CC8520 app

Hi Daniel,

I will try to help you get rid of your nightmares ;-). However, I cannot see your schematic. Can you attach the schematic as a *.pdf or something?

Cheers,

Per H

Hi Per,

Thank you so much for that quick reply!!

I'm attaching my schematic as a *.pdf now, let me know if you can see it =]

0474.rx.pdf

Hi Daniel,

I see a few very serious things here.

Decouplings caps (I think this is why it does not work):

Decoupling of the different supply pins! It is not only important to have the correct values the placement(and layout) of them are critical for proper device operation.

Inductors on IO lines:

Check out the our ref design in terms of inductors on some on the IO lines. (explanation can be found here:

http://e2e.ti.com/support/low_power_rf/f/155/p/41401/145148.aspx#145148))

Pad on the backside of the chip:

Is it connected to GND?

What crystal are you using?

Finally, if you update your schematic a can review it and after you have done a layout you can also post it and I will have a look at it!

Cheers,

Per H

Hi Per,

About the decoupling caps and inductors on the I/O lines, i already have a transmitter, which design is very similar to the one i posted for the receiver and that one is working just fine, as i mentioned i already tested that one with the PPWDK... I still can figure out why the receiver i built can't connect properly with the PPWDK whilst the transmitter can, if their designs are practically the same... =S I did connect the pad on the back side of the chip to GND, i'm using a 48MHz cystal.

Summing up: 1) i managed to program both transmitter and receiver boards i designed

2) i tested and verified my transmitter with the PPWDK

3) receiver still doesn't work, and it's design is practically the same as the transmitter that is already working

4) i'm still testing, unsuccesfully, the receiver; every once in a while, after reseting my board it transmists for a second or two and stops even though the LED indicates that is connected, despite the PPWDK board shows the ALONE pattern....

I'm thinking the problem might be in the actual communication between the two boards, and i believe neither the decoupling caps or the inductors have to do with that, but i'm just guessing here....

I'm attaching the layout designs (only images of a two layer design) for that schematic i sent earlier. 

2021.rx_layout.pdf

Regards,

Daniel Jimenez

Daniel Jimemez,

I agree that your problem probably is with your decoupling capacitors.  I see that they are all on the board, but they are spaced very far away from the supply pins which they are supposed to be supplying.  In general, a thin trace has about 10nH of inductance, and when a supply pin is pulling current from this trace, this equates to about 150 Ohms of reactive impedance per centimeter at 2.4GHz, and 3 Ohms per cm at 48MHz (Z = j2*pi*f*L).  The fact that the board can be programmed and sometimes works for a very short period of time indicates that all the connections are in the right place, but something is disturbing the normal operation, and I think it is probably an issue with the supply decoupling.

I also see on the antenna impedance matching network that the component placement does not match the reference design, and that the antenna traces actually go from top layer to bottom layer and then back to top layer.  Making such layout changes like this pretty much ensures that you need to start all over with you matching network design.  I can guarantee you that for the reference design, an antenna specialist spent a considerable amount of time first calcuating what component values were needed, and then designing, assembing, testing, and measuring the resulting layout, and then making component value corrections to account for parasitics associated with the layout.

If you would like some more background information, there is a well written book called "High-Speed Digital Design: A Handbook of Black Magic,"  and it does a good job of describing what to consider when designing high frequency designs.