LED Panel powered by Battery

Other Hardware General Electronics
1

Hi.

I have been getting a ton of 3w SMD LEDs (see link) that are very bright. I soldered about 25 of them onto a PCB and attached a large battery with a charging circuit. I am impressed with the results, but now I want to make more.

I know cheap Lithium Ion batteries from Ali and similar sites are very scary, and one almost exploded on me a few years back. Ratings are also off very often. I also have had excellent experiences ordering Lipos from those sites, since they sell a lot of drones they need to have accurate listings and safe Lipos available.

Therefore, I want to power my future LED panels with Lipo batteries. The problem is that they don't sell high capacity Lipo batteries in 1S configurations. I think I could either step down the battery voltage, or set up my circuit so it has two rows of LEDS in parallel, so the Vf would be 6.4 (which is close enough). The remaining problem is to find a protection circuit for a 2S Lipo, or find a large capacity 1S.

What are your thoughts? How should I design this?

2

Can you provide a drawing? Placing LEDs in parallel will not increase your Vf (Forward Voltage). Placing them in series will. This would go much better with the planned battery voltage and a link to the LED data sheet.

Ron

3

Right now they are all in parallel, all 25. As I said before, I could make "two rows of LEDs in parallel" which means the two rows would be in series, to increase Vf. I linked to the LEDs listing on the website I got them from, which is as close to a datasheet as possible :confused:. Do you know of any 2S lipo protection circuits?

I can upload an image but not a schematic at the moment, if you would like.

4

Can you provide a drawing? Placing LEDs in parallel will not increase your Vf (Forward Voltage). Placing them in series will. This would go much better with the planned battery voltage and a link to the LED data sheet.

Vforward for each led is intrinsic and you cannot increase it.
Putting leds in series doesn't change the forward voltage, it only increases the total voltage drop accross
the series string. If that's what you meant then you that's what you should say, because the forward voltage is what it is and is not going to change.

You can see from this chart that the blue and violet leds have a higher forward voltage.

5

Alright, my terms were off. Having LEDs in series doesn't change Vf, it increases the voltage required to run them all because by the time the voltage reaches the second LED, if the remaining voltage is less than the forward voltage, no current will run. I understand that.

I'm trying to find a good 2S lipo protection circuit, or a higher capacity 1S lipo, ideally 2Ah but at least 1.2Ah. Or any alternative configuration ideas you might have would be appreciated.

6

The link for the leds doesn't work. It goes to an Aliexpress login page.
I don't see any voltage information in your posts.
What is the led forward voltage per led ?
What is the voltage you are applying across the three leds ?
What is the battery you are using ?

As you probably know the 1S charged voltage of single lipo cell is 4.3V.
that makes 2S 8.4 to 8.6V.

7

raschemmel:
The link for the leds doesn't work. It goes to an Aliexpress login page.
I don't see any voltage information in your posts.
What is the led forward voltage per led ?
What is the voltage you are applying across the three leds ?
What is the battery you are using ?

As you probably know the 1S charged voltage of single lipo cell is 4.3V.
that makes 2S 8.4 to 8.6V.

The link for LEDs works, I just tested it in 5 different browsers using their incognito mode equivalent - no login page. 3.4-3.6V is the stated range, and I am currently powering my 25 LED board with a 4.2v max Lithium Ion cell from an old power bank. Therefore, each LED has 3.2-4.2V across it at any given time during normal operation, and that's plenty to function with. I haven't tested, but as you know battery voltage drops under load; the board has been measured to draw 3A at 3.5V with a bench PSU, but I'd imagine the battery isn't able to source nearly as much.

Yes, I am aware of the voltages of a 1S and 2S lipo. If I got a 2S battery, I'd need to have two sets of 12 LEDs in parallel, with set 1 being in series with set 2. My series voltage would be around 7.2V, which is just below the nominal voltage of a 2 cell lipo, and is generally safe for use. Since the board is drawing around 3A peak, a 1500mAh 2S cell would be discharging at a rate of 2C, which I suspect would reduce the voltage enough for the LEDs to have a decent lifespan. But at that price, they don't need to last more than 100 or 1000 hours anyway

8

My forward voltage would be around 7.2V

Again, total series voltage drop is not forward voltage. Forward voltage is the term used to describe the intrinsic voltage drop of a single led. I realize you think that it means voltage drop but it doesn't mean
what you are saying . There are no leds with a forward voltage above approximately 4V. What you should
be saying is not "my forward voltage" but "my series voltage drop". While this may seam trivial to you ,
continued use of the term "forward voltage" when you mean "series voltage" will just lead to more confusion
in the future. Nomenclature is important.

Also, 3W/3.6V = 0.8333A per led.
12 leds in series (or even 24 leds in SERIES) is STILL 0.833A.
12 leds in parallel is 12*0.833A = 10A
Set1 in series with set2 is still 10A.
Why it is only drawing 3A instead of 10 must be due to the power source not being able to source 10A.

Since the board is drawing around 3A peak, a 1500mAh 2S cell would be discharging at a rate of 2C

1.5Ahr (1500mAh) battery powering a 3A load would last 1.5Ahr/3A = 0.5 hours.

1.5Ahr battery powering a 10A load would last approximately 1.5/10 = 0.15 hours (0.15*60min/per hour = 9 minutes.

I would be interested in knowing how you arrived at 3A for 12 leds in parallel .
Do lthese leds have a series current limiting resistor ?
Have you tried powering 12 leds in parallel with a supply that can source 10A ?

The math doesn't seem to add up. I've never used high power leds like that so maybe someone who is
familiar with them can shed some light on the above. At this point I have no choice but to conclude you
are unable to source the amount of current they can draw.
There's no way 12 times 0.833A (3W/3.6V) winds up being 3A , unless you've found a way to bend the
rules of physics. The number of strings in series is not relevant to this discussion since it doesn't change
the current.
I may not be the best person to address this post since I have no experience with high power leds but
I can tell you the math doesn't add up. Maybe someone else can explain why.

9

2,200 or 2,800 mAh LiPo (or even bigger - over 3,000 mAh exists) is easy enough to find in 18650 size.

Those high brightness LEDs normally have multiple LEDs on a single die, to get to the total brightness. They normally need proper heatsinking to keep from overheating (and the cooler an LED is kept, the more efficient it is).

10

You haven't indicated what your target run time is so we don't know how long you need it to run without
recharging the battery.
Obviously if you wanted 10A you would probably need three 3300mAh rated batteries in parallel.
I think you'll find that you have not even begun to see the maximum intensity of those leds if you're only
running 3A through 12 of them in parallel. I assume you aware of cooling requirement for high power leds.

11

raschemmel:
Also, 3W/3.6V = 0.8333A per led.
12 leds in parallel is 12*0.833A = 10A
Why it is only drawing 3A instead of 10 must be due to the power source not being able to source 10A.
1.5Ahr (1500mAh) battery powering a 3A load would last 1.5Ahr/3A = 0.5 hours.

1.5Ahr battery powering a 10A load would last approximately 1.5/10 = 0.15 hours (0.15*60min/per hour = 9 minutes.

I would be interested in knowing how you arrived at 3A for 12 leds in parallel .
Do lthese leds have a series current limiting resistor ?
Have you tried powering 12 leds in parallel with a supply that can source 10A ?

The math doesn't seem to add up. I've never used high power leds like that so maybe someone who is
familiar with them can shed some light on the above. At this point I have no choice but to conclude you
are unable to source the amount of current they can draw.
There's no way 12 times 0.833A (3W/3.6V) winds up being 3A , unless you've found a way to bend the
rules of physics. The number of strings in series is not relevant to this discussion since it doesn't change
the current.
I may not be the best person to address this post since I have no experience with high power leds but
I can tell you the math doesn't add up. Maybe someone else can explain why.

Yes, I tested the board with one of those $50 Amazon bench power supplies, they are weak. But 3A at around 3.5V produced sufficient brightness for me. I understand that I have never even given it full power yet. And the batteries I've used so far tend to max out around 2 or 3 amps, I'm working with that as my number. A 1.5Ah battery would therefore give me thirty minutes of runtime, which I think is sufficient.

I will not be putting a current limiting resistor in parallel, since that would require large or powerful cooling solutions, and I do not have access to a psu that can properly power this board to it's fullest extent.

Your conclusion is correct. I do not have the resources to power the board fully, and that is fine. As far as what I HAVE done for cooling is concerned, I've found that using large solder lines to connect each LED is more than sufficient, especially after the first couple minutes of usage when the battery's ability to power the circuit is diminished.

wvmarle:
2,200 or 2,800 mAh LiPo (or even bigger - over 3,000 mAh exists) is easy enough to find in 18650 size.

Those high brightness LEDs normally have multiple LEDs on a single die, to get to the total brightness. They normally need proper heatsinking to keep from overheating (and the cooler an LED is kept, the more efficient it is).

Thanks for the reply. As per my original post, I have had 18650s from Aliexpress explode on my in the past (the batteries kindly gave me enough warning so I could put them outside each time), so I am trying to stay away from Lithium Ion. Lipos are much safer from these websites in my experience, which is why I am looking for a 1s Lipo with at least 1.5Ah of power, or a good solution to make a 2S work.

Also, I just realized, while Lithium Ion batteries usually discharge around 1 or 2C, Lipo batteries are often rated for 25-65C discharge rates. Should I be concerned when using a lipo that they will send too much current, and either kill my LEDs or have shorter battery life?

12

Get quality batteries. Those so-called 4700 mAh from Aliexpress et.al. give you maybe 800-1000 mAh, and improper charging may lead to serious problems.

Of course a quality LiPo costs about 10x what you pay on Aliexpress... but the difference is more than obvious when you start actually using them!

That "discharge" you give is the maximum possible. Your components only take what they need. Or how else do you think a 220V 7W LED bulb doesn't blow when connected to the net on a group with 32A breaker?

13

wvmarle:
Get quality batteries. Those so-called 4700 mAh from Aliexpress et.al. give you maybe 800-1000 mAh, and improper charging may lead to serious problems.

Of course a quality LiPo costs about 10x what you pay on Aliexpress... but the difference is more than obvious when you start actually using them!

That "discharge" you give is the maximum possible. Your components only take what they need. Or how else do you think a 220V 7W LED bulb doesn't blow when connected to the net on a group with 32A breaker?

Right, the Lithium Ion batteries (especially 18650s) on Aliexpress don't come close to their ratings.

As for the Lipos from Ali and Banggood, I've tested them to be usually no less than 10% off the mark, but some thought is required before purchasing one because often they are sold in the same size with different markings. I've learned which brands to trust mostly.

To your last point, if you've taken apart an LED light bulb you would see a SMPS or something similar. But I will trust that you are still correct in principle.

14

Incandescent tungsten filaments have infuctactance
(since they are tiny coils) which impedes current
inrush to some extent.

15

bobdabiulder:
To your last point, if you've taken apart an LED light bulb you would see a SMPS or something similar. But I will trust that you are still correct in principle.

True - it's actually a constant current driver (normally a power supply is configured to be constant voltage), that's how they're run. You will probably use a resistor as current limiter, same principle.

16

what sort of battery capacity do you want? maybe look at cell phone batteries. cheap, ubiquitous, and safe.