Space requirements for 600w of CREE CXA 3070

SupraSPL

Well-Known Member
Yessir my math agrees with your numbers and brushing up on this stuff with you guys helps me big time. I think it will perform great for you at 530 PPFD but if you want a bit more punch you could crank them up to 1150mA :eyesmoke:
 
Wow, i didn't expect all this when i posted the thread!

Thanks for everything guys, you've made everything endlessly easier!

Now just to try and source some Z4 3070s! Easier said than done i guess! With the Z2 it drops from ~204 PAR W (Z4) to ~191 PAR W (Z2), so i guess if i can't find any Z4's, an extra Z2 would fill the void so to speak.

Thanks for the efficiency information on the driver; i'll have to carefully reconsider my options there! Ideally i'd prefer not to spend the extra cash and introduce HV into the room, but depending on the efficiency, it may be the way to go.

Thanks again everyone!
 

mauricem00

Well-Known Member
has anyone experimented with transformerless power supplies in order to improve driver efficiency.these are the type of drivers used in led light bulbs and a larger version would drive 50 watt cobs. the one I built runs much cooler than the chinese CC drivers and showed an efficiency of over 97%
 

captainmorgan

Well-Known Member
Hmm thatsa good question Cap, that would be 61.3 PAR W for 6 sq ft, 10.2 PAR W/ sq ft. That is right where I am at and it works well.

For comparison, if you went with 5 COBs on the HLG-185H-1050A as OTIMT mentioned you would get 84 PAR W = 14 PAR W/sq ft. (I calculated at 1150mA because that driver actually puts out up to 1150mA).
Thanks Supra,do you think it would be worth the added expense of two more COB's to gain 3% efficiency and and the added PAR watts for a 30"x30" space and what size wire would you need for the COB's for both those options?
 

SupraSPL

Well-Known Member
Yes that 3% gain yields 7% more bud/W and 5 COBs gives a more even spread, plus of course the overall increase in PAR W means more bud as well. So if the money was available for the build, I do think that would be worth it and would pay for itself in 1 week.

I like 18 gauge wire because it is easy to work with and is cheap enough, but in this case I would recommend making sure it is 600V rated. I get spools from elecdirect.
 

Bueno Time

Well-Known Member
has anyone experimented with transformerless power supplies in order to improve driver efficiency.these are the type of drivers used in led light bulbs and a larger version would drive 50 watt cobs. the one I built runs much cooler than the chinese CC drivers and showed an efficiency of over 97%
I haven't and know nothing about them but 97% efficiency sounds amazing.
 

mauricem00

Well-Known Member
you can google them and find information on designing one for your application. they use a series capacitor on the ac line to shift the phase relative to the load and drop the ac voltage to the load voltage but unlike a resistor they do not dissipate power in the form of heat.they simply but the load out of phase with the ac line so that it only draws the power the load needs.they run much cooler than a cc driver but if you do not know how to work with ac line voltage safely they can be dangerous.
 
I like 18 gauge wire because it is easy to work with and is cheap enough, but in this case I would recommend making sure it is 600V rated. I get spools from elecdirect.
I am planning to use 18 AWG 4 conductor CL2 rated for 300V. This way I can carry my LED and fan power into the tent on one line without the need to wire loom my own cable.

The price is very right, but what are the benefits of 600V in our applications? Or is it more just a measure of safety?

http://www.monoprice.com/Product?c_id=102&cp_id=10239&cs_id=1023904&p_id=4043&seq=1&format=2#specifications
 

SupraSPL

Well-Known Member
has anyone experimented with transformerless power supplies in order to improve driver efficiency.these are the type of drivers used in led light bulbs and a larger version would drive 50 watt cobs. the one I built runs much cooler than the chinese CC drivers and showed an efficiency of over 97%

you can google them and find information on designing one for your application. they use a series capacitor on the ac line to shift the phase relative to the load and drop the ac voltage to the load voltage but unlike a resistor they do not dissipate power in the form of heat.they simply but the load out of phase with the ac line so that it only draws the power the load needs.they run much cooler than a cc driver but if you do not know how to work with ac line voltage safely they can be dangerous.
Which LED bulbs have these type of drivers? Do you have a pic of the one you built?

I found this one but it is only 30mA and does not mention efficiency
 
Last edited:

MrFlux

Well-Known Member
has anyone experimented with transformerless power supplies in order to improve driver efficiency.these are the type of drivers used in led light bulbs and a larger version would drive 50 watt cobs. the one I built runs much cooler than the chinese CC drivers and showed an efficiency of over 97%
Can you post more info on this? Can it handle closed circuit, open circuit, and AC spikes?
 

mauricem00

Well-Known Member
Can you post more info on this? Can it handle closed circuit, open circuit, and AC spikes?
these are very simple circuits that are designed for a specific load think of them as a resistor on the ac line connected to a bridge rectifier and filter capacitor to convert the ac voltage into dc voltage.now replace the dropping resistor on the ac side of the bridge rectifier with a motor run capacitor. still in series with the powerline.by using reactance instead of resistance to drop the voltage you are matching the impeadance of the load to a source impeadance that would draw the same aount of power as the load.spikes are filtered out by the filter capacitor on the dc side.it can not handle short circuits and would need a fuse for that. open circuit would just draw 0 current. this is a very old technology that has been adapter for use in led light bulbs. but phase shift technology for power management dates back to the earliest AC power grids and is used on electric motors as well. you will probably find motor start or run capacitors in your refrigerator or ac units. impedance matching is commonly used in ham radio so any ham could explain the principle behind it. http://www.engineersgarage.com/contribution/transformerless-power-supply-circuit this site will give you information and formulas for designing one and this site will gige you diagrams and information https://www.google.com/search?q=designing+transformerless+power+supply&rlz=1C1PQHB_enUS569US569&es_sm=122&tbm=isch&tbo=u&source=univ&sa=X&ei=uU92U4LoK8b7oATtsIL4BQ&ved=0CEUQsAQ&biw=1366&bih=667. you are simply matching the impedance of the AC side of your bridge rectifier to a virtual impedance that would pull the same amount of power as your load .current regulation depends on how consistent your line voltage is but unless leds are pushed to there maximum power levels a high degree of current regulation is not needed I am currently testing one of these supplies on a 50 watt led panel because I was troubled by how much heat the CC driver produced the three components in the power supply stay cool even after twelve hours of operation but the only source for large AC capacitors are motor run capacitors that are designed to handle a few thousand watts of power do not use motor start capacitors since they are not designed for continuous operation motor run capacitors are this is the unit I am using http://www.ebay.com/itm/Run-Capacitor-40-MFD-440v-Round-AC-Electric-Motor-HVAC-440-vac-v-volts-40-uf-/261104075876?pt=LH_DefaultDomain_0&hash=item3ccb040464 but the value of the capacitor has to be calculated for the load you are running. it may look simple, old fashion and low tech but it is used in every led light bulb on the market and helps them achieve high efficiency with cheap low powered leds driven a little to hard for maximum efficiency.bottom line its cheap and efficient and it works well
 

mauricem00

Well-Known Member
Which LED bulbs have these type of drivers? Do you have a pic of the one you built?

I found this one but it is only 30mA and does not mention efficiency
I built a larger version to power a homemade array of 576 smd 5630 ww/cw leds wired in a series/parallel array that runs on 77.6 volts at 600 ma (25ma per diode) based on the philips data sheet it is putting out 5600 lumens (120 L/W 114 L/W with driver loss) and the total power from the wall outlet is 49 watts for led and driver combined.I like the (wall of light) approach. that is similar to what I get with T5s for some reason my camera lense acted like a prism and created a ghost image in this picture the light is bright enough to cause me to see sunspots for a few minutes after looking directly at it but since it is in effect a giant discrete COB. this kind of power supply should work well on high powered diodes. it is amazing how well "matched" those 5630 are. this discrete array shows better current sharing than many chinese COBs that I have tested ( no ballast resistors used on array)
 

Attachments

Hey i need help. I was wondering if there a difference between the meanwell hlg 185h-36b and the meanwell hlg 185h-1400b. I have the 185h-36b and was planning trying to hook it up to 4 cree cxa3070 and then i read i can only run 3 instead of 4 like the cxb3070. Fuck! Anyways I tried hooking up just 3 of them and i still got nothing. Im running in series positive to negative and the only thing i havent hooked up yet was the potentiometer but i thought if you didnt hook that up then it would just run at full power. Am i missing something. Ill take pictures and post them
 
Last edited:

nevergoodenuf

Well-Known Member
The 36v driver will require you to run them in parallel and yes you can run 3 or 4. The 1400B is the driver you would run in series.
 
The 36v driver will require you to run them in parallel and yes you can run 3 or 4. The 1400B is the driver you would run in series.
Thanks. I just found that out. I wish i knew that the drivers were different and i was mislead by the sales person. Is running them in parallel any less efficient?
 
Top