HI,
I want to make a 40 X 3W led array (28 Red, 5 Blue, 5 UV, 2 Green) ,
35 of the LEDS has a forward voltage of 3.6V and other 5 UV Leds are 3.2-3.4V. All of the LED's max currrent is 700mA.

I want to make the array 10 leds in a series string and 4 sets of series strings in parallel.

Each 10 LED series string has 9 x 3.6fV LED's and 1 x 3.4fV LED's. Is it possible to run the different forward voltages together at 35.8V per string?

The power is coming from a 38Vdc power supply to a 5A buck module.
When i try power each string at 34-35.8V 600mA. Im finding the ends are bright but some are dim and few not working, (which are the 3.4v ones and some 3.6v ones). The String seems to flicker on and off,sometimes all the LEDs in the string,sometimes only a few.

If anyone is able to assist, id be very appreciative.

 

Welcome to AAC!

Show us a schematic of how you have the LEDs connected and indicate the ones that are dim or not working.

 

Hope this might help, I Do Not have any resistors after the last led in the strings, Is that needed if i am running Constant Current and Constant Voltage?

Under led is max forward voltageand working condition under that
W =Working
D = Dim (can see chip is lit up but maybe running at 25-30%)
VD = Very Dim (chip is lit up barely)
NW = Not Working

1st string
+34V 600mA __

3.6V 3.6v 3.4V 3.6V 3.6V 3.6V 3.6V 3.6V 3.6V 3.6V

D NW NW NW VD VD VD VD D D

2nd string (This string had more light up at start but 4-5 leds went dead after few seconds while checking which ones work)
+34V 600mA __

3.6V 3.6v 3.6V 3.6V 3.6V 3.6V 3.4V 3.6V 3.6V 3.6V
D D NW NW NW NW NW NW VD D

3RD string
+34V 600mA __

3.6V 3.6v 3.6V 3.4V 3.6V 3.6V 3.4V 3.6V 3.6V 3.6V
D D NW NW NW NW NW NW NW D

4th String (This string actually works great and very bright compared to other leds that are lighting up in other strings forgot this one has 2 x 3.4Vf)
+34V 600mA __

3.6V 3.6v 3.6V 3.4V 3.6V 3.6V 3.4V 3.4V 3.6V 3.4V
W W W W W W W W W W

 

Hope this might help, I Do Not have any resistors after the last led in the strings, Is that needed if i am running Constant Current and Constant Voltage?

You need resistors.

Under led is max forward voltageand working condition under that
W =Working
D = Dim (can see chip is lit up but maybe running at 25-30%)
VD = Very Dim (chip is lit up barely)
NW = Not Working

Using code tag with fixed font to line things up...

1st string
(35.8V nominal)
3.6V  3.6v  3.4V 3.6V  3.6V  3.6V 3.6V 3.6V 3.6V  3.6V  
D     NW    NW   NW    VD    VD   VD   VD   D     D
 
[B]2nd string (This string had more light up at start but 4-5 leds went dead after few seconds while checking which ones work)
(35.8V nominal)
3.6V  3.6v  3.6V 3.6V  3.6V  3.6V 3.4V 3.6V 3.6V  3.6V
D     D     NW   NW    NW    NW   NW   NW   VD    D

[B]3RD string
(35.6V nominal)
3.6V  3.6v  3.6V 3.4V  3.6V  3.6V 3.4V 3.6V 3.6V  3.6V
D     D     NW   NW    NW    NW   NW   NW   NW    D

[B]4th String (This string actually works great and very bright compared to other leds that are lighting up in other strings forgot this one has 2 x 3.4Vf)
(35.2V nominal)
3.6V  3.6v  3.6V 3.4V  3.6V  3.6V 3.4V 3.4V 3.6V  3.4V
W     W     W    W     W     W    W    W    W     W

I'd expect the string (4th) with the lower total nominal voltage to work best.

But, the bottom line is that you should use resistors in each of the strings to set the current in that string to be close to the others. Your supply can supply enough current to burn out all of the LEDs.

 

Constant Current and Constant Voltage?

Those can't both be controlled as constants at the same time, only upper limits. How, exactly, is this power being controlled? Are you trying to set a voltage regulator to exactly the expected sum of Vf values? Are you setting a current limit?

Also, are your 4 strings all in parallel with each other, hooked to the same power source, with no resistors? Or are you testing the strings one at a time right now? (when we ask for a schematic, we're looking for ALL the connections!)

 

At The Moment im testing one strip at a time, hence only 600mA.

Once all the strings work i will add the strings in parallel and up the current to 2.4A. I was going to get the 1Ohm 1Watt resistor according to led array calculator for each string tomorrow when joining the strings in parallel.
At the moment i was wanting to test the LED Strings individually

I Have 230VAC to 36VDC 200W PSU (Running at 38V), From the PSU it goes to a 5A CC CV Buck Controller. This sets the Voltage to constant 34V and constant current of 600mA which can be adjusted by a pots.

 

As ebeowulf said, you can't have both constant current and constant voltage. They are mutually exclusive.

Using your specified forward voltage drops across the full string of ten, you would need at least 36 volts available. But LED's are not voltage driven devices, they are current driven devices. You need to have enough voltage to push the desired current through them. Right now your available voltage is not sufficient to push 600 mA through the string. Measure it.

For your proposed strings, you would need a power supply capable of 50-55 volts and a separate current limiter for each string.

 

As ebeowulf said, you can't have both constant current and constant voltage. They are mutually exclusive.

Using your specified forward voltage drops across the full string of ten, you would need at least 49.6 volts available. But LED's are not voltage driven devices, they are current driven devices. You just need to have enough voltage to push the desired current through them. Right now your available voltage is not sufficient to push 600 mA through the string.

For your proposed strings, you would need a power supply capable of 55 - 60 volts and a separate current limiter for each string.

Can you explain how you get those totals?
would would it not be 9 x 3.6V = 32.4V + 3.4V= 35.8V? are the max forward voltages not just added?

 

The Only difference between the 4th string that works and the other 3 that dont is 1 led forward voltage.

4th string is 8 x 3.6Vf and 2 x 3.4Vf
other 3 are 9 x 3.6Vf and 1 x 3.4Vf

so there is only 0.2 max forward voltage difference in the strings, surely running the other 3 at 0.2 more voltage and same current should light the strings the same as the 4th string?

 

so there is only 0.2 max forward voltage difference in the strings, surely running the other 3 at 0.2 more voltage and same current should light the strings the same as the 4th string?

Not necessarily. We're using nominal voltages for the LEDs. If you check the datasheet, you can see what range of voltages you could have.

You get better results (brightness matching) if you drive LEDs with a current source.

 

You got to my post quickly - I edited it to adjust the numbers. Please read it again.

To get the 36 volts, I simply took the worse case of 10 3.6 volt diodes in a string. The suggested power supply voltage is so that you have enough headroom to allow the 4 constant current regulators to work properly.

 

Ok im thinking the soldering is the issue,

if i give 3.4v 600mA to each led directly on the 20mm heat sink the leds work, but there is one in each of those strings if i touch the positive and negative tab on the star heat sink they dont come on.
So seems the problem is between the heatsink connection solders and the led to heatsink solders on those leds. will try fix those and try again

 

Ok im thinking the soldering is the issue,

if i give 3.4v 600mA to each led directly on the 20mm heat sink the leds work, but there is one in each of those strings if i touch the positive and negative tab on the star heat sink they dont come on.
So seems the problem is between the heatsink connection solders and the led to heatsink solders on those leds. will try fix those and try again

Really struggling with the voltage/current relationship. If your buck converter is set up for constant current at 600mA, you shouldn't be setting a voltage limit, at least not one that's so close to the voltage you need. You could set a 4 or 5V limit for the single LED, just as an extra safeguard in case you touch the wrong thing accidentally, but for the sake of driving your LED(s) properly, set your desired current where you want it and don't set a voltage limit right at your expected forward voltage.

Maybe I'm misunderstanding you and everything is fine, but the way you're describing things makes it sound like you're trying to set both at once, which would be a mistake.

 

I Understand what you saying and i probably am explaining it wrong, this Buck converter just gives me the option of running a stable voltage and a stable current. As to if its a set Voltage or Max voltage im not sure

As far as i know i could run the Constant current at 36V, then the led will only draw whats needed ie. 12v, 24v, 33v ect.. but the converter wont let more than 36v through. same with the current.

Ive used them with 50w 12v and 36v LED cobs with no issue. set to max voltage and max current and make sure you got good cooling. but i always set the voltage to what is max required. ie. running 2 x 12V 3A cobs in series i just run at 24V 3A

Here is a link with the buck controllers info, you can see there is a pots for voltage and current
https://www.hobbytronics.co.za/p/1053/5a-lithium-charger-cv-cc-buck-step-down-power-supply

But the soldering was the issue, fixed four led's and now each string lights up the same. and thats when the controller is set at 34V 600mA

 

Glad you found your issue and got things working. Congrats!

If I were in your shoes, I wouldn't be cutting it so close on those voltage limits, but as long as it's working now, it's none of my business!

 

Have you measured the V of individual LEDs @ 600 mA ?

 

Glad you found your issue and got things working. Congrats!

If I were in your shoes, I wouldn't be cutting it so close on those voltage limits, but as long as it's working now, it's none of my business!

its my first 3w Build but been learning. The 36v PSU is humming slightly so i do think im overrunning it a bit with the voltages. Hopefully get resistors tomorrow for putting the strings in parallel. (I did join them without resistors for a few second test and could hear the psu a bit more)
I may remove a led from each string. to bring the max forward voltage down to about 32.6 and see how that helps.

Appreciate all the help though.

Have you measured the V of individual LEDs @ 600 mA ?

Not when running the strings, only tested each led indvidually at 3.4V 600mA

 

For anyone interested in what it looks like

 

Freshwater planted tank light?
Or "herb" garden

 

supplementary light for indoor aeroponics garden. the cobs will be lined in the frame