After a bit of a break in the project while I was away doing some other things… we re-convened and decided to have a trouble-shooting session.. even though things had been running more smoothly, there were still some issues with the Pi crashing… Matt had a version that had been running pretty smoothly, but there were a few things missing from it, so we decided to add these new elements one by one and Tarim had prepared a few tests and techniques to work out what the glitch was…. 
 
It was quite a long day, and  we had to make a few tricky decisions… basically – we all think that the USB is a bit of a problem on the Pi – (and this is also backed up with reports of this online).. which means that because we are sending all the sensor date via USB on the Arduino – AND all the audio input and output via the the soundcard on the other USB port, it is causing problems that do seem a bit difficult to solve. 
 
So – for now, the only way to really make it completely stable is to take off some of the controls and a bit of functionality… so – goodbye to manual controls for delay and reverb : (  which is a bit of a shame as these gave the performers a nice amount of control to really shape their sounds… but – we still have the basic functionality modelled on echo-location. The proximity sensor changes the echo-delay rate and we still have the presets to introduce extra notes at close proximity. We also left ‘note length’ as a manual control, so the length of the sample can still be adjusted manually.  I think there is still enough to make an interesting group piece – especially when there are a lot of singers using them together… progress was made! – hurray!... IT also means we can hopefully expand the bat-box out again in the future to include the controls we have just removed.. hopefully!