Author Topic: mcModules goes cold (Read 2465 times)

kristofferis · « **on:** June 30, 2016, 04:55:50 pm »

Monitor a freezer is something that many people is interested in.
So now it's time for a mcModule to go for a ride in the cold.

I have a simple code that reads the temperature and the battery voltage once every minute and publish to MQTT.
The data is then logged to a text file so that i could see the battery voltage with a time stamp.
For this test case the battery is LS14250 3.6 Voltage, a Lithium Thionyl Chloride 1/2 AA battery that I think should work really good in the freezer.

I will update this post with some images tomorrow.

Right now (20 minutes in the freezer) the voltage of the battery is 3.589 and the temperature -21.3

kristofferis · « **Reply #1 on:** July 01, 2016, 02:05:51 am »

Here is some result from the first night in the freezer.

Start: 3.596 - 26.1
1h: 3.564 - -20.6
2h: 3.557 - -20.5
3h: 3.554 - -20.5
4h: 3.554 - -20.5
5h: 3.554 - -20.4
6h: 3.550 - -20.3
7h: 3.550 - -20.1
8h: 3.550 - -20.0
9h: 3.550 - -19.9

You could see that the first hours the battery was getting colder and colder and it was losing some voltage at higher speed but it seems to have stopped and it looks like that the battery is handle this very well.
I will continue to monitor this and post the updates here.

Right now the battery and the module is just soldered together and is laying in the freezer, if this works out well then I will make some sort of case for this so that it is protected in the freezer.

mc-John · « **Reply #2 on:** July 01, 2016, 06:44:23 pm »

This is great.
We are doing a long running test with the "beer" fridge but that is much easier on the batteries.

Good to know that the LS14250 is working

kristofferis · « **Reply #3 on:** July 02, 2016, 03:42:23 am »

Ahh that sounds interesting, its always good to make a couple of long runs to see what happens.

Here is some new result from my freezer.

12h: 3.547 - -20.6
16h: 3.540 - -21.0
20h: 3.540 - -19.6
24h: 3.540 - -19.8
28h: 3.540 - -20.9
32h: 3.540 - -20.9
36h: 3.540 - -21.0

It still look very good.

But i had something strange going on last night. The module is reporting every minute but for some reason the module did not send anything for 12 minutes
23:20:04 - 3540 - -19.8 and next report where
23:32:11 - 3213 - -20.8 and there were 4 reports at same time so the module had filled the send buffer but if you look at the voltage the module have done some strange thing, after this then the battery voltage is stabilizing.
23:32:11 (second message from buffer) 23:32:11 - 3526 - -21.0
23:33:10 - 3529 - -21.1
...
23:40:10 - 3533 - -21.3
...
23:48:11 - 3536 - -19.8
...
00:17:12 - 3540 - -20.0

So my guess is that the module must have been stucked in some kind of mode that was consuming much power.

plains203 · « **Reply #4 on:** July 03, 2016, 08:45:08 am »

I have logged the same behavior tonight with one of my modules. Was sitting stable around 2.949 and reporting every 2 minutes then no messages for 8 minutes and suddenly volts drop to 2.777 and sits stable.

kristofferis · « **Reply #5 on:** July 03, 2016, 09:54:44 am »

Thanks for reporting that, then there is some issue related to that.

kristofferis · « **Reply #6 on:** July 03, 2016, 12:16:06 pm »

I have now checked my log file to verify the battery voltage, then i noticed that last post was yesterday at 22:06 when i looked at the mcGateway the status light did not blink anymore so the mcGateway was the first thing to freeze even that its not in the freezer

Everything worked again after a reboot of the mcGateway.

This might be important, after the reboot of mcGateway then i did only receive one message not several from the mcModules queue, so i guess that mcModule had contact with mcGateway and send messages all the time but it was the ethernet/network part of the mcGateway that was stucked.

After 52 hours the battery is still on 3.540V

plains203 · « **Reply #7 on:** July 03, 2016, 07:39:36 pm »

I haven't re implemented QoS on my modules yet so not sure if the message is getting to the gateway at this point or if it is not leaving the module. I will try to update all my modules tonight and then log again with QoS and the later firmware.

kristofferis · « **Reply #8 on:** July 05, 2016, 08:09:31 am »

After 108 hours 3.543 - -21.2 still going strong

I have noticed several times every day that the mcModule is not sending the MQTT message to the mcGateway and misses a couple of minutes and every time I could see that the voltage has a drop, so something strange is happening on the mcModule.
Almost every time when the issue occurs then I could see that the mcModule is sending multiple messages from its queue but sometimes not? maybe the queue could be filled with something else and the messages gets lost?

kristofferis · « **Reply #9 on:** July 17, 2016, 07:35:00 am »

Now the mcModule has almost been in the freezer for 17 days and the battery is now reporting 3.547v

And now I am sending MQTT once every minute and maybe once every 5 minute will do in a real life scenario, so this battery is working very good.

mc-John · « **Reply #10 on:** July 17, 2016, 05:59:29 pm »

That is really good. The voltage should be stable now for a long time. The sleep current of the device is about 5 to 6. We are still working to improve the current consumption by switching off some of the memory. When the device connects to the gateway it uses much more memory. Some of our customers want the temperature every minute and they write the temperature in the beacon. The beacon requires about 20 to 40 times less energy than connection to the gateway. The beacon timing and the connection timing can be controlled in code.
The new device mcMod120 will use less than 2uA. The chip itself is better and we have more ways to limit the current of accelerometer.

-John-

mc-T2 · « **Reply #11 on:** August 02, 2016, 11:36:12 am »

@Kristofferis - Can you give us an update on how the battery is performing after a month in the freezer?

kristofferis · « **Reply #12 on:** August 02, 2016, 11:59:07 am »

Here is the requested update

The battery is right now at 3.533 Volt.
The mcModule is running the latest firmware available and Iam still using the same code that sends MQTT updates every minute.

I also have one module inside my refrigerator I use the same battery for that module and same code with updates every minute, it has been in the refrigerator for 26 days and the battery is still at the same voltage as when I inserted it to the module 3.596 Volt

mc-T2 · « **Reply #13 on:** August 02, 2016, 02:11:28 pm »

That is so great! This exact setup is where people could start to use the mc-Things platform for cold-chain monitoring and management (Startup idea!). Imagine a retailer or a cold-storage company starting to use these devices for showing their temperatures in real-time and alert if a temperature threshold is broken. Very cool!

Can i also ask what you are using for displaying the data (screenshots)?

kristofferis · « **Reply #14 on:** August 02, 2016, 04:14:33 pm »

Yes that's true, this could be used for many different temperature monitoring.

The screenshot is from Node-Red and a plugin named node-red-dashboard.

News:

Author Topic: mcModules goes cold (Read 2465 times)

kristofferis

mcModules goes cold

kristofferis

Re: mcModules goes cold

mc-John

Re: mcModules goes cold

kristofferis

Re: mcModules goes cold

plains203

Re: mcModules goes cold

kristofferis

Re: mcModules goes cold

kristofferis

Re: mcModules goes cold

plains203

Re: mcModules goes cold

kristofferis

Re: mcModules goes cold

kristofferis

Re: mcModules goes cold

mc-John

Re: mcModules goes cold

mc-T2

Re: mcModules goes cold

kristofferis

Re: mcModules goes cold

mc-T2

Re: mcModules goes cold

kristofferis

Re: mcModules goes cold