Can't delete, just please ignore post. Thanks.

Hey guys,

I'm new to growing, not electrical engineering, and had some questions about building my own LED strands.

I have a couple of hot air rework stations, and I figured it would be cheaper for me to make my LED lights than buy them, but I have to ask where are you guys buying your PCBs from?

Also, does anyone have a compiled list of LED Diodes, their specs, spectrums they hit, costs, etc.

The more I read about these LED strips compared to my cheap blurples, the more I'm starting to wonder if the LEDs/COBs are considered so much better simply because of efficiency, but it looks like everyone eventually supplements them with reds/UV/IR/etc that my blurps already have. So why not build high efficient LED blurps from the start? Every 5th led toss in an red and every 12th an IR, maybe a blue or two, not sure on the frequency yet, need to read some studies on cannabis photosynthesis spectrum requirements.

Anyways.. any help would be appreciated in finding info on the best LED Diodes to use as of today, NOT LED strips, individual LED Diodes you solder to a strip, and where to find prefabbed PCB boards of various dimensions in order to build them. I'd appreciate it.

Thanks,

-sunsp0ts

Mostly because of the time and effort involved in working with individual diodes.

People are getting 1.5-2 GPW using plain jane white, so it's an enthusiast thing to add UV and IR. Slightly more trichomes and yield is worth taking a look at for those who have an interest.

The recent trend using XP-E photo red might have something to do with cost coming down, but mostly having companies that are making pre-built products available. XP-E2 are more efficient than XP-E so that's what I would be looking at. The 630nm diodes don't have the same efficiency as 660 glancing at the datasheet, but may still be a better deal than getting it from phosphor. Would need to analyze the performance curve and see how efficient you can get them while keeping cost at an acceptable level.

A Bridgelux 1' strip has ~56 diodes on it. I can solder really well really fucking fast. I can't out do a 56 diode strip for $3 and I damn sure can't solder 56 diodes for cheaper than what it'd cost me in bench labor.

For a fun project, more power to you. Financially economical, not so much.

So many pcb mfrs, design a board and start farming out bids to OEM houses.

Custom pcbs at www.PCBway.com

But you want to build your own custom fixture, but using pre made boards.
Strips ar your best bet.

CannaBruh said:

A Bridgelux 1' strip has ~56 diodes on it. I can solder really well really fucking fast. I can't out do a 56 diode strip for $3 and I damn sure can't solder 56 diodes for cheaper than what it'd cost me in bench labor.

For a fun project, more power to you. Financially economical, not so much.

So many pcb mfrs, design a board and start farming out bids to OEM houses.

Click to expand...

I'm looking at samsung f3 series strips as a basis for comparison, they run about $13 a piece, I haven't even heard of a 1' strip with 56 diodes on it, seems like overkill. I could solder an f3 equivalent in about 5-10 mins with a hot air rework station including prep time. NOT a soldering iron. It's closer to a rework oven except you direct the heat and it takes about 2-3 seconds per SMC. Plus the benefit of customized panels.

Airwalker: I'm talking about PCBs prefabbed just like these samsung strip pcbs are, minus the chips, should only run 2-3 dollars a piece tops. Then you just add your own LEDs. No one has designed something comparable yet or found ones you can just order straight away? Wasn't looking to design my own, not like these are complicated, but the spacing has to be right, could just copy a samsung I suppose. Don't own one yet, however.

sunsp0ts said:

I'm looking at samsung f3 series strips as a basis for comparison, they run about $13 a piece, I haven't even heard of a 1' strip with 56 diodes on it, seems like overkill. I could solder an f3 equivalent in about 5-10 mins with a hot air rework station including prep time. NOT a soldering iron. It's closer to a rework oven except you direct the heat and it takes about 2-3 seconds per SMC. Plus the benefit of customized panels.

Airwalker: I'm talking about PCBs prefabbed just like these samsung strip pcbs are, minus the chips, should only run 2-3 dollars a piece tops. Then you just add your own LEDs. No one has designed something comparable yet or found ones you can just order straight away? Wasn't looking to design my own, not like these are complicated, but the spacing has to be right, could just copy a samsung I suppose. Don't own one yet, however.

Click to expand...

ooohhh, inch in width, 56 across. It sounded like there were 56 diodes in one inch of space. That's no big deal with a hot air flow device. Either an oven or wand. I've done this for TV's that have LED panels blow out, the painful part is taking the TV apart and the diffusers glued on, the soldering is the fastest part. If you took any longer than a few seconds you'd destroy the plastic.

It would take maybe 10 minutes to design one of these pcbs. It's two landings and some interconnects over and over.

If you're looking to design something superior or proprietary I commend your efforts.

Look into panelization.

56 diodes at nSeconds @ $x.xx labor rate + facility costs =/= $3/ea efficiencies... not going to happen without great scale.

For a fun project or something to do, wish ya the best.
" == inches
' == feet

CannaBruh said:

It would take maybe 10 minutes to design one of these pcbs. It's two landings and some interconnects over and over.

If you're looking to design something superior or proprietary I commend your efforts.

Look into panelization.

56 diodes at nSeconds @ $x.xx labor rate + facility costs =/= $3/ea efficiencies... not going to happen without great scale.

For a fun project or something to do, wish ya the best.
" == inches
' == feet

Click to expand...

Okay, but you have to make sure you have the right sized traces between LEDs, and you have to make sure the spacing is perfect for heat dissipation and for maximum luminosity. If that's a word.

56*3 seconds each is a little under 3 minutes. The longest part is making sure you have the negative in the right direction. Labor rate? I'm doing it myself for my own grow, not trying to start a business. I'm using my own equipment that I already have. The perks being customized panels. I don't understand why this is getting argued against.

Oh I think it was miscommunication on my part, I'm meaning that it would be cheaper for me to make these customized than anywhere else. And I would enjoy doing it. Like I said, I'm not trying to start a business. And NO ONE would want to fork out the money for a hot air rework station to save a few bucks a panel, even if I think it would be cheaper if they already had one, that's my opinion, plus they'd have to get proficient at it, etc.

So again, I wasn't trying to suggest you all are wasting money by buying these panels. I just assumed there were a few like minded people out there on this forum that had gone down this path as they already have the equipment/proficiency/inclination to do it.

As an FYI, I've attached a pic of my station in case any of you are curious as to what they look like. It's the one on top.

Just to add another note... I know CannaBruh thinks it would only take 10 minutes to design a panel PCB, and he very well may have the know-how to do it that fast, but I personally know it would take me longer, especially if you're not just copying a professionally designed one meant for those LEDs. And the only reason I bring it up again, is because if you simply just tossed in two landings, and guessed as to the gauge of traces between them, especially when you start running these panels in series, you COULD EASILY cause a fire if you don't use the right math. Just putting this here as a warning in case someone decides to just throw one together. Not to mention you'd lose out on lighting and cooling potential if they're too far or too close together.

Good luck with the build. Sorry for sounding negative. Your original post tho was asking for lists of diodes (little research) and offering that you can build for less cost (highly unlikely) unless you work for free?

Instead of hot air I think you'd be better off with a fluid dispenser solder paste and an oven, pick n place would be faster

pcb design assumes you know what you're doing.... if you don't then please spend a few dollars and buy something safely designed by someone who did and take advantages of the cooling and electrical and cost efficiencies.

CannaBruh said:

Good luck with the build. Sorry for sounding negative. Your original post tho was asking for lists of diodes (little research) and offering that you can build for less cost (highly unlikely) unless you work for free?

Instead of hot air I think you'd be better off with a fluid dispenser solder paste and an oven, pick n place would be faster

pcb design assumes you know what you're doing.... if you don't then please spend a few dollars and buy something safely designed by someone who did and take advantages of the cooling and electrical and cost efficiencies.

Click to expand...

Yes, I was asking for a list of diodes that work well for growing weed. Which is specifically what this thread is best for. Otherwise I have to dig through thousands of diodes and I don't have the equipment to test for spectrums/PAR. So yes it makes perfect sense to ask you guys for this info. Otherwise I guess I should just go through the panels you guys use, but I assumed since there are panel DIY posts you guys had a list somewhere, I don't see them. Except for the F3 Diodes. I was looking for a variety. Not just one type. For every spectrum.

I don't understand your point about not using hot air, but using solder paste? You use both. That's why it literally takes about 5 minutes total to do 60 LEDs, I do it all the time. Ovens are slower for a number of reasons.

I do know what I'm doing, that's why I pointed out a bunch of issues with just picking some landings like you suggested. Also, I'm not just going to be using one type of diode, so it's more complicated than you think.

Eh this is getting frustrating, I'm just going to delete the thread if I have the option. Thanks anyways guys.

One of my builds was using the LT-Q562A strps and they are rated from 450 to 900ma. They are adhered to a sheet of aluminum that is 0.080 inches. They have oscillating fans blowing over and under to keep cool and air movement to plants and everything is very cool to the touch running 600watts on the four by four foot with 28 LT-Q562A driven by HLG-600H-24B. Samsung has a cool calculator to help you out. Just google search samsung engine calculator.(of course its for using there products) I bought 200 strips to make 7 boards and I paid $6.52 a strip(40 diodes of LM301B) So each board costed me$182.56 for strips, can't see it being easier and or cheaper. Pretty much the same as QB but a much better spread of light and easier heat management. But that's the great thing about DIY- it suits me and my needs and makes me happy. Everyone is different.

sunsp0ts said:

Eh this is getting frustrating, I'm just going to delete the thread if I have the option. Thanks anyways guys.

Click to expand...

Naw, man, no need to bail on the conversation! Just let it roll, and you may get the feedback you're looking for. And if not, no harm done. You could also post over on the forum at ledgardener.com - some clever people are doing interesting things over there.

Relax and give people a chance! This thread is only a few hours old and most people haven't seen it yet, including some of the most knowledgeable. This is probably the best grow lighting forum on the net, no need to bail because you haven't been promptly handed a list of grow diodes and their properties.
Besides, this place needs an EE.

sunsp0ts said:

The more I read about these LED strips compared to my cheap blurples, the more I'm starting to wonder if the LEDs/COBs are considered so much better simply because of efficiency, but it looks like everyone eventually supplements them with reds/UV/IR/etc that my blurps already have. So why not build high efficient LED blurps from the start? Every 5th led toss in an red and every 12th an IR, maybe a blue or two, not sure on the frequency yet, need to read some studies on cannabis photosynthesis spectrum requirements.

Click to expand...

Yes, LED is so popular due to the high efficiency/efficacy. Also, the efficiency is often compounded by reducing or eliminating the need for air conditioning.
Some people supplement with reds and UV, but it's nowhere near everyone. White light works just fine, but many feel adding reds or UV are worth the cost/effort. I believe UV is best done with a fluorescent reptile light.

As for high efficiency blurple, I don't think such diodes exist? Several people and companies have added reds to boards and strips.

sunsp0ts said:

Anyways.. any help would be appreciated in finding info on the best LED Diodes to use as of today, NOT LED strips, individual LED Diodes you solder to a strip, and where to find prefabbed PCB boards of various dimensions in order to build them. I'd appreciate it.

Click to expand...

Look at LM561C and LM301B. I believe Nichia also has some very efficient diodes. I don't know about prefabbed boards, but several people have had custom boards made in china. Look for the threads on custom boards or maybe start a thread asking about custom board makers. The knowledge is here.