Arduino Forum

Using Arduino => LEDs and Multiplexing => Topic started by: Boopidoo on Jan 18, 2013, 10:47 am

Title: Strange LED problem (faulty RGB LED on eBay)
Post by: Boopidoo on Jan 18, 2013, 10:47 am
I'm guessing the more experienced ones will be able to answer this in a heartbeat.

I have bought this set of common cathode RGB LEDs: http://www.ebay.com/itm/321029238303?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

Despite what the ad says they are not clear and maybe they're not even common cathode?

Anyway I've made a sketch in Fritzing which shows my connections. The resistor values for Red is 1k and for Green and Blue is 470 each. The problem is that like this it only light up green and blue, not red.

When I connect it the exact same way but connect to 5V instead of ground, like a common anode, the red lights up but not green and blue.

I have a common anode RGB LED which works perfect when connected like a common anode.
Title: Re: Strange RGB LED problem
Post by: Grumpy_Mike on Jan 18, 2013, 11:49 am
Yes it looks like that those LEDs are from the reject bin, either that or they were a special. I have never come across one like that before.
Title: Re: Strange RGB LED problem
Post by: Boopidoo on Jan 18, 2013, 12:50 pm
Maybe they were indeed wrongly assembled. I guess this behaviour would be explained if the somehow mounted the RED LED the wrong way.
Title: Re: Strange RGB LED problem
Post by: Hippynerd on Jan 18, 2013, 05:43 pm
Those resistor values sound a bit high for 5v. I have CC LEDs with the same basic specs, and I use 150,100,100 ohm resistors, I think they keep my LEDs under 20ma, or at least close to 20ma.

I have some SMT LEDs that are marked with the cathode mark, but they clearly work the opposite of the markings, so I can only assume its a manufacturing error, I didnt even know that you could buy parts that are mismarked, I just never heard of that happening.

Calcualting red for 2v@20ma with a 1k resistor, you will need to feed it 20 volts, i imagine at 5, it wouldnt even light up.

If you re-do your math, I think you will find that with 5v, all your resistors will be well under 200 ohms.
Title: Re: Strange RGB LED problem
Post by: Boopidoo on Jan 18, 2013, 07:38 pm
As I wrote the LEDs does light up but only blue and green when connected as a common cathode and red when connected as a common anode.

I know the values are too high, I calculated 75ohms for G&B and 150ohm for R but I didn't have those right there and then so I used what I had using twice as high value for R then G&B. I can't imagine that using too high resistor values would invoke this behaviour, if so please explain how?

Also for testing they are indeed quite bright as is so I guess it would be close to unbearable using R that are 7 times lower. :)
Title: Re: Strange RGB LED problem
Post by: Grumpy_Mike on Jan 18, 2013, 08:10 pm
Quote
Also for testing they are indeed quite bright

Yes a lot of people are fixated on using 20mA when that is the maximum but some LEDs are just too bright at this current, especially if you are viewing them directly as opposed to through some sort of diffuser or using them to illuminate something.

Quote
I can't imagine that using too high resistor values would invoke this behaviour,

To be fair I don't think he was suggesting it would.

Quote
I have some SMT LEDs that are marked with the cathode mark, but they clearly work the opposite of the markings, so I can only assume its a manufacturing error,

I have some of those as well. It could be however a data sheet error, or in fact another type of LED that is being passed off at that type number. That is the problem with China eBay components, you never know what you will get.
Title: Re: Strange RGB LED problem
Post by: Boopidoo on Jan 18, 2013, 08:19 pm
Quote from: Forrest Gump
Buying LEDs from China is like a box of chocolate, you never know what you're gonna get.
Title: Re: Strange RGB LED problem
Post by: MAS3 on Jan 19, 2013, 12:07 am
So your common now needs to be connected to an output pin instead of GND.
You would then need to alternate that pin to get either red or the other colors to light.
Would that work ?

Not what you bought that LED for, but interesting to get it to work.
Title: Re: Strange RGB LED problem
Post by: Grumpy_Mike on Jan 19, 2013, 12:09 am
Yes but you then can't have the red on at the same time as any of the other two colours.
This LED is a mistake and is a bit useless, try and send it back.
Title: Re: Strange RGB LED problem
Post by: Boopidoo on Jan 19, 2013, 08:37 am
Below my messange also the reply from the seller.

Quote from: boopidoo
These LEDs are not common cathode. Well the green and blue is common cathode but the red LED seem to be mounted to other way so it only works if current and ground is reversed.

I'd like a new shipment of correct, 100% working, LEDs.


Quote from: dasetn
Hi, Simon I'm very sorry, I know this problem. At present I have corrected the mistake, the reason is I suppliers provide goods wrong to me, and I didn't see. Don't worry, I will resend new item to your address, water clear RGB led... Hope to understand, sincere thanks! Any question can email to contact me


So lets see how the new ones work out. However the question is how many newbies like me think they themselves make an error and doesn't bother to send a message, get stuck with bad ones and looses interest because it just doesn't work?

But they're 10pcs incl int shipping  for $2.99 so I don't complain. :)
Title: Re: Strange LED problem (faulty RGB LED on eBay)
Post by: Grumpy_Mike on Jan 19, 2013, 09:20 am
Quote
However the question is how many newbies like me think they themselves make an error and doesn't bother to send a message, get stuck with bad ones and looses interest because it just doesn't work?

Hopefully more than beforee you posted this thread.
Thanks for posting.
Title: Re: Strange RGB LED problem
Post by: Hippynerd on Jan 19, 2013, 08:54 pm

Quote
I have some SMT LEDs that are marked with the cathode mark, but they clearly work the opposite of the markings, so I can only assume its a manufacturing error,

I have some of those as well. It could be however a data sheet error, or in fact another type of LED that is being passed off at that type number. That is the problem with China eBay components, you never know what you will get.


The data provided in the sale was insufficient, so i emailed the seller, and asked about the specific forward voltages. this was the reply:
"Dear customer Hello, very sorry, but I am now because of some things, in other cities, can not send you some data, you can consult the professionals"

it was clearly translated poorly. The diagram looked like zener diodes i think.

On 5050 SMT LEDs, does the triangle mark ever identify the anodes?

reversing one diode in a set of 3 diodes with common leads is almost like they were trying to charlieplex within the package.

Heres the schematic from my diodes.
(https://lh6.googleusercontent.com/-IpnrriyaxOc/UJ24Aocyi0I/AAAAAAAAC9A/G9j8ICHZVNA/s400/RGBLED5050.jpg)

It said the forward voltage was 2.9v-3.6v ?
The LEDs work fine, but the cathode mark identifies the anodes not the cathodes.
Title: Re: Strange LED problem (faulty RGB LED on eBay)
Post by: Grumpy_Mike on Jan 19, 2013, 09:46 pm
Quote
On 5050 SMT LEDs, does the triangle mark ever identify the anodes?

No but:-
1) They might not be 5050 LEDs but some other type.
2) It could be a manufacturing error and that is why your supplier got them cheap. He obviously doesn't know what he is selling.