Power Management for a RS 232 sensor

Dev Kit Support
1

Hey there,

I am currently integrating a Pro Oceanus CO2 Pro CV sensor using the dev kit. The sensor has pretty high power requirements during the warm up phase (very close to 10W ) and I can’t get it to power on using the spotter.
The sensor is set on continous measurements and automatically powers on when it receives power. The sensor powers on without any issue when I connect the dev kit to a power outlet but not when I connect it to the spotter directly.

I have tried to put full batteries on the spotter, let it be powered by a power outlet.

I have turned off the spotther blinking light ( Another way to save power is to turn off the visibility LED, or lengthen the portion of its blink pattern when it is OFF.

cfg vle 0
cfg save

I have enabled Power Save Mode, and set the dwell time between wave sampling to 23 hours as described in External power to spotter buoy

cfg glpmen 1
cfg dwl 1380
cfg save

To keep things simple I have put the stock serial_payload_example code. I get this on the bm mote serial when connected directly by a power outlet :

When i power the sensor using the direct power to the bm mote i get this message and i can clearly hear the sensor going on and the message is that it is warming up :

[payload-line] | tick: 33152, rtc: 0, line: CO2-Pro Warmup - W 33.5
[payload-bytes] | tick: 34146, rtc: 0, bytes:43 4F 32 2D 50 72 6F 20 57 61 72 6D 75 70 20 2D 20 57 20 33 33 2E 37 0D 0A
[payload-line] | tick: 34156, rtc: 0, line: CO2-Pro Warmup - W 33.7
pressure | tick: 34198, rtc: 0, temp: 25.049734, pressure: 1023.571899
power | tick: 34205, rtc: 0, addr: 67, voltage: 24.060801, current: 0.331750
power | tick: 34207, rtc: 0, addr: 65, voltage: 24.100800, current: 0.339000
[payload-bytes] | tick: 35150, rtc: 0, bytes:43 4F 32 2D 50 72 6F 20 57 61 72 6D 75 70 20 2D 20 57 20 33 33 2E 38 0D 0A
[payload-line] | tick: 35170, rtc: 0, line: CO2-Pro Warmup - W 33.8
[payload-bytes] | tick: 36164, rtc: 0, bytes:43 4F 32 2D 50 72 6F 20 57 61 72 6D 75 70 20 2D 20 57 20 33 34 2E 30 0D 0A
[payload-line] | tick: 36174, rtc: 0, line: CO2-Pro Warmup - W 34.0
[payload-bytes] | tick: 37168, rtc: 0, bytes:43 4F 32 2D 50 72 6F 20 57 61 72 6D 75 70 20 2D 20 57 20 33 34 2E 32 0D 0A
[payload-line] | tick: 37188, rtc: 0, line: CO2-Pro Warmup - W 34.2
[payload-bytes] | tick: 38182, rtc: 0, bytes:43 4F 32 2D 50 72 6F 20 57 61 72 6D 75 70 20 2D 20 57 20 33 34 2E 33 0D 0A

We can see indeed a significant power draw : voltage: 24.100800, current: 0.339000

When i tried the same thing using the spotter i get this :

[payload-bytes] | tick: 418523, rtc: 2024-08-30T18:00:06.699, bytes:53 65 6E 73 6F 72 20 6F 6E 2E 2E 2E 61 63 74 69 76 65 20 73 61 6D 70 6C 69 6E 67 20 64 65 74 65 63 74 65 64 0D 0A
[payload-line] | tick: 418543, rtc: 2024-08-30T18:00:06.718, line: Sensor on...active sampling detected
Set RTC to 2024-08-30T18:00:10.777
hum_temp | tick: 422189, rtc: 2024-08-30T18:00:10.367, hum: 47.468410, temp: 25.033833
pressure | tick: 424198, rtc: 2024-08-30T18:00:12.375, temp: 25.036638, pressure: 1023.640869
power | tick: 424205, rtc: 2024-08-30T18:00:12.382, addr: 67, voltage: 19.795200, current: 0.012250

or this :

power | tick: 34205, rtc: 2024-08-30T18:12:52.121, addr: 67, voltage: 23.961599, current: 0.000750
power | tick: 34207, rtc: 2024-08-30T18:12:52.125, addr: 65, voltage: 23.953600, current: 0.007750
[payload-line] | tick: 37831, line: Sensor on...active sampling detected
Set RTC to 2024-08-30T18:13:00.636
hum_temp | tick: 42189, rtc: 2024-08-30T18:13:00.109, hum: 45.880245, temp: 26.160017
[payload-line] | tick: 42831, line: Sensor on...active sampling detected
pressure | tick: 44198, rtc: 2024-08-30T18:13:02.117, temp: 26.156275, pressure: 1023.467285
power | tick: 44205, rtc: 2024-08-30T18:13:02.125, addr: 67, voltage: 23.959999, current: 0.001250
power | tick: 44207, rtc: 2024-08-30T18:13:02.125, addr: 65, voltage: 23.953600, current: 0.007500

There is some power draw but very little : could it be simply a message from the mote whenever it detects something connected (there must be a tiny bit of resistance just due to there being wires from the sensor to the bm mote right )

The message " [payload-line] | tick: 42831, line: Sensor on…active sampling detected " disappears when I disconnect the RX or TX or the power to the sensor. It does not mean that the sensor indeed is on as it doesn’t make the sound of it warming up.

Regarding the sensor’s power budget :
I am really trying here to find a solution without using an internal battery and making the system work like this.
The average power consumption during warmup for the CO2-Pro CV is 7.3 W
(610 mA @ 12 VDC) with no pump attached.

here is a graph of the power demands during warm up, the peak power demand at the beginning is very close to the 10W limit.

image

Also on the Sofar API, I get this : Could changing this to more conservative value make the application work ?

Satellite Spotter Data

||Mode|WAVES: STANDARD 
|Update Rate|1 Hour 

Smart Mooring

Sampling Configuration
Samples Per Message 2 Samples
--- ---
Sample Interval 20 Minutes
Sample Duration 5 Minutes

Sensor 1 Unknown sensor

Cellular Spotter Data : HDR enabled, Lake/Bay, processing rate 15 min

Is there a way to make this work ? i have tried to simplify the serial_payload code to have as little power requirements as possible (disabling the led, disablimg the transmitting of data etc) as I have read that the Spotter + BM system will consume 250mW to 300mW in default configuration. As it is for a prototype, I can just use the spotter as a power source and the sensor logs the data internally and handles everything internally. Is it possible to go past the 10w limit ?
I am aware that there are issues arising whenever there is continuous sampling but I want to test it for now for short duration testing.

@zack_j there seems to be similarities to the topics you answered : Sonde drawing too much? and Power management questions

Thank you very much in advance for your help !

Renan

2

Hi @renanm - You have the wall wart plugged into the orange Dev Kit board to one of the black Molex connectors, and then the sensor being powered from the Dev Kit board, correct?

  • Are you using the VOUT power pin, or the VBUS power pin from the Dev Kit?
  • What is the minimum and maximum input voltage of your sensor?

I think it is likely that you’re exceeding the Spotter’s 3W max continuous power delivery to the Bristlemouth bus (Bristlemouth is rated for 10W, but Spotter can only do brief 10W peaks < seconds, not for minutes).

We can advise on troubleshooting to confirm this is the case, but if it is I don’t have any immediate ideas to work around it.

I’m tagging @sldiehl in who may be able to fill in more details.

Here’s a PDF she presented at Bristlecon that explains the details about power monitors 65 and 67, and has other related deeper technical info as well: Notion – The all-in-one workspace for your notes, tasks, wikis, and databases.

3

Hi, thank you very much for the in-depth reply.
I am using the VOUT power pin.
The input voltage is between 10-18 VDC

Given that the power requirements are not met and the sensor ideally requires a pump that draws additional power, there doesn’t seem to be any way currently to power it directly.
I managed to make it work using a small 18V battery that I fit inside the pressure housing, and the sensor can come with an external battery.
My aim was to investigate how to use the spotter to deploy the sensor for long duration missions. This would probably require to do it properly using the Spotter 3W continuous power delivery to charge the battery, but it seems a bit complicated at the moment given the inability at the moment, if I understand correctly, to transmit to the mote information on battery status and solar charge from the spotter.

I will share in a few days days more info regarding the integration of the Pro Oceanus C02 Pro CV sensor in a separate post for anyone interested in carrying it out !