TAS5717EVM reliability

Hi, Dalibor,

    We apologize for the difficulties you are having. This is an unexpected behavior, if the GDE indicates that the EVM is connected, it means that the I²C communication with the amplifier is correct. If the amplifier is not working, I think it may be damaged. You can try reading register 0x02 and check if any fault is occurring. you need to clear this register before reading it to perform a correct lecture. The IC could be damaged by different situations, in my experience, a possible cause will be the operation at PVDD above 20V when the modulation index is not set as minimum while using loads below 8 Ohms.

 Best Regards,

  -Diego Meléndez López
   Audio Applications Engineer

Hi Diego,

Could you please try and start the GDE but without connecting the EVM to your PC.  If you select Target -> connect, you will find the the output window states that the target is connected.  If you then go to Tools -> I2C Memory Tool and do a read from address 0x00, it will return data of 00 and the bottom of the screen will find a message 'Read Successful'.

I don't understand this since the target is physically disconnected so could you kindly verify that GDE is functioning as expected?  The version I am using is 3.2 build 7.

Also, could you point me somewhere in literature where this PVDD dependency is explained for EVM.  I have 5V at the Vin, 24V at the PVCC input and loads are 4ohm 3W, so I need to know how to set the GDE in order to protect the EVM from blowing up again.

Is there a correct order to connect USB, 5V and 24V?

Many thanks,

Dalibor

Hi, Dalibor,

    I performed the test you mention and was able to replicate your issue. Looks like this GDE does not verifies the I²C communication with the amplifier as other similar GDEs. The I²C memory tool is the best way to verify if the communication with the amplifier is correct. Even if  "Read successful" message appears, you need to check if the Memory tool is receiving the information from the IC. As an example, when reading register '0x00', you shalll receive a '0x6C' value. Can you please try again with the EVM connected to the MC57xxPSIA controller board and check if the communication with the IC is correct? If you don't get the '0x6C' when reading register '0x00', please verify the environment variable is set as indicated in this document. If this settings are correct and the communication is not established, the IC may be damaged or the controller board is not working correctly. You may try testing with a different combination of EVM or controller board to find the issue. You can test this with only 5V power supply turned ON; PVCC is not required for testing the I²C communication.

You need to change the modulation index manually with I²C Memory tool, by writing '0x07' to register '0x10'. I suggest doing this before exit shutdown. 

There is no order required to connect the power supplies or USB. 

Best Regards,

  -Diego Meléndez López
   Audio Applications Engineer

Hi Diego and thanks for your reply.  I am sorry for my runt in advance but my frustration with TI is simmering so I need to off-load.

I will try as you suggest once I receive the new EVM and report back...another $200... :-(

However, here again there seems to be an issue with TI releasing half-baked stuff to the community.  As we have confirmed here, GDE just does not seem to be reliable enough to be sold as a debug tool and one cannot trust what it produces.  I am supposed to use the EVM kit to tune my TAS5717 filters and it does this very nicely, but when you try and get the data out from the GDE, you realise that it is mainly garbage and does not give you all of the values reliably.

The literature for the EVM kit is clear that it supports PVCC up to 26V and there is no suggestion of risk to the EVM?  Should I try and guess all of the other ways I could damage the $200 kit?  Please give me a chance.  

In my frustration I have taken to discussing TI as a supplier with my colleagues and contractors.  I am afraid that the consensus is not good.  The lack of quality and support is the theme with time and effort wasted waiting for E2E support and poor product quality.  If I had time I would probably opt to change the vendor at this stage...and TAS5717 is not the only TI device we are having issues with...lesson learned.

Again, I am sorry as I know you are only doing your job and this is not a reflection on you.  However, I wonder if there are other TI customers experiencing the same kind of frustration.

I will be posting my results once the new kit is here next week.

Best wishes,

Dalibor

Hi, Dalibor,

Thanks for your honest comments. We apologize for the problems you are having, unfortunately some of devices are quite old and the documentation is not as good as we would like it to be... We are working hard to make the documentation and support better for our devices.

I will wait for your results, please let me know if I can help you further with the EVM evaluation.

Best Regards,

  -Diego Meléndez López
   Audio Applications Engineer

Hi,
The latest EVM kit I purchased has just stopped working...it was expected I guess. Could you please tell me if is possible to upgrade the DFU firmware on the kit?

Best wishes,
Dalibor

Hi, Dalibor,

If the DFU firmware of the MC57xxPSIA board got corrupted, you can try the steps mentioned in the next Wiki page:

How to Revive a MC57xxPSIA Board 

Best Regards,

  -Diego Meléndez López
   Audio Applications Engineer