@martin There is little information about the JRF on his website. It kind off looks like a piezo element with some airspace behind it, some weight and "plastic dip" to seal it. I have no idea if that is the whole construction. This construction would create a high impedance microphone so there could be an impedance transformer included. Maybe you can try to measure ohms between the tip and sleeve (MIX and GND) contacts. If it is infinite or higher than 10k ohm it is wise to mount a resistor across the MIC and GND contact so the bias voltage is lowered, otherwise audio might not pass the analog switch circuit.

@terry.dawson @jaybarlow33 I do not want to be clogging up this forum but I am looking for some information on how to test a hydrophone, maybe you can help me. I made my own very simple hydrophone it seems to work but I can not tell how well it works. I have no reference materials around me (I think) maybe you can e-mail me so we do not abuse the forum. I usually use gmail with my initial an last name (ehouwertjes). If anyone else has tips one how to test a hydrophone they are also very welcome to send me an email (or still reply here perhaps). Kind regards, Edwin

Hi Koen, I have several diy devices to playback bat sounds. I like to use ribbon tweeters which have a wide frequency range. I use a fairly cheap 6W model like used in the Vrignault batplayer project. Be aware that the sounds are much weaker that the bats can emit so it is no comparison to test the "range" The range mainly depends on the volume the bet emits, the direction and the emitted frequency. Oh, and you would not want to work on 16kHz where you can hear only a social call and maybe something of the larger species. Al lot of USB soundcards have higher samplerates, up to 384Khz so you can easily play the higher frequencies, but you need to build or modify the amplifier as these usually are not intended for frequenties above 22kHz.

It is nice to hear that you can see some sound on the microphone input. The resistor you placed helps lowering the voltage on the analof input circuit. In the Audiomoth it seems that there are two resistors in series creating the bias voltage for an electret element. If there is no load this voltage is 3.3V on the input of the analog switch IC. I drew an image showing the pass band of the switch which is between 0V and the supply voltgage. If the bias voltage still is too high you can get into trouble. If your microphone, hydrophone or other audio source does not lower that bias voltage you will probably not hear a lot. Without the added resistor over the microphone input you can only hear the bottom halfs of the sine wave of the stronger signals. Adding a 10 or 12k resistor will lower the voltage to about 2.8V or so, which is low enough for most small audio signals. Ideally one would like to see this around half the supply voltage (1.65V) but that might take a resistor of about 2700 Ohms as load to reach that voltage which might give additional load to the microphone and consume a little more energy.

There are some ideas in this external microphone thread. https://www.openacousticdevices.info/support/device-support/external-ultrasonic-microphone

@hinojosalangosto Maybe this makes things easier to understand. The 3.5mm jack connector is a 3-conductor connector wit a switch contact. Tip Ring and Sleeve, Sleeve is ground (0v) Tip is the microphone signal. The Ring contact is not used. I put 3V (via a 1000 ohm resistor) on that pin. The connector on the PCB has 6 pins, two are on the Tip contact, two are the Sleeve, one is the Ring contact and one is a SWitch contact. The switch contact senses the external microphone, when you place the connector the switch contact pulls to GND. The FG elements mic wire (usually yellow) does not like the bias voltage that is on the tip contact so it needs a capacitor. 1uF is fine but this is not critical 100nF oe 100uF will also work. The bias volatge needs to be lowered, for that I use a 10000 ohms resistor between the mic lin (Tip contact) and the GND (Sleeve contact) . This should make your 3-wile microphone work.

@hinojosalangosto The problem with the Knowles FG and SPU microphones is that they need a separate voltage supply where the Primo and Panasonic WM61a and others use a 2-wire system. The Audiomoth input is made for this two wire input. Simply tying mic output and mic-power together will not work (I tried and it did not work for me) I only got it to work with a power supply as in the picture and a load on mic line. The signals are much lower as with the SPU microphone. I try to add a picture of a recording of a finger rub in fromt of the microphone with both the SPU and the FG element. (they both look quite weak because I desabled internal amplifiers in the audiomoth to prevent ugly harmonics). Maybe the Primo works well enough on high frequency, I never tried it because I was already discouraged by a friend that did not have very good results. I guess I will stick with the SPU microphones as external mic even though they are less waterproof.

these FG microphones are widely used in batdetectors. The datasheet from manufacturer does not show how wel it works in the ultrasonic range but it does work fine. It is al bit hard to get this to work, you need a power supply, which you might be able to connect to the ring contact. Ik used a 1k resistor connected to a pin that should have 3.3v on it. The resistor is to prevent high currents in case of a short circuit. The bias circuit probably also needs a load. (Tip contact/mic signal) otherwise the DC voltage is too high for the audio to properly pass the switch/amplifier circuit. I dus not fry my Audiomoth with these experiments but it is no guarantee that it will not break yours. Just be carefull! (The mic in the picture is one of my SPU0410LR5H-QB experiments)

A while ago I suggested hooking up an SPU0410LR5H-QB with just a simple resistor to steal some power from the mic/bias pin. I have the feeling the load of the microphone is not enough and we need to create a little voltage drop in order to lower the bias voltage. Otherwise the audio might not pass through the microphone switch circuit. If you tied and found it was not working, try to add a 10k resistor over mic and gnd connection. (10k is not critical, a higher value might be a bit better for less influence on the audio) Oh, and if I ever did sent any of you a 7mm round PCB with SPU0410LR5H-QB microphone on it, it probably has 1nF capacitors and no 1uF capacitors. 1nF creates a highpass filter effect so these only work on ultrasound. Sorry for that mistake. (8mm round versions were fine)

It would be nice to have the schematics, it looks like there is a SN74LVC1G3157 SPDT switch or something similar to switch between internal and external microphone and a MAX4466 opamp to amplify the audio. The small finger on the jacklug socket seems to connect directly to the switching line of the SPDT switch. Somehow it seems that U5 could be switching the power for the external microphone, but I can not see how what is actually switching U5, or maybe I just got confused switching back and forth between layers.

The microphones have different footprint and different signal levels. The SPU0410LR5H-QB needs an extra amplifier like MAX4466 or something. (Not sure if this is the exact type used in v1.2.0 board). The footprint is quite different, you woud not be able to fit the SPU on the SPM pads. So the mest solution is finding some old leftover stock of the SPM microphones if you want to fis the V1.0.0. or V1.1.0 audiomoth but soldering is difficult.

It is a tricky microhone to solder since it has all pads and audio port on one side, it takes quite some skill to do that. If you have v.1.0.0 or 1.1.0 it is also hard to find a microphone since that type of microphone is no longer produced.

The Elecon batlogger microphones have an internal amplifier and temperature sensor. +3.3V is on the tip, mic output (+ temp, possibly DC) is in the ring contact. It won't work on an audiomoth unless you make an adapter that also supplies 3.3 volts. The batscanner microphone looked like it had no amplifier and is quite weak. That one seems to have the 3.3V on the tip and audio output on the ring. So both wil not work without applying supply voltage and rewiring.

Audiomoth only uses two pins, GND and Output. You need to connect O (output) and + using a resistor of about 4.7 kilo ohms. Connect GND to the sleeve and O to the tip of the 3.5mm connector. Like in the pictures below.

I just had a look at the jackplug I can see it should not make a diffrence if one uses a 2-conductor (mono) 3-conductor (stereo) or four concductor jackplug. There are two contacts on ring 2, pin 4 connects to GND which makes the audio circuit switch o external microphone. The internal micropone seems to work for David, the external does not, but does the internal mic actually switch off? Maybe you could try and measure if there is a small DC voltage (like 3V) on the microphone signal pins (2 and 5). The voltage should (also?) be present when the microphone jack is in place. If there is no voltage there might be a short in the mic, but as the mic seems to work on an other device it cold ly deeper in the audiomoth. The picures below com from the applicationnnote and datasheet. I added some lines to explain the pin-connections.

I sent some of my elements to David in Belgium, he tested the microphone and compared to the internal (which is te same element) It seems to be working about just as well. I don't know if he has compared the recordings to something that uses an FG element. I wish I had the 1.2.0 version so I could do some test. I just tried to mod my 1.1.0 so it might look more like te 1.2.0. I recorded a sweep from 150-20Khz which I played with a homebrew speaker, so we can only compare the two (maybe the mics work on higher frequency that te speaker). The SPU was used recording on middel gain setting, the FG on the highest setting. Unfortunately connecting the FG did not work on my modded 1.1.0 board, so I attached V+ of the microphone to 3v3.

Connecting the Knowles SPU and FG seems possible by using the circuit I described earlier. The FG element however is very much less sensitive than the SPU so the FG might need like 20-30db more amplification. (I don't know any exact numbers) A recording will be a whole lot softer when you use an FG element. I have no idea of the gain setting in the audiomoth can compensate for that, maybe running on the higest gain setting will be good enough. The FG has a huge price compared to the SPU, I think the FG is around half the price of an audiomoth. It is also difficult to mount and you would probably double the price of the audiomoth. The knowles SPU element seems to be doing quite a good job on the ultrasound range.

Adaudacity is a nice program to look at Spectrograms, but also wavesurfer looks nice. Both are free to also batexplorer can be used as free version. Something like Batclassify might be of help preselecting audiofiles. https://bitbucket.org/chrisscott/batclassify/downloads/ I think it is based on Tadarida, or maybe you could use the acoustic pipeline if you need some auto ID-ing.

Oh, it seems there could be a small error in the picture I posted. The actual part name is SPU0410LR5H-QB, the versions 1.0.0 and 1.1.0 use the SPM0408LE5H-TB like the uMoth. These elements don't have info in the datasheet above 10kHz, but have a look at the popular FG elements, these also have no specs in the ultrasound range but are very well known for their ultrasonic response. The SPM0408LE5H-TB has a built in amplifier, the pinout for SPM0408LE5H-TB and SPU0410LR5H-QB is also different so it is no drop-in replacement. The SPU0410LR5H-QB seems to work a little bit better on the higher ultrasound frequencies.

How many devices did you test of each type? It kind of seems that people are reporting differences even between the same models. I think it is hard to tut any conclusions to testing just 1 or a few pcs per model. I do not own 1.1.0 and 1.2.0 versions but only a few of an adapted version 1.1.0 with SPU0410LR5H microphone. It seems this microphone has a bit more sensitivity on the higher frequencies. This could explain why a pigmaeus can have a better score than the kuhls but if I see correctly ppip is performing much better? I think it would be nice to look at the sonograms to try and undersand the difference. Maybe some noise is in the way?