SNV Write Fails CC2541

Hello. That is a very strange place to fail as it essentially means the page erase is failing. Have you modified the linker / board config files at all? Is this custom hardware? Do you see this error with the default simpleBLEPeripheral project? Another debugging step would be to read out the .hex file using Flash Programmer to verify if the page erase is actually failing.

I finally got my had back onto a working dev kit and tested the code with no issue with the write. What would cause custom hardware to fail? The custom hardware is a copy of the reference design.

Hello. Please see the hardware resources on the wiki page:

processors.wiki.ti.com/.../Category:BluetoothLE

Hi Tim, I'm already aware of the hardware section and have been using it heavily. Are you able to point me to something more specific and suggest some reasons on what would cause this error? I have not run into any other issues with programing / any other functional issues with the custom hardware.

Hello Ansgar,

That's good that you are following the HW recommendations. Regarding the erase/write fail, is it 10% repeatable? Are you powering by a CR2032 coin cell battery or other means? What is the batt voltage during erase/write?

Best wishes

I'm currently powering the system through the cc debugger. The write is currently failing every time, due to the fact that failF is being set to True see my first post for additional details on that. Are their any hardware design considerations that I should be double checking that could effect it's ability to write?

can you post your schematic and layout so that we can review them? If you do not want to share them here, send me a friend request, so you can send them to me directly instead.

Great, I have send you a friend request. Thank you for reviewing the files.

Hi,
First of all, sorry for the delay. I did look at your layout (if you want us to look at the schematic as well, send me a new message), and I do not see a reason for this kind of failure here. The power routing is not optimal, but it should not cause this kind of error. Do you see any other issues, does it work if you use other SW? What is the voltage over RBIAS? Any more info you can provide?

What is less than optimal about the power routing? What should be done to improve it. 

So far I have seen no other issues. I'm not sure what you mean by "does it work if you use other SW?". Can you please clarify?

RBIAS 56k 1% - 1.266 Volts

What type of information would be helpful to debug this issue? 

The "other SW" you can test with is the default simpleBLEperipheral project included with the installer.

Also, please see my original comment for debugging ideas.

I am already testing this withe the supplied simpleBLEperipheral which works on the dev board and not on the custom hardware.

Are you able to share a link with me on how to use the cc debugger to read out the hex?

Thank you

You'll want to use the Flash Programmer (version 1.2.7): www.ti.com/.../flash-programmer. You can use the "Read Flash into File" function. See the User's Guide for more info: www.ti.com/.../swru069g.pdf

If simpleBLEPeripheral is working on the dev board and not on your board I think that pretty much rules out a software issue. But it would be good to double check by reading the hex file.

Hi,

I do not have the design in front of me now, but if I remember correct, it was the placement of the decoupling caps.  They should be placed close to the vdd pin. The order should be VDD, decoupling cap and then the VDD pin on the CC2541 to prevent noise. This can cause poor RF performance, but it does not explain your case. Have you tested several boards and see the same failure on all boards? Have you tried swapping chip from a known good board to a known not-confirming board (if you have any good boards to spare)? Have you tested when you power the board with a good power supply? If you send me the schematic as well I will look again on Monday (I am out of the office a couple of days), if you want.

Hi,

I have sent you my schematic so that you may review it. Thank you for the notes on the caps.

I have tested this on several boards that were assembled by a professional company with all of the same results. On each board all other functionality that we are using works properly. As per you suggestion I tested the device with a power supply and received the same results.

The schematic looks ok. Have you tried separating the power to the CC2541 from the rest of your comp? In our ref designs we use a bead and a large cap to filter power and prevent noise on power between comp. Have you tried proping on power and your control signals with an oscilloscop? Do they look as expected?

I've also responded to your email. I've separated the power with no change in behavior. Are you able to point me to some information on what the expected signal is?