Post History
This is an expansion on Biasing a differential opamp input post. Olin pointed out that gotchas can be in the implementation, so I will provide the circuit and design specifications as well as my re...
#1: Initial revision
by
Elleanor Lopez
·
2025-06-10T10:01:55Z (3 days ago)
Geophone amplification circuit behavior
This is an expansion on [Biasing a differential opamp input](https://electrical.codidact.com/posts/294083) post. Olin pointed out that gotchas can be in the implementation, so I will provide the circuit and design specifications as well as my research so far. The geophone is [HG-6](https://hgsindia.com/www.hgsproducts.nl/Pdf/210HG-6%20V1.1.pdf) B coil, shunt for 0.7 damping is specified to be 9090 ohm. The opamp I had is [INA188](https://www.ti.com/lit/ds/symlink/ina188.pdf?ts=1749469015824). The goal was to have 0V to 5V single ended output that goes to ADC. Therefore, the plan was to reference the signal to 2.5V so that the common mode input requirements of INA188 would be satisfied. Reading application note of TI [Importance of Input Bias Current Return Paths in Instrumentation Amplifier Applications](https://www.ti.com/lit/an/sboa503/sboa503.pdf?ts=1749544236254#:~:text=One%20of%20the%20common%20application,circuit%20does%20not%20function%20properly.) chapter 4 mentions that the reference and bias voltage has to be able to both sink and source current, so for that I selected [MAX6225](https://www.analog.com/media/en/technical-documentation/data-sheets/max6225-max6250.pdf) voltage reference, capable of sinking/sourcing 15mA, albeit drawing current from it reduces stability quite rapidly. Perhaps it would have been wise to buffer it. The simulation itself:   It does what I expected it to in simulation. Yes, the circuit in the previous post is slightly different, but since we are getting down to details, this is what is actually there. V4 is a charge pump producing -0.23V to achieve true 0V on the output of the INA188. # The actual circuit   The only added extra things are [PESD5V0U1BB,115](https://assets.nexperia.com/documents/data-sheet/PESD5V0U1BB.pdf) TVS diodes with 5V working voltage and R33,R42 to protect the opamp in case the geophone is dropped while handled, since it can produce high voltage spikes. # Testing Before connecting a geophone to the circuit I wanted to verify it by applying a differential signal in a controlled manner first. By what I saw on the scope I feel like this did not work as I had imagined in my mind, there may be a fundamental misunderstanding on my part, so give me some rope here. I used [Keithley 3390]((https://www.tek.com/en/documents/specification/model-3390-arbitrary-waveform-generator-specifications)) function generator. I measured the BNC housing against PE, and it was not connected, so as far as I could figure, if I connected it to GEO+ and GEO-, the circuit would see it as a differential signal. I applied a sine wave to it, left side of the image are the GEO+ and GEO- in reference to circuit ground, and right side is a subtraction of the channels:  So despite the massive noise on input, it seemed there is a voltage produced on the input. However, the output of INA188 was just 2.5V - the reference around which it was centered. I am not sure why this was the case or how it is possible. What I decided to do next was to connect the actual geophone to the circuit, since I felt that there was something off about how I connected the function generator to the circuit. When shaking the geophone, there was output produced, but only from 2.5V down to 0V:  So far I have no concrete thoughts on why this does not work, I guessed perhaps the 2.5V reference has too low impedance in relation to the geophone input, so I thought perhaps something like this would be more suitable:  Simulation results are the same, and the coil resistance specified is 375Ω, so I don't think that is it. Seeing that the signal never goes towards 5V, I have a thought now that perhaps one of the TVS diodes has a short circuit, but I have not checked that yet (Will update post after I check it). The questions are: 1. Is it reasonable to use function generator as differential input the way I tried? 2. If 2.5V is the only reference for the signal, is that sufficient or does there need to be a direct GND reference somewhere? 3. What may be the reasons you can see or think of that the circuit does not behave as expected? I have provided all the information I can think of at the moment, if you need anything extra, let me know.