Other Parts Discussed in Thread: TMP468
Tool/software:
Following is an email we sent to a contact we had at TI, but she recommended to ask this on the E2E forum...
We have several designs that use the TMP9R00HKT/EM and TMP468 where the D- pin is connected to the corresponding D+ pin through a capacitor (1000pf) only. We have a new board version with essentially the same circuitry as the earlier versions using the TMP9R00HKT/EM but it seems to only work if D- is connected to GRD. We have been trying to find what is different from the other layouts as the schematic/circuit design is identical to the others as far as the temp circuit portion, but as of today we are unable to pinpoint any difference that explains the need for grounding D-. In reviewing your documentation for these parts I don’t see anything suggesting the need for grounding D-.
My question is if you are familiar with this situation and the possible need to ground D- in some situations/circumstances, and if you are not aware of this situation and we do have to ground D- if the root cause cannot be determined why it needs to be grounded, will the grounding of D- cause any long term detrimental effects on the chip itself or are there any adverse performance issues that need to be considered?
We are in a bit of a pinch as this was just discovered after we had a new board built for a customer and are needing to ship this ASAP, but we are leery of making this “patch” without knowing if there is a risk of degradation after the board is shipped. I would like to ultimately know what the specific reason is that requires this patch as none of our other boards need this mod, and if you or your team have any thoughts for our further exploration, it would be appreciated, but if you could help us determine if we are at least “safe” to make this mod until the real solution is discovered, we would really appreciate that also as soon as you are able to.
Thanks for any help you can give...
