DS90CR285 output jitter

Hi Vincent,

The 2015 production you refer to has a marking that is the standard TI format, and I was able to verify that they came from a lot in October 2013 from Melaka, Malaysia. The lot from your 2013/2014 production are from approximately September-October 2012 from Melaka Malaysia, and its number is based on the National Semiconductor format, so that is why they are a little different. Despite the difference in marking convention, I do not believe there was a difference in test coverage that would cause the jitter to increase in the different production.

Can you describe the increase in jitter and provide scopeshots to help us compare the previous production and current production?

Thanks,

Michael

Hi Michael,

the jitter measurement is made on the 8th rising edge of the MCLK+ output.

On 2013/2014 production, this jitter is about 1ns but it is about 2-3ns for 2015 production (see images below).

This jitter is too important for our customer who uses a FPGA to receive data and clock. The FPGA doesn't lock with such a jitter.

Regards,

Vincent

Hi Vincent,

I can check with our test team to see if changes have been made between these two different lots. Please help answer a few background questions for me.

1. Is this an issue that you see for every DS90CR285 in the new lot? 

2. For this new production part, did the customer make any changes to other parts of the board, such as connectors or FPGA setup?

Thanks,

Michael

Hi Michael,

1. Yes the jitter is increased for every part in the new lot but the jitter is not always the same. It varies from 1,5ns to 3ns (it was constant at about 1ns for the old lot)

2. there is no change in the camera board we have developed for our customer. He discovered the problem because his new acquisition board has a new FPGA which is less tolerant to jitter. But the problem comes from our camera. We tested it in different configurations (cable length, acquisition board,..) and the difference between old and new cameras is the same.

We have swapped the DS90CR285 on two cameras from 2014 and 2015 production and we noticed the jitter changed. The 2015 camera (now with an old lot) has a less jitter (1,5 instead of 2,7) and the 2014 camera has a more large jitter (1,6ns instead of 1ns). The DS90CR285 affects clearly the jitter but it seems that this is not the only source of the problem.

Best regards,

Vincent

Hi Vincent,

Thanks for the response. I checked with our product test team and no changes have been made to the test program during the time you switched to the new lot. I therefore think this issue may be application-related.

1. Can you explain more about what you did when you swapped the DS90CR285 ICs? It is interesting that the 2014 camera with the new lot DS90CR285 did not show as high a jitter as it did with the 2015 camera.
2. Can I assume that the board layout and schematic have not changed between the 2014 and 2015 camera boards.
3. What operating frequency are you running at?
4. Would you be willing to share your schematic and/or layout file for the camera board? We can take a look to see if there is anything amiss.

Thanks,

Michael