In order to enhance 0/1 ratio compensation, I did some improvements on the MCU firmware. The 0/1 ratio is now trimmed at the end of every cluster. The flash program memory address Z=0x00AE contains a constant (now set at 0x24A3) that enables very precise 0/1 ratio trimming.  After all the adjustments of S5,S6 and S7 with help of BSA, the precise trimming of 0/1 can be done by changing the 16-bit value at 0x00AE - increasing this constant increases the number of ones, decreasing it increases the number of zeroes. For even better NIST testing results!

I think you have missed one factor that can make the illogical logical: the law. There is a legal requirement for facilities where food is produced to control pests, and to make sure that dead vermin are removed within 24 hours. Non-compliance is expensive, a facility could potentially be locked down for a period of time after a violation. This is what makes the business case for the 'smart' mouse traps, and indeed it is already being commercially exploited, see for instance https://www.xignal.com/products/xignal-mousetrap.  On a tangent, it is probably more 'humane' to kill a mouse than to catch it alive. The mouse will experience a lot of stress, starve quite quickly, and will probably die shortly after being released anyway...

From the text: “IC3B is configured as a Schmitt trigger which produces a square wave output from the input signal. The Schmitt trigger adds a variable amount of hysteresis to the input signal via R23 and P2, and this prevents small fluctuations (mainly noise) of input signal (around the reference voltage) causing false triggering.”   So, if the signal form the microphone is quite large P2 will have no effect. A weak signal can be too low for the hysteresis and there’s no square wave or is not entirely characteristic for the bat signal. The modulation is processed separately. P2 is not a linear volume control, but its setting can be the reason weak bat signals are not audible. No or a too small hysteresis, set by P2, will cause noise to cloud and distort the audible signal. The hysteresis should be set as small as possible. Without a signal the speaker than only occasionally makes a sound. Looking back at the circuit, maybe putting an automatic gain microphone amplifier before creating the square wave would have been a better solution? And maybe it’s control level can give some information about the distance of the bat? Just a thought, food for thought…  

That is a bit strange.   The first time I tried it the bat call came in loud and clear, and when I turned towards it the call got louder.  I then saw it quite high up (8m?) at at least 10m distant, probably 15m. (I'm still curious about the ineffective P2 adjustment -- or maybe I misunderstood that paragraph.)

Changes: Prototype construction 1 Variant. Component LM336_5 used for IC2.  When putting into operation / voltage check at the IC Sockets, was not AREF. At LM336_5.0V (there's no datasheet) in TARGET 3001 V19 but in the database???  Component corrected changed in LM336Z-5.0V and on board. AREF is now exactly adjusted to 5V. For those who already have boards of version 1 in progress, you can rotate by 180 ° IC2, then functionality is available.   For all users who want to directly insert the BOB ACS712 (20A) in K9, I've moved K9 slightly to the North.   For S7, S8, S9 MEC must be soldered must switch 5GTH935xx. Pin numbering on the back of un due note, it's also about 180 degrees twisted to solder. built-in LED XX stands for the color. Adjust to R7, R28, R29! XX = 22 = green value would then be 470 Ohm.   100 K = voltage divider for Umess1 R11, R12 was changed from 18 K to 20 K and tested with Multisim blue. Measuring range 0 now... 30V. Bascom program now changed due to changes in the value of R12 from 18 K to 20 k. Measuring range is now 0... 30V.   The IC TS912IN (rail-to-rail) is hard obtained a Dil version. New type is now MCP602-I/P.   It follows now first start-up with injection of 0... 30V from the laboratory power supply unit and Bob ACS712 with closed briefly entrance examination to 0A in the display (is). The setpoint values for voltage and current are with multi meter to 0... Checked 5V. Is that all right is connected with the University Lab. There is solder jumper on the circuit board. Once you can use the 5V in the circuit board. Jumper the other way around used the 5V from the University Lab. That's only because yes my project with other power supplies should go. All images, display of actual value voltage and actual power using an external lab power supply were. In the next step we a Unilab 2 PCB prototype connected to full functionality to test.   Dieter Aschmann