Electronics > Beginners
Hooking up 4 x 3v LED string lights to a 6v PSU
gf:
--- Quote from: bob21 on February 02, 2020, 05:39:40 pm ---Question is, is it wise to run it at it's limit? Would it not be better to take it down a notch and say, for example:
Each LED 7mA
--- End quote ---
Well, it's a matter of the desired brightness. You can even go down further if you don't need that much brightness. Just try different currents and check visually. But don't exceed the maximum current given in the datasheet of the LED chips! And, since the thermal conduction of the bulbs is likely unknown, I'd even stay quite below the maximum current given in the data sheet, in order to prevent overheating the LED chips. Overheating is one of the major causes for a premature death of LED chips.
For the calculation of the current limiting resistor I would also measure the actual forward voltage of the LEDs at the desired operating point, and I'd also measure the PSU voltage under the actual load (it may drop when loaded, and even drop significantly if it gets overloaded and a current limiter happens to joins in).
--- Quote ---The PSU is a 6V DC and I checked it by switching my Multimeter to '20' on the DC side and the reading was 6.01V.
--- End quote ---
Is it a SMPS or a traditional PSU?
I.e. does it provide clean DC (filtered, or even stabilized), or just pulsating DC directly from the rectifier?
Do you read any voltage with the multimeter switched to AC mode?
[ Note, in case of pulsating DC, the calculation of the value and power dissipation of the current limition resistor is no longer that simple, but requires integration of the rectified sine wave over the time interval where the waveform is larger than the LED's forward voltage. ]
bob21:
--- Quote from: gf on February 02, 2020, 08:14:25 pm ---Well, it's a matter of the desired brightness. You can even go down further if you don't need that much brightness. Just try different currents and check visually. But don't exceed the maximum current given in the datasheet of the LED chips! And, since the thermal conduction of the bulbs is likely unknown, I'd even stay quite below the maximum current given in the data sheet, in order to prevent overheating the LED chips. Overheating is one of the major causes for a premature death of LED chips.
--- End quote ---
Is it daft to say, as bright as possible? I obviously want them to look good, but if they're going to burn out then they will of course have to come down a bit. As mentioned before, I don't so much care if they die at this point (as this is a learning curve for me) but what I would care about is if they caught fire. So if by 'death' you mean they go off and never come on again ergo 'you've lost £3' then I'm annoyed, but not that much vs them overheating catching fabric and setting a blaze to the place because I didn't keep a constant eye on them with a fire extinguisher in hand.
--- Quote from: gf on February 02, 2020, 08:14:25 pm ---Is it a SMPS or a traditional PSU?
I.e. does it provide clean DC (filtered, or even stabilized), or just pulsating DC directly from the rectifier?
--- End quote ---
The PSU is a traditional PSU. It's an old Jabra brick Model: ACGN-22B
The label says: Output 5-6V max 5W (6V 300mA)
--- Quote from: gf on February 02, 2020, 08:14:25 pm ---Do you read any voltage with the multimeter switched to AC mode?
--- End quote ---
Yes, I have just checked and I read 13V AC when red probe is to pos and black/common probe is to neg. I don't get a reading if I swapped them round (in DC I would see a negative value ie. -13, I didn't see this).
--- Quote from: mariush on February 02, 2020, 06:58:52 pm ---Yes, well that's no load.
Repeat the measurement with 20 leds connected to it and see if you still get 6v. You may find out that the closer you are to its maximum advertised output of 300mA, the voltage may go down a bit. So you may see 5.5v at 250mA or something like that.
--- End quote ---
Using a friends meter, which cost a bit more than mine, but is still no Fluke I did this:
1. Took a voltage reading with nothing attached to PSU: 6.19V
2. Took an Amp reading by connecting the red probe to the unfused 10A port and switching the dial to 10A: it fluctuated .44-.49 so WAY over the quoted measurement if I understand that .3 would equate to 300mA.
I then repeated, by hooking up all 41 LEDs for max load as the above diagram (Circuits A and B comprised of 1+2 for A and 3+4 for B):
1. Took a voltage reading by bearing the cable at the point the purple dot is shown and equivalent on the neg wire. The reading was: 5.93V
2. Took an Amp reading using the same method as above, all the LEDs go out upon probe contact and the reading is the same: .44-.49
--- Quote from: mariush on February 02, 2020, 06:58:52 pm ---Reuse the formula just put R instead of current, for example
Circuit B - 21 LEDs: 6v - 3v = 0.21 x R => R = 3 / 0.21 = 14.28 (so safe to just use 15?)
Rewrite it like this
Circuit B - 21 LEDs: 6v - 3v = Current x 15 => Current = 3 / 15 = 0.2A (200mA / 21 leds = 9.52mA per led)
--- End quote ---
Got it, thanks, that makes sense.
--- Quote from: mariush on February 02, 2020, 06:58:52 pm ---The power supply may be conservatively specified, you may find out that it can actually output 500mA while the voltage goes down to around 5.5v ... not saying it would a good idea to use such a power supply for long time above its advertised capabilities... it's up to you.
--- End quote ---
Well, it seems from the testing I have just done that you have hit the nail on the head. It's way over stated spec. I don't want to use something or create something unsafe.
--- Quote from: mariush on February 02, 2020, 06:58:52 pm ---Yes, you can use resistors rated for higher wattage than needed. You can also parallel resistors to spread the heat across a larger surface. When you parallel two resistors, you get half the resistance but 2x the area to dissipate heat.
So for example, instead of using a 10 ohm resistor, you could have two 22 ohm resistors in parallel ending up with 11 ohm
--- End quote ---
Will this generate more heat than a single 3W?
--- Quote from: mariush on February 02, 2020, 06:58:52 pm ---You can use 3w resistors but they're big and ugly. I'd look into paralleling 2 1w resistors first.
--- End quote ---
This may be less of a problem, because I could end up housing this in a small ABS box that's non-offensive. But I will keep both methods in mind.
So what's the verdict on this PSU? It's not fit for this task....? Any recommendations for a replacement?
Damianos:
--- Quote from: bob21 on February 02, 2020, 01:52:22 pm ---Wow, thanks so much for such a time consuming write up Mariush, and thank you to you also Brumby for all the info. I totally appreciate the time people are taking to help me with this.
I am still re-reading everything and trying to get my head round it, but, I have added a 4th line to the circuit like this:
So if I understand correctly, each parallel circuit (let's call them A and B) has almost the same amount of LEDs.
If I understand correctly, a 300mA 6V PSU should be dishing 150mA to A and B. 150mA / 20 = 7.5mA, so each LED is receiving 7.5mA - or am I totally off target here?
I think I am definitely going to grab some resistors to make this safer, so I have also marked the diagram with coloured dots.
Would I just need one resistor where the purple dot is? Or would I need 2, where the green dots are? I am thinking the answer to this is that both A + B need one?
Would the aforementioned resistors from eBay / Rapid still be suitable?
I'm really sorry if it seems like I'm not getting this. But I'd like to say again how much I appreciate you putting up with me ;D
Update:
Ok, I have re-read all of this and I think I understand I need a resistor where the green dots are, at the beginning of each series.
I have (lets's say) 20 white 3V LED's on each series chain (A and B), and they should get no more than about 20mA.
The PSU is 6V 300mA running 2 parallel circuits, so 150mA a circuit. So looks like I only have 7.5mA available to each LED with this PSU assuming all are working, so lets limit to that.
6V - 3V = 0.0075 (7.5mA) x R ===> R = 3V / 0.0075 = 400 ohm
So I could use the 'common' 330 ohm resistors, place where each green dot is in the diagram and still be ok?
--- End quote ---
From this point and next, it cannot be assumed that the LEDs consume the same current.
Explaining this as a kid: in the midpoint of the branch B "arrives" a current equal to 6*ILED but "departs" a 15*ILED, how is this possible?
It is late at night and I am not able to post now some thoughts about possible solutions.
gf:
--- Quote from: bob21 on February 02, 2020, 09:35:40 pm ---Took an Amp reading using the same method as above, all the LEDs go out upon probe contact and the reading is the same: .44-.49
--- End quote ---
Hmm, did you hook up the Amp meter in parallel to the LEDs, i.e. shorting the LEDs with the meter?
In this case you would of course measure a higher current than flowing through the LEDs w/o the meter attached.
For current measurements, the meter must be connected in series (and for voltage measurements in parallel) to the DUT.
Brumby:
--- Quote from: gf on February 03, 2020, 12:01:28 am ---
--- Quote from: bob21 on February 02, 2020, 09:35:40 pm ---Took an Amp reading using the same method as above, all the LEDs go out upon probe contact and the reading is the same: .44-.49
--- End quote ---
Hmm, did you hook up the Amp meter in parallel to the LEDs, i.e. shorting the LEDs with the meter?
In this case you would of course measure a higher current than flowing through the LEDs w/o the meter attached.
For current measurements, the meter must be connected in series (and for voltage measurements in parallel) to the DUT.
--- End quote ---
Oh dear....
For your typical multimeter:
Voltage measurement - the meter is just peeking in on the circuit:
Current measurement - the meter becomes part of the circuit
Yes, this means you have to break the circuit to do a current measurement - which is one of the things nobody likes. This also introduces a voltage drop - called "burden voltage" - in the circuit which would not be there in normal operation. This will affect the numbers and it can cause problems. Just be aware of the burden voltage for your meter and the range it is set on. (It is usual that this varies between ranges on a given meter - and, of course, it can vary between meters.)
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version