4

I'm working on a lighting project for a van build out. I'm using 12 V car batteries that are solar-powered. I know that I can use WS2811 strips that are 12 V strips and power them directly off the car batteries, but I'm curious if I can instead use two 5 V WS2812 strips, and wire them in parallel to split the voltage across two LED strips.

Since 12 V/2 = 6 V, I guess it should be close enough to not blow out the 5 V LED strips. Am I understanding it correctly?

Peter Mortensen
  • 1,676
  • 3
  • 17
  • 23
JackKalish
  • 177
  • 1
  • 5

6 Answers6

11

Nope. 5V things in parallel still needs 5V. You must mean series which would require 10V.

And 12V divided to two series LED strips will be 6V per strip, which will be beyond absolute maximum voltage rating for a WS2812 LED so they will blow up.

Even if it did not blow up, one of them would reference the data pin to 0V and the another one to 6V so you can't control them with a single controller, you would need two separate controllers. And they would only have 6V per strip if they always consume the same amount of current, which won't happen in real life.

It is not possible. And I just realized this is a product usage question which has nothing to do with electrical engineering.

Justme
  • 127,425
  • 3
  • 97
  • 261
9

WS2812B is a digitally addressable LED strip, it will use different current depending on brightness and color. It is not possible to wire them in series for this reason: the one that uses less current would get all the voltage, and die.

Note WS2812B is very inefficient and these strips draw a lot of current even when all the LEDs are set to off because the chips are still running and waiting for the digital control signals, and they are absolutely not optimized for low power. So it's not ideal for battery power. WS2812B uses about 1mA per chip when it is off. So, a 2m strip with 200 LEDs will use 200mA. If you don't turn it off with an actual switch that cuts the power supply, that will discharge your battery completely in a few days.

Besides, being RGB LEDs and not white LEDs, the white light quality is terrible.

If you want lighting, use 12V LED strips with good quality white LEDs. The 2200K 95CRI from this manufacturer is super cozy, add some 4000K or get the dual color strip if you want variable CCT.

If you want decoration, you can use WS2812B, but you will need a 12V to 5V buck converter. These LEDs use a huge amount of current, maximum 5A for every 100 LEDs, so that's a problem. Please do not get the counterfeit LM2596 from aliexpress. You could use a 12V equivalent to WS2812B but it is even less efficient.

bobflux
  • 70,433
  • 3
  • 83
  • 203
4

No. Strips connected in parallel will have the same voltage across them...12V in your case.

You could try connecting them in series and if they are perfectly identical then they would split the 12V into 6V across each strip. Will the strips work at 6V and not be damaged? Without the datasheet we don't know.

But an automotive 12V battery can be charged with voltages greater than 14V and your strips are probably not perfectly identical. One of the strips could see more than 7V across it. You would be much better off if you provide a true regulated 5V for the strips.

Elliot Alderson
  • 31,192
  • 5
  • 29
  • 67
  • WS2812 using PWM to control brightness. So the current can not be identical if it changes randomly. – user263983 Dec 26 '21 at 19:30
  • @user263983 I don't see any statement by the OP that the two strips would be illuminated in different ways. In any event, that wouldn't change my bottom line advice. – Elliot Alderson Dec 26 '21 at 19:45
  • 1
    Even the strips are illuminated in same way, inside each ws2812 chip is PWM , there is on/off moments can not be predicted due to difference in internal generator frequencies and other factors. – user263983 Dec 26 '21 at 19:51
  • In series is NOT a good idea. "if they are perfectly identical" is highly unlikely -- when this is actually needed, we are talking high-end parts selection and trimming, which will not be occurring for these. And then as @user263983 pointed out, there is the question of the PWM (on/off cycling) for each individual chip that will not be synchronized between the two strips. – Technophile Dec 28 '21 at 16:45
  • @Technophile Sorry, I thought it was obvious to the most casual reader that two strips could not possibly be "perfectly identical". – Elliot Alderson Dec 28 '21 at 18:50
  • 2
    @ElliotAlderson we have different understandings of how casual readers think. It seems to me that someone (especially inexperienced in electronics) could look at two rolled-up LED strips and see the same size, package, part number, same-appearing tape, same terminals, and same rows of little multi-colored bumps on the tape. The LED strips may be 'black boxes' to them -- e.g. they may not know what "PWM" or "component tolerances" mean. They could easily conclude "perfectly identical". You have to understand quite a bit in order to know about the complex internal structure, etc. – Technophile Dec 28 '21 at 19:02
2

Your idea is to synthesize 6V by having two "banks" of LED in series with each other. There is a critical flaw in that strategy.

North American AC power does the same thing, with 240V being split in the middle to make 120V twice. However, it has something that holds neutral in the middle - the transformer. Look up what happens when AC power has a "lost neutral" - the two legs still add up to 240V, but one is below 120V and the other is above 120V. This is destructive.

You would need something to hold DC power in the middle, and that is impossible to do with passives. When it was done in ancient times, e.g. for "control voltage for multi-unit tram, interurban and subway trains", it was done with a resistor ladder. The voltage varied quite a lot with use, possibly 100V open-circuit, but 40V with the worst-case of a full length train with 3 contactors energized on each car. The equipment was designed with a high tolerance for that. But this is a terrible solution as it is very wasteful of electricity.

So you are stuck using a silicon solution! Might as well use a DC/DC converter.

1

No, use a 5V regulator, like 7805 and then connect in parallel. If you want to divide the voltage, the strips must be connected in series. That means when fully charged (12V) the voltage across each would be 6V. However, most 5V strips (including yours) will blow up when applied above this rating. I won't recommend connecting directly to 12V.I

No Man
  • 347
  • 2
  • 9
  • 4
    It is better to use a DC-DC converter rather than a linear regulator like the 7805. A linear regulator "wastes" power - it acts like a series resistor - and will get very hot powering an LED strip. – Peter Bennett Dec 26 '21 at 17:38
  • 4
    LED strips generally consume more current than a 7805 can provide. Even if it could provide the current, it can't handle the power dissipation required to provide that current. So, bad advice. – Justme Dec 26 '21 at 18:22
  • @Justme The above-mentioned LED strip draws much less current than the 7805 can provide and can easily power up the 2 such LED strips and thus producing no considerable heating effect. – No Man Dec 27 '21 at 04:30
  • @Noman "much less current" for the WS2811 is true -- for a single LED module. 3 LEDs at 20mA / LED = 60mA. However for a lighting project, let us assume at least 100 LEDs: 60mA x 100 = 6,000mA (6A). Worst-case supply voltage is around 14V (engine running, alternator is charging the battery) plus load-dropping spikes, however ignoring those, 14V - 5V = 9V drop. 9V x 6A = 54W. Except the 7805 has built-in thermal protection, so it will be in shutdown most of the time. – Technophile Dec 28 '21 at 16:52
0

How large of a lighting project, and when and how long do you plan to run it with the engine off?

For a WS2811, a single LED module has 3 LEDs and runs at around 20mA per LED, total 60mA. If you have an LED strip with 100 LEDs, that's 60mA x 100 = 6,000mA = 6A at 5V.

This will mean 6A at 12V if you use a linear regulator, which will discard 6A x (12V - 5V) = 42W as heat. With the engine running, expect 14V at the battery, so 6A x (14V - 5A) = 54W.

Using a switch-mode regulator, let's say 90% efficient, will reduce draw closer to 6A x 5V / 12V / 0.90 = 2.78A.

Either is going to run your battery down fairly rapidly.

You might want to look for high-efficiency LEDs that will provide more light per mA.

Technophile
  • 441
  • 2
  • 9