TIDM-TM4C129POE: TIDM-TM4C129POE DC/DC Converter Issue

Gergely Bacso

Part Number: TIDM-TM4C129POE

Hi,

I had two boards made based on TIDM-TM4C129POE. One still works, one not.
The second one used to work, but after a week of rest, it just stopped working when I tried to turn it on again.

The PSE is continuously retrying to add power, but it fails.
I compared impedances to the other board's but they seemed to match.
I narrowed the problem to the (48V --> 5V) DC/DC part by supplying 48V to J7 and measuring this voltage between POE_NET05 and GNDX. And by measuring 0V between POE_NET12 and GND. Note, the circuit consumed only mAs from J7.

I couldn't find any comprehensive documentation about this DC/DC converter part describing how it works.
Can someone please help me narrow the problem more? I would be also thankful for a functional description about this part for dummies (as I am originally a software developer).

Thank you,
Gergely

over 8 years ago

0 Charles Tsai over 8 years ago

TI__Guru**** 191886 points

Hi Gergely,
So powering from either J6 or P7 don't work, right?

The DC/DC controller datasheet can be found at: www.ti.com/.../tps23753.pdf

Can you also check TP6, TP8 and TP10 and compare between the working and the non-working board?

0 Gergely Bacso over 8 years ago in reply to Charles Tsai

Intellectual 545 points

Hi Charles,

Thank you for your answer. I've been looking at the design and the document you mentioned since then and I think I mostly understand what happens, but not everything is clear...

But I just found what the problem is: U3 wasn't soldered well and by pushing it a little the circuit began to work. (Huh...)

Thank you for your help,
Gergely

0 Charles Tsai over 8 years ago in reply to Gergely Bacso

TI__Guru**** 191886 points

Glad your problem is solved!

0 cb1_mobile over 8 years ago in reply to Gergely Bacso

Guru 117855 points

One note of caution - depending upon the "pin(s) w/failed board connection" - and the "duration of operation" while "failed"- that IC may have been damaged. The fact that it (now) appears to work may not mean that it has escaped (some) damage - which may give rise later - & downstream - when "far from & outside" your control.

Is it not possible that some "internal chip issue" caused the "lifted pin" due to "localized" chip heating? (firm/I have noted such - especially w/power switching/converting ICs.) You (really) do not know if yours was a "board manufacturing issue" - or - a chip issue - do you?

Best practice for such boards is to tag them - identifying the IC & pin numbers w/"issues" - and (ideally) replace that IC - or confine board's use to "local & uncritical" applications, only.

0 Gergely Bacso over 8 years ago in reply to cb1_mobile

Intellectual 545 points

Thank you for the warning. These PCBs are only prototypes so they are not safe-critical at all.
Yes, you cannot say for sure that it was a manufacturing issue, but it works now, and the chip isn't and wasn't heating (I've checked that), so I suppose a manufacturing issue has bigger probability.

0 cb1_mobile over 8 years ago in reply to Gergely Bacso

Guru 117855 points

Good for you - hold (that) board close - and do not speak of "manufacturing issue" too loudly... (I just barely - heard you...)

0 Gergely Bacso over 8 years ago in reply to cb1_mobile

Intellectual 545 points

It was soldered by hand, so it's not that critical I guess ;)

Arm-based microcontrollers

Arm-based microcontrollers forum

TIDM-TM4C129POE: TIDM-TM4C129POE DC/DC Converter Issue