x100 LEDs - Is a Resistor needed @ 3.0v ?

[spondootre]

Once again thanks,

I have used various resistance calculator etc etc which is what origonally brought me to ask the question is a resistor needed when running within the allowed supply range 3.0v - 3.6v in this case. As the calculations come to 0 at the correct supply voltage ie -

3.3v - 3.3v = 0v

0v/30mA = 0 ohms

Some calculator automatically round the value up to 1 Ohm as being the nearest value to 0, other just say 0 !!

I have been assured by one of the suppliers of the LEDs that no resistor is needed for the LEDs if they are to be operated strictly between 3.0 - 3.6 v . But again this is just another view!!

 

[audioguru]

Each LED comes with a free current-limiting resistor. Then why bother about blowing them up in parallel?

I wonder how long it will take for them all to blow up in parallel? The one with the lowest voltage will blow up first, then the rest in the sequence of their voltage. They might even stop blowing up if a bunch of them have the same voltage.
Maybe some will blow up just one colour or two instead of blowing up all 3 colours.

Try connecting 100 in parallel then tell us how long it took for them to blow up.

 

[spondootre]

Well they have been connected to the PSU in parallel for about a month or so, turned on for apprx 6 hours a day and none have blown yet? Is this because they are running within their operating voltage?

Let me get this straight, I don't want to connect this up and have it blow up, I have no wish to midlessly keep changing the blown LEDs forever more, which is why I would like to wire it safely before installing it. I am just not understanding the reason why they would blow up (although they haven't yet) when operating well within their supply range of 3-3.6v . I understand that LEDs current draw is exponentially related to voltage across it, but @ 3.3v this current draw shouldn't be higher than 30mA and granted if the voltage was to go to say 3.8v then it would be at the peak operating current of around 75mA but why would it go that high? It's regulated supply, measured with a multimeter at the highest of 3.5v and lowest of 3.35v during the colour change cycle. What point am I missing??

Please don't give me anymore scenarios of how the LEDs will 'blow up' or how the whole thing is doomed!! Just please enlighten me as to what I can do to make it run safely, the options available are -

1) install resistors in series with each LED @ 3.3v (would this be 1ohm then?
2) Up the voltage to 5v and install resistors with each LED
3) Leave it as is.

 

[Gorgon]

Measure the current through one led connected to the 3.3V via a resistor of 10ohm. You may want to do this on several leds. Use that to calculate the resistor to get the current you want. Without resistors you have no control with the current consumption whatsoever.

TOK

 

[Dr.EM]

Option 2 is your best bet. No problem doing that, resistors of 68ohm can be used with each LED for about 25mA current.

 

[audioguru]

You don't have LEDs, you have an IC circuit that is driving LEDs that probably have different voltages.
You don't have a detailed datasheet for them to see how high is the current at a voltage higher than typical. It doesn't even say a typical voltage, just, "less than 3.6V" and not saying at what current.

If the driver IC has outputs that saturate, then the voltage depends on which colour is lighted.
The max continuous current rating is 30mA. Some will draw 30mA with a voltage of 3.6V and others at 3.0V or less. The ones that draw 30mA at 3.0V will draw maybe 100mA at 3.3V and will quickly burn out. Then others in parallel will have more current than before.
Did you try all 100 to see if they light with only 3.3V? Did you measure the current of the brightest ones?

Use a 5V supply. Then a 75 ohm resistor in series with each one. Ones that operate at 3.0V will get 27mA. Ones that operate at 3.3V will get 19mA and ones that operate at 3.6V will get 19mA. None will burn out if they are any good.

 

[Hero999]

You don't have LEDs, you have an IC circuit that is driving LEDs that probably have different voltages.

Exactly, thay behave more like flashing LEDs. Quite often no series resistor is required because the IC limits the current. Some flashing LEDs work from 5V others from 12V, the type I have will work from 4.5V to 16V, no current limiting resistor is required!

Let's look at the specification, because it contains some contradictions:

Emitted colour: Slow RGB change.

How does this work?

I assume it slowly change from one colour to the next by blending one colour into another. What would be a simple way of doing this? You could use PWM to vary the duty cycle of each chip or just vary the current in a linear fashion.

Forward voltage: <3.6V

I assume this is because it varies depending on the colour. I suppose this could indicate you need a variable resistor.

Reverse Current (?A): 30

What does ?A mean? Does it mean they're not sure and 30 what, amps? The normal acceptable reverse current for an LED is normally microamps or even nannoamps, never miliamps or amps!

Maximum power dissipation: 80mw

Is that continious or instantanious. I've noticed they've used a lower case w for Watts, is this a mistake? In electrical enginneering lowercase units mean instantanious imits whilst upercase is continious.

Maximum continious foward current: 30mA
Maximum continious forward current: 75mA

Either these forward current specs are bs or the power dissipation is. 30mA*3.6V = 108mW and 70mA*3.6=252mW, what a load of utter rubbish!

Quantity : 100 and Free Resistors (Work for 12v)

What does work for 12v mean? Note the lower case v again, so does it mean they will run from 12V for short periods?

Approximately 5 second routine with colour merges and transitions.

I suppose this partly answers the first question regarding the emitted colour.

My conclusion
This specification is total crap. I wouldn't buy this rubbish as they obviously don't know what they're bloody talking about!

 

[spondootre]

Emitted Colour
RGB Slow Colour Change

Size (mm)
5 mm

Lens Colour
Water Clear

Forward Voltage (V)
< = 3.6

Reverse Current (?A)
< = 30

Luminous Intensity Typ Iv (mcd)
4000 - 5000

Life Rating
100,000 Hours

Viewing Angle
20 Degree

Absolute Maximum Ratings
( Ta = 25°C )

Max Power Dissipation
PM = 80 mw??

Max Continuous Forward Current
IFM = 30 mA

Max Peak Forward Current
IFP =75 mA

Reverse Voltage
5 ~ 6 V

Lead Soldering Temperature
240 ° C ( < 5 Sec )

Operating Temperature Range
- 25 ° C ~ + 85 ° C

Preservative Temperature Range
- 30 ° C ~ + 100 ° C


Hero999 -

I couldn't speak in defense of the supplier other than to say that the notation of capital and lower case letters are probably a result of the language difference between there and here and not negligence or incompetence on their part.
I can say that they work with the supplied resistors (570 Ohm) @ 12VDC + as I have had them installed in my van for several months.

audioguru -

Thanks for your reply, That clarifies the issue for me much better!! It hadn't clicked into place why I may need to limit the current with a resistor but now I am beginning to see why. I will take your suggestion and up the Voltage to 5V with a resistor in series with each one.
I haven't measured the current drawn by any of them, I went to do that this evening and have found my multimeter doesn't want to register current through it so i'll have to get another one. There isn't any noticable difference in brightness at all between any of the LEDs.

 

[Hero999]

I can say that they work with the supplied resistors (570 Ohm) @ 12VDC + as I have had them installed in my van for several months.

Why didn't you say so in the first place?

Tha worst case current consumption assuming 13.8V for the battery and 1.9V for the red LED is:
[latex]I = \frac{13.8-1.9}{570} = 21mA[/latex]

So use a 12V power supply and connect them in groups of three in series.

Now let's calculate the series resistor using 25mA for the maximum current through the 1.9V red LED, with a 13.8V supply:

[latex]R = \frac{13.8-1.9 \times 3}{0.025} = 324 \Omega[/latex]

Use the next value up 330R.

 

[audioguru]

570 ohms is not a standard value. 560 ohms? Then at 3.3V the current is only 15.5mA.
They have an IC in them so they shouldn't be wired in series.

 

[Hero999]

There's no harm in trying it though, if they don't work properlty in series then connect them in parallel each with their own series resistor and use the 5V supply.

 

[spondootre]

All 200 LEDs now installed in the bathroom ceiling. They are operated via 2 capacitive switches which are hidden behind the tiling. So far so good. I'll let you know in 6 months if there have been any losses!!

**broken link removed**
**broken link removed**
**broken link removed**
**broken link removed**

 

[audioguru]

I thought the LEDs each fades 3 colours. Then the colours should be random but they aren't. Some LEDs aren't lighted. Burnt out allready?

Your bathroom looks nice (?). I was in a restaurant that looked like that.

 

[Sceadwian]

Keep in mind, those aren't strictly speaking LED's they're an integrated circuit. The colour changing effect has to come from some kind of built in driver. That's going to include some level of curent regulation if for nothing else internal biasing. If you really want to know what's going on, ask for a schematic of the internal workings of the 'units'. Seeing as how you effectivly have three diodes built into the module, even if you burn one of them out the other two will still work. You're more likley to burn out the drive circuit's input. As much as a pain in the ass getting that information may be, they're going to be significantly more forgiving than a standard LED.

P.S.
Won't after a significant amount of time the colour change rate fluctuate to the point where they don't sync anymore? Or do you 'reset' the diodes periodically by turning power off for a blink?

 

[spondootre]

No, happy to say so far none have burnt out, there are two sets of LEDs colour fade ones and white light ones, they are place randomly. The colour fade ones have 3 colours and 7 cycles. What you are seeing on the picture is the first 3 cycles Red-blue-green it then does combinations of 2 and 3 colours together. After a few cycles they go out of sync and are a bit more randomly coloured.
The jacuzzi has an underwater colour change light and a thick perspex panel to view it, but that doesn't run the same cycle as the ceiling ones.

**broken link removed**

 

[spondootre]

They do re-start the cycle everytime they are switched on but to be honest when they fluctuate out of sync completely the effect is still as good if not better. I'll ask for the schematics from the supplier and see if he can provide them.

 

[Sceadwian]

Keep in mind the time required to be off to reset the circuit is probably outside of the range of persistance of vision. If the LED's start and stop their colour change cycle at the same colour you can reset it each cycle to keep the colour change nearly perfectly in sync. I imagine it starts to look like 'stars on acid' after a long period of them drifting but you could add another 'mode' to the lightning. People like options =)

 

[Oznog]

The color changing ones CANNOT be put in series. Their current draw and voltage drop will be inconsistant and I'd expect it to blow LEDs right away.

 

[spondootre]

They are in parallel, supplied from a 3.3v supply.

 

[matt_d82]

That looks really cool!!! You should take a video if you can and post it!