Shutdown RTC current

Status
Not open for further replies.
C

cinderblock

Member
Hey all,

I'm wondering if anyone has an idea of the RTC battery current consumption when the rest of the teensy is completely unpowered. (I'm using a Teensy 3.2 currently but if someone has this information for the other chips, I'm sure someone would find it useful)

I'm trying to size a coin cell to keep the RTC alive for some period of time.

I have some measurements but I don't trust them as my meter's accuracy at such low levels is questionable at best. It also has like no resolution at that low range.

I've found the power consumption operating behaviors table from the K20 datasheet. However, it is not clear to me which numbers are valid for vbat. The two relevant rows:

Teensy low power.png
Teensy low power.png

I think it's pretty clear that the "with RTC [...] disabled" is not what I'm interested in. Is this just leakage current on VBAT when RTC is not running?

I'm guessing the "Average current when CPU is not accessing RTC registers" is the rating I'm interested in. However, the "when CPU is not accessing RTC registers" is unclear to me. Is this the datasheet trying to say, if VDD is unpowered, this is the expected draw on the RTC battery?
 
I did a quick test just now on Teensy 3.6, which has the 32.768 kHz crystal. I programmed it with the Time library TimeTeensy3 example.

It's drawing 3.2uA on the VBAT pin when I unplug the USB cable. This is the current needed to keep the 32.768 kHz crystal running and the RTC actually tracking time. Part of that 3.2uA might be leakage through the schottky diode to the rest of the board. I tested with a multimeter connected in series with the 3V power to Teensy's VBAT pad, not the actual current flowing into the chip's VBAT pin.

I plugged the cable back in and confirmed it has kept track of the time.

This test was done on a single Teensy 3.6 at room temperature.
 
Quick update... I repeated this test. Looks like the current is actually 4 uA right after unplugging the cable. It quickly decays to about 3.2 uA, which seems to correspond to the chip cooling down. Then it keeps very slowly decreasing and seems to eventually end up at 2.5 uA.
 
Interesting. This is at 3V?

If the chip cools down to room temperature, I'd expect closer to, if not under, 1uA of draw, according to my understanding of the datasheet. Is the chip internally getting that hot and making it draw more current? I guess it could just be leakage from the diode... I don't have a part number for that so cannot check what it claims.

For reference, here are the RTC current draw table rows for the new Teensys (3.5, 3.6)

Teensy low power.png

Teensy low power.png
 
I have measured 2.54uA for Teensy 3.6 and 2.33uA for Teensy 3.5. Yes, that seems quite a bit more than the datasheet says. They may have used a crystal with less power consumption, but I still wouldn't expect that big a difference...
 
Status
Not open for further replies.
Back
Top