noob about to make a final decision. Please help.

[Chuff420]

@Chuff420

If you're cooling passively, bottom choice is the way to go. If you take actively cooled route, with fan, then either or will more than likely work for you. Lots of factors that could sway the final decision though, so second one is most likely suitable, although the first gives you a possibly better spread potential depending on cob output; gotta make sure middle ground is covered before advancing outwards.

Alas. The root cause of my insomnia since I started down this rabbit hole.

 

[AquariusPanta]

@Chuff420

You just gotta make a decision at some point and go with it and then your knowledge over this matter will take root and before you know it, you'll be sharing a thread on your DIY accomplishments!

 

[Rahz]

Alas. The root cause of my insomnia since I started down this rabbit hole.

http://www.ullrich-aluminium.co.nz/extrusions/heatsinks - if you don't mind drilling and tapping, ordering from here might be the best option for shipping costs to AU. They have several interesting profiles depending on which cob set you decide on.

As far as spacing, figure out your cob grid and then work the heat sinks in, 2-3 inches on the end from the center point of the cob.

 

[Chuff420]

I've heard that Jerry at Kingbrite is the man to go to for cheap CREE COBs from China.

Is there an equivalent for cheap Mean Well drivers from China?

Thanks

 

[Airwalker16]

I've heard that Jerry at Kingbrite is the man to go to for cheap CREE COBs from China.

Is there an equivalent for cheap Mean Well drivers from China?

Thanks

He's got those too

 

[Chuff420]

He's got those too

Thanks.

Email sent to Jerry and reply received.

He said he only sells drivers to people who purchase COBS from him, but his minimum quantity for sale is 5 COBS. So I replied asking if he was prepared to sell less COBS on the proviso that I also purchase the matching driver. Hopefully he says "Yes", and the price is good.

After all the considering I have done, I have decided to eliminate CXB2530s simply due to the number of COBS and the fart-assing around needed to wire them up and cool them all.

I've also concluded that running any combination using passive heat sinks, which was my preferred option, was just too expensive. If I could get HeatsinkUSA prices in Australia, then passive would definitely be the go. Instead, CPU coolers will be used.

So these are the combinations I am currently considering for my area of 0.43m² or 4.6285 Feet².



Price & availability of drivers will probably be the determining factor.

On the subject of CPU coolers, I've read that the Arctic 64 Plus & Arctic 11 are popular for the CXB3590 & CXB3070 respectively. Will these CPU coolers handle the higher Heat Watts?

Thanks

 

[Chuff420]

Generally speaking, would a CXB3590 stuck to a good CPU cooler, such as the Arctic 64 Plus, run cooler than being stuck to a suitably sized passive heatsink, and to what degree would the performance of the COB differ as a result?

Thanks.

 

[Chuff420]

Just got prices from Jimmy and I've narrowed it down to 2 choices, but I am very much leaning towards the 1.75A option.



The price difference between the 1.4A & 1.75A drivers is only AU$32 and my PPFD will jump from [email protected] to [email protected]?

Plan is to buy a Model B Driver and have a remote dimmer set up. Everything I've read has led me to believe that the CXB3590 3500K CD BIN works perfectly well for vegging, so I will use the same box for both and adjust the light intensity accordingly.

That should work, yes?

Any reason I should consider the 1.4A over the 1.75A, or can we call it, 4 @ 1.75A FTW?

Thanks everyone.

 

[Danielson999]

I would go with the 1.4ma option. It produces considerably less heat and is already overkill for PPFD. People are doing amazing grows with 800PPFD so I'm sure you'll be just fine with 1155!

 

[Chuff420]

I would go with the 1.4ma option. It produces considerably less heat and is already overkill for PPFD. People are doing amazing grows with 800PPFD so I'm sure you'll be just fine with 1155!

That exact argument has been going through my head for a few days now, but I keep thinking with a 7 foot high cabinet, and not growing a SCROG, maybe it could get better penetration at 1378 PPFD, or is my thinking wrong and 1378 PPFD is just overkill?

Apologies for all the questions. When I started this thread 10 days ago, I had absolutely no understanding of any of this and had a completely different set-up in mind.

I've learnt so much since then, and still have so much more to learn.

Thanks again everyone.

 

[Danielson999]

Right, makes sense if you're just letting the plant grow naturally. I didn't take that into consideration.

 

[Rahz]

I'm not sure I've seen any posters running cob grows above 1200 PPFD come to think of it. I suspect 1378 could produce a slightly higher yield, which could objectively be a good deal but I've also read it's more challenging to grow under near 1500, FWIW.

 

[voon]

I need some additional information in order to continue my decision process.

My original design was to limit the flowering chamber to a height of 4 feet. After having a few people mention that 4 feet was quite restrictive, I changed the design of the cabinet so I would end up with 5 feet of height, and it is this 5 feet that the discussion has been based on since.

If I was to utilise the entire cabinet to flowering I would end up with a base dimension measuring 860 * 500 mm (33.8 x 19.6 inches) given me an area of 0.43m² (4.62852 Feet²) and the height of the chamber would increase from 1500mm (4.92 feet) to 2200mm (7.22 feet).

Given the new height limit, does this open up the possibility of better utilising the space I now have?

This is my latest spreadsheet. I've added 2 lines in the CXB2530 grouping labelled 0.7T2 & 0.7 witch mathematically calculates the 0.7 values & 0.7T2 values based on the average difference between the T2 & U2 Bins for the CXB2530. My mathematical calculations assume a linear relationship, which obviously isn't the case in the real world, but it gives you the ability to quickly and easily compare figures at a glance.

View attachment 3619408

Also, in opening up the cabinet to purely flowering, I have made some savings as I no longer need to build an internal cabinet to segregate the chambers as originally planned, so there's a bit more cash available for lighting.

I'd also like to know what the ramifications would be if I were to bump up the amperage of the 3 x 3590 to 2.1A in this area. Would the higher cabinet allow for it, and would it be beneficial or detrimental?

Also, my initial plan was to go SCROG due to my 5 foot height limit. Having almost 8 feet of height now available, would I have enough space to eliminate SCROG and do a normal grow, or would I be better of sticking with SCROG?

Thanks.

hello .. allow me to respond to your post .. but the last part of Table W / ft2 is wrong .. your values gives W / the entire area of 4.63 square feet rather than one square feet .. In addition, it is important for you to take into account the PARW / ft2 want If you or PARW / cover area ft2 (4.63 square feet)
a simple calculation:
3x CXB3590 @ 0,7A total wattage watt 68.71 / 4.63 = 14.84 (9,56PARW / ft2) total
10x CXB2530 @ 0.3A 101.88 / 4.63 = 22 (11.77 PARW / ft2)

 

[Chuff420]

hello .. allow me to respond to your post .. but the last part of Table W / ft2 is wrong .. your values gives W / the entire area of 4.63 square feet rather than one square feet .. In addition, it is important for you to take into account the PARW / ft2 want If you or PARW / cover area ft2 (4.63 square feet)
a simple calculation:
3x CXB3590 @ 0,7A total wattage watt 68.71 / 4.63 = 14.84 (9,56PARW / ft2) total
10x CXB2530 @ 0.3A 101.88 / 4.63 = 22 (11.77 PARW / ft2)

Fixed. Thanks.

At some point during this thread is was pointed out that the the W/ft² standard was more for the HID/Sodium type lighting and for COBs we were more interested in the PPFD value, so I lost focus on W/ft².

I'm not sure I've seen any posters running cob grows above 1200 PPFD come to think of it. I suspect 1378 could produce a slightly higher yield, which could objectively be a good deal but I've also read it's more challenging to grow under near 1500, FWIW.

That now makes two posts that have commented on my solution.

If there is a flaw in my thinking I think I see it. I've been using 1200 - 1500 PPFD as my minimum/maximum target. Due to my limited floor space of 860mm x 500mm (2.8' x 1.6') combined with the fact that I now have 7 foot of height, meaning I no longer needed to go SCROG, I was aiming for the higher PPFD values and driving the COBS harder and further away as opposed to softer and closer.

Now I don't think there's anything wrong with my logic, but I think my minimum/maximum target should probably be revised to 800 - 1200 PPFD, and if that's the case then running them at 1.75A is overkill.

If all that sounds about right, then we have a final decision.

4 x [email protected]

 

[Chuff420]

Driving the CXB3590s softer @ 1.4A means less heat to dissipate which has now opened up the possibility of eliminating the CPU fans and going for passive heatsinks, my preferred choice, but I need a few more pieces of the puzzle.

I have 860mm x 500mm of floor space to work with and I'd like 2 rows of passive heatsinks with 2 x CXB3590 COBS per sink.

At 1.4A each COB is generating 21.33 Heat Watts so each heatsink is required to dissipate a minimum of 42.66 Heat Watts.



Should A = 286mm evenly spacing the COBS out along my 860mm width?

Should B = 166mm evenly spacing the COBS out along my 500mm depth?

With passive cooling in mind, what value would be considered a minimum for length "C"?

Thanks.

 

[Rahz]

Assuming you're using the 5.88 profile 20" bars would be good. Anything over that is gravy. I think B is fine, but I would probably make A around 400mm, basically splitting the length in two and centering the cobs. 3 inches on each end and the bars would be 22".

 

[Chuff420]

Assuming you're using the 5.88 profile 20" bars would be good. Anything over that is gravy. I think B is fine, but I would probably make A around 400mm, basically splitting the length in two and centering the cobs. 3 inches on each end and the bars would be 22".

I like your idea about splitting the length in two and spreading the COBSs further apart. That would definitely give better light distribution

Before discounting CPU coolers, could someone please tell me what would happen if I had a [email protected] generating 21.33 Heat Watts attached to a CPU cooler and the cooler fan stopped working due to mechanical failure or a failed power supply. Would the COB survive this or would it overheat and fail?

Thanks.

 

[klx]

I like your idea about splitting the length in two and spreading the COBSs further apart. That would definitely give better light distribution

Before discounting CPU coolers, could someone please tell me what would happen if I had a [email protected] generating 21.33 Heat Watts attached to a CPU cooler and the cooler fan stopped working due to mechanical failure or a failed power supply. Would the COB survive this or would it overheat and fail?

Thanks.

IF your fans go out they will overheat after an unknown period of time but the probability of that happening is low. The problem in Oz is the stupid price of heatsinks. It even works out cheaper to import heatsinks even with the weight and shipping. Crazy. For me, CPU coolers are the go here until I find a heatsink supplier with realistic prices. I have checked em all, fastron, conrad, ullrich etc. Hundreds of dollars per bar. Joke. If you are on a budget passive is going to be tricky. If you find a good source please let us know.

I must ask, if you are in Oz, why are you ordering your COBs from China and not getting from cutter.com.au? I am chiming in late sorry if you already mentioned why.

 

[Chuff420]

IF your fans go out they will overheat after an unknown period of time but the probability of that happening is low. The problem in Oz is the stupid price of heatsinks. It even works out cheaper to import heatsinks even with the weight and shipping. Crazy. For me, CPU coolers are the go here until I find a heatsink supplier with realistic prices. I have checked em all, fastron, conrad, ullrich etc. Hundreds of dollars per bar. Joke. If you are on a budget passive is going to be tricky. If you find a good source please let us know.

I must ask, if you are in Oz, why are you ordering your COBs from China and not getting from cutter.com.au? I am chiming in late sorry if you already mentioned why.

Yep. My concern was loosing a COB if the fan went out.

Agree about heatsink prices in Oz. I've considered the same manufacturers as you and I've been focusing my preference towards the Fastron H42 profile, which I'm now thinking will be overly efficient for the lengths I need.

Calculations tell me that as little as possibly 680mm is required to passively dissipate the total Heat Wattage of the 4 COBS and this adds up to to $125, however, in order to achieve optimum placement of the COBS within my space I will need to buy about 1100mm, thereby bumping the cost up to $203.50. 1100mm of 5.88 at HeatsinkUSA is almost half that at AU$105. If I could get 1100mm of H42M for $105, this discussion would be over.

So, buying approximately 400mm more of H42M than required just to better space out the COBS doesn't make economic sense, especially at the prices we pay, so I'm now considering the smaller H4 profile. I'll need a longer length of the H4 which will be better with COB placement, but with only 860mm of cabinet width, I may not have enough space for the length to dissipate the Heat Watts..

As for buying from China, Cutters don't have the HLG-240H-C1750 drivers I needed to run them at 1.75A. I even emailed Mean Well Australia and their reply was:

"The HLG-240H constant current LED driver has only recently been released. As of yet there are no distributors with stock."​

Now that I need an HLG-185H-C1400, this may change.

 

[pop22]

I came here looking for information related to using the CXA 2530 for my build also. Great information here, thank you all! I'll be reading thru the rest of this for sure!

My plan is to build 3, 39" long heatsinks with 5 COBs each. haven't decided yet whether to do passive or active cooling.

So I walked away and educated myself a bit more and discovered that SupraSPL had uploaded a copy of his spreadsheet, and for a guy like me, that's the Holy Grail.

I made some modifications based on what I have learnt here and hopefully discovered all the answers I wanted, only to then find an error in SupraSPL's spreadsheet which impacted directly on my work.

The snippet of the CXB2530 from the COB spreadsheet thread below which I posted previously in this thread is at this stage an unknown.

View attachment 3617064

Notice the underlined CXB2530 above the spreadsheet is not the same as in the top row in the spreadsheet, which is CXB2350. I've put up a post in the COB spreadsheet thread regarding this discrepancy in order to clarify what the correct labelling should be. You can read my post here for a detailed explanation if you wish. https://www.rollitup.org/t/cob-efficiency-spreadsheets.865238/page-18#post-12360078.

Needless to say, I am keen to find out exactly what COB it represents as will become apparent further down.

So with that in mind, I won't be sure what data is being represented in the samples below were references to CXB2530 are made until SupraSPL or someone else can clarify which COB the data really represents. Regardless I'll press on, as all the other information is correct, assuming my calculations are correct.

I modified the original spreadsheet and added the following information based on the formulas provided:

• Green area where cabinet dimensions added for later calculations.
• Red areas indicating the number of COBs to evaluate, as well as there Unit cost & Totals.
  
• Blue area:
  • PAR(W)/COB = 50c dissW * 50c %
  • PAR (W) Total = PAR(W)/COB * # COBS
  • PPF = PAR (W) Total * 4.51
  • PPFD = PPF * Area (Metres²)
  • W/Feet² = ((# COBS * 50C %) / Area (Feet²)) * 100

The first thing that became apparent was the area I had to work with. In my first post I noted the chamber size as 860mm wide x 450mm deep and it's 1200mm high (approximately 3.5' x 1.8' & 4' high). My mistake was using "round about" figures for the dimensions in feet, and as such all further calculations were based on those figures equalling about 6 feet². Well, a proper conversion shows that I actually have 4.165668 feet².

Anyway, this is what I have found.

If I give myself a budget of around $150 for COBS this is what I get:

View attachment 3617109


If I give myself a budget of around $200 for COBS this is what I get:

View attachment 3617110


I haven't costed a driver into the equation, but my budget for COBS & Driver is around $300, so with the 2 spreadsheets above I've allowed enough room to get close or slightly above my $300 limit.

Excluding the CXB2530 figures for now as I'm not sure what the data represents yet, and assuming my formulas are correct, then the 3 * CXB3590 @ 0.7A & 4 * CXB3070 @ 0.35A are getting close to the magic 50W/foot² figure based on the new area of 4.165668 feet² if I spend $200 on COBS. The figures for the CXB2530 look absolutely brilliant, and based on the price would definitely be the way to go, but I'm kind of stuck until I know what COBs that data represents.

Does all this look about right?

Thanks.