CCS/TMS320F28069: JTAG Connection with TMS320F28069 with XDS100V2

Hi Siddharth,

Two comments:

1) Is J3.9 connected to J3.11?  There is a connection, but no dot.  It's probably not an issue, but thought I'd mention it.
2) Is there isolation in your system between the SMPS and your computer?  If not, I think this may explain part of what you are seeing.  I might recommend taking a look at something like the following (namely the BH-ADP-ISO14):
http://www.ti.com/tool/tmdsadp


Thank you,
Brett

Hi Brett,

1. There is a connection between J3.9 and J3.11.

2. Brett Please explain what exactly could be happening and how isolation will work.

There is one more observation: If I power the control card from separate DC Supply. The JTAG works Fine. And I power the rest of the system without adding the control card to it. Now I join the grounds of Control Card DC supply and rest of the system, this leads to blinking of JTAG. Some rare times the JTAG LED stabilizes for some time, but randomly connection breaks in CCS.

Regards,

Siddharth Gupta

Hi Siddharth,

Let's say that you have an inverter on the 'power board' which your C2000 is controlling.  When the C2000 turns a lower leg FET off, you assume that the current going through the FET will go to board ground, then through the rectifier and make a loop through the AC grid. 

Note that without isolation it is possible that some percentage of the FET current finds that the path of least resistance is from the lower-side FET, through the board ground, to your custom control card, through the emulator, across the USB cable, into your PC and then complete the circuit with the AC grid. 

Adding isolation between the board and the computer will keep the computer and emulation circuitry from carrying circulating currents from the power electronics circuitry.

If the above could match your situation, I suspect that adding a JTAG isolation adapter will improve the signal integrity of your connection.

===

If what I am explaining doesn't match your circumstance, I'd look more deeply at the connections between the power card and your control card.  Make sure there's adequate ground connection between the two boards, make sure that they are secure.  Make sure there is adequate input capacitance for the TPS526601 on the control card side and make sure that this power rail is clean.  Etc...

Thank you,
Brett

Hi Brett,

Thanks for the reply.

1. Please find below connections between Power and Control Card.

Brett Kindly Note:

1. SMPS Outputs 7V to control Card and Control Card DC DC converter supplies 3.3V to Controller from that.

5V is derived from 7V in Power Card and used to translate PWMs in Control Card to 5V.

12 V is used for Opamps in Control card. (12V,7V,5V,3.3V have same grounds).

24 V is isolated supply .

Inverter Legs are supplied by Isolated 15V from SMPS.

2. The JTAG problem occurs when I have not given any PWMs to the inverter. (No Inverter Leg is on).

3. Checked the TPS562201 Output 3.3V, and its clean with less than 100mV ripple.

4. Input Caps of TPS562201, Please confirm if its enough.

Input caps- Two 10uF Ceramic

Output Cps- Two 22uF Ceramic.

Kindly advice on the same.

Regards,

Siddharth Gupta

Hi Siddharth,

Is the 7V that the SMPS generates isolated from the grid (you've marked the inverter power rail is, but not this one). 

If it is not, I'd definitely recommend using a JTAG isolator.
If it is, using an isolated JTAG isolator will not hurt things.  It will merely better protect your computer and improve signal integrity.

I don't see anything obvious that is wrong with what you've done.  Perhaps there is something you'll need to debug, but this sounds like the kind of issues that a JTAG isolator will help.


Thank you,
Brett

Hi Brett,

7v,12v are generated from one winding, Inverter rails from other, 24V also has its separate winding. SMPS is Isolated from Grid.

I am not able to comprehend why this situation is arising, even when inverter is in off state (No Leg is on).

Regards,

Siddharth Gupta

Hi Siddharth,

Based on your last response, I also do not have a good explanation for what you are seeing. 

I'm not going to be able to debug your full system, however if the inverter's winding and power card's winding share the same ground, I could potentially see this enabling some path - but it may not be obvious (if one is there).  Things are not very clear because it appears that the issue occurs even when the inverter is off.

Something that might help you debug may be to flash some program into the device's flash with the external supply.  And then try to run the code with the SMPS.  This might help you determine if the issue is purely with emulation or whether the SMPS is creating larger issues for the C2000 device.


Thank you,
Brett