Clive, the way you can draw a circuit with no eraser and call it "very straight forward" is pretty brilliant. And the fact that you get the praise of AvE from his channel is testament to your awsomeness. Keep up the great work.
A few years ago the local council installed solar-powered speeding alarms, including two near my house. When they detected traffic they would flash up the speed - in green if you were below the speed limit, in red if you were above. They looked like robust constructions, but since the winter they've stopped working, and from stories in the local paper I gather that's a problem with all the ones they put up. I suppose, as Clive would put it, water has found a way.
The council put the same thing in my road. Cars and motorbikes used to compete for who could clock the highest speed. It came down after a month or two.
These solar modules are made in the exact same way as big panels for roof-top or field installations: - a front glass - cover the front glass with a sheet of EVA (ethyl-vinyl-acetate) film, a kind of hot-melt glue - lay on the already interconnected strings of solar cells - put on a second sheet of EVA - put on the back material, which can be either a special multi-layer, moisture barrier plastic film (like here) or a second sheet of glass The whole stack is then put into a laminator and heat-treated under slight pressure. This not only melts the EVA but also cross-links it to give better long-term and environmental stability.
They did so much right on this one...except the plastic case...kind of disappointing in that regard. I do like that they are starting to put LiFePO4 in more things these days...I pretty much use them exclusively for my backup systems.
Those plastic containers on the back of the solar panel are very attractive for little critters and are sensitive for moisture build-up. Resulting in broken wire connections, which will cause an early dead of lithium cell(s) due to complete discharge. Be sure to spray the inside with a protective coating or fill them with hot glue or epoxy. Tell me how I know...
I got one of these lights and I wanted to re watch this video because I wanted to see what it is supposed to be doing when charging. I can't put it outside to charge it with the solar panel because it is winter in NE and the light sucks. I am trying to determine a way to charge it with a wall wart. I am now going to watch the video again. As always, I trust Clive's information implicitly! 🙂😌
I just got an email from them notifying me of their policy changes. Not only will they allow other parties to sell on Temu, sharing your data with others, but they'll start charging tax for the country they're shipping to starting June/1. This scam is kinda new. I've seen it a few times already. Since they're not even registered with Revenue Canada, they can't be collecting any tax, unless they wanna pocket it. Demand they provide taxation number you can verify from your end.
Just chiming in here to note Temu also doesn't seem to be (as of right now) registered in the Netherlands with the revenue service (Belastingdienst). This seems a bit scammy...
The fact you have the app is already a concern. Did you just press "yes" to everything, or did you stop to consider whether they really needed to track your location?
I've tested quite a few of these. they all come in every size imaginable, they all tend to work very good for what they are and their price level, some comes with rubber gasket sealing, some comes without, they all have different sets of Led lights but tends to be built the same way. I use them as indoor lights in my workshops, it's free light so can't complain there. The solar panels I've placed in my windows and it works just fine.
Pretty typical for a hard-framed solar panel, though note that the little plastic boxes on the back of the panel are not environmentally sealed, which is actually a bit of a problem for a streetlight application (even though it is underneath the panel). Larger hard-framed panels seal their boxes and have bypass diodes in the boxes. Most of these sorts of devices do use LiFePO4, not NMC. Solar-powered outside path lighting has used LiFePO4 cells for over a decade in fact, because the batteries are very robust in variable environmental conditions and can tolerate being held at near 100% capacity almost indefinitely. LiFePO4 is used all over the place these days, particularly in power stations and stationary battery storage. It is completely replacing lead-acid in sailboats, RVs, and other mobile applications (but not e-bikes yet). LiFePO4: Discharge no lower than 2.5V. Charge >= 3.45V for 100% charge, target
@@vkristof1 Several ways the MCU can figure it out. First is from the 2K/10K resistor divider, at least while the sun is out. The solar panel will be forced to BAT + 0.3V or so due to the battery voltage forcing the solar voltage down through the Shottky diode (or the Zener + Schottky diode if the battery protection circuit has disconnected it). Second, from the 100 ohm resistor part of the circuit, you see that goes straight into the MCU. The Zener only caps the voltage, so the actual voltage on that pin is in fact very close to the actual voltage of the battery. Third, those 5 LEDs through the 220 ohm resistor going into the MCU. The MCU is using that ping to turn the LEDs on and off by setting the pin to an output and then logic 0 (very close to ground)... perhaps as a power dump, not just for pretty LEDs. But there is nothing stopping the MCU from changing the pin mode to an input and using the ADC to read the voltage. It will be the BAT - (LED-drop @ 5uA) or something like that.... that will be very close to the battery voltage too. And finally, the supply voltage of the MCU itself (if it isn't that pin in #2) can usually be internally connected to the ADC, allowing the supply voltage to be read that way (again with an internal reference to calibrate it). #2 or #4 is the most likely point. I didn't look at the MCU data sheet but most of these sorts of MCUs can internally connect their ADC's to any of the pins, and many have internal voltage references to calibrate the value against to calculate the actual voltage. Note that the LiFePO4 battery has a *very* *low* impedance. A little 5V solar cell even with full sun can't pull up its voltage very much above the voltage on the charge curve for LiFePO4 at any given SOC. -- If they needed another pin they could in fact use method #3... I use method #3 all the time with really tiny 8-pin PIC chips. Basically multiplex two functions onto one pin, an input function and an output function. It only takes about 20uS to switch the pin mode and do an ADC reading, and switch it back, so a person wouldn't even notice the flicker on the LEDs. The varying supply voltage on the MCU is not an impediment either. Most MCUs in this class have very wide supply ranges. Like 1.5V to 5.6V (for example). So as long as it has an internal voltage reference (which most do), the program can figure out the actual voltages. -Matt
@@junkerzn7312 Thanks: I hadn't though of changing the direction of the SOC LED I/O pins and don't have current experience (& pin mixing) on low-end/low-pin count MCUs.
Any place I can go to just buy the solar panel? It's the best part of the whole thing 😂 If you buy any more street lights you're going to have to start putting them up if only as a place to keep them!
@@sihamhamda47the problem is for one good panel there is at least 10 claiming the same specs and failing. We rely on people like Clive to point us to reliable equipment.
In my experience, these lights do quite well in our equally pretty dark and long winters here in Germany, but if the weather gets really bad for 3-4 days (not only some rain, but really dark all day one), these lights tend to be completely discharged.
It looks half decent. Are the charge indicator LEDs phosphor converted, or did you colour them in on your pictures? The sun in April is a similar strength to late summer, so it's a reasonable test.
You know what happened the last time the president went against the grain. Don't think it wouldn't happen again, assuming we have any say over who gets elected. That's a big assumption.
i scored a pv panel of similar construction to that one you showed though about half the size. it is very reliable. it has never failed in about 10 years of service. i havent opened the little black box attached to it for quite a while (like i said, its very reliable) but from memory, i'm sure it has a decent sized capacitor in it. it is beyond my understanding of electronics why it is there so i did not bypass it when i connected a light to it.
circuitry wise that looks just like my very first arduino experiment .. it was around 7 bucks in total cost also.. well okay, the led matrix was a little smaller, but it was a lot more fun to build and install ! and the 1 € tupperware it lives in is still holding up to the weathers after 7 years.. probably could have used an old water bottle too like in this example here... had to change the street salvaged lipos twice though. Solar panel were salvaged from old garden lights (they only made it two years..) As long as the honored customer does not change behavior we will be flooded with this E-crap that is electronic waste right from the conveyor .. Would be nice if Clive could show us how to build such a device with what can be found on the street and in the bin. :)
This similar type of china solar light is actually pretty affordable in Asia (about $10). I got a 6.4w Monocrystalline type solar panel (0.85A, 8.1v), the volt was somewhat high and non standard. It might cause damage if pair with mismatch solar lamp. It has around 200led powered with a single 26650 LiFePo4 (claim to be 4Ah), well at least with BMS attached. I then upgraded it with a spare 12w poly solar panel and changed the battery with a used prismatic LiFePo4 22Ah, works like a charm, bright from dusk till dawn. Then I salvaged the spare for another upgrade.
The charge indicator is interesting, but as it's only going to be active in sunlight, I wonder how visible those red LEDs will be a bunch of feet up a pole in daylight.
Mine just landed yesterday. Not gonna tear it down but did a test on the solar output which was pretty good considering there's no regulation inside the little breakout box. I can't remember the output in full sun but I think it was 9v
Looks like it aught to be better than that to me, clearly not top shelf last forever quality but still (and surprisingly for a retailer of Chinesium grade) not terrible. Add in a desiccant pouch in the battery compartment and glue it up all around on that glass so its got a reasonable seal, or make darn sure it will shed any moisture that builds up inside and I'd think you'd get a few years out of it. The solar cell looks pretty good, though I wonder if the glass is really up to being hailed on or taking a snow load, so assuming that plastic isn't the sort that dies after minimal UV exposure or something...
Not likely to be much use in our winters. Not just because the general low light levels will not provide much charge but on those bright sub-zero mornings the charge current is likely to damage the cell, they don't like charging at low temperatures.
That red button might be motion detection on/off. I've been using similar Chinese solar street lights like this one for years. I highly recommend it too
I’ve often thought that in a few years it will be possible to have solar street lights in remote areas where power would be impractical to get to otherwise. Like long remote stretches of highway.
I've got one of those lights mounted near my driveway. It uses a microwave motion detector to switch between low and high-brightness. The LED array is quite large and it does use the LiFePO₄ battery--solar panel is on top. It's been plugging away for about 5 years, sub-freezing and 40C temp extremes. Original cost was about USD $50.
@@emilen2 Less honesty. More scammers. Cheaper, shittier products that AliExpress sellers would be too ashamed to sell. Serious security and privacy concerns in the app.
I used one of those protection boards for charging a lithium cell, the board died within about 3 uses, then it became hot👀 I wouldn't trust them just to charge a cell
I stumbled a little over the heat sink fins on the plastic housing, but otherwise it does seem like a nice light/solar panel. Curious about the 6AH claim on the battery
I have a few similar floodlights that run off one LiFePo4 battery just like this one. The downside is that the tiny BMS/protection board mounted on the side of the battery is poorly deisgned and doesn't cut off the charging until 3.75v (it should be no more than 3.65v). My batteries last a year or two, but they should last a decade. Also, the replacement batteries are listed as 5Ah yet they have tested only 3.4 to 3.6AH. I have no doubt the 6AH in that light is greatly exaggerated.
they prolly planned a twin 3Ah cell model for more sunny places but sell those halfed capacity variation to like UK and other low sun irradiance places because the PV wouldn't stand a chance to charge both cells.
Anything reliant on LFP will fail in regions where the temperature remains below 0C for any length of time. You cannot charge those cells when it's that cold without causing irreparable damage.
Reminds me of self-driving cars in Michigan where in winter during snowstorms the entire environment is a featureless white. Good luck in winter when the high temperature is below 0° C for weeks at a time.
Wouldn't Lithium iron phosphate batteries be potentially bad for a solar light if you live in a cold climate? I thought they couldn't charge when cold and I doubt that thing has any kind of battery warming
Probably around half the power supplied to the emitters is making it out as light, so only 2 watts dissipated as heat over such a large area seems perfectly fine with a plastic housing.
How are the LEDs configured please? As it’s an odd number divisible by 3, are they lots of parallel groups of 3 in series (not sure that would work at 3.6v) or are they all in parallel?
I bought a solar powered shed light that came with a smaller solar panel but identical in construction including the plastic bit on the back. Got it from Amazon.
I think it's a really strange mix of stuff. The circuit looks to be really cost optimized, biggest example of that is the lack of any charger IC and it relying purely on the protection IC. Presumably they get assembled and then sit in their box at 0% SoC in under-voltage lockout until bought? Surely that's not good for the cell? Makes me confused as to why they bothered spending the extra time and money designing in a LiFePO4 battery, kinda doubting it will last between the 500 to 2,000 cycles for it to even matter? But why on earth does it have 357 discrete LEDs? That must be a pretty significant part of the BOM, a COB would be much cheaper! 2835 LEDs can do 0.2 W each easy, but no way this thing is doing 70W so they must be massively under driven? I can only assume they just added more LEDs in parallel to get more light output with a single LiFePO4 cell and not need a boost converter to make each LED brighter (and more efficient).
AND: it's expensive. temu is cheapest crap besides ali ect. none of the chinese garbage sellers is putting proper brain in their products. an alu/zinc cast housing with (deeeep) fins is the best you can get when it comes to led lighting.
For the battery test, drain the battery to flat and charge it with a 5v 1a input where the solar panel connects and put current sensing on the battery as it charges, then do the same with the 3ah battery and see which one charges faster. You'll figure out pretty quickly that Chinese manufacturers tend to over-exaggerate capacity ratings.
Draining LFP to "flat" will ruin the cell. 2.5v to 3.65v are the limits for these. When they first came out, 2.0v to 4.0v were common BMS settings, but the cells will rapidly degrade using those values.
Lead voltages are a bit lower but LiFePO4 are constant current charged at, ideally, 0.5C so the actual charge cycle is different and current critical as anything over 1.0C is too high from the data I've been using from EVE (who now quote P - power rating not C - capacity on newer cells). So a 10Ah cell charges at 5A 3.6V CC and can discharge at 20A down to 2.5V I tweaked my lead charger to work with lithium but it's an electronic charger, not an old school transformer, rectifier with current limiting resistors - these are a bit risky for LiFe cells/batteries.
you need to serial gang up at least 2 LiFe's to match lead acid voltage. so 3,2V ×2 = 6,4V is about same as 3x 2,1V lead acid cells. so 6V, 12V, 24V ect. could be charged off of lead acid charger but you need a balancer for LiFePo because they can't pass-through current to self-equalize all cells like lead acid can. so 4S LiFePo at 12,8V nom. could also be 1V + 1V + 0,8V + 10V =12,8V. that'd go horribly bang in notime. everything lithium in series needs a balancer AND if not in lab conditions a BMS to avoid destruction or battery fire.
Check position RA2 again on that board. It's populated with an upside-down resistor. Most likely they're tuning that 62 ohm with a much higher value resistor in parallel.
Interesting that they would use a 3.2V(nom) battery to light White LEDs that typically have a 3.2V(min) Drop. I think you can only get about 50% of the battery capacity that way.
Testing LiFePO4 cells: max charge voltage 3.6V, 3.55V is adequate in reality, they don't gain much charge after this point. OCVP 3.65V Min discharge voltage 2.6V. UCVP 2.5V Most charge at 0.2C to 1C with 0.5C being optimal & discharge up to 2.0C max but with smaller cells I'd be monitoring temperature all the way at higher loads. [C =capacity] Cell output is very linear from 10% to 90% so voltage is not a good indicator of SOC. A fully charged cell will sit at 3.34V after resting. I discharge test at 0.5A for consistency of data using a prebuilt 5A board unless I have >100Ah cells to test then I use 2A to the full 5A depending my patience. If I was regularly testing large cells I'd buy a purpose built unit, but a 5A or better a 10A board from AliX is cheap, flexible and reliable enough for most smaller li-ion, lifepo4 etc
My landlord got one of these for the yard and it stopped working after a few months I took it apart and it's similar in concept. The batter is 6 cells in parallel I assume as the voltage is that of a single cell. I am trying to figure out a good way to calculate the capacity but it says 20000mAH
Your video was very helpful because the plastic cover and silicon adhesive used to mount the batter are extremely similar. So it let me guess that there is a protective circuit in the plastic wrap of the battery. I don't need to worry about it as much
7 месяцев назад+1
You know... I'd buy a few but they are all at a disgusting 78000K Also, I'm curious how many winters would that cell withstand... edit: BTW, solar panels output their best in the winter. The colder the better plus reflections off the snow. That if they are cleared and facing the sun.
@@BromideBride that is the theory. You and I will go out of our ways to protect our lipo/life batteries but _the people_ just buy these and plop them in the yard. Just next door to me, a neighbor put one of these at their entrance. It almost certainly has a li-ion cell but it's been several years and the bugger still works...
Are the LEDS on an aluminum PCB or just an ordinary one? Seems like a bad design to have that many LEDs in parallel. Won't some hog most of the current?
That’s almost certainly a 6Ah cell. The 32650/32700’s are almost all 6Ah with the high energy ones being 7Ah. The 6Ah cells can typically support a 3C continuous discharge and sometimes a 5C burst. Any 3Ah 32650 cell is likely something like a Full River high discharge cell. For 26650 you have the 2.5Ah A123 M1B. You can even find some high discharge 5Ah 26650’s these days.
Even just the fact that it's a LiFePO4 with a decently good solar panel is likely worth it! Seems nifty to do some DIY stuff to like increasing the size of the cell for less sunny environments or Zigbee/Z-Wave control to integrate with home automation! Don't believe that IP65 rating for a second but maybe with a re-application of silicone sealant all the way round could save it in wetter environments without too much fuss.
Would it improve the longevity or stability of the circuitry if you filled in the empty spaces? Or would that rather be for a different setup entirely, e.g. when using multi-chip LEDs that need more power?
If these have 5.5Ah cells and 6th highest brightness is 1.1A does that mean that it'll run out of power after 5hrs? Likewise with if it charges at 1.2A, in April, that'd mean it'd take less than 5.5hrs to recharge? It seems pretty pointless for a streetlight to only have 5hrs of power, 10hrs being possible on the 3rd lowest brightness but possibly too dim to be useful
I do like LiFePo cells. But i doubt 6Ah. I prefer A123 cells, which have 2,5 Ah of that size. I build motorcycle starter batteries out of it. They are not suitable to be charged below 0°C, but for A123 i ignored that with success. But they will have much less power in winter.
@@bigclivedotcom So it will last longer. Given the current you are getting from the solar panel, adding another battery or three will pay dividends at least here in Australia.
@@MiniLuv-1984 I suppose calling it a "remote controlled personal street light" is too wordy. Or perhaps the name in Chinese makes no distinction between the various types.
Hmm, this looks like a half decent unit. I would happily have some of those if I had a pressing need to light up somewhere that couldn't be connected to mains (e.g. at an allotment).
If they had sandwiched the solar cells between glass and put that in the frame, then the output could have been a few volts more and an amp or two more. Sad how they do things cheaply that cause the products they manufacture to not work at there the greatest potential, and they way they laminate the back of the cells it is impossible to remove the white backing. Damn even putting a mirror on the back, a two-way mirror allowing light in not out would have improved the output of the panel. what you need with solar panels is a way to get as much light to the cells as possible covering the back just prevents so much light entering and a mirror back reflects the light that would not have been used.
Clive, the way you can draw a circuit with no eraser and call it "very straight forward" is pretty brilliant. And the fact that you get the praise of AvE from his channel is testament to your awsomeness. Keep up the great work.
12:20 the lightweight construction is by design - when the mild steel bracket rusts and fails, it won't hurt so much when it falls on your head!
Good point. If we glue paper airplane wings on it, what are the odds it will do a left bank turn into the trash can?
Plastic where it should be metal. Metal where it should be plastic. Peak China.
A few years ago the local council installed solar-powered speeding alarms, including two near my house. When they detected traffic they would flash up the speed - in green if you were below the speed limit, in red if you were above. They looked like robust constructions, but since the winter they've stopped working, and from stories in the local paper I gather that's a problem with all the ones they put up. I suppose, as Clive would put it, water has found a way.
The council put the same thing in my road. Cars and motorbikes used to compete for who could clock the highest speed. It came down after a month or two.
Ask the council nicely if Clive can have them for an autopsy.. might be an interesting video.
👍
Or the Lifepo4 battery in them have died from the cold temperature.
@@KuntalGhosh Doubt they had LiPO batteries
"Much better than I expected" = the BigClive seal of approval.
These solar modules are made in the exact same way as big panels for roof-top or field installations:
- a front glass
- cover the front glass with a sheet of EVA (ethyl-vinyl-acetate) film, a kind of hot-melt glue
- lay on the already interconnected strings of solar cells
- put on a second sheet of EVA
- put on the back material, which can be either a special multi-layer, moisture barrier plastic film (like here) or a second sheet of glass
The whole stack is then put into a laminator and heat-treated under slight pressure. This not only melts the EVA but also cross-links it to give better long-term and environmental stability.
That sounds like WAY too many steps for Chinese manufacturing. I would rather presume they just hot glue gunned everything like they usually do.
They did so much right on this one...except the plastic case...kind of disappointing in that regard.
I do like that they are starting to put LiFePO4 in more things these days...I pretty much use them exclusively for my backup systems.
Those plastic containers on the back of the solar panel are very attractive for little critters and are sensitive for moisture build-up. Resulting in broken wire connections, which will cause an early dead of lithium cell(s) due to complete discharge.
Be sure to spray the inside with a protective coating or fill them with hot glue or epoxy.
Tell me how I know...
How i know?
@@jayyydizzzle You said "I" meaning "me" - that is incorrect. the correct answer is "you" meaning "I"
@@asedcopf I'm aware. Op asked for somebody to tell them: "how I know"
It was a joke haha
There's nothing quite like DC to eat wet connections and wires.
I got one of these lights and I wanted to re watch this video because I wanted to see what it
is supposed to be doing when charging. I can't put it outside to charge it with the solar panel because it is winter in NE and the light sucks. I am trying to determine a way to charge it with
a wall wart. I am now going to watch the video again. As always, I trust Clive's information
implicitly! 🙂😌
Is there a component in that RA2 position? It looks like an upside down resistor with the white side upwards.
There is! I didn't actually see that. I'll update the description.
I'm glad you spotted it too. I thought I was seeing things
I think BC forgot to update the description?
@@Pieman93 I did forget. It's updated now.
I find it truely amazing how much you still response and read your comments as a bigger RUclipsr! Greetings from Germany :)
It's getting harder, but I still try to read them all.
Clive's a good feedback boi. Setting standards! ✌🏽
It would be interesting to get a thermal image of the light running to see how hot those LEDs get without a heatsink.
Yeah I would be curious how well they hold up for cooling if they are sealed water tight?
they've got to under drive those LEDs or that thing would just melt away right?
I really enjoy your teardowns of cheap stuff. Thanks.
I just got an email from them notifying me of their policy changes.
Not only will they allow other parties to sell on Temu, sharing your data with others, but they'll start charging tax for the country they're shipping to starting June/1.
This scam is kinda new. I've seen it a few times already. Since they're not even registered with Revenue Canada, they can't be collecting any tax, unless they wanna pocket it.
Demand they provide taxation number you can verify from your end.
Just chiming in here to note Temu also doesn't seem to be (as of right now) registered in the Netherlands with the revenue service (Belastingdienst).
This seems a bit scammy...
Don't buy from Temu.
They add VAT on in the UK.
The fact you have the app is already a concern. Did you just press "yes" to everything, or did you stop to consider whether they really needed to track your location?
@@tin2001 The fact that you assume people have the "app" is concerning.
Your comment was as useless as it gets.
I've tested quite a few of these. they all come in every size imaginable, they all tend to work very good for what they are and their price level, some comes with rubber gasket sealing, some comes without, they all have different sets of Led lights but tends to be built the same way. I use them as indoor lights in my workshops, it's free light so can't complain there. The solar panels I've placed in my windows and it works just fine.
Pretty typical for a hard-framed solar panel, though note that the little plastic boxes on the back of the panel are not environmentally sealed, which is actually a bit of a problem for a streetlight application (even though it is underneath the panel). Larger hard-framed panels seal their boxes and have bypass diodes in the boxes.
Most of these sorts of devices do use LiFePO4, not NMC. Solar-powered outside path lighting has used LiFePO4 cells for over a decade in fact, because the batteries are very robust in variable environmental conditions and can tolerate being held at near 100% capacity almost indefinitely. LiFePO4 is used all over the place these days, particularly in power stations and stationary battery storage. It is completely replacing lead-acid in sailboats, RVs, and other mobile applications (but not e-bikes yet).
LiFePO4: Discharge no lower than 2.5V. Charge >= 3.45V for 100% charge, target
I don,t see from Clive's schem how the MCU even gets an accurate voltage reading of the cell voltage to guesstimate the cell's SOC.
@@vkristof1 Several ways the MCU can figure it out.
First is from the 2K/10K resistor divider, at least while the sun is out. The solar panel will be forced to BAT + 0.3V or so due to the battery voltage forcing the solar voltage down through the Shottky diode (or the Zener + Schottky diode if the battery protection circuit has disconnected it).
Second, from the 100 ohm resistor part of the circuit, you see that goes straight into the MCU. The Zener only caps the voltage, so the actual voltage on that pin is in fact very close to the actual voltage of the battery.
Third, those 5 LEDs through the 220 ohm resistor going into the MCU. The MCU is using that ping to turn the LEDs on and off by setting the pin to an output and then logic 0 (very close to ground)... perhaps as a power dump, not just for pretty LEDs. But there is nothing stopping the MCU from changing the pin mode to an input and using the ADC to read the voltage. It will be the BAT - (LED-drop @ 5uA) or something like that.... that will be very close to the battery voltage too.
And finally, the supply voltage of the MCU itself (if it isn't that pin in #2) can usually be internally connected to the ADC, allowing the supply voltage to be read that way (again with an internal reference to calibrate it).
#2 or #4 is the most likely point. I didn't look at the MCU data sheet but most of these sorts of MCUs can internally connect their ADC's to any of the pins, and many have internal voltage references to calibrate the value against to calculate the actual voltage.
Note that the LiFePO4 battery has a *very* *low* impedance. A little 5V solar cell even with full sun can't pull up its voltage very much above the voltage on the charge curve for LiFePO4 at any given SOC.
--
If they needed another pin they could in fact use method #3... I use method #3 all the time with really tiny 8-pin PIC chips. Basically multiplex two functions onto one pin, an input function and an output function. It only takes about 20uS to switch the pin mode and do an ADC reading, and switch it back, so a person wouldn't even notice the flicker on the LEDs.
The varying supply voltage on the MCU is not an impediment either. Most MCUs in this class have very wide supply ranges. Like 1.5V to 5.6V (for example). So as long as it has an internal voltage reference (which most do), the program can figure out the actual voltages.
-Matt
@@junkerzn7312 Thanks: I hadn't though of changing the direction of the SOC LED I/O pins and don't have current experience (& pin mixing) on low-end/low-pin count MCUs.
Any place I can go to just buy the solar panel? It's the best part of the whole thing 😂
If you buy any more street lights you're going to have to start putting them up if only as a place to keep them!
There are many of those panels being sold on aliexpress. I checked that out and it cost 7 bucks each
@@sihamhamda47the problem is for one good panel there is at least 10 claiming the same specs and failing.
We rely on people like Clive to point us to reliable equipment.
If you put this up somewhere it would be cool to see how well it copes with a long british winter.
Please film it with stop motion animation for the light to stop working and then falls off the wall.
Haha 😂❤ @@doctoroctos
In my experience, these lights do quite well in our equally pretty dark and long winters here in Germany, but if the weather gets really bad for 3-4 days (not only some rain, but really dark all day one), these lights tend to be completely discharged.
@@doctoroctos Like Solar Fucking Roadways did with their installation? :)
I'd like to see how it goes in a hot Australian summer - discolourisation of the solar panel in particular.
It looks half decent. Are the charge indicator LEDs phosphor converted, or did you colour them in on your pictures?
The sun in April is a similar strength to late summer, so it's a reasonable test.
They are phosphor based.
Clive, please run for president here in the United States. We really need someone sane over here
You know what happened the last time the president went against the grain. Don't think it wouldn't happen again, assuming we have any say over who gets elected. That's a big assumption.
i scored a pv panel of similar construction to that one you showed though about half the size. it is very reliable. it has never failed in about 10 years of service. i havent opened the little black box attached to it for quite a while (like i said, its very reliable) but from memory, i'm sure it has a decent sized capacitor in it. it is beyond my understanding of electronics why it is there so i did not bypass it when i connected a light to it.
circuitry wise that looks just like my very first arduino experiment .. it was around 7 bucks in total cost also.. well okay, the led matrix was a little smaller, but it was a lot more fun to build and install ! and the 1 € tupperware it lives in is still holding up to the weathers after 7 years.. probably could have used an old water bottle too like in this example here... had to change the street salvaged lipos twice though. Solar panel were salvaged from old garden lights (they only made it two years..)
As long as the honored customer does not change behavior we will be flooded with this E-crap that is electronic waste right from the conveyor ..
Would be nice if Clive could show us how to build such a device with what can be found on the street and in the bin. :)
This similar type of china solar light is actually pretty affordable in Asia (about $10). I got a 6.4w Monocrystalline type solar panel (0.85A, 8.1v), the volt was somewhat high and non standard. It might cause damage if pair with mismatch solar lamp. It has around 200led powered with a single 26650 LiFePo4 (claim to be 4Ah), well at least with BMS attached. I then upgraded it with a spare 12w poly solar panel and changed the battery with a used prismatic LiFePo4 22Ah, works like a charm, bright from dusk till dawn. Then I salvaged the spare for another upgrade.
Had one of these up for over a year with no problems. Very coastal area, so salty atmosphere
can’t beat a solar or LED review video.
The charge indicator is interesting, but as it's only going to be active in sunlight, I wonder how visible those red LEDs will be a bunch of feet up a pole in daylight.
because of the flat discharge curve of LiFePO4 it should have a fairly stable brightness until close to empty.
Mine just landed yesterday. Not gonna tear it down but did a test on the solar output which was pretty good considering there's no regulation inside the little breakout box. I can't remember the output in full sun but I think it was 9v
That gem looks like it should last three or four weeks outside.
Looks like it aught to be better than that to me, clearly not top shelf last forever quality but still (and surprisingly for a retailer of Chinesium grade) not terrible. Add in a desiccant pouch in the battery compartment and glue it up all around on that glass so its got a reasonable seal, or make darn sure it will shed any moisture that builds up inside and I'd think you'd get a few years out of it. The solar cell looks pretty good, though I wonder if the glass is really up to being hailed on or taking a snow load, so assuming that plastic isn't the sort that dies after minimal UV exposure or something...
Winter or summer?
@@rimmersbryggeri First one, then the other. Though not necessarily in that order.
@@zybch In the order of your liking? From what I have heard summer doesnt last that long in britain though. That's might be apocryphal though.
@@rimmersbryggeri Might or might not. Just about every British RUclipsr I watch says it’s two, occasionally three days long.
Not likely to be much use in our winters. Not just because the general low light levels will not provide much charge but on those bright sub-zero mornings the charge current is likely to damage the cell, they don't like charging at low temperatures.
I guess that's why Tesla cars, which switched to this type of cell a couple years ago, charge so slowly in the snow belt.
That red button might be motion detection on/off.
I've been using similar Chinese solar street lights like this one for years. I highly recommend it too
Got 3 Watt Cycle 100AH LifePo5 batteries for less than $100 each. Our utility car practically flys now!
I’ve often thought that in a few years it will be possible to have solar street lights in remote areas where power would be impractical to get to otherwise. Like long remote stretches of highway.
Interesting. Just bought one, I'm planning on adding a motion sensor and use this as a light for my back yard.
thnx
I was thinking it would be useful for me in that exact setup. How are you going about implementing the PIR?
I bought one with a P.I.R. though the solar panel is much less impressive.
It does a decent job of lighting the driveway.
Looks like there is room for two more LiFeP04 battery cells, maybe for use with a larger panel and charging controller.
I've got one of those lights mounted near my driveway. It uses a microwave motion detector to switch between low and high-brightness. The LED array is quite large and it does use the LiFePO₄ battery--solar panel is on top. It's been plugging away for about 5 years, sub-freezing and 40C temp extremes. Original cost was about USD $50.
Changing the "designed to dissipate heat" case from aluminium to plastic is so typical for Chinese cost cutting 😂
Its only 5W so its fine. Pointless over design just wastes resources and money.
No O-ring or peripheral sealing. But, it has heat sink fins? Maybe it’s a carbon filled plastic😂.
@@backgammonbacon Plastic is better than metal for the environment how? Don't just make something up either.
Plastic heatsinks are a waste of resources.
I'm just thinking of a comment, "One moment please!"
Watch your eyes, the light comment is coming, "Very illuminating, Big Clive!"
Temu Addiction Kicking in...lol
What can Temu provide that AliExpress cannot? Genuine question.
@@emilen2 a slick interface
@@tyttuut Temu wolf warrior: All the better to distract you with, my dear.
@@emilen2
Less honesty.
More scammers.
Cheaper, shittier products that AliExpress sellers would be too ashamed to sell.
Serious security and privacy concerns in the app.
Perhaps I missed it, if it was said… How much did it cost?
Thanks for the video!
£32.
Cheerz Clive. Looks to be of reasonable quality.
Street light with remote!
Now there's a sudden rush on Universal Remote Controllers.😆
Or magic wands with IR emitters - Luminos!
Yes, I get your pointer😆
@ 8:58 Question. What would happen if the unused spaces for components were filled?
I used one of those protection boards for charging a lithium cell, the board died within about 3 uses, then it became hot👀 I wouldn't trust them just to charge a cell
It's supposed to be the last line of defense before a battery fire, not an actual battery management system.
Crikey. Clive and DonnyBoy certainly love their Temu stuff.
I stumbled a little over the heat sink fins on the plastic housing, but otherwise it does seem like a nice light/solar panel. Curious about the 6AH claim on the battery
Plastic heat fin's, awesome....cheers.
I have a few similar floodlights that run off one LiFePo4 battery just like this one. The downside is that the tiny BMS/protection board mounted on the side of the battery is poorly deisgned and doesn't cut off the charging until 3.75v (it should be no more than 3.65v). My batteries last a year or two, but they should last a decade. Also, the replacement batteries are listed as 5Ah yet they have tested only 3.4 to 3.6AH. I have no doubt the 6AH in that light is greatly exaggerated.
they prolly planned a twin 3Ah cell model for more sunny places but sell those halfed capacity variation to like UK and other low sun irradiance places because the PV wouldn't stand a chance to charge both cells.
i love your oeuvre bc
Anything reliant on LFP will fail in regions where the temperature remains below 0C for any length of time. You cannot charge those cells when it's that cold without causing irreparable damage.
Reminds me of self-driving cars in Michigan where in winter during snowstorms the entire environment is a featureless white. Good luck in winter when the high temperature is below 0° C for weeks at a time.
Wouldn't Lithium iron phosphate batteries be potentially bad for a solar light if you live in a cold climate? I thought they couldn't charge when cold and I doubt that thing has any kind of battery warming
It will be interesting to see how long the light lasts at full brightness with a full battery. Also, it looks hackable with space for more batteries.
Probably around half the power supplied to the emitters is making it out as light, so only 2 watts dissipated as heat over such a large area seems perfectly fine with a plastic housing.
Curious if it will work for long, without any metal to cool the LEDs
Hey Clive did you mean to be zoomed in from the start?
It was just the size of the stuff being a bit too big for the filming area.
How are the LEDs configured please? As it’s an odd number divisible by 3, are they lots of parallel groups of 3 in series (not sure that would work at 3.6v) or are they all in parallel?
They're all in parallel.
I bought a solar powered shed light that came with a smaller solar panel but identical in construction including the plastic bit on the back. Got it from Amazon.
I think it's a really strange mix of stuff. The circuit looks to be really cost optimized, biggest example of that is the lack of any charger IC and it relying purely on the protection IC. Presumably they get assembled and then sit in their box at 0% SoC in under-voltage lockout until bought? Surely that's not good for the cell? Makes me confused as to why they bothered spending the extra time and money designing in a LiFePO4 battery, kinda doubting it will last between the 500 to 2,000 cycles for it to even matter?
But why on earth does it have 357 discrete LEDs? That must be a pretty significant part of the BOM, a COB would be much cheaper! 2835 LEDs can do 0.2 W each easy, but no way this thing is doing 70W so they must be massively under driven? I can only assume they just added more LEDs in parallel to get more light output with a single LiFePO4 cell and not need a boost converter to make each LED brighter (and more efficient).
Would be cool to see what the panels output actually is.
1.2A on a Sunny April day and about a tenth of that on a cloudy day.
Thermally conductive plastic tends to be heavy. I suspect this will overheat.
AND: it's expensive.
temu is cheapest crap besides ali ect. none of the chinese garbage sellers is putting proper brain in their products.
an alu/zinc cast housing with (deeeep) fins is the best you can get when it comes to led lighting.
For the battery test, drain the battery to flat and charge it with a 5v 1a input where the solar panel connects and put current sensing on the battery as it charges, then do the same with the 3ah battery and see which one charges faster. You'll figure out pretty quickly that Chinese manufacturers tend to over-exaggerate capacity ratings.
Draining LFP to "flat" will ruin the cell. 2.5v to 3.65v are the limits for these. When they first came out, 2.0v to 4.0v were common BMS settings, but the cells will rapidly degrade using those values.
....maybe the mystery red button is a panic button like you find on the key fobs of some yankee cars, and possibly a siren can be fitted? (😉)
Hmmmm.... is it just me, or it seems like there's two resistors, but the one in the middle (RA2) is just upside down?
april sunshine? when was this filmed please ? its mid may lol
Tested in April. Video made late April to early May.
From my understanding (and implementation) a standard lead acid float charger will charge LiFePO4 without damage.
Lead voltages are a bit lower but LiFePO4 are constant current charged at, ideally, 0.5C so the actual charge cycle is different and current critical as anything over 1.0C is too high from the data I've been using from EVE (who now quote P - power rating not C - capacity on newer cells). So a 10Ah cell charges at 5A 3.6V CC and can discharge at 20A down to 2.5V
I tweaked my lead charger to work with lithium but it's an electronic charger, not an old school transformer, rectifier with current limiting resistors - these are a bit risky for LiFe cells/batteries.
you need to serial gang up at least 2 LiFe's to match lead acid voltage. so 3,2V ×2 = 6,4V is about same as 3x 2,1V lead acid cells. so 6V, 12V, 24V ect. could be charged off of lead acid charger but you need a balancer for LiFePo because they can't pass-through current to self-equalize all cells like lead acid can.
so 4S LiFePo at 12,8V nom. could also be 1V + 1V + 0,8V + 10V =12,8V. that'd go horribly bang in notime.
everything lithium in series needs a balancer AND if not in lab conditions a BMS to avoid destruction or battery fire.
How did the remote work? Was there and antenna trace on the board going into the chip?
Infrared. The sensor is on the LED array.
Check position RA2 again on that board. It's populated with an upside-down resistor. Most likely they're tuning that 62 ohm with a much higher value resistor in parallel.
Interesting that they would use a 3.2V(nom) battery to light White LEDs that typically have a 3.2V(min) Drop. I think you can only get about 50% of the battery capacity that way.
White LEDs will start conducting at around 2.5V. The fully charged cell will be around 3.6V.
Testing LiFePO4 cells:
max charge voltage 3.6V, 3.55V is adequate in reality, they don't gain much charge after this point. OCVP 3.65V
Min discharge voltage 2.6V. UCVP 2.5V
Most charge at 0.2C to 1C with 0.5C being optimal & discharge up to 2.0C max but with smaller cells I'd be monitoring temperature all the way at higher loads. [C =capacity]
Cell output is very linear from 10% to 90% so voltage is not a good indicator of SOC. A fully charged cell will sit at 3.34V after resting.
I discharge test at 0.5A for consistency of data using a prebuilt 5A board unless I have >100Ah cells to test then I use 2A to the full 5A depending my patience. If I was regularly testing large cells I'd buy a purpose built unit, but a 5A or better a 10A board from AliX is cheap, flexible and reliable enough for most smaller li-ion, lifepo4 etc
My landlord got one of these for the yard and it stopped working after a few months I took it apart and it's similar in concept. The batter is 6 cells in parallel I assume as the voltage is that of a single cell. I am trying to figure out a good way to calculate the capacity but it says 20000mAH
Your video was very helpful because the plastic cover and silicon adhesive used to mount the batter are extremely similar. So it let me guess that there is a protective circuit in the plastic wrap of the battery. I don't need to worry about it as much
You know... I'd buy a few but they are all at a disgusting 78000K
Also, I'm curious how many winters would that cell withstand...
edit: BTW, solar panels output their best in the winter. The colder the better plus reflections off the snow.
That if they are cleared and facing the sun.
Answer: *one winter*
at
I didn't know snow was from the Sun.
@@BromideBride that is the theory. You and I will go out of our ways to protect our lipo/life batteries but _the people_ just buy these and plop them in the yard.
Just next door to me, a neighbor put one of these at their entrance. It almost certainly has a li-ion cell but it's been several years and the bugger still works...
Do you have any update on how well it worked outside? Was considering buying a model like it since it seemed well made.
I've not tested it yet.
Are the LEDS on an aluminum PCB or just an ordinary one?
Seems like a bad design to have that many LEDs in parallel. Won't some hog most of the current?
Aluminium.
Wonder if those charge fully on winter and how long the battery holds - whole night? Probably not
I’m looking for a decent solar powered battery 🔋 backed up light 💡 for my porch. Any suggestions for a decent priced unit???
It depends on where you are. None of them are great in typical dull British winters when they're needed.
That’s almost certainly a 6Ah cell. The 32650/32700’s are almost all 6Ah with the high energy ones being 7Ah. The 6Ah cells can typically support a 3C continuous discharge and sometimes a 5C burst.
Any 3Ah 32650 cell is likely something like a Full River high discharge cell. For 26650 you have the 2.5Ah A123 M1B. You can even find some high discharge 5Ah 26650’s these days.
Even just the fact that it's a LiFePO4 with a decently good solar panel is likely worth it! Seems nifty to do some DIY stuff to like increasing the size of the cell for less sunny environments or Zigbee/Z-Wave control to integrate with home automation! Don't believe that IP65 rating for a second but maybe with a re-application of silicone sealant all the way round could save it in wetter environments without too much fuss.
Would it improve the longevity or stability of the circuitry if you filled in the empty spaces? Or would that rather be for a different setup entirely, e.g. when using multi-chip LEDs that need more power?
Just wondering, shouldn't the solar panel have a diode included in the box on the back to stop it hogging power at night, or is the diode elsewhere?
It's on the PCB.
An infrared sensor on a board that is likely to get hot. How does that work?
Infrared sensors like that cancel out ambient heat by default.
A question: How does such a tiny MOSFET survive the current of 350+ LEDs? On that note, how much current does the Lamp take?
The tiny MOSFETs have a very low on state resistance that means they dissipate very little heat.
Hi Clive, got a link to this Lamp??
It's from Temu and the listings change all the time.
Did I miss the brightness test ? is it brighter than a candle?
It's OK. But I didn't test it in a real world scenario.
If these have 5.5Ah cells and 6th highest brightness is 1.1A does that mean that it'll run out of power after 5hrs? Likewise with if it charges at 1.2A, in April, that'd mean it'd take less than 5.5hrs to recharge? It seems pretty pointless for a streetlight to only have 5hrs of power, 10hrs being possible on the 3rd lowest brightness but possibly too dim to be useful
The intensity and current draw will gradually reduce in a curve.
Noice looking plastic heat sink lol 😂
Would you be interested in doing a video of a syntron chipping hammer with working tungar rectifiers from the 1920's?
Maybe a bit too big and heavy for shipping and the video area.
How much does the whole thing cost
It cost £32.
One has to ask if the plastic is UV resistant...
I'm in Scotland. In my experience solar powered anything doesn't last long outside.
My solar garden lights work fabulously during the summer, in the winter it’s clearly been a nice day if I get an hour of illumination from them.
Do you have a link to the same or equiv on Temu or AliExpress pse Clive?
LEDs 23x16(-11 for surface mounted components)= 357. Big number, Huge!
Could you install both of those batteries in parallel and double the runtime?
Theoretically yes. (Making sure the cell voltage were very close first.)
and then you had to double the solar module to actually charge the 2nd cell.
Clive you have the link for this one?
They change listings regularly. Try searching for solar street light.
Happy Weekend, Clive!! You Big Lug! 😊 (And I'm Straight! 😳Nobody is perfect.)
Love You, Clive!! ❤Hey! What happened to the 💥's? 💣🤣😂🍻🥃
Unfortunately they will not work at freezing or below. If you put a charge to LiFePo4 at or below 0°C it will damage them.
I do like LiFePo cells. But i doubt 6Ah. I prefer A123 cells, which have 2,5 Ah of that size. I build motorcycle starter batteries out of it. They are not suitable to be charged below 0°C, but for A123 i ignored that with success. But they will have much less power in winter.
But why use 357 LEDs for a 5 W fixture? They must be so under-driven, and the bottleneck ought to be the total heat dissipation anyhow?
Using more ans under-running them spreads the heat dissipation and also increases efficiency.
What's the point of this solar street light if it lasts somewhere between 20 - 50 minutes on a charge? (given battery capacity and led load)
The remote controller not give you a clue to their use?
The current will reduce progressively as the battery discharges and its voltage drops.
@@bigclivedotcom So it will last longer. Given the current you are getting from the solar panel, adding another battery or three will pay dividends at least here in Australia.
@@two_tier_gary_rumain Sure, but its described as a street light. In my neck of the woods, a street light is on from dusk to dawn.
@@MiniLuv-1984 I suppose calling it a "remote controlled personal street light" is too wordy. Or perhaps the name in Chinese makes no distinction between the various types.
But think of the deluxe parking lot at home you could build with a bunch of these.
Hmm, this looks like a half decent unit. I would happily have some of those if I had a pressing need to light up somewhere that couldn't be connected to mains (e.g. at an allotment).
Looks like red LEDs are phosphor based but is yellow phosphor on green LED purely for aesthetics?_
I think they may all be phosphors on blue chips.
If they had sandwiched the solar cells between glass and put that in the frame, then the output could have been a few volts more and an amp or two more. Sad how they do things cheaply that cause the products they manufacture to not work at there the greatest potential, and they way they laminate the back of the cells it is impossible to remove the white backing. Damn even putting a mirror on the back, a two-way mirror allowing light in not out would have improved the output of the panel. what you need with solar panels is a way to get as much light to the cells as possible covering the back just prevents so much light entering and a mirror back reflects the light that would not have been used.