ANC hardware selection

Hello,

I'm working on implementing an active noise control algorithm. I am currently working on an ADSP-221489 EZKIT LITE, because we had one lying around. I am quite new to these things, but I am beginning to suspect that this board and the AD1939 codec on it are not quite suited to my application. I got the algorithm to work, but I feel the system is not working as well as it could be, mainly due to latency issues. I am wondering which hardware would be better for me. I hope you can help. Below I have listed some information about my application:

- Active noise control system with adaptive controller and adaptive plant model.

- Preferred sample frequency: 8 kHz.

- 2 ADCs/input ports.

- 2 DACs/output ports.

- Low latency (currently multiple ms's)

If you need more specific details to answer my question, please let me know.

Regards,

Thijs

  • Hello Dave,

    thank you for the explanation and all your help so far. I was looking at the documentation for the adau1772 board. If I am reading I correctly the wires should be soldered to the pins on header J4 on that board, is that correct?

    I was wondering about how I would use the boards once they are connected. Could I run my code from the 21489 board like I am doing now and just use the 1772 board as a codec? I know I should use the 1372 if I just want a codec, but if I'm not mistaken the codec parts of the 1372 and 1772 are basically the same. Since the evaluation boards of these parts are equally priced, I think I might as well use the 1772 board since this has additional functionalities that might be usefull in other applications.

    If I would want to use the filter banks of the 1772 board while using the 21489 as an external processor for adaptive filtering, would I need to run a code from the 1772 board, and would that change the way the boards would need to be connected?

    Thijs

  • 0
    •  Analog Employees 
    on Mar 30, 2016 5:01 PM

    Hello Thijs,

    Yes, you connect to header J4 on the 1772 eval board. Now let me suggest something. I use these cables around the lab: Item # 9110, .025" Socket Jumper On E-Z Hook, A Div. of Tektest, Inc.

    They are available through Digi-Key but I did not see them on the Radio Spares site. There may be another distributor in the EU that carries them. What I have done in the past is to take one of these cables and cut it in half. Then you could solder one side to the DIP switch on the EZkit  and then you just insert the connector onto the header J4. No need to solder to J4 and it gives you flexibility to make changes and to easily disconnect the two boards. But I constantly have to reconfigure my boards so it is great for me but you may be fine with just soldering on both sides.

    Now on to your next set of questions...

    I would take this in stages. The software is setup to control the AD1938 codec on the EZkit and you will be interrupting the serial data and clocks so that the DSP thinks it is getting and receiving data from the 1938 but it will be the 1772 instead. So setting up the initial communication will be easy and I think it may only be I2S but it is fine for step one.

    Then to go further you could connect the SPI communications port to go to the 1772 instead of the 1938 and then write the software to control the 1772 from the DSP. This is a little more involved since I am rather sure there are no pre-written defines and code to do this.

    A simpler way would be to let the 1772 selfboot from the EEPROM. If you end up using a 1772 in the product then this may be a good final solution for you. If you decide to use the 1372 then you will eventually have to move over to the DSP to setup the codec.

    So for now you can use SigmaStudio and control the 1772 directly. Then you can write the program to the EEPROM and run without a computer.

    Now to use the filters on the 1772, and the low latency when using the on-chip converters, you will need to connect some control lines from the DSP to the 1772. This can be I2C or SPI. The SPI uses more pins and so it depends on what you need for GPIO on the 1772. Then you have a choice. You can still let the 1772 selfboot, but there are some details to watch out for. Then once it is booted just send updated filter coefficients etc. The 1772 has two banks to allow you to instantly change all the filter coefficients and avoid instabilities. The other way would be to load in the entire program to the 1772 from the DSP.

    So my suggestion is to start simple and move up incrementally.

    More than enough info for now...

    Dave T

  • Hello Dave,

    thank you for your explaination. I have received my EVAL-ADAU1772. While I was preparing to connect to boards together by soldering some wires, I found something in the manual of the ADSP-21489 EZ KIT that made me wonder if soldering is necessary at all.

    On page 2-8 of the manual it is stated that setting SW1 and SW2 to off disconnects the DAI pins from the peripherals and makes them available to the expansion II interface, which if I am reading correctly is header P2.

    On sheet 15 of the schematics it is stated that the signals from all DAI pins are available at this header. I would think that I could connect the two EVAL boards by just running a cable from the proper pins of header P2 on the EZ KIT to header J4 on the 1772 board, which means I would not have to solder any wires.

    Could you tell me if I interpreting this correctly, or if there is a mistake in my thought process?

    Thanks in advance.

    Thijs

  • 0
    •  Analog Employees 
    on Apr 5, 2016 4:52 PM

    Hello Thijs,

    I just looked at the schematic and you are correct. P2 does have those signals present. Like I said earlier, this is not an eval board I directly support so I am learning about it with you! So those jumper wires I gave you the link to would be perfect to jumper the signals over.

    Thanks,

    Dave T

  • Hello Dave,

    I am going to try that, since that would make it much easier to disconnect the boards again.

    I am going to see if I can get my hands on some of the wires you linked to. However, the specific

    wires you linked to have .025" connectors so I think I would need slightly larger ones, since

    the headers have 0.1" pins, right?

    Thijs