Yet another UV thread

[canadian1969]

I thought the same thing last week.
The legend shows times a # for a few plots. So I think the amplitude has been multiplied by a constant t bring it up for plotting reasons.

I was looking into it a bit more and the Doctors seem to suggest that you can still get upwards of 80% UV exposure even on an overcast day.

So if these assumptions are correct...
5% or less of solar radiation falls as UV
Of that 5% (95% UVA,5%UVB)
The max intensity is at noon is 250 microwatts/cm2 of UVB
then for a 95 watt white grow light you need 5 watts of UV
presumably (4.75watts UVA and 0.25watts UVB)
However if 250uw/cm2 is high noon, we should average that out over a 12 hour period. So I am thinking a bell chart, split half way, 125uw/cm2 as per the equation 0.1161 watts per square foot. So for a 4.5sq' tent would only equal half a watt.

If I try to work it out the other way and say 200 watts for a 4.5sq' tent, *5% = 10watts, 9 watts UVA , 1 watt UVB
It is not wildly off.

 

[iHearAll]

I was looking into it a bit more and the Doctors seem to suggest that you can still get upwards of 80% UV exposure even on an overcast day.

So if these assumptions are correct...
5% or less of solar radiation falls as UV
Of that 5% (95% UVA,5%UVB)
The max intensity is at noon is 250 microwatts/cm2 of UVB
then for a 95 watt white grow light you need 5 watts of UV
presumably (4.75watts UVA and 0.25watts UVB)
However if 250uw/cm2 is high noon, we should average that out over a 12 hour period. So I am thinking a bell chart, split half way, 125uw/cm2 as per the equation 0.1161 watts per square foot. So for a 4.5sq' tent would only equal half a watt.

If I try to work it out the other way and say 200 watts for a 4.5sq' tent, *5% = 10watts, 9 watts UVA , 1 watt UVB
It is not wildly off.

Sounds about right along with the graph. Where the multipliers were only ~1.5x and ~3× . I see there could be a problem within penetration of canopy if you only use 5w distributed overall he canopy in a few chips. I know my actually power draw is very high for my setup but it averaged about 7w per triplet. I think the best route would likely be to use 1w diodes, power 60 or so of them at various wavelengths at a low power, and just get more distribution for a short amount of time. I imagine that once the needed energy from the UV range has been absorbed, the rest becomes detrimental like we have seen with synged and burnt leaves in grow rooms and outdoors.

How many diodes did you purchase for experimentation???

 

[xX_BHMC_Xx]

I'm running about 5-6% of total wattage in those SemiLEDs UVA diodes from Rapid in the cab. Just barely UVA at about 395nm, but they are for reef tanks after all. Assuming the mid-level NF3 bin, that's about 693mW/sq.ft., which correlates to about 37mW/sq.ft. of UVB using the 95-5 ratio. Seems I'm running my UV a bit higher than I thought. I haven't noticed any negative effects, no popped trichs or burnt pistils (other than the ones I damaged myself). I'll try to run a comparison when I get the new tent set up but it won't be a straight side by side.

 

[iHearAll]

i had to short out a triplet so my driver would work in the voltage range. For now i am only using 27 of 30 of the UVa diodes. I ordered another driver to split the UVa in groups of 15 diodes later. So, half the panel could have UVa and the the other half may not. Or even time them to come on at different times for whatever reason.

 

[canadian1969]

How many diodes did you purchase for experimentation???

none yet, I needed to work out the proportions and power, I am still not sure I have a working formula. You definitely seem to have found the best deal on the diodes, ebay, go figure. So thanks for that. What drivers did you buy? I am interested in how you wired it up too if you dont mind giving me a lesson on those LEDs.

The panel I am building is made with EB strips 22" there is space for 3 more strips in the configuration, but they are linear, so I cant get great distribution. The sinks will be 1'x24' aluminum flats ,with 22" of workable space. I will have to work out how many UV/ far red (and anything else I can think of) and build those 3 flats with the specialty diodes only.
I suppose distribution would look like:
1=5700K , 2=3000K 3=Special
212321232123212

 

[iHearAll]

none yet, I needed to work out the proportions and power, I am still not sure I have a working formula. You definitely seem to have found the best deal on the diodes, ebay, go figure. So thanks for that. What drivers did you buy? I am interested in how you wired it up too if you dont mind giving me a lesson on those LEDs.

The panel I am building is made with EB strips 22" there is space for 3 more strips in the configuration, but they are linear, so I cant get great distribution. The sinks will be 1'x24' aluminum flats ,with 22" of workable space. I will have to work out how many UV/ far red (and anything else I can think of) and build those 3 flats with the specialty diodes only.
I suppose distribution would look like:
1=5700K , 2=3000K 3=Special
212321232123212

They peak at 1a and are advised too use .7a by the supplier. I am using .6a and around 3.8v to 3.9v. The drivers fortunately ship from the us but are Chinese made. They're not bad, their power factor is 95% or more. I am using two drivers each has 15 diodes. You can use up to 27 diodes per driver but not less than 15. Unless of course you get a different model. Ill just link you the webpage. Take their "number of leds" count with a grain of salt because I found it was inaccurate and the seller wouldn't listen when I said that I had measured the characteristics of the diode and that they were making an ill suggestion by saying it can power 30diodes and apparently I was WRONG (which im not) that the leds require 3.8v at .6a. Ebay salesmen for ya smh. Go with the waterproof model

http://m.ebay.com/itm/High-Power-Supply-3W-10W-20W-30W-50W-100W-Waterproof-Constant-Current-LED-Driver-/122480543106?var=&hash=item1c8468c582:m:m_9_Ce2c08G3bkEQe1zCeuQ&_trkparms=pageci%3A6fb62b0c-6c81-11e7-8404-74dbd180df70%7Cparentrq%3A5aea318715d0aa411d558440ffff2cb4%7Ciid%3A1

 

[canadian1969]

How would you power 4 of these?
https://www.digikey.ca/product-detail/en/RVXE-280-SB-071004/1807-1003-ND/7363758
They are star boards 150ma 6.5v

 

[iHearAll]

How would you power 4 of these?
https://www.digikey.ca/product-detail/en/RVXE-280-SB-071004/1807-1003-ND/7363758
They are star boards 150ma 6.5v

So, this will be goofy but cost effective: if you parallel the four of these and place them in series with the .6a chips I listed, they will be perfectly happy the catch however is your parallel running wires will be spaced throughout your panel so the 280nm leds are evenly spread over the canopy. I can sketch a pic if you are confused by what I mean.

I may buy some of these if theyre already mounted to pub!! Thanks!

 

[iHearAll]

Woah 61$ for four ??? Yikessss. Ill be looking into 300-360nm diodes first i think

 

[canadian1969]

Woah 61$ for four ??? Yikessss. Ill be looking into 300-360nm diodes first I think

yes but I cant find any others that are even close to that price at 280nm, min order is 4, I thought wtf why not. I bought 10 of the ebay 365nm, so now I have to get them to work along side the ebay ones without creating a birds nest of wiring. A sketch of what you mean would be helpful, but I think i get it 4 in parallel takes you to 600ma , but these are 6.5v and the others are 3.8v, confused on that part

 

[iHearAll]

yes but I cant find any others that are even close to that price at 280nm, min order is 4, I thought wtf why not. I bought 10 of the ebay 365nm, so now I have to get them to work along side the ebay ones without creating a birds nest of wiring. A sketch of what you mean would be helpful, but I think i get it 4 in parallel takes you to 600ma , but these are 6.5v and the others are 3.8v, confused on that part

Only the 280nm are in parallel, they go in series with the 3.8v diodes. So the voltage of the 280nm does not matter as long as your driver can handle it. Voltages add up in series and current is constant. Current adds up in parallel and voltage is constant.

Here's an schematic example

And a layout example of just the 280nm. Your wire can be as long as you like to not be in the face of your EB strips

 

[canadian1969]

As per your schematic (thanks!) you can take the parallel 280's and consider it one diode in series with the other 2, and as a series circuit they will share 1/3 of the voltage each and the same current. So the 280's are 6.5v/150ma, so we need to present 6.5 volts a the top of each diode, so you would need a constant current driver to supply 600ma and 19.5v (6.5x3) , in that example. The min vf on those 280's is 5v max 8v and a max of 200ma. Seem to me you would have to drive the ebay diodes at 3 watts (5v*600ma) to get compatible output on the 280s. Having said that my 4 semesters of electronics is 25 years old, so real rusty. lol. Of course you would lose some output on the 280s. I think.

 

[iHearAll]

As per your schematic (thanks!) you can take the parallel 280's and consider it one diode in series with the other 2, and as a series circuit they will share 1/3 of the voltage each and the same current. So the 280's are 6.5v/150ma, so we need to present 6.5 volts a the top of each diode, so you would need a constant current driver to supply 600ma and 19.5v (6.5x3) , in that example. The min vf on those 280's is 5v max 8v and a max of 200ma. Seem to me you would have to drive the ebay diodes at 3 watts (5v*600ma) to get compatible output on the 280s. Having said that my 4 semesters of electronics is 25 years old, so real rusty. lol. Of course you would lose some output on the 280s. I think.

Let me use your original numbers. I think you wanted to use 10 off the 365nm correct? So, these each require 3.8v. Ten would be 38v. Then you add the four 280nm diodes at, lets say, 6.5v. They're only going to need 6.5v to power all four at once and will split the 600ma current into four. So, you need .6a and roughly 44.5v to power the entire circuit.
Like such,

 

[canadian1969]

And to my understanding with that power supply 44.5v @600ma you would in fact be driving each diode (including the parallel set) at 44.5/11 = 4.05v afaik you cannot have unequal voltages across a series circuit.

 

[iHearAll]

And to my understanding with that power supply 44.5v @600ma you would in fact be driving each diode (including the parallel set) at 44.5/11 = 4.05v afaik you cannot have unequal voltages across a series circuit.

Untrue. You're thinking of the situation where you have eleven equal resistances Im series. Here you have ten equivalent resistances and one different. 3.8v ÷ .6 = 6.333ohm x 10= 63.33 ohms and in series with a parallel bank of four 6.5÷.15=43.33ohm resistors . the parallel bank has a total resistance found by the formula (r1^-1 + r2^-1 + r3^-1 + r4^-1)^-1= Rt. Plug in 43.33 for each and you'll get 10.83ohms.

So its perfectly safe to wire this way and your leds will power perfectly

 

[canadian1969]

Untrue.

Then I definitely need a lesson as to how/why etc.

 

[iHearAll]

Then I definitely need a lesson as to how/why etc.

I edited with an explanation

 

[canadian1969]

I edited with an explanation

Thanks, clear now.

 

[canadian1969]

hey have you ever tried using something like this?
http://www.ebay.com/itm/New-AC-DC-Power-Supply-Buck-Converter-Step-Down-Module-5V-12V-3-3V-9V-24V-500mA/122311100637
was thinking the 150ma one 24v would about work to power them in series (the 280's alone I mean)

this blew my mind today too, hence the new question lol

http://www.ledsmagazine.com/articles/print/volume-6/issue-2/features/led-design-forum-avoiding-thermal-runaway-when-driving-multiple-led-strings-magazine.html

or this
https://www.aliexpress.com/item/10Pcs-Isolation-5W-AC85-277V-LED-Driver-3-10x0-5W-150mA-DC9-34V-LED-PowerSupply-ConstantCurrent/32670741180.html

I should have been checking Alibaba/aliexpress , dunno, maybe I am barking up the wrong tree

 

[iHearAll]

hey have you ever tried using something like this?
http://www.ebay.com/itm/New-AC-DC-Power-Supply-Buck-Converter-Step-Down-Module-5V-12V-3-3V-9V-24V-500mA/122311100637
was thinking the 150ma one 24v would about work to power them in series (the 280's alone I mean)

this blew my mind today too, hence the new question lol

http://www.ledsmagazine.com/articles/print/volume-6/issue-2/features/led-design-forum-avoiding-thermal-runaway-when-driving-multiple-led-strings-magazine.html

or this
https://www.aliexpress.com/item/10Pcs-Isolation-5W-AC85-277V-LED-Driver-3-10x0-5W-150mA-DC9-34V-LED-PowerSupply-ConstantCurrent/32670741180.html

I should have been checking Alibaba/aliexpress , dunno, maybe I am barking up the wrong tree

All depends on your goals in the end. You'd have a bare circuit. There is a waterproof 300ma 3v minimum driver in the link,I provided a few comments stringIts 6$ and some change. That'd be 2 in series and the series repeated in a second parallel circuit. It'd work like your link to ledmagazine suggested essentially