DIY leds Discussion Thread for all your how tos and doubts and anything related

[Aristoned]

Im not entirely sure what your situation is or if i understand you correctly.

Your mixing gen7 and gen 8 vero cobs?

Youre seeing higher light intensity coming out of the gen8s?

Yes.

The Gen 7 run at higher voltages, 3 volts at nominal. Putting in-line with an 8 will push a higher voltage through the 8. The 97CRI is a tough one on PAR. I wish I would have gone with the 90CRI instead, but using the 3000K 80CRI offsets the 97CRI.

 

[Rocket Soul]

Yes.

The Gen 7 run at higher voltages, 3 volts at nominal. Putting in-line with an 8 will push a higher voltage through the 8. The 97CRI is a tough one on PAR. I wish I would have gone with the 90CRI instead, but using the 3000K 80CRI offsets the 97CRI.

I edited above; i think this means they are running on unbalanced load. More current in low voltage cobs.

 

[Aristoned]

I edited above; i think this means they are running on unbalanced load. More current in low voltage cobs.

From what I understand, the higher voltage COB’s will demand a higher forward voltage, the driver will drive the current and voltage until 200W is achieved. If that means less 80CRI and more 97CRI then I’m ok with that. As we add more COB’s the forward voltage increases due to voltage drop between COB’s.

From what I’m seeing and reading from the PAR meter is that the gen 8 is getting kicked hard.

I’m up to 700-800 ppfd in the ~16-18” range. I’ll have to measure height and PAR tonight.

 

[Rocket Soul]

Depends on how you got them connected, series or parallel. What driver for the cobs?

From my understanding current over each string in parallel is calculated as if the led was a resistor; in that higher voltage of the individual cob, lower current passing thru it.

If you have them in series all cobs should have the same current.

My understanding maybe rusty. @Absorber youre a sparky iirc, can you clues us up? Is this a case of the 1/R for each string equation? Am i getting this wrong?

 

[Aristoned]

Depends on how you got them connected, series or parallel. What driver for the cobs?

From my understanding current over each string in parallel is calculated as if the led was a resistor; in that higher voltage of the individual cob, lower current passing thru it.

If you have them in series all cobs should have the same current.

My understanding maybe rusty. @Absorber youre a sparky iirc, can you clues us up? Is this a case of the 1/R for each string equation? Am i getting this wrong?

I’m running everything in series to prevent fires and destruction of diodes. I can build a few panels that can lose half of the string and not blow up though.

In this case the current should be constant while the voltage changes to achieve 200W.

XLG-200-L 350-700 (1050 over-drive)

 

[Neferhotep]

push a higher voltage through the 8

You can't "push" voltage through something.
You can however push current through something.

With a CC driver that has high enough voltage the following will happen when you push 1 amp though both cobs in series.
The Vero 18B needs 35.4 volt; the Vero 29B needs 49.3 volt.
The driver will thus have to provide 84.7 volt.
Total power P = UI = 84.7 × 1 = 84.7 watt.

 

[Aristoned]

You can't "push" voltage through something.
You can however push current through something.

View attachment 19193601

With a CC driver that has high enough voltage the following will happen when you push 1 amp though both cobs in series.
The Vero 18B needs 35.4 volt; the Vero 29B needs 49.3 volt.
The driver will thus have to provide 84.7 volt.
Total power P = UI = 84.7 × 1 = 84.7 watt.

View attachment 19193680

I have the 8 on the end of two 7’s, I was to the understanding the forward voltage would carry over the 8 to achieve the wattage required by the 7’s. Given that each COB would require slightly higher voltage to get through to the next, ending up with higher voltage on the 8 at the end of the series.

I’m still trying to understand these things.

 

[Neferhotep]

If you send 1 amp though a cob the current-voltage scheme tells you what voltage that would take.
If that current goes through two cobs in series you have to add up the voltages.

 

[Drop That Sound]

^ So, I think I can get away with using a 20a rotary changeover switch like this, to use one single ammeter (a PZEM-031, in my case) for the 3-4 different variable output channels from my atx supply..
As long as whatever type of rotary switch breaks contact between the positions. Not a "Make-before-break" type apparently, which would arc and short out the rails when I switch from one to the other, and make the psu shutdown..
Not sure if it will actually work properly though, and give correct readings with so many connected grounds. Or even let me use the other channels at the same time and not just the one i'm metering.

Hmm..

 

[Aristoned]

If you send 1 amp though a cob the current-voltage scheme tells you what voltage that would take.
If that current goes through two cobs in series you have to add up the voltages.

I understand that much, what is foggy from years ago is mixing lower voltage with higher voltage. I remember guys building them always said to make sure you don’t mix voltages otherwise it will burn out the lower voltage chip. I figured 3V delta wasn’t tough and I have a fan blowing on the heatsinks.

I will say without any doubt, cooling your COB is the best option for getting the most out of them. I could feel notable warmth from the heatsinks before the fan, now I don’t feel any heat.

 

[Aristoned]

View attachment 19193816
^ So, I think I can get away with using a 20a rotary changeover switch like this, to use one single ammeter (a PZEM-031, in my case) for the 3-4 different variable output channels from my atx supply..
As long as whatever type of rotary switch breaks contact between the positions. Not a "Make-before-break" type apparently, which would arc and short out the rails when I switch from one to the other, and make the psu shutdown..
Not sure if it will actually work properly though, and give correct readings with so many connected grounds. Or even let me use the other channels at the same time and not just the one i'm metering.

Hmm..




View attachment 19193822

Is there a simple way to have a LCD display the voltage for each circuit? A simple five digit display would suffice, amperage would be relatively low, voltage would be under 300V I’m sure.

 

[Drop That Sound]

 

[Absorber]

Depends on how you got them connected, series or parallel. What driver for the cobs?

From my understanding current over each string in parallel is calculated as if the led was a resistor; in that higher voltage of the individual cob, lower current passing thru it.

If you have them in series all cobs should have the same current.

My understanding maybe rusty. @Absorber youre a sparky iirc, can you clues us up? Is this a case of the 1/R for each string equation? Am i getting this wrong?

Sounds correct to me

 

[Absorber]

I use an old computure power supply to power my shutters on my automatic air extraction setup for my two tents and it also supplies power for my light lifters in each tent

 

[Absorber]

View attachment 19193816
^ So, I think I can get away with using a 20a rotary changeover switch like this, to use one single ammeter (a PZEM-031, in my case) for the 3-4 different variable output channels from my atx supply..
As long as whatever type of rotary switch breaks contact between the positions. Not a "Make-before-break" type apparently, which would arc and short out the rails when I switch from one to the other, and make the psu shutdown..
Not sure if it will actually work properly though, and give correct readings with so many connected grounds. Or even let me use the other channels at the same time and not just the one i'm metering.

Hmm..




View attachment 19193822

You will only have 3 choices with that rotary switch

 

[Drop That Sound]

You will only have 3 choices with that rotary switch

I know. They have ones in that style that have more positions\terminals too, like 5 or 6 settings including the off postition. Not even sure yet if that is the non shorting kind though. Should be good.
I think even postition 0 can be utilized on some of those rotary switch models, and also has terminals. They have ON/ON/ON/ON, etc..

I'll make sure I get the right kind. Hopefully something more compact, because those 20a or higher switches are huge, lol. Ah well.

You have motorized light lifters in your tents?!!

I was planning on adding that feature directly to my fixture\s at some point. Probably with 2 seperate lifts on each end, so I can tilt the light to whatever angle from side to side if needed.

I also need to make an auto chicken coop door opener sometime this summer, and maybe could use some of the same components

 

[Drop That Sound]

 

[Absorber]

You have motorized light lifters in your tents?

Yea i designed and 3d printed them

Ive since added a switch to it this was the prototype video

Not even sure yet if that is the non shorting kind though.

Id be confident in saying it is the non shorting kind as you put it

 

[Neferhotep]

mixing lower voltage with higher voltage

As you can also see, it's not possible to put these two chips in parallel. Lighting up the 29B would require some 47 volt. That voltage would imediately kill the 18B.
Parallel is not possible here.

 

[Aristoned]

As you can also see, it's not possible to put these two chips in parallel. Lighting up the 29B would require some 47 volt. That voltage would imediately kill the 18B.
Parallel is not possible here.

Ah!

So, that’s what they were talking about. That makes perfect sense, each channel will receive the necessary voltage for the highest channel. That means, if I take the Gen 8 out of series and put it in parallel it will get +3V.

I’m more interested in running four cobs over the three right now, that pattern will give me the center section to add the largest IR bulb available.