nRF9160 bottom current difference

[Jayant Tang]

Hello DevZone,

One of our key customers is preparing to move their 9160 project into the mass production phase. Recently, they have raised a couple of questions regarding the eDRX bottom current of the 9160 module:

1. They observed individual variations across different SiPs, with some samples showing a bottom current that is approximately 2 µA higher than others.

2. They also noticed that the bottom current increases as the temperature rises. While I understand that this is typical behavior for semiconductor devices, they would like to know if we have more precise data or characterization results to share on this temperature dependency.
   

Best regards,

Jayant

[Simon D-M]

Hi Jayant,

They observed individual variations across different SiPs, with some samples showing a bottom current that is approximately 2 µA higher than others.

The nRF9160 consumption is less than 10uA when idleing in eDRX (6uA -> LTE-M, 9uA -> NB-IoT). You can find the documentation about it here.

They are seeing this 2uA deviation, but what is the actual floor current? To evaluate whether this 2uA is a lot or not, the context is really important here.

This deviation has plenty of possible sources. Without more context, the most plausible cause may be the SIM card. Please check if your SIM card supports supply shutdown or if it has a low-power mode for eDRX. Also you can try to switch the SIMs between 2 devices that show a high deviation and see if the extra consumption has switched boards.

They also noticed that the bottom current increases as the temperature rises. While I understand that this is typical behavior for semiconductor devices, they would like to know if we have more precise data or characterization results to share on this temperature dependency.

Here are some temperature characterizations. These are not focused on eDRX floor current but with a bit of extrapolation you can see what to expect from a change in temperature.

And again, most likely the biggest consumption here is not the nRF9160 itself, but probably the SIM. So maybe the difference is caused by the SIM and not the nRF9160.

Best regards,

Simon

[Jayant Tang]

Hi Simon,

Thanks for your reply. The bottom current of these two 9160 SiPs are 6uA and 8uA.

I‘ll ask them to exchange the SIM card and test again.

Best regards,

Jayant

[Jayant Tang]

Hi Simon,

The customer has exchanged the two 9160 SiPs on 2 PCBAs. The bottom current follows the SiP.

So this may be individual deviation. Do we need Failure Analysis? Or this is normal?

Best regards,

Jayant

[Simon D-M]

Hi Jayant,

I'm not sure actually... What ratio are affected ? Out of all the devices, do you see a gradiant of current consumption between those 2 values ?

I made a Jira about it, but I didn't get any answer yet. It probably takes a bit longer than usual because many people are in holiday right now.

I'll keep you updated as soon as I get a reply (I'll keep you updated in the first half of next week)

Best regards,

Simon

[Simon D-M]

Hi Jayant,

I don't really have a proper answer yet, but it looks like the first feeling of R&D is that you shouldn't be too worried about it. It seems like unit to unit variation is normal.

Keep in mind that this is just a first guess, and we still need to check further.

Best regards,

Simon 

[Jayant Tang]

Hi Simon,

The customer replies that most of their devices are between 7uA to 8uA.

If they are individual deviation, the customer wants to know how much will the deviation be. What is the largest deviation.

Best regards,

Jayant

[Simon D-M]

Hi Jayant,

We don't have official deviation data, as we don't have such test in production.

However, it looks like the unit to unit variation is still within the expected range. So you can tell the customer that this is OK and as expected.

The only test we have that is in the same uA range is the production is concerning "Sleep_Cuerrent" where the variation is typically from 0.8 to 3uA (from a mean value of 1.4uA. So much more than the 33% difference they are seeing).

Best regards,

Simon