TMP1075: No response from sensor on I2C

Makarand -

Welcome to E2E and thanks for posting! 

In this case, we would need to see (to confirm) the oscilloscope or logic analyzer capture of the I2C traffic to comment further. If you have schematic to share, that would also be appreciated. 

Josh, 

Thank you for your reply.

I have attached the picture of the initial I2C signal and a snippet from the schematic for your inspection.

Thanks!

Hi Makarand,

We tested I2C on a good part and have a few questions to clarify:

1. In the schematic, the P3v3 line provides power to the TMP1075. Is this line active with 3v3 while I2C communication is being attempted ?
2. Has visual inspection been conducted for proper electrical connection ? Is it possible to reflow the board to reflow the solder in case there is a bad connection ?
3. How many devices has this been observed on ? Is it only occurring on one board or a few other devices ?

The timing on your oscilloscope capture looks fine. Here is a reference for a good waveform:

Thank you,

Chris

Hi Chris, 

Thank you for your reply.

I have got about 7 boards the sensor has been mounted on. 

To answer your questions, I could check the 3v3 flowing to the pad and that looks okay on all the boards. Checked the comms on all the boards and all seem to have the same issue. 

Soldering looks okay to me, here's a picture attached if your are able to find anything, I apologize though; it is a bit unclear for the package. 

Thank you!

Makarand

Makarand - 

Thanks - is it possible to take a more focused picture? From what it looks like from what I can see, it appears this circuit has been touched up manually, and may have cold or open solder joints - a clearer image would help double check that. Did you already Ohm out all the connections, to the top of the parts to ensure the connection through the pads and solder fillets, including the passive parts?

Josh, 

Thanks for your reply. 

I agree that I have made some additional connections on to this particular circuit (probing the I2C signals) which is why you can see some manual rework. 

However, I am observing the same behavior on 5 other cards. All have been machine assembled and reflowed as we have got other fine pitch parts on the board as well. 

I will still try to get a better picture for you and the ohm values as well. 

Other than that, what are the other probable reasons that you can think of for the sensor not replying?

Makarand

Makarand- 

no connection would be a great reason - please check it and send over another image, if possible - thanks!