DIY with Quantum Boards

Chip Green

Well-Known Member
@Stephenj37826 Or anybody who has done this.... I have three 133mm pin heatsinks that came from CobKits. These are pre drilled of course, what I'm wondering, is there an issue with these holes, if I wanted to retrofit the QB 96 Elites onto them? I'm looking for no more than 150W each on these.
Fill the holes with thermal paste, before the pad goes on?
If this is a bad idea all together, I'll just get the engines, just trying to use up parts collecting dust.
 
Last edited:

2com

Well-Known Member
The best method is to attach each board to the power supply. Check its vf. Do this without adjusting anything for each board. If they are within .1 volts you'll be really good to go. Also running 3 boards on the 120h in parallel will of course not give more output but rather divide it amongst the boards.
Anyone interested...

I tried this earlier with eight QB288 V2 3000K, just for a test of the method. They were/are paired and wired in series on 240h-c2100b drivers, prior.

The setup: Driver used was an HLG-240h-c2100b, positive and negative output leads each going to three-port wago, with a length of 18g solid leaving the wago and going to the (single) QB288 pos and neg. The third wago port had the DMM test probes inserted. Attached to the dimming leads was a basic, cheap 100K ohm pot (shit, just trying to remember if that should only be for two boards right now..). All boards except the last two (#7 and #8.) were "cold" - as in not in use for days (~68*F). The last two boards/fixture had been in use earlier, and were allowed to cool to what felt like the same temp as the first six, but was measured at ~73*F. The driver/pot was dimmed all the way down before these tests (and this is a B type, not A type) and wasn't touched at all between boards/tests. So, I'd hook up a single board, and then plug in the driver, get an instant voltage reading, give it about 5-10 seconds and then unplug the driver and move to the next board.

Results:

#1: 45.83v (flashed .84)
#2: 45.80v (flashed .81)
#3: 45.91v
#4. 45.84v
#5. 45.85v
#6. 45.89v
#7. 45.81v
#8. 45.79v (flashed .80)


Thoughts on how I did this, or the results, or anything at all is welcome. For example, would there be more accurate or appropriate results performing this test at higher/closer to full power? As I need to do this for several QB96s and 288Rspecs as well (for someone wanting a parallel build) even though the Rspecs were one order - I think, I'll have to double check. I also like an excuse to use the DMM.

Thanks.
 

ChiefRunningPhist

Well-Known Member
Anyone interested...

I tried this earlier with eight QB288 V2 3000K, just for a test of the method. They were/are paired and wired in series on 240h-c2100b drivers, prior.

The setup: Driver used was an HLG-240h-c2100b, positive and negative output leads each going to three-port wago, with a length of 18g solid leaving the wago and going to the (single) QB288 pos and neg. The third wago port had the DMM test probes inserted. Attached to the dimming leads was a basic, cheap 100K ohm pot (shit, just trying to remember if that should only be for two boards right now..). All boards except the last two (#7 and #8.) were "cold" - as in not in use for days (~68*F). The last two boards/fixture had been in use earlier, and were allowed to cool to what felt like the same temp as the first six, but was measured at ~73*F. The driver/pot was dimmed all the way down before these tests (and this is a B type, not A type) and wasn't touched at all between boards/tests. So, I'd hook up a single board, and then plug in the driver, get an instant voltage reading, give it about 5-10 seconds and then unplug the driver and move to the next board.

Results:

#1: 45.83v (flashed .84)
#2: 45.80v (flashed .81)
#3: 45.91v
#4. 45.84v
#5. 45.85v
#6. 45.89v
#7. 45.81v
#8. 45.79v (flashed .80)


Thoughts on how I did this, or the results, or anything at all is welcome. For example, would there be more accurate or appropriate results performing this test at higher/closer to full power? As I need to do this for several QB96s and 288Rspecs as well (for someone wanting a parallel build) even though the Rspecs were one order - I think, I'll have to double check. I also like an excuse to use the DMM.

Thanks.
If you're mixing different model boards then I think you'd want to test Vf at operating temps as Vf will be effected differently for different chips and different topologies (if individual fixture Vf is important to you). I don't think matchimg Vf is important when using a C.V. and wiring loads in parallel, except to maintain equal intensity across your fixture. As long as your parallel fixtures are consuming currents which they're rated for, matching the load Vf's is less than important. I've been a contributor to the thermal runaway fear but I'm really lacksing my position. If you're not worried about your board enduring thermal runaway in series, you shouldn't have any fear when in parallel. When in parallel all loads will be under the same V. As long as the current consumption per fixture is within ratings and your intensity matches, you'll be good regardless what the Vf of the individual load may be. An LM561C fixture will be less efficient than a LM301B fixture and thus a lower Vf on the LM561C (less chips in series) may help offset the intensity difference, this would be an example of matching fixtures that don't have the same Vf but flow current at rated amounts while matching other parallel fixtures intensities.

The biggest fear with DIY LED imo is if you're wiring parallel but using a C.C. driver. If you lose a parallel fixture, your remaining load will absorb the extra current. The smaller the parallel string, ie the less loads in parallel, the worse the damage when losing a single parallel fixture. I recommend inline fuses before each parallel fixture to mitigate this possible damage. This isn't thermal run away though.
 
Last edited:

Airwalker16

Well-Known Member
@Stephenj37826 Or anybody who has done this.... I have three 133mm pin heatsinks that came from CobKits. These are pre drilled of course, what I'm wondering, is there an issue with these holes, if I wanted to retrofit the QB 96 Elites onto them? I'm looking for no more than 150W each on these.
Fill the holes with thermal paste, before the pad goes on?
If this is a bad idea all together, I'll just get the engines, just trying to use up parts collecting dust.
The holes aren't a problem, I just worry they won't fit on the 133's or if the base is thick enough for higher wattage. At least the heats spread better than a cob.
 

2com

Well-Known Member
This was all from a couple pages back, when the concern was (re-)raised by someone, and this was pointed out:
"PARALLEL WIRING
Multiple Boards can be wired in parallel. Different batches of boards may have slightly different voltages. Use only boards from same batch for parallel connection. Do not use different spectrum boards in parallel as they may have slightly different voltages."

-HLG QB Guide.

I remarked how I can't believe I'd missed or forgot seeing that on the guide. And of course, once I heard it, I had to be "worried" and go with the advice :)
Then asked how best to either match batches, or as it ended up, individually test them for Vf, and the answer I got from Stephen/HLG was as long as they were within .1v then it's all good.
...I think you were involved in this discussion, lol! Take that beer helmet off, man. :D

If you're mixing different model boards then I think you'd want to test Vf at operating temps as Vf will be effected differently for different chips and different topologies (if individual fixture Vf is important to you).
Noooo, I'm not wiring different models together, in parallel or otherwise.
I don't think matchimg Vf is important when using a C.V. and wiring loads in parallel, except to maintain equal intensity across your fixture.
See above. But I came across something saying about what you're saying here - again, maybe it was you, haha. That the only difference *might be* that one board would be brighter/dimmer (not even perceivable?) but still, I'd rather match them as close as possible if that's the advice from HLG. But still, as long as the test I outlined that I did above were done correctly (and I'd appreciate feedback on that) then it really was pretty simple.
I recommend inline fuses before each parallel fixture to mitigate this possible damage. This isn't thermal run away though.
I asked about this a few pages back as well, I think you said the ones I'd linked were good to use. The "automotive" style ones, the only ones I know of. I kinda wish there were a more aesthetically pleasing inline fuse to use, but as long as the ones I posted will safely do the job I'd expect them to, then that's fine.

Thanks for chiming in bud. I appreciate the help.

EDIT: Duno how I missed this part of what you said "
...then I think you'd want to test Vf at operating temps as Vf will be effected differently for different chips and different topologies (if individual fixture Vf is important to you)...
 

ChiefRunningPhist

Well-Known Member
This was all from a couple pages back, when the concern was (re-)raised by someone, and this was pointed out:
"PARALLEL WIRING
Multiple Boards can be wired in parallel. Different batches of boards may have slightly different voltages. Use only boards from same batch for parallel connection. Do not use different spectrum boards in parallel as they may have slightly different voltages."

-HLG QB Guide.

I remarked how I can't believe I'd missed or forgot seeing that on the guide. And of course, once I heard it, I had to be "worried" and go with the advice :)
Then asked how best to either match batches, or as it ended up, individually test them for Vf, and the answer I got from Stephen/HLG was as long as they were within .1v then it's all good.
...I think you were involved in this discussion, lol! Take that beer helmet off, man. :D


Noooo, I'm not wiring different models together, in parallel or otherwise.

See above. But I came across something saying about what you're saying here - again, maybe it was you, haha. That the only difference *might be* that one board would be brighter/dimmer (not even perceivable?) but still, I'd rather match them as close as possible if that's the advice from HLG. But still, as long as the test I outlined that I did above were done correctly (and I'd appreciate feedback on that) then it really was pretty simple.

I asked about this a few pages back as well, I think you said the ones I'd linked were good to use. The "automotive" style ones, the only ones I know of. I kinda wish there were a more aesthetically pleasing inline fuse to use, but as long as the ones I posted will safely do the job I'd expect them to, then that's fine.

Thanks for chiming in bud. I appreciate the help.

EDIT: Duno how I missed this part of what you said "
Haha ya my memory is fading! Sorry man, if they are all the same model then no need to include operating temps. :)

They also have, idk what they're called, barrel inline fuses? But it's just a more streamlined look, I think I used to use for my car stereo amp, but they should have smaller sizes and stuff for our applications. Ill try to find a link. Sorry if I get confused sometimes I blend people together an such lol
 

Airwalker16

Well-Known Member
They definitely will fit, physically, based on the dimensions provided.
They're 3.75"x2.77". Sink is 5" diameter circle. The holes will be directly under the center of the board.
I kinda thought, one of the driving factors behind making these 96s' was drop in cob replacement.
You could spend a bit of time splaying your pins like I did for my 4 square ones to up the wattage capabilities.
 

2com

Well-Known Member
Haha ya my memory is fading! Sorry man, if they are all the same model then no need to include operating temps. :)

They also have, idk what they're called, barrel inline fuses? But it's just a more streamlined look, I think I used to use for my car stereo amp, but they should have smaller sizes and stuff for our applications. Ill try to find a link. Sorry if I get confused sometimes I blend people together an such lol
No sweat at all, just joking. I don't expect you to remember allll the goings on of the forum interactions. Good pointing out the operating temps anyways, I'd thought of it but I didn't actually know the "why" of why I thought of it. So thanks.
AGC fuses with a holder would look WAY more athstetic.

Something like this.

Thanks dude, this is something like what I was hoping for. They're much more streamlined and less clunky looking.
Thanks for the link.
 

Jqwerty1

Well-Known Member
My veggie grows with the 4x v2 132s 185w in 2x4. is going amazing. I now get a harvest of something everyday. Im trying few methods at the same time. One im using my tested cannabis grow style to create some "mothers" with heavy topping/harvesting coco mix, grow bags light watering, no runoff. The other im trying growing just many individual plants and using a flooding style. The pots are 2g and i have like 14 of them. I have cilantro, dill, basil, i harvest one everyday and use them for cooking, im hoping for them to get super bushy and im going to put it outside in the spring. The other set up uses 3inch pots, then i put the pots in trays and switch them out every 12 hours. This gives me 75 pots in a 2x4, then i have 25 each in 3 trays. I simply flood the trays every 4-5 days. Im getting crazy harvests from this after only a few weeks and with the HLG lights, amazing quality and the temperatures are low enough in the tent and my house to keep everything in a veg state, possibly year round. This one gives me a salad harvest everyday. In the photos you can see
purple bok choy
Purple broccoli
Purple brussel sprouts
Purple kale
Purple basil
Purple spinach
Rainbow swiss chard
Red mustard
Purple cayenne pepper
My best producer is bokchoy by far. Kale is probably second. Dill, cilantro, thai basil also great. The only problem ive seen so far is my peppers developing flowers, which I didn't expect this early at 16hrs light. Need to research that.
I really doubt this would be possible with any other lights. 1$per watt, high efficiency, good spectrum, great coverage. I've tried this with high end HID and still had too much IR,UV and heat. The plants were almost baked and went to flower instantly, with blurples shit efficiency and low quality.
 

Attachments

fragileassassin

Well-Known Member
AGC fuses with a holder would look WAY more athstetic.

Something like this.

  • 304pcs Samsung 561C bin Diodes and Epistar 660nm leds
  • Dimensions 6.85" x 11.26
  • Max current per board 1200mA
  • 1.0/W conductivity, 1OZ,50-137W, 102-114V, 0.48-1.2A
Thats the shit im using.
These should work right?
Untitled.png
 
Last edited:

Airwalker16

Well-Known Member
Top