Forum Rule: Always post complete source code & details to reproduce any issue!
Page 18 of 18 FirstFirst ... 8 16 17 18
Results 426 to 439 of 439

Thread: Bat detector

  1. #426
    Senior Member CorBee's Avatar
    Join Date
    Jun 2018
    Location
    Netherlands
    Posts
    203
    Well the Mic is PCB mounted and thus that central pad is not something I can or will try to fiddle with ... its way to small for programmers I am used to buttons instead of a microscope

  2. #427
    Senior Member DD4WH's Avatar
    Join Date
    Oct 2015
    Location
    Central Europe
    Posts
    409
    well, then we will never know or solve the problem with the noise that you have . . .

  3. #428
    Senior Member CorBee's Avatar
    Join Date
    Jun 2018
    Location
    Netherlands
    Posts
    203
    Well we have Edwin who assembles these modules and he might have a go at trying one without the central pad connected. So never ... say never

  4. #429
    Member
    Join Date
    Nov 2018
    Location
    Netherlands
    Posts
    75
    In single ended output the -out pad is floating.

    I do have to mention the codec gain used is quite high,


    This is what I see with codec gain on 10

    Click image for larger version. 

Name:	codecgain 10a.jpg 
Views:	7 
Size:	149.7 KB 
ID:	17229

    I also made a few noises....

    The spectrum of the silent part looks like this.

    Click image for larger version. 

Name:	codecgain 10.jpg 
Views:	3 
Size:	51.2 KB 
ID:	17230


    When I have codec gain set to 20 I see this

    Click image for larger version. 

Name:	codecgain 20.jpg 
Views:	3 
Size:	54.8 KB 
ID:	17231

    Sorry for the different scales....

    Somehow the peak near 23Khz seems a lot larger than the rest of the noise.....

  5. #430
    Member
    Join Date
    Nov 2018
    Location
    Netherlands
    Posts
    75
    Frank,

    Maybe you could use a SD-card extender like this to be able to seperate the SD powersupply from the Teensy. On the teensy it is very hard to do that but here you can easily cut a track.


    Click image for larger version. 

Name:	tfextender.jpg 
Views:	0 
Size:	15.2 KB 
ID:	17233

    There are also versions with white plastic ribbon cable but I think those could be a bit more difficult to solder on to.

    Edwin

  6. #431
    Senior Member DD4WH's Avatar
    Join Date
    Oct 2015
    Location
    Central Europe
    Posts
    409
    Hi Edwin,

    thanks a lot! I will think more about that!

    However, adding additional hardware for features already implemented (built-in SD card) is a bit annyoing, and I think I read that the built-in SD card is much faster, which could be important for high speed ultrasound logging.

    All the best,

    Frank

  7. #432
    Member
    Join Date
    Nov 2018
    Location
    Netherlands
    Posts
    75
    The externder was just an idea to be able to cut a power supply track and put a separate 3v3 source on the SD card to try and get the writing noise out of the teensy.



    I did some experiments again with the SGTL gain settings microphones and amplifiers. We already knew that using high gains settings on the SGTL adds a lot of noise.

    Both the ICS and SPU microphines seem to have some noise around 23Khz, with the SPU it is presented as a wider lower hump but the ICS presents a larger spike.

    The amplifiers help to suppress the low frequency sounds and enhance the high frequency sounds.

    Using these, or connecting the microphones directly seem to show the same raised levels on or around 23Khz.



    Although it seemed the noise is coming form the microphone the SGTL5000 also seems to have a big part in it.



    Of course this is not a huge problem but I just wanted to get rid of the vertical bar on the waterfall screen when I am searching for bats on high gain.
    Once one has located bats and they are closer one can reduce the gain and the bar becomes barely noticeable.



    The main problem is introduced by the user I guess, when there is no bat around, one tends to turn up the gain in the hope and hear some bats.


    If one wants great signal to noise level I guess the SGTL is not the best thing to use.

    Your stationary bat recorder is probably a lot better at recording but for running around the field using the detection modes and de graphical presentation dit approach is just great.


    Edwin

  8. #433
    Member
    Join Date
    Nov 2018
    Location
    Netherlands
    Posts
    75
    I just updated some documentation.

    I don't quite remember if I shared the link to these files before but any eBay buyer of the V0.2 board als got the link form the item description.
    The link now holds the new files.

    Anyway, if you still have to start working on the V0.2 board here are the updated files.

    The most important changes:

    Display update noise is reduced by quite a bit. (100 ohm resistors in all data lines)

    Added instructions on how to build a transistor amp on this board if you like that better.

    Beside that there was some minor parts optimization.



    The V0.2 version was designed to use a low noise rail-to-rail opamp that would give us a bigger dynamic range and less noise.
    Unfortunately the opamp als picked up some display update noise. This noise is nearly gone by using the resistors in the datalines the opamp version now is just slightly better than te transistor version. No huge differences so if you have the transistor board version keep it. The biggest improvement is in reducing the data noise, of course one can also use the resistors in the data lines with the transistor version.


    If one wants to use standard headers, it is better to use a transistor amplifier.


    Kind regards,

    Edwin


    Almost forgot, there is some extra information on LCD revisions, some versions use direct LED drive, others have a transistor on the LED pin. One can determine the difference if there is a Q1 part next to the U1. It makes a huge difference in the resistor used on the LED pin.

  9. #434
    Senior Member
    Join Date
    Dec 2016
    Posts
    101
    Quote Originally Posted by pe1pwf View Post
    I did some experiments again with the SGTL gain settings microphones and amplifiers. We already knew that using high gains settings on the SGTL adds a lot of noise.

    Both the ICS and SPU microphines seem to have some noise around 23Khz, with the SPU it is presented as a wider lower hump but the ICS presents a larger spike.

    The amplifiers help to suppress the low frequency sounds and enhance the high frequency sounds.

    Using these, or connecting the microphones directly seem to show the same raised levels on or around 23Khz.



    Although it seemed the noise is coming form the microphone the SGTL5000 also seems to have a big part in it.


    If one wants great signal to noise level I guess the SGTL is not the best thing to use.

    Your stationary bat recorder is probably a lot better at recording but for running around the field using the detection modes and de graphical presentation dit approach is just great.


    Edwin
    Edwin

    Have you tried to use the tl972 with a higher gain and using the SGTL at a lower gain?


    Alain

  10. #435
    Member
    Join Date
    Nov 2018
    Location
    Netherlands
    Posts
    75
    Quote Originally Posted by AlainD View Post
    Edwin

    Have you tried to use the tl972 with a higher gain and using the SGTL at a lower gain?


    Alain
    Yes I did, somehow it seems I did not post my reply yesterday.

    I set the amplification close to 40dB on the TL972, these are the results i got in a quit room with SGTL gain set to 0

    Click image for larger version. 

Name:	40-0db.jpg 
Views:	1 
Size:	53.2 KB 
ID:	17251


    I also did the same with SGTL gain set to 10

    Click image for larger version. 

Name:	40-10db.jpg 
Views:	1 
Size:	52.9 KB 
ID:	17252


    With the ICS microphone I still get that peek although it is very low when using 0dB on the SGTL the SPU microphone also gives me a bit of signal near this frequency but it is much wider and a lot lower.
    I would still like to gave some control over the gain, If i do set the gain on the opamp very high I have only little control over the total gain.
    An extra control could be an option but I do not really want an extra switch or pot.

    Other experiments showed the amplification in the SGTL adds quite a bit of noise when it is at a high level. This 23Khz signal also proves we should better have more old fashioned analog amplification.

    For that reason I already changed the resistors a bit in the last builds.

    Since the SPU microphone has a wider and lower hump than the ICS, I still choose that in combination with this bat detector. Maybe I do not have the lowest noise level microphone but I think with the rest of the hardware one will hardly notice the difference.

    Kind regards,

    Edwin

  11. #436
    Junior Member
    Join Date
    Jun 2019
    Posts
    13

    Query regarding resistors in documents for build 0.2a

    Quote Originally Posted by pe1pwf View Post
    I just updated some documentation.

    I don't quite remember if I shared the link to these files before but any eBay buyer of the V0.2 board als got the link form the item description.
    The link now holds the new files.

    Anyway, if you still have to start working on the V0.2 board here are the updated files.

    The most important changes:

    Display update noise is reduced by quite a bit. (100 ohm resistors in all data lines)

    Added instructions on how to build a transistor amp on this board if you like that better.

    Beside that there was some minor parts optimization.



    The V0.2 version was designed to use a low noise rail-to-rail opamp that would give us a bigger dynamic range and less noise.
    Unfortunately the opamp als picked up some display update noise. This noise is nearly gone by using the resistors in the datalines the opamp version now is just slightly better than te transistor version. No huge differences so if you have the transistor board version keep it. The biggest improvement is in reducing the data noise, of course one can also use the resistors in the data lines with the transistor version.


    If one wants to use standard headers, it is better to use a transistor amplifier.


    Kind regards,

    Edwin


    Almost forgot, there is some extra information on LCD revisions, some versions use direct LED drive, others have a transistor on the LED pin. One can determine the difference if there is a Q1 part next to the U1. It makes a huge difference in the resistor used on the LED pin.
    Hi Edwin
    Can you clarify where the 100 ohm resistor bodies go between teensy and audio board please. Do the main resistor bodies go between the teensy and the audio board or under the audio board and the main pcb with just a wire link between the teensy and the audio board. Photo is not too clear. Many thanks.

  12. #437
    Member
    Join Date
    Nov 2018
    Location
    Netherlands
    Posts
    75
    Quote Originally Posted by seriesman71 View Post
    Hi Edwin
    Can you clarify where the 100 ohm resistor bodies go between teensy and audio board please. Do the main resistor bodies go between the teensy and the audio board or under the audio board and the main pcb with just a wire link between the teensy and the audio board. Photo is not too clear. Many thanks.
    The resistors are all soldered on the teensy so these are between the teensy and audioboard.

    I hope this helps.

    Click image for larger version. 

Name:	IMG_20190807_175740765.jpg 
Views:	7 
Size:	102.8 KB 
ID:	17253


    On the last boards I soldered I cut the pins like this so that mounting was easier.

    Click image for larger version. 

Name:	IMG_20190811_101209075.jpg 
Views:	5 
Size:	192.0 KB 
ID:	17254


    Yes I know, the second picture are not 100 ohm resistors, I ran out of 100 ohm and used 150 which also work.

    Edwin

  13. #438
    Junior Member
    Join Date
    Jun 2019
    Posts
    13
    THanks Edwin, yes that helps.
    Regards
    Barry

  14. #439
    Senior Member CorBee's Avatar
    Join Date
    Jun 2018
    Location
    Netherlands
    Posts
    203

    next software update: requests ??

    Hi,

    Behind the screens I am working on a rather large change/update of the current sourcecode. If anybody has specific wishes/ideas/bugs I would like to hear them to consider adding them to the sourcecode. For me this is more or less the last week that I can spend more than average time on this project. From next week on my daytime job will be more demanding than in the past weeks and that means the progress will slow down.

    regards
    Cor

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •