Remote Battery Voltage Monitoring
Good Morning,
I've got my Kiwi setup at a remote location and was wondering if there was an inexpensive way to monitor the battery voltage which is currently charged via solar. I know there are out of the box solutions for this as I used work in the cellular industry but those boxes were quite expensive for what I need. Unfortunately, my solar controller doesn't have this ability.
I saw a video online where you can read voltage from the BBB but I'm unsure if I can use those pins because of the daughter board location.
Does anyone have a similar setup that could offer any advice?
Thanks!
Comments
I designed and built an APRS/solar_charger for monitoring the PS at a remote KiWi site. It's not quite what you are asking for but since the Kiwi data is IP over microwave, the use of out-of-band monitoring means I can see the power situation and termperature even if/when the microwave link is down.
You can see the present condition of N6GN/K2 this way
If you are interested, write me.
Glenn n6gn
I have some FrSky Smart Port voltage and current sensors, they output 3.3V serial data (inverted).
https://code.google.com/archive/p/telemetry-convert/wikis/FrSkySPortProtocol.wiki
The only thing stopping me using them is the dense bit of wet carbon between the keyboard and chair.
Perhaps you can build an LC-oscillator using a varicap, so its frequency becomes voltage dependent. Then you could monitor its frequency on the kiwi receiver, for example using the 27 MHz ISM band.
Although intended for monitoring mains voltage, I did a somewhat similar thing using the ESP8266, described here:
The source code is linked, and making it read supply voltage rather than mains voltage would be trivial. The '8266 has only a single A/D input, but the addition of a simple analog MUX chip like the 4051 would easily expand this to 8 inputs - although one would probably want to level-convert the MUX drive to 5 volt logic (e.g. an NPN and two diodes).
* * *
If a network connection isn't in the cards (the ESP8266 has 2.4 GHz WiFi only) this also includes a Morse beacon that telemeters the information - in this case, on 10 meters: This was added so that telemetry could be seen by anyone interested, but also so that we could still receive it should the WiFi connection drop for some reason.
* * *
Finally, the same, basic low-power TX could be driven by a PIC or Arduino, of course to produce Morse telemetry.
To get back to an aspect of your question: I suppose that a BBG GPIO pin with an A/D input would be available, but I chose to leave the KiwiSDR unit unmodified - both for practical reasons (a spare could be swapped out) and also to avoid the possibility of conducting undesired signals in/out of the chassis (e.g. static, RFI, lightning, etc.)
73,
Clint, KA7OEI
Thanks Clint,
That's similar to what I'm looking to do and being able to monitor the shed temp would be a bonus as well. I have an ethernet connection but I'm not really interested in setting up wifi although I could if it boils down to it and have been looking at charge controllers that have RS232 output that can be converted to eth via a converter/server. That seems to be more my speed than Arduino at this point in time. Haha I've never done anything with the ESP8266 but it doesn't seem too difficult. It would be handy to be able to measure the bank voltage, solar array input voltage, current readings, etc... So I would need more than just the one input.
I'll check it out.
73!
Steve KU4BY
System telemetry is as old as spaceflight, probably older. Surely the low-bandwidth data involved could be multiplexed onto any data channel, or on a special admin-only dedicated channel?
Given the low cost of sensors, some key electrical parameters would seem to be a worthwhile fit. For a few dollars more, we could have temperatures (multiple), pressure & humidity too - not a bad idea for any kind of equipment mounted outside the normal domestic environment, or even within.
That's why I chose APRS as the data channel. It stays up when the Kiwi system is down, has built-in worldwide distribution and can be done with cheap hardware within amateur radio. With the Arduino as controller a whole host of sensors are similarly cheap and easily added, should they be desired. As it was I only added voltage and temperature sensors but an upgrade is almost trivial. It transmits only one short APRS packet every few minutes.
Also because the whole thing was Kiwi-centric and meant for a quiet remote site, the solar charge controller is very low-rate PWM so doesn't generate RFI virtually all commercial solar charging systems seem to do, even the 'RF quiet' ones.
Initially there was a CW ID on a local 10m crystal sending the battery voltage (only) but that took almost as much hardware as the APRS/charge-controller solution.
"Perhaps you can build an LC-oscillator using a varicap, so its frequency becomes voltage dependent. Then you could monitor its frequency on the kiwi receiver, for example using the 27 MHz ISM band." (HB9TMC)
Haha, that principle was also my very first thought while reading KU4BY's opening post. But I didn't dare to post it as it seemed too weird to me... 🤣
Of course, it would be quite simple, but probably not that accurate as digital telemetry. And if the kiwi should stop working due to undervoltage, it would be no more possible to get any information at all. Another disadvantage would be the need of manual reading; it would be difficult to implement some kind of automatic warning. So overall only of limited use - but still a nice and ham-like idea!
That's an interesting idea but I don't think it would be accurate enough for what I'm wanting to do and I'm also wanting to monitor multiple sensors. Plus the addition of even a small amount of RF near the Kiwi seems to dampen the receive. I had the receiver about 200 feet from my house and when I keyed up 5W of JT65 at the house it killed the Kiwi's RX altogether.
I just found out that a friend of mine is doing something similar to what I am attempting by using a raspberry pi and an MCP3008 ADC along with some sensors. That should give me 8 channels if I read correctly. So I'm seeing what I can do with something like that now.
73!
Steve KU4BY
All these ideas for external sensors, processors and data links are surely possible (for you clever guys), but my sloppy contribution was meant to refocus on the existing BB hardware and Kiwi IP channels with the minimum of additional sensors.
I'd love to get some comment from John, firstly whether he could adapt the software to provide a comprehensive and flexible service, then whether existing BB IO has capacity enough, finally how interested he would be advising on sensors and interfacing.
Just one other thing that occurs to me and has been discussed on here before, what other kit is used on the solar battery voltage and does any of it already have voltage monitoring?
If you have a router or wireless bridge it might be possible to swap that out for one that runs a wide voltage range and has voltage monitoring.
For example if you had say a wireless bridge running from a fixed supply replace it with one that runs the required range with monitoring. I'm sure there are a bunch but Mikrotik or Ubiquiti would be good to look at. A 30-second search gives Mikrotik RBGrooveGA-52HPacn, 11-30V with monitoring. The UBIQUITI BULLET5 is good for up to 24V and I would be very surprised if they don't show voltage (I've got an old 2.4GHz one I'll try and check).
On the previous thread I was looking at the advantages of POE out control for rebooting network devices but on the "auto" feature I did run into some issues with the input voltage swinging widely due to solar disrupting the output current monitoring (on a POE fibre to Ethernet switch). As long as you only want to manually turn devices off and on it works fine.
Stu
Yes Stu, there are solar chargers out there already that do what I am trying to do but I am being cheap about it as the sole purpose of my solar system at this point is to power my kiwisdr.
I opted to go with my Raspberry Pi B model and an MCP3008 ADC. There is a ton of information online showing the many different types of sensors. I've got it running now and when I run the python script I get output like this:
***********
Battery Voltage: 0.00V
Solar Input Voltage: 0.00V
Sensor 3 Voltage: 0.00V
Sensor 4 Voltage: 0.00V
Sensor 5 Voltage: 9.47V
Temperature Reading: 1023.00
Sensor 7 Voltage: 3.09V
Sensor 8 Voltage: 2.82V
***********
Sensor 5 and 6 are the only 2 sensors in use right now. 5 is hooked to a 9V battery and I haven't figured out the code for the temperature sensor yet. I'm no programmer. Channels 6 & 7 are open pins not connected to sensors and I suspect that is why I am seeing bogus info on those pins.
Eventually I might even figure out how to send the data over to a self hosted webpage but for now I'm gonna get a few current sensors, figure out the temp sensor, and maybe even try to learn Python...😂
Steve
@n6gn Can you give any insight on power usage/needs for your Kiwi setup? I can easily measure what my Kiwi uses, but curious about solar panel size, battery size, charge controller, microwave tx power usage, etc.
What I'm after is trying to figure out actual needs (and then costs) for a fully off-grid/remote kiwi. I'm still kicking around that thought after living with my Kiwi for a while at my transmitting site... I could move it 3/8-1/2 mile away, which would be away from all RFI, too...but there's a lot of guessing on actual power consumption.
-Nate
N8BTR
Nate,
I have a long running project where I am trying to power my two FCC Part-15 beacons from Solar/Battery Power. Until this year, I have had little success getting thru December, let along the rest of winter. This year I am having better success after switching to LiPoFe4 batteries.
I have learned that the most important factor in doing this is to determine if you have enough sun power at your site to even consider Solar. In my location I have immovable trees and forests that block the direct line to the sun until 10AM or so, and another block to the west where at 2-3 PM the sun gets cut off. If your thinking of Lead-Acid chemistry that is simple not enough time to fully recharge that type battery. Even with unlimited AmpHours.
My "Hifer Beacon" transmitter outputs a little less than .004 Watts. The Exciter, GPSDO, Solar controller, and battery monitor consume about .070 mA (24/7). I am running a 12 Ah battery to provide some backup for cloudy days and a 25 Watt solar panel to supply the power. It's doing ok for now, but I may need to add more solar watts. My Battery monitor reports to a webpage where I keep track of it's performance. It is open to the public at https://thingspeak.mathworks.com/channels/2729463. Today (12/20/24) has provided no sun power, and my power reserve is being used up.
The investment needed in the solar/battery equipment, to supply the 1.5A (24/7) needed by the kiwi, will be several magnitudes over my outlay. I hope that gives you some idea of a small scale, stand alone, power system.
Feel free to contact me directly if you have more questions.
Mike N8OOU 73
If I remember right, my remote+active_antenna+uwv_link is around 12W total with kiwi a bit less than half of it.
Lots of places I think total storage can become the biggest deal but depends on days of sun/year etc.
Nate,
A Solar Power Controller will add an additional load to your estimate. If we round your 12 to 13W for total load, times 24 hours per day means you will need a battery capacity of 312 Wh, to run for one day without sun. If you need to run multiple days, the battery capacity jumps accordingly.
Now, the next day your Solar panels will need to provide 321Wh in the window that you have sunshine, to replenish the power removed from the battery the previous day. Plus, it will need to supply the power to run the Kiwi during the recharge window. It does not take many clouds to quickly degrade power out. A 1/4 inch layer of snow on the panels, stops all power generation. If you have a 3Hr window, a single 100W panel (300Wh) is not enough capacity. You likely will need 2 or 3 times that size.
My 100W panel on my Lowfer beacon meets it's specs in full sun at noon during the summer. However in December sun, it is only making 50W at noon. My Lowfer beacon uses about 75 Whr per day. The 100W panel is barely keeping up on full sun days. On consecutive cloudy days the battery does not fully recoup it's charge and after a couple days it will shut down.
I'm sure I have left something out of this summary. I hope is has helped you in deciding your starting configuration.
Mike N8OOU 73
@n8oou and @n6gn, Thanks for your thoughts and info! 12w is good to know. I figured 4.5 watts to keep the Kiwi going all the time, plus any voltage conversion losses.. My current Wifi link for it is known to be low power, 1.3 watts is one measurement claim I've seen, 0.85 for another. (TP-Link TL-WR802N ) But to get 2000 feet through some trees, I'm thinking I'd be better off with one of the new 900mhz wifi links, and no idea what power consumption on those looks like.
Fair point in the huge variance in storage needed, and panels required depending on location. I have a fan that I've had running 24/7 with a lead acid battery, and powered by solar since 2016. My cabin I'm still slowly building was almost was off grid, so I had been doing a lot of calculations and experiments on power usage, needs, etc 5-6 years ago before I was able to get the property on grid...so I'm fairly familiar with the math, and sun amounts I get in winter, etc.
Math time: If we're looking at say 10-12w 24/7, that's 264 wh/day.
In my area, I think I'd want at LEAST 5 days of battery capacity. I'd want at least 20-30% extra capacity beyond that to reflect what's usable... So something like 160AH of 12v battery.
In my area, I can count on maybe an average of 1 peak sun hour/day production average in the worst time of year, or even a bit less. Plus charging inefficiencies... So I need at least 300w of panel... But then I'd likely use a PWM controller, @n6gn, looks like you've got a Morningstar setup. I'd likely use something from them, possibly one where you can move the PWM switching frequency to under 1hz. So since the PWM controllers just cut the voltage, and many "12V" panels are something like 18v rated, I'd probably need a minimum 400w of panel, maybe 500.
400 watts panels: $300 USD
Controller: MS Prostar 30A PWM: $170
Battery: something like $200? 250?
Wifi: $100
WIfi Antennas: $150
New Kiwi receive Antenna: $200
Enclosures, mounts, cables, wires, voltage conversion, ferrites, etc etc etc: $500?
Something like $1600 USD to get my Kiwi completely remoted to a site 1/2 mile away, plus the time and effort. OK...not cheap, but not insanely expensive either.
-Nate
N8BTR
A couple of addenda:
Solar controller power seems not to be a big deal as it only requires energy during periods of charging when there is a lot to be had.
Our experience with snow at 7500' elevation in the Colorado Rocky Mountains is that while it can occasionally indeed cause a problem, by the time that sun returns the tilt angle of the panels along with the presence of sun seems to solve the problem on its own in moderately short order. Altogether it hasn't been too much of a problem.
Adequate storage which can withstand very cold temperature is more of an issue and expense.
Keeping critters and lightning strikes from doing major damage is also a real concern where our remote is located.
Lower charger switching rate could help but moving to a high CMRR system like the Single Antenna could probably avoid the bit of QRN we see during sunny days. This is usually only present when absorption is so high that the lower portion of the spectrum is much less interesting to us.
I've seen similar things on the charge controller side of it...power consumption of the controller seems to be minor in the scheme of multi-watt loads and panels above 20 watts or so.
Same with snow and batteries... Lithium is great, but not much of it is optimized for sub freezing temps yet still on that side..I'm still using lead.
Interesting on the controller QRN..hadn't thought about it much, but makes sense it'd be worst during absorption, which generally the "best" time to interfere with the low end anyway.
-Nate