function.  If you need to preserve the high frequency information as well, then we need a bypass pathway around the lowpass filter for those high frequency signals. In the below schematic I inserted op amp U3 to act as a bypass pathway around the lowpass filter. The two outputs are summed through the 100 ohm resistors. 

    By matching the corner frequencies and flat transfer function over a wide bandwidth can be produced:

  • Thank You, John

  • John: I am looking for a zero drift instrument amplifier like INA 114 or INA128/129 that has a max voltage drift less than 50 micro V. does TI has anything like that?
  • Check out the INA188, it is a zero drift instrumentation amplifier.