How Solar Spotter Learns Your Rig
Every rig is different, and the differences are the ones nobody publishes: what your blanket really does at the angle you actually set it, what your fridge really pulls on a warm night, what a fortnight of dust costs you. Energy History is the app watching your rig do its job, and getting better at describing it.
What it is actually watching
Once a compatible Bluetooth battery is paired, Solar Spotter keeps a quiet record of each day: what went into the battery, what came out, and how that sits against what the sky could physically have delivered where you were. There is nothing to fill in and nothing to remember to press.
It is part of Solar Spotter Pro, and it runs off the Battery Bluetooth link — also Pro, and still in BETA while the supported list grows. Without that link there is nothing to record: every figure comes from the battery's own counter, not from anything you type in.
That last comparison is the part worth having. Plenty of things can tell you how much charge you got. The useful question is what you got against what was on offer - and that is the number that can turn into advice.
Two checks a day is the whole habit
Your battery does the counting in its own hardware, all day and all night, whether or not a phone is anywhere near it. So you are not being asked to monitor anything — you are marking the start and the end, and the battery fills in everything between.
That is why a day away costs you nothing. Check at breakfast, walk out of range for nine hours, check again back at camp — those nine hours are fully measured. The battery counted them on its own; your two readings tell the app where the tally started and where it finished.
What actually costs you data is the ends. The hours before your first check and after your last one is the part nobody can account for, so the wider apart the two checks, the better the day counts. First thing and last thing beats twice at lunch.
Two deliberate checks also beat leaving the app open. Both phone systems stop taking readings the moment the screen locks, and leaving the app running behind that lock changes nothing — so a phone in a pocket records almost nothing and flattens itself trying. And if charge arrived while you were away, the main screen's battery row says so when you reconnect. On a 100 Ah bank it reads +12 Ah (+12%) charged since your last reading · ended 3h ago. Nothing goes missing.
Days you camp, not days on the calendar
Solar Spotter learns your rig from the days you actually use it — not from the calendar. Camp twice a month and it takes longer to build a picture, but you get the same answers in the end.
This matters more than it sounds. Count calendar weeks and a full-timer gets thirty readings a month while a weekend camper gets four - and the weekend camper is handed conclusions built on almost nothing. Counting the days the rig was actually out means you both get the same quality of answer. One of you just gets there sooner.
It appears after seven days of real use: roughly a week if you are touring, three or four trips if you get away on weekends. Seven different dates with a recording on them, counted across your whole record. Nothing counts down at you while you wait — the first you hear of it is Your energy history is ready, and once it is open it stays open.
Honest from day one
In its first days the record hasn't seen enough full days to quote a daily average — so it could stay silent for a week, or it could guess. It does neither. From day one, the header shows the one number those early days can honestly support — a rate, avg draw ~14 Wh/h while connected — computed from exactly the hours the battery was watched, never stretched into a daily figure it did not see.
Once most of your recorded days are measured right through — about day five of steady use — it hands over to the real thing: avg draw 307 Wh/day. The unit is the promise. While connected tells you how much of the day the app actually saw, and a per-day figure prints only when the record can stand behind it. Days it only partly watched say so on their own row — battery connected 16h of 24 — and sit out of the daily averages, because under-counted draw flatters your battery.
Those part-days show rates instead of bars — draw ~65 Wh/h beside charge ~18 Wh/h — because a six-hour total drawn on the same scale as a whole day would read as a catastrophe. The rate is exact for the hours it covers; it just is not a day. And where a day can still be rescued, the row says so.
Today's short row carries → connect again before you turn in. Yesterday's, if the closing reading never arrived, carries → pending — connect to your battery to finish this day — the hours are sitting in the battery's own counter and one reconnect hands them over. Older rows get neither, because a day that has closed short cannot be repaired and a line saying so would just be a reproach.
Why it stays quiet
Some readings stay quiet until there's enough to compare against. That's deliberate: a guess off three days is worse than no answer, so the app would rather say nothing than tell you something it can't stand behind.
A blank where a number could have gone is a decision, not a fault. The alternative is a confident figure built on one unusual evening - and out where it matters, that figure has you planning a night the battery cannot actually cover. Silence is the honest answer until the evidence is there, and it fills itself in as you use the rig.
The days that sit out
Days when the panels weren't out, or when you were driving and topping up from the car, sit out of the solar comparison — there's no way to tell what your panels did on those days. They're still recorded, and still count for everything else.
A battery monitor reports what is happening at the battery. It cannot tell you which wire the amps came down. So on a day you drove four hours with the DC-DC running, the charge is real and it is recorded - it just cannot be used to judge your panels, because most of it was not theirs. Rather than guess at a split, the app leaves those days out of that one comparison and keeps them for everything else. How charge gets credited to the right source is its own story.
Same for a day the blanket stayed rolled up. Nothing to measure, so nothing claimed — the row simply reads No panels out, and the day still counts toward the seven, because the battery was measured either way.
Where your charge came from
Above the daily rows sits a strip headed Where your charge came from: one thin bar split three ways, with the shares written underneath — 62% sun, 31% driving, 7% other charging. At camp that answers a real question. Do I need to move the car today, or will the panels cover it?
The split is worked out against the same ceiling as everything else here: the most a cloudless sky could have put into your panels, where you were, at that moment. Charge up to that line could have been the sun, so it is counted as the sun. Charge above it could not, so it is counted as something else. Cloud only ever takes away, which is what makes that line safe to lean on.
Of the charge that was not the sun, the app separates out only the part it can prove: the rig was moving, so a DC-DC charger was running. Everything left over is other charging, and that vagueness is deliberate. A battery monitor reports at the battery. It cannot tell a generator from a mains hookup from a drive it never saw, and naming one would be inventing it.
It stays silent until three days qualify, and hedges until seven. The count sits at the end of the line — 4 days so far early on, plain 9 days once it is settled — so you can weigh the figure yourself rather than being told how much to trust it.
Which days qualify — a different rule from the one above
A day joins the split when every reading in it could be tested against that ceiling, and the ceiling needs two things: a position fix, and your saved panels. One reading without them and the whole day drops out. That matters, because the alternative would be to credit charge nobody could account for to the sun, which is the flattering direction and the one you would never see.
So this is not the same rule as the solar comparison above. A day you told the app the panels were packed away is in the split — a packed panel's ceiling is nothing, so everything that arrived lands under driving or other charging, which is exactly right. It only sits out of the comparison that judges your panels, where there was nothing working to judge.
Telling driving apart needs one permission
Proving you were driving comes from Motion & Fitness on iPhone, or Physical activity on Android. It reads whether you were in a vehicle and nothing else — no location, no route, no speed, and nothing that could be turned back into where you went.
Turn it down and nothing breaks. The charge is still real, still recorded and still counted. The app simply can rarely prove the drive, so what the car put in lands under other charging with everything else it could not name. If your driving share looks lower than your trip felt, that permission is the first thing to check.
Each rig learns on its own
Each rig is learned separately, so swapping setups never mixes one rig's history into another's.
Run a 200 W blanket off the ute one weekend and the full setup off the van the next, and those are two different rigs with two different histories. Swap between them as often as you like - neither drags the other's numbers around.
The history opens on the rig you are running. Two buttons sit at the top: All rigs for the whole record, and one carrying a rig's name. Tap the named one to read that rig on its own; tap it again and it drops down a list of your saved rigs, with in use beside the one you are running now.
The days-recorded count follows whatever you are looking at, so a rig you have only just built can read three days while the seven that opened the screen were counted across everything. Try a different battery for an hour mid-day and that hour is filed under the setup you were actually running, so your rig's day reads a little shorter and carries a note: also recorded under another setup that day. Nothing is lost and nothing is mixed.
The one thing a forecast cannot tell you
Weather data can tell you how much sun is going spare. It cannot tell you that your panels are filthy, that the tree you park under has grown, or that a connector has started to corrode - because you were never getting one hundred per cent of what was on offer anyway, so there is no clean line to fall below.
Your own history can. Once there is enough of it, the app can see that you used to pull a certain share of the available light and lately you have been pulling less. When that gap gets big enough to be real rather than noise, it says so once:
Power generation has dropped recently — even allowing for the weather.
Worth a quick wipe over the panels; dust and road grime are the usual culprits.
Those last four words are doing real work. Without them the first thought is "well, it has been overcast" - and the one time the message fires would be wasted. The weather is already out of the comparison before anything is said.
What it will not do
It will not diagnose your rig. The app can see that less is coming in than used to; it cannot see whether that is dust, shade, a tired cable or a panel that has had a hard life. Telling you where to look is useful. Pretending to know is not.
It will not nag. Readings that would answer "yes, that's normal" almost every day were deliberately left out - a nudge you learn to ignore is worse than no nudge at all.
It will not tell you what your charger really delivers. The app notes the strongest non-solar charge rate it sees, but nothing reads that back to you and nothing is ever checked against the rated amps you typed in. What is on screen is what reached the battery, never a verdict on the hardware that sent it.
And it will not build a picture of you. There is no account, no tracking and no analytics of any kind, and the record itself holds no coordinate — your position is read live to work out the sun and then let go. The one time your location leaves the phone is the weather request, so the forecast is for where you are, and nothing is stored.
Getting the most out of it
Tell it when the panels are out. The packed-or-out row on the main screen is how the app knows which days can be used to judge your panels. If nothing is out, say so - the day still counts, it just sits out of that one comparison. The full story of what declaring changes is in the panels out guide, and how the whole chain fits together — declaring, crediting, recording — is laid out in the main-screen guide.
Check morning and evening. Two deliberate checks bracket the day, and the battery fills in the rest — see the habit above. It is the single cheapest thing you can do for the quality of your record.
Open the app while you are out there. Some of what a day tells the app - including how clear the sky actually was - can only be captured while that day is happening. It cannot be filled in afterwards.
You do not need a signal for the record itself. What the sky could have delivered is worked out on the phone from your position and your saved panels, so the day still counts out of range. Only the sky-clarity reading needs a connection, and that is the finer of the two ways the app spots a drop in output.
And if your battery won't pair, don't let it fail silently. Report a Bluetooth issue in the app writes you a pre-filled email with the exact diagnostics in it — you send it, nothing goes on its own — so a reply can land straight back with you — and telling us your setup at solarspotter.app/bt is how batteries get added to the supported list.
Common questions
What is Energy History?
A record of what your rig actually did, day by day: what went into the battery, what came out, and how that compares with what the sky was offering. It builds itself from your Bluetooth battery readings, so there is nothing to fill in. It is part of Solar Spotter Pro.
Why can't I see it yet?
It appears after seven days of actually using your rig. Fewer than that and any pattern in it would be noise dressed up as a finding, so the app would rather not show you one.
Do I have to leave the app open all day?
No, and it would not help if you did — both phone systems stop taking readings when the screen locks, so an app left open in a pocket records very little. Two deliberate checks a day, spread as far apart as your routine allows, gather far more. The battery itself does the counting in between.
Does a quiet month set it back?
No. Solar Spotter learns your rig from the days you actually use it — not from the calendar. Camp twice a month and it takes longer to build a picture, but you get the same answers in the end.
Why do some readings stay blank?
Some readings stay quiet until there's enough to compare against. That's deliberate: a guess off three days is worse than no answer, so the app would rather say nothing than tell you something it can't stand behind.
Why does the average say "while connected"?
Because that is what was actually measured. Early on, the record knows which hours the battery was watched — but not much more — so it gives you an honest rate over those hours rather than pretending it saw whole days. Beside it, readings in ~9 of 24 hours says how much of the day those readings were spread across. Check the battery before bed and at dawn and that figure climbs. Once most of your recorded days are measured right through, it switches to a real per-day average on its own.
What about days the panels stayed packed away?
Days when the panels weren't out, or when you were driving and topping up from the car, sit out of the solar comparison — there's no way to tell what your panels did on those days. They're still recorded, and still count for everything else.
Why is my driving counted as “other charging”?
Because the app could not prove you were driving. Telling the alternator apart needs the motion permission, called Physical activity on Android and Motion and Fitness on iPhone — it reads whether you were in a vehicle and nothing else, no location and no route. Without it the charge is still real and still recorded, it just cannot be named, so it sits under other charging.
I run more than one setup. Does that confuse it?
Each rig is learned separately, so swapping setups never mixes one rig's history into another's. The history opens on the rig you are running. Two buttons sit at the top: All rigs for the whole record, and one carrying a rig's name — tap that one to read that rig alone, tap it again for a list of your saved rigs. The days-recorded count follows whichever you are viewing.
What does the dropped-generation message mean?
It means your rig is harvesting noticeably less of the available sun than it used to, once the weather has been accounted for. Nine times out of ten that is dust and road grime on the panels. It is the one thing a forecast cannot tell you, because it is about your gear rather than the sky.
Does any of this leave my phone?
The record itself, no — it is stored on your phone, for you, and it holds no coordinate at all. There is no account, no analytics and no tracking. The one thing that does leave the phone is the weather request, which carries the location being forecast so the forecast is for where you are. Nothing is stored.
Let your rig tell you what it can do
Energy History is part of Solar Spotter Pro, alongside the Battery Bluetooth link that feeds it — in BETA, and growing as more batteries are confirmed. No account, no tracking, and your record stays on your phone.
