Cutter Electronics: Complete DIY COB kits

Redoctober

Well-Known Member
Looks like the 93cri 3000K has a good amount of blue already and indeed does hit that 660. I'd definitely use it if you do decide this route. But Growmau5 will most likely tell you as he tells everyone. He keeps coming back to 3500k CXB3590's.

In my opinion, I'd probably use 3000K 93CRI mixed with 3500k more so than 4000. RED is what you want most of and those 93 CRIs with 3500's spectrum will give ya pleeeeeenty of Blue.
Yeah funnily enough, the 90cri 3000k has more blue AND more red than the 80cri (or at least more red in the 630-700 range). I was looking up CRI and I can't find reference illuminant spectra for anything but daylight.

"For CCTs less than 5000 K, the reference illuminants used in the CRI calculation procedure are the SPDs of blackbody radiators"
....Which is not terribly helpful.

{a small aside about CRI...}
What this means to those not familiar with CRI (someone asked about it a few pages back) is that for a given color temperature say 3000k in this example, all light sources i.e. light bulbs, LED etc, are given a value or Color Render Index for how closely they reproduce the spectral signature of a perfect blackbody radiator (object that perfectly radiates energy but absorbs none) at that particular temperature. So if just for the sake of argument, a star was burning at 3000 degrees kelvin, the LED with the CRI of 80 would be pretty similar to the light signature coming from that star, but the LED with the CRI of 90 would reproduce it even more accurately. And anything above 95 is so accurate, you'd probably need photo meters to tell the difference.

{/small aside about CRI over}

So yes I agree, I think the 3500K mixed with the 90cri 3000K would also be an intriguing experiment. I really wish to god they could have just thrown in the 3500K spectrum on that chart to make it easy.

My rationale in going with the extra 4000K blue, is this:
In the very very first grow I ever did a few years back, I ran a chemdawg clone. The only light I had at the time was a 400 watt Sun System III metal halide 4200K given to me by a friend. I didn't really know what I was doing but went straight to aero and shockingly produced a really fantastic plant. I broke all the rules of drying, I cut all the branches off immediately, I hung them in a closet where I was vegging another plant with a blue T5 exposed to light the entire time, and I probably trimmed and jarred before adequate drying had taken place.....yet....to this very day....I have never produced buds with the same smell as in that first run. When you opened the jar, it was like smelling the strongest Pine Sol ever, made from pine trees that grew on Krypton; it really was THAT strong. Stings the nostrils (like Sex Panther cologne). I have reproduced nearly every variable of that first grow with the exception of the light, in an attempt to achieve that smell but to no avail. I now realize that immediately after that grow, I switched to an HPS (cause that's what you're supposed to flower with obviously), then I went through an experimental early generation LED phase, then a T5 phase, then back to an HPS phase, now in a COB phase!

I dug that old Sun System III MH out of storage and am going to run it again, this time flanked by 5 of the 3500K COBs from the mau5-V kit. Then adjacent, I will run this new 12 COB fixture....all the same original Chemdawg strain (which is now admittedly a few years older) and see what happens. My theory, which happens to fit the facts, (as Spock would say) is that it was the MH that was responsible for that amazing result that first time, and is the reason I wanted to include the 4000K COBs.

Maybe I'll go bonkers and do 2x 12-COB fixtures one half 3500K half 3000K and the other half 4000K half 3000K.......SCIENCE!!!!!!!
 
Last edited:

Airwalker16

Well-Known Member
Yeah funnily enough, the 90cri 3000k has more blue AND more red than the 80cri (or at least more red in the 630-700 range). I was looking up CRI and I can't find reference illuminant spectra for anything but daylight.

"For CCTs less than 5000 K, the reference illuminants used in the CRI calculation procedure are the SPDs of blackbody radiators"
....Which is not terribly helpful.

{a small aside about CRI...}
What this means to those not familiar with CRI (someone asked about it a few pages back) is that for a given color temperature say 3000k in this example, all light sources i.e. light bulbs, LED etc, are given a value or Color Render Index for how closely they reproduce the spectral signature of a perfect blackbody radiator (object that perfectly radiates energy but absorbs none) at that particular temperature. So if just for the sake of argument, a star was burning at 3000 degrees kelvin, the LED with the CRI of 80 would be pretty similar to the light signature coming from that star, but the LED with the CRI of 90 would reproduce it even more accurately. And anything above 95 is so accurate, you'd probably need photo meters to tell the difference.

{/small aside about CRI over}

So yes I agree, I think the 3500K mixed with the 90cri 3000K would also be an intriguing experiment. I really wish to god they could have just thrown in the 3500K spectrum on that chart to make it easy.

My rationale in going with the extra 4000K blue, is this:
In the very very first grow I ever did a few years back, I ran a chemdawg clone. The only light I had at the time was a 400 watt Sun System III metal halide 4200K given to me by a friend. I didn't really know what I was doing but went straight to aero and shockingly produced a really fantastic plant. I broke all the rules of drying, I cut all the branches off immediately, I hung them in a closet where I was vegging another plant with a blue T5 exposed to light the entire time, and I probably trimmed and jarred before adequate drying had taken place.....yet....to this very day....I have never produced buds with the same smell as in that first run. When you opened the jar, it was like smelling the strongest Pine Sol ever, made from pine trees that grew on Krypton; it really was THAT strong. Stings the nostrils (like Sex Panther cologne). I have reproduced nearly every variable of that first grow with the exception of the light, in an attempt to achieve that smell but to no avail. I now realize that immediately after that grow, I switched to an HPS (cause that's what you're supposed to flower with obviously), then I went through an experimental early generation LED phase, then a T5 phase, then back to an HPS phase, now in a COB phase!

I dug that old Sun System III MH out of storage and am going to run it again, this time flanked by 5 of the 3500K COBs from the mau5-V kit. Then adjacent, I will run this new 12 COB fixture....all the same original Chemdawg strain (which is now admittedly a few years older) and see what happens. My theory, which happens to fit the facts, (as Spock would say) is that it was the MH that was responsible for that amazing result that first time, and is the reason I wanted to include the 4000K COBs.

Maybe I'll go bonkers and do 2x 12-COB fixtures one half 3500K half 3000K and the other half 4000K half 3000K.......SCIENCE!!!!!!!
Same chem Dawg as in you've kept the cut all these years?
 

welight

Well-Known Member
Yeah funnily enough, the 90cri 3000k has more blue AND more red than the 80cri (or at least more red in the 630-700 range). I was looking up CRI and I can't find reference illuminant spectra for anything but daylight.

"For CCTs less than 5000 K, the reference illuminants used in the CRI calculation procedure are the SPDs of blackbody radiators"
....Which is not terribly helpful.

{a small aside about CRI...}
What this means to those not familiar with CRI (someone asked about it a few pages back) is that for a given color temperature say 3000k in this example, all light sources i.e. light bulbs, LED etc, are given a value or Color Render Index for how closely they reproduce the spectral signature of a perfect blackbody radiator (object that perfectly radiates energy but absorbs none) at that particular temperature. So if just for the sake of argument, a star was burning at 3000 degrees kelvin, the LED with the CRI of 80 would be pretty similar to the light signature coming from that star, but the LED with the CRI of 90 would reproduce it even more accurately. And anything above 95 is so accurate, you'd probably need photo meters to tell the difference.

{/small aside about CRI over}

So yes I agree, I think the 3500K mixed with the 90cri 3000K would also be an intriguing experiment. I really wish to god they could have just thrown in the 3500K spectrum on that chart to make it easy.

My rationale in going with the extra 4000K blue, is this:
In the very very first grow I ever did a few years back, I ran a chemdawg clone. The only light I had at the time was a 400 watt Sun System III metal halide 4200K given to me by a friend. I didn't really know what I was doing but went straight to aero and shockingly produced a really fantastic plant. I broke all the rules of drying, I cut all the branches off immediately, I hung them in a closet where I was vegging another plant with a blue T5 exposed to light the entire time, and I probably trimmed and jarred before adequate drying had taken place.....yet....to this very day....I have never produced buds with the same smell as in that first run. When you opened the jar, it was like smelling the strongest Pine Sol ever, made from pine trees that grew on Krypton; it really was THAT strong. Stings the nostrils (like Sex Panther cologne). I have reproduced nearly every variable of that first grow with the exception of the light, in an attempt to achieve that smell but to no avail. I now realize that immediately after that grow, I switched to an HPS (cause that's what you're supposed to flower with obviously), then I went through an experimental early generation LED phase, then a T5 phase, then back to an HPS phase, now in a COB phase!

I dug that old Sun System III MH out of storage and am going to run it again, this time flanked by 5 of the 3500K COBs from the mau5-V kit. Then adjacent, I will run this new 12 COB fixture....all the same original Chemdawg strain (which is now admittedly a few years older) and see what happens. My theory, which happens to fit the facts, (as Spock would say) is that it was the MH that was responsible for that amazing result that first time, and is the reason I wanted to include the 4000K COBs.

Maybe I'll go bonkers and do 2x 12-COB fixtures one half 3500K half 3000K and the other half 4000K half 3000K.......SCIENCE!!!!!!!
Hi Red
Do you see any value in a 90 CRI 3500K chip in what your trying to achieve? a 3500 Base with more 660
Cheers
Mark
 

JorgeGonzales

Well-Known Member
Yeah funnily enough, the 90cri 3000k has more blue AND more red than the 80cri (or at least more red in the 630-700 range).
That's because it's a 3500K 90CRI chip. I think it's a mistake in the data sheet, or at least somebody should put it under a spectrometer before claiming Cree is the only LED manufacturer who can't hit a 3000K target.
 

Redoctober

Well-Known Member
Same chem Dawg as in you've kept the cut all these years?
Yep, same one. Senescence may have diminished its vigor a bit, but I'm still hunting for that holy grail aroma I got the first time round! If I had to guess, I'd say its between 3-4 years old at this point. I'm not exactly sure how long the guy who gave it to me had it before I got it. Could add a few years on as well
 

Redoctober

Well-Known Member
Hi Red
Do you see any value in a 90 CRI 3500K chip in what your trying to achieve? a 3500 Base with more 660
Cheers
Mark
Wow exactly, a 3500 base with more 660nm, perfect! I would definitely be very interested, is that available?
Do spectral graphs of the 80 CRI and 90 CRI 3500K exist?
They're not on Cree's datasheet
Makes it tougher to evaluate without the spectra, like flying blind-ish (we do have a ballpark estimate of what it might look like)

Thanks Mark

That's because it's a 3500K 90CRI chip. I think it's a mistake in the data sheet, or at least somebody should put it under a spectrometer before claiming Cree is the only LED manufacturer who can't hit a 3000K target.
Is that true? How certain are you? Maybe Mark can chime in.
I'm now super confused.
Would Cree really make an error that egregious?
I suppose I can email Cree and ask
What you're saying is that the spectrum in this graph labeled 3000K 90CRI is really a 3500K 90 CRI?
and the 3000K 90CRI doesn't exist?
What did you mean they can't hit a 3000K target?
 

Attachments

Last edited:

AtterStiga

Well-Known Member
Ok, so I realized sometime during the night that I've made a little mistake. After scanning the Cutter shopping page, I was going to be a smart arse and buy the parts instead of the kits to save myself some money. And I did, several hundred dollars on a k + shipping. At least that is my story and I'm sticking with it. But I ended up with the 100w heatsinks instead of the new 50w. Which seems like a good thing for me, more of a buffer and maybe some day I'll run them harder and so on. Except it says that they use 4M screws instead of 3.5M. Sometimes half a mm can mess things up. Thing are shipped and I started a thread for my next project.https://www.rollitup.org/t/project-new-cob-veg-light-planning-thread.921656/
Writing walls of text sure pass time lol.
 
Last edited:

JorgeGonzales

Well-Known Member
Is that true? How certain are you? Maybe Mark can chime in.
I'm now super confused.
Would Cree really make an error that egregious?
I suppose I can email Cree and ask
What you're saying is that the spectrum in this graph labeled 3000K 90CRI is really a 3500K 90 CRI?
and the 3000K 90CRI doesn't exist?
What did you mean they can't hit a 3000K target?
Data sheet errors happen all the time, the only thing I am sure of is that the graph in the CXB3590 data sheet is 3500K color temp. Maybe that's how they label their 3000K, or maybe it's just a typo. I'm guessing typo, but it's only a guess.
 

Redoctober

Well-Known Member
Data sheet errors happen all the time, the only thing I am sure of is that the graph in the CXB3590 data sheet is 3500K color temp. Maybe that's how they label their 3000K, or maybe it's just a typo. I'm guessing typo, but it's only a guess.
Wow! That actually changes a lot if that's the case. I didn't order anything yet thankfully.

So here are two spectral graphs, the first is the one I posted from Cree with the various spectra of CXB3590's overlayed with PAR numbers, the second is an independent graph from another thread courtesy of @Malocan in {this post}. It appears that the independent graph of the 3500K seems to match the spectrum of the one on the Cree graph labeled "3000K 80CRI." So it does seem to confirm what you're saying @JorgeGonzales that Cree mislabled the 3000K 80CRI.
I guess we are to believe that the 3000K 90CRI is correct?
Both the supposed "3000K 80CRI" and the 3500K both have the same red peak at 600nm and the same blue peak at 455nm, coincidence?
However the 3000K 90CRI has its red peak at 630nm, and a 20% more intense blue peak at 455nm.
So we are assuming that the 3000K 90CRI is accurate, and the 80CRI is in fact a graph of the 3500K spectra?
What do you guys think?cxb3590 PAR spectra.jpg cxb3590-3500K spectrum.png
 

Qjay420

Well-Known Member
Hello :) I'm growmau5. I contacted Mark over at www.cutter.com.au a few weeks ago to ask if he could package together some high quality components and offer us a 1-click solution to ordering the parts necessary to assembly the single most popular configuration for most growers here: Four of Cree's flagship 3590s @ 1400ma.

As a paid supporter and sponsor of RIU, Mark was happy to help us out. He mentioned that he had been thinking about doing this for some time and that he would fast track the only missing component from his site: the Heatsinks.

A lot of us have been using extruded bar heatsinks with great success for sometime now. And I am sure many of you will continue to enjoy the process of measuring, marking, drilling and tapping these type of heatsinks for your custom COB setup.

View attachment 3630067
In my estimation, the heatsink machining and preparation process is the most intimidating part of a DIY COB build for most beginners, accounting for 60-70% of the time it takes to complete a build. To streamline and make this process easier for all of us, the guys at Cutter really went out above and beyond to source a Passive Anodized Pin heatsink for us :with a thermal resistance of 1.07 C/w. The increased heat emissivity of this design allows for smaller, lighter, more efficient thermal management at a comparable price.

OH, and did I mention they come pre-drilled and tapped for popular COB models including the Cree CXB3590!

If you are interested in learning more about this kit, follow this link directly to the kit page: http://www.cutter.com.au/proddetail.php?prod=cut2900

Here is what is included:
4 pcs Cree CXB3590, top bins only
4 pcs Ideal 50-2303CR CXB Holder
4 pcs Ideal 50-2300AN Adapter for the Ledil ANGELINA
4 pcs Ledil Angelina 90 Degree reflector
4 pcs Graphite Thermal pad(TIM) 1 side adhesive for simple assembly on the COB

1 pc LGSU-1400D, Dimmable driver allows the COBS to be run at just under 50 watts, the sweet spot of wall plate efficiency

1 pc 0-10V dimmer compatiable with the LGSU Driver

4 Passive Pin fin Coolers suitable for the 50 watt drive( based on eff% calculations), predrilled and tapped for the Ideal Holder and adapter( you will only need a screw driver not a drill press)

16 pcs M3 6mm screws for the Holders

16 pcs M3.5 12mm Screws for mounting to side rails



The kit is called the MAU5 : IV so that it is easy to find & so that newer hobbiests know they are ordering the same components I showcase on my YouTube Channel. I will also be offering installation support and instructional videos on wiring, assembly, & configurations.

Additionally, these heatsinks and Cutter Kits will be the foundation of a new series of videos that I am producing called: the
Community OpenSource DIY LED grow light

the goals of the series are :
-to cooperatively design and build a completely modular, adjustable, infinitely configurable system with standard components .
-affordability, global accessibility, high performance
-applicable to anyone from the closet grower to the warehouse grower.
-ease of assembly and use.
To complete the project I will be creating all of my files in a collaborative affordable software called: Autodesk Fusion 360 (free 30 day trial, $40/mo license). Publishing will go directly to my profile over at GRAB CAD, a free website were anyone can download files and send them to a local shop for fabrication.

I will also be assisting our community by reaching out to fabricators that will cut out parts and ship anywhere in the world. Places like Big Blue Saw can take one of these files , produce the part and ship to your doorstep for less than you might think. (more on that later).

In the meantime, I will be designing and fabricating prototypes that will hold the heatsink, allow for modularity, and allow flexibility for mounting other components like reds and UVs.

One note to USA buyers, the bold typeface prices on Cutter.com are in Australian dollars, US dollar conversions are listed below that.
The RIU discount code: CUTANDROLL will work for this kit!!!
Pre-orders are going on now (mid-march) for April 2016 availability.

Cheers
View attachment 3630098
View attachment 3630099
20160922_180408.jpg Thank you GrowMau ,You inspired me to build a canopy 10. all CXB 3590. 3500K. With Two HLG-240 1400MA. 20160922_180327.jpg20160922_180408.jpg
 

Attachments

Qjay420

Well-Known Member
Oh and to be honest the cutter pin fin heatsinks are junk, All my side mount holes stripped out on me. So I drilled them out and used rivets instead. I grabbed ten more of the 140mm from PLC for my next light.
 

welight

Well-Known Member
Oh and to be honest the cutter pin fin heatsinks are junk, All my side mount holes stripped out on me. So I drilled them out and used rivets instead. I grabbed ten more of the 140mm from PLC for my next light.
dude we have sold over 3000 of these here, Junk is a little harsh, what size screws were you using?
Cheers
Mark
 

welight

Well-Known Member
Wow exactly, a 3500 base with more 660nm, perfect! I would definitely be very interested, is that available?
Do spectral graphs of the 80 CRI and 90 CRI 3500K exist?
They're not on Cree's datasheet
Makes it tougher to evaluate without the spectra, like flying blind-ish (we do have a ballpark estimate of what it might look like)

Thanks Mark



Is that true? How certain are you? Maybe Mark can chime in.
I'm now super confused.
Would Cree really make an error that egregious?
I suppose I can email Cree and ask
What you're saying is that the spectrum in this graph labeled 3000K 90CRI is really a 3500K 90 CRI?
and the 3000K 90CRI doesn't exist?
What did you mean they can't hit a 3000K target?
I will get some 3500k 90 cri in stock and will do a spectral analysis on 3500k 80+90 CRI and then do the same on 3000k 80 + 90 CRI be an interesting check
Cheers
Mark
 

Redoctober

Well-Known Member
No, the 3000K 90CRI chart is objectively 3500K, the 3500K 80CRI is 3500K, and accurate, as measured by spectrometer.
God Damnit Cree!!!! Get you're shit together...you're supposed to be the best!

Okay how about this graph...even though its a spectral graph of the Citizen CLU048 COBs spectra (all 90CRI), could it apply to the Cree CXB COBs as well? Or is that wrong to assume that?
I assume that a 90CRI 3500K chip is a 90CRI 3500K chip regardless of who makes it since color temperature spectrum is objective not subjective?
Very interesting to note that all the 90CRI COBs (besides the 5700K) have their blue peak at pretty much the same 450nm spot, just different intensities, and the red peak is mainly 630nm with a smeared out green peak for the 4000K.
This is what the Cree chart SHOULD look like


citizenCLU-90CRIgraph.jpg
 
Last edited:

Redoctober

Well-Known Member
I will get some 3500k 90 cri in stock and will do a spectral analysis on 3500k 80+90 CRI and then do the same on 3000k 80 + 90 CRI be an interesting check
Cheers
Mark
Mark you are the best!!! I wish I could buy you a beer, or fax you one!

I'm sorry if I'm nitpicking, but when making a potentially substantial investment, I like to make sure I have my information straight so as to make a theoretically informed decision.

If it helps to know, I will be ordering a ton of COBs from you as soon as I figure this whole spectrum thing out!
 

Qjay420

Well-Known Member
I say we make the spectrum one chip one color one Cobb at a time. I have 22 Cobbs CXB3590 and almost 100 far red 3watt Cree chips that will cut up any spec chart you can throw at me. I call it once apon a Cobb my friends.
 

Qjay420

Well-Known Member
dude we have sold over 3000 of these here, Junk is a little harsh, what size screws were you using?
Cheers
Mark
actually not harsh at all no disrespect! I use the screws that are speced out. if you built a light with the cutter heat sinks than you absolutely know that the side screws really don't work out. I really didn't want to bore out a threaded two holes per side of each heatsink then go back and buy rivets to do the job tight and right but I did because otherwise they would be flopping around. the heatsinks itself isn't bad. they are the same screw you use in the face of the heat sinks but longer.
 
Top