[C6655] checking power up/down sequencing

Naoki Kawada

Guru 19890 points

Hello,

My customer is checking the power sequencing for C6655. They have now two versions of PCB(A and B) on which a C6655 is mounted. I briefly checked them and had some concerns about them. Could you please double-check ?

Best Regards,
Naoki

4087.pcb_a_power_sequence.doc

3482.pcb_b_power_sequence.doc

over 8 years ago

0 Raja over 8 years ago

TI__Guru* 81335 points

Naoki-san,

Thank you for the post. We will get back to you as soon as possible.

0 Senthilkumar Srinivasan over 8 years ago

TI__Mastermind 43660 points

Naoki,

We do not have known leakage path from CVDD to DVDD15. Please review the design for another leakage path. The errata Usage Note 10 shows there is a known leakage path from VDDT to VDDR.

Regards,
Senthil

0 Naoki Kawada over 8 years ago in reply to Senthilkumar Srinivasan

Guru 19890 points

Senthil,

Thanks for your quick reply. My customer is now confirming for another leakage path with errata description. In the meanwhile, do you have any comments on the ramping in non monotonic manner ?

Best Regards,
Naoki

0 Naoki Kawada over 8 years ago in reply to Naoki Kawada

Guru 19890 points

Senthil,

They tried to refine the waveform to suppress the leak on 1.5V power lane. I've attached them. Do you have any comments on this ?

Best Regards,
Naoki

pcb_ab_refined_power_sequence.doc

0 Naoki Kawada over 8 years ago in reply to Naoki Kawada

Guru 19890 points

Hello Senthil,

Do you have any comments on the revised waveform ? Thanks for your help in advance.

Best Regards,
Naoki

0 Senthilkumar Srinivasan over 8 years ago in reply to Naoki Kawada

TI__Mastermind 43660 points

Naoki,

I am checking on this. Will update you soon.

Regards,
Senthil

0 Tom Johnson 16214 over 8 years ago in reply to Senthilkumar Srinivasan

TI__Mastermind 46460 points

Naoki-san,

What did the customer change between the first set of captures and the second? The first set of captures appear to show the 1.5V supply ramping slowly as soon as CVDD starts ramping. This is not expected. The second set of captures in the "refined" file shows the expected leakage from VDDT (1.0V) into VDDR (1.5V). This is known and characterized. As stated in Usage Note 10, this will not cause an issue.

They also highlight the non-monotonic ramping of the 1.0V supply. This is caused by the power supply. You may need to adjust the parameters in the power supply such as ramp rate or compensation to make this ramp monotonic.

Tom

0 Naoki Kawada over 8 years ago in reply to Tom Johnson 16214

Guru 19890 points

Hi Tom,

Thank you for your reply, and sorry for my late response. I've forwarded your answers to them, but there is no feedback from them. I'll update you once I got a response.

Best Regards,
Naoki

0 Naoki Kawada over 8 years ago in reply to Naoki Kawada

Guru 19890 points

Hi Tom,

I got some feedback from the customer. as for their fixes I heard, they added a resistance for electrical discharge to their power circuit. Also, as you suspected, they found another leak path to 1.5V and fixed it.

So you mentioned the refined power up/down sequence would be expected in the point of view of Usage Note 10, but it might be better to fix the non-monotonic ramping of 1.0V. If you have any additional comments, please let me know.

Best Regards,
Naoki

0 Tom Johnson 16214 over 8 years ago in reply to Naoki Kawada

TI__Mastermind 46460 points

Naoki-san,

The non-monotonic power supply ramp is an issue with the power supply design. That is not our expertise. Please contact an analog apps engineer for this support.

Tom

0 Naoki Kawada over 8 years ago in reply to Tom Johnson 16214

Guru 19890 points

Ok, thanks, Senthil. Now I close thread.

Best Regards,
Naoki

Processors

Processors forum

[C6655] checking power up/down sequencing