I'm used to have ready-made microphones with XLR plugs which are easily connected to the input of my audio interface. Now, I'd like to buy a vibrometer (Knowles BU-21771-000) which is sold as an electronic component. I'm not so strong in electronics and I don't find a proper associated documentation to design the whole electronic circuit. What should I do to connect it to my audio interface (XLR or jack input)?

On the datasheet, it is said that it was tested as follows:

enter image description here

  1. Does it mean that I should buy a 5.6 kOhm resistor, a 1.5V power supply, and solder the outputs to the + and - of an XLR cable which will be connected to the sound interface?

  2. Should I add a connection between the vibrometer 0 (ground?) to the XLR cable ground? If yes, why don't they put one on the "test diagram"? If not, what is this connection for?

  3. How do you deal with impedance mismatch?


2 Answers 2


I put together a vibration sensor using a similar Knowles BU-23173-000 accelerometer (which has a built in pre-amp) fairly recently. There are 3 solder points on the sensor, so to solder together a circuit:

POS --> + battery, I used a 1.5V;

NEG --> negative/ground of audio connector (in my case, a BNC connector); and to negative battery terminal.

OUTPUT--> positive of audio connector (BNC).

It works fine, with low SNR (signal to noise ratio), when plugged into my data acquisition board. So, I would recommend trying the same for your sensor. Probably there are ways to improve the signal if you wanted to, but for me it was fine.

As for the diagram, the spec of my sensor also has the same diagram, I think that is just a test condition but honestly I do not know- if anyone else knows the answer on that I'd be curious!


While I'm not familiar with audio cable terminology, the circuit shown indicates that the sensor (as nearly all microphones?) has an internal Field-Effect Transistor (FET) for pre-amplification and impedance transformation.

The resistor is indeed relevant as it determines the current flowing through the FET and therefore the amplification. If you use the values given in the schematic you are not wrong.

Concerning to connections to audioboards, you must consider that hot side (top) in schematic carries the signal and must be AC coupled using good and large capacitors. The cold side (bottom) is connected to ground (GND).

Concerning size of AC coupling, I admit I have no experience with commercial products, as I build my own and I'm working with underwater hydrophones without build-in FETs. However, the only Audioboard on my desk uses 2,2 uF.

Note further, if sensor has its own external power you would connect it to line-in and not to the microphone port of your audio card. The microphone port typically provides the current into the FET, but it is hard to tell if the current provided by the audio card is suited for the FET, without internals of sensor and audioboard.

Edit: changed terminology from microphone to sensor as this particular sensor is a vibration sensor and not a microphone.

Edit2: from the schematic it seems that this is also directional sensor.

Edit3: from the datasheet, which is not great, this seems to be 3 pin sensor (positive, negative, output terminal). A three-pin sensor is typically a sensor with complete pre-amplifier, not with phantom voltage, as discussed above. So, there are discrepancies between test set-up and pin descriptions.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.