CONNECTING AND DRIVING MULTIPLE LEDS

Thread Starter

Hugh Riddle

Joined Jun 12, 2020
23
CONNECTING AND DRIVING MULTIPLE LEDS
The LEDs in strings of party lights appear connected in parrallel and some strings are powered by dry batteries (typically 2 or 3 AA cells). Yet we are strongly advised not to parrallel LEDs or to drive them from a constant-voltage source. So what's going on in those strings and their LEDs?
 

Dodgydave

Joined Jun 22, 2012
9,195
Those are relying on the Internal resistance of the battery, they are 4.5 V with all LEDs in parallel.

Solar LEDs are the same they use a NiMH battery and buck booster to 3V , but the current is limited to 20mA, .

All LEDs are Current driven, and usually have a Series resistor per led.
 

Thread Starter

Hugh Riddle

Joined Jun 12, 2020
23
Those are relying on the Internal resistance of the battery, they are 4.5 V with all LEDs in parallel.

Solar LEDs are the same they use a NiMH battery and buck booster to 3V , but the current is limited to 20mA, .

All LEDs are Current driven, and usually have a Series resistor per led.
Thanks Dodgydave. As you indicate, the total current must be (partly) limited by the battery's internal resistance. However, on my 4.5V battery-powered string of 40 white LEDs, the wiring goes directly to the LED pins and connects them all in parrallel and they all look equally bright. So I'm still a tad mystified.
 

peterdeco

Joined Oct 8, 2019
181
We've been paralleling LEDs without individual resistors for years. As long as they're all exactly the same, we have never had a problem. However, using a different manufacturer's LED with a slightly lower voltage drop will reduce the brightness of the others.
 

MisterBill2

Joined Jan 23, 2018
6,271
There may be current limiting resistors that are very well hidden, OR they may be LEDs with an internal current limiting circuit. Just because you do not see the rest of the circuit does not mean it is not there.
 

Thread Starter

Hugh Riddle

Joined Jun 12, 2020
23
We've been paralleling LEDs without individual resistors for years. As long as they're all exactly the same, we have never had a problem. However, using a different manufacturer's LED with a slightly lower voltage drop will reduce the brightness of the others.
Thank you peterdeco for confirming what I had suspected. May I ask what type of product/application and whether each set needs to come from the same production batch?

I reckon there's a bit more worth investigating here as one might expect current sharing between LEDs to be unstable due to LED turn-on voltage decreasing as LEDs heat themselves up ('current hogging') - so back to the bench for me.
 

Thread Starter

Hugh Riddle

Joined Jun 12, 2020
23
There may be current limiting resistors that are very well hidden, OR they may be LEDs with an internal current limiting circuit. Just because you do not see the rest of the circuit does not mean it is not there.
There may be current limiting resistors that are very well hidden, OR they may be LEDs with an internal current limiting circuit. Just because you do not see the rest of the circuit does not mean it is not there.
Or could it be that LED V-I characteristics allow good current sharing between parrallelled connected LEDs?
 

peterdeco

Joined Oct 8, 2019
181
We buy LEDs in large quantity from China. Yes, they all come from the same production batch and are used in different configurations. On one 9V product, we put three 3V LEDs in series, and parallel them with multiple strings of three. On others, they are just paralleled and run off of 6V with one series dropping resistor.
 

Audioguru again

Joined Oct 21, 2019
1,562
When you buy a few LEDs, they might not all be from the same batch or even from the same manufacturer. Then if you put some in parallel their differences will cause one or two to be very bright and burn out soon and cause the others to be dim.
It is easy to for you to measure them and make matched forward voltage sets of them.
 

Tonyr1084

Joined Sep 24, 2015
4,687
I tested all 100 of my LED's of each given color. Listed them in an Excel spread sheet and did the averages and the standard deviations. So I know those LED's used in the video were within 10mV of each other.

What @Audioguru again is saying is that if you have six parallel LED's with similar but not close forward voltages (Vf), the one with the lowest Vf will draw the most current. In my video notice that the red (lower Vf) drew all the current. If I hadn't set the resistance at 100Ω but at some lower level, assuming all six LED's would share current alike, then there would have been too much current for the red LED. It would have burned out quick. That, in turn, would leave excess current for the rest of the LED's. With more current available, the next lowest Vf will be the next to blow out. Then with the third, the fourth and so on. So it's not advised to put LED's in parallel, but the video shows it IS possible as long as you don't rush too much current through a single LED. Remember, the red LED had a Vf of 1.95 volts. With a supply of 4.5 volts, subtract the 1.95 volts and you're left with 2.55 volts to calculate current with. 2.55V ÷ 100Ω = 25.5mA. That's a lot for an LED, but survivable. Notice how bright it was at the start of the video. The blue LED's had a much higher Vf, so they shared a lower current. And when removing one from the circuit, the rest didn't blow out because there was never enough current to blow them all out.

One other experiment I did (by accident): I had some coin cells I wanted to completely drain of power. So I slipped LED's on to them. Two things happened: First, the LED didn't burn out because the coin cell could not deliver more current than the LED could handle. Second, even with dead (extremely low voltage) the LED's were lit for weeks before they became too dim to distinguish. Because the cell could not drive more current, the LED was safe from harm. With multiple parallel LED's, a double A battery isn't that big a threat. There is resistance in the wire, so some how someone has figured out how to drive a bunch of LED's from an AA battery pack. It's doable. Both the string you have in hand and the video prove it.
 

Thread Starter

Hugh Riddle

Joined Jun 12, 2020
23
We buy LEDs in large quantity from China. Yes, they all come from the same production batch and are used in different configurations. On one 9V product, we put three 3V LEDs in series, and parallel them with multiple strings of three. On others, they are just paralleled and run off of 6V with one series dropping resistor.
Here's an experiment I did. This is my video by the way.

Really nice little video with a great practical feel - and wisely keeping clear of the much trickier business of understanding why LEDs of same type can be parralleled, which involves other factors (incl spreads of threshold ('turn-on') voltage, slope resistance(s) and thermal resistance of the LEDs used).
 

Thread Starter

Hugh Riddle

Joined Jun 12, 2020
23
When you buy a few LEDs, they might not all be from the same batch or even from the same manufacturer. Then if you put some in parallel their differences will cause one or two to be very bright and burn out soon and cause the others to be dim.
It is easy to for you to measure them and make matched forward voltage sets of them.
Best to try it (with any old handful of LEDs of the same type code) rather than asserting without evidence that there will be large brightness differences or damage.
 
Top