-
ICMag with help from Landrace Warden and The Vault is running a NEW contest in November! You can check it here. Prizes are seeds & forum premium access. Come join in!
Can you grow with cold cathode computer lighting?
- Thread starter ScrubNinja
- Start date
Hey kidz, awesome posts. My buddy has a light meter that measures in lux, I may be able to borrow it if I ask nicely. Is lux useful to us? My camera has some kind of lighting stuff in it. I've seen a thread where you can use a decent camera as a dodgy light meter...I think.
I know you said no diagrams, but I live on the edge
Look at where the positives and negatives are going. In parallel, all the positives come back to one terminal, as do the negatives. If you study how the wires go in the standard molex setup you'll see this. In series they're all over the place. So it's easiest to just describe parallel. Let me know if you still don't get it. You're probably going to beat us all to actually grow something, lol. Really nice progress mate
Stealth, did you or someone have the cooltube idea already? I think I remember the airpump being mentioned already. Only thing about the airpump is they're so noisy for how many cfm they pump.
Your cab is looking bomb dizzle fo scnhizzle! Loving that louver!!! You did oversize it to allow for the 1/3rd blockage right?
That bare wire inside is called the return wire. You are right, it's external to the *real* tube by the looks. Yeah I wouldn't want plants contacting it.
My bro has a clamp meter, like a killawatt, I can use that too but it's a long way away. There are some figures already as I'm sure you're aware too, but I also would love to measure it myself.
I think that's all. Big up CCFL massive
Oh also, all I can think of is to stick a thin sheet of styrofoam behind your power adaptor.
I know you said no diagrams, but I live on the edge
Look at where the positives and negatives are going. In parallel, all the positives come back to one terminal, as do the negatives. If you study how the wires go in the standard molex setup you'll see this. In series they're all over the place. So it's easiest to just describe parallel. Let me know if you still don't get it. You're probably going to beat us all to actually grow something, lol. Really nice progress mate
Stealth, did you or someone have the cooltube idea already? I think I remember the airpump being mentioned already. Only thing about the airpump is they're so noisy for how many cfm they pump.
Your cab is looking bomb dizzle fo scnhizzle! Loving that louver!!! You did oversize it to allow for the 1/3rd blockage right?
That bare wire inside is called the return wire. You are right, it's external to the *real* tube by the looks. Yeah I wouldn't want plants contacting it.
My bro has a clamp meter, like a killawatt, I can use that too but it's a long way away. There are some figures already as I'm sure you're aware too, but I also would love to measure it myself.
I think that's all. Big up CCFL massive
Oh also, all I can think of is to stick a thin sheet of styrofoam behind your power adaptor.
apples
Active member
Looking good stealth. I have some progress aswell.
I build a cab out of 1/2 inch plywood I had layin around.. 26tallx12widex13deep(inner dims)
I only used 12 bulbs until I get all this wiring business sorted out. I still don't understand the parallel thing. I need a visual with actual molex connections.
3 holes on each side will be made to run the wires out to the inverters. A hole will be cut in the top for a nice Aspire pc fan. I might need two.
There is a gap at the bottom for passive intake. All seams will be sealed and the front door will get hinges.
I think I'll line the walls with reflective tape.
still much to do.
I build a cab out of 1/2 inch plywood I had layin around.. 26tallx12widex13deep(inner dims)
I only used 12 bulbs until I get all this wiring business sorted out. I still don't understand the parallel thing. I need a visual with actual molex connections.
3 holes on each side will be made to run the wires out to the inverters. A hole will be cut in the top for a nice Aspire pc fan. I might need two.
There is a gap at the bottom for passive intake. All seams will be sealed and the front door will get hinges.
I think I'll line the walls with reflective tape.
still much to do.
Science, my good fellow, science.
And I'm hoping you hadn't seen my post when you posted, apples. Looks wicked. Some thoughts: Do you need lights that low? Have you calculated the area of your intake slot? It doesn't look very big to me.
And I'm hoping you hadn't seen my post when you posted, apples. Looks wicked. Some thoughts: Do you need lights that low? Have you calculated the area of your intake slot? It doesn't look very big to me.
apples
Active member
yeah, why try doggy style when you can just do missionary....pffft.
Anyway...after looking at alienbaits pictures I think I finally get it. I should just chop all the wires off the inverters and wire all the reds together and all the black together to 1 main molex. Correct?
edit: No, I duno how to calculate that. Its about 1/2 inch tall 12in wide
Anyway...after looking at alienbaits pictures I think I finally get it. I should just chop all the wires off the inverters and wire all the reds together and all the black together to 1 main molex. Correct?
edit: No, I duno how to calculate that. Its about 1/2 inch tall 12in wide
Hey apples, nice cab can't wait to see how it works!
There seems to be some wiring confusion, I looked at the pic of your power supply and you have 2 12v output rails. Each one runs some of those molex connectors like a chain. To spread the load across the power supply evenly, you need to find out which string of molex conns is on each rail and half of the lights on each chain, let me know if you need help with that, it will require an ohm meter. If you hook all of the lights to one of them, you will be using half the power supply. Half of your power supply is capable of running 30 or so of those lights so it doesn't really matter if you want to run them just like they are. I'm not a big fan of molex connectors as they fail alot. Ideally, ditch the molex conns and hard wire half of the lights to each 12v rail on the PSU.
There seems to be some wiring confusion, I looked at the pic of your power supply and you have 2 12v output rails. Each one runs some of those molex connectors like a chain. To spread the load across the power supply evenly, you need to find out which string of molex conns is on each rail and half of the lights on each chain, let me know if you need help with that, it will require an ohm meter. If you hook all of the lights to one of them, you will be using half the power supply. Half of your power supply is capable of running 30 or so of those lights so it doesn't really matter if you want to run them just like they are. I'm not a big fan of molex connectors as they fail alot. Ideally, ditch the molex conns and hard wire half of the lights to each 12v rail on the PSU.
I will try to make up a diagram of how I plan to do it. I haven't thought too hard about it so it may take a while.
What is a formula people can use to add up their amperage to check if their power supply is suitable? I get confused because there's all these figures for the lights, and the inverters. Thus I bought a PSU
which arrived today!
Edit: Yes, that is basically how you should wire it, apples. Well, it's how I plan, except break half the lights into one molex on one rail, and the other half to the other molex and rail, as mentioned.
What is a formula people can use to add up their amperage to check if their power supply is suitable? I get confused because there's all these figures for the lights, and the inverters. Thus I bought a PSU
which arrived today!Edit: Yes, that is basically how you should wire it, apples. Well, it's how I plan, except break half the lights into one molex on one rail, and the other half to the other molex and rail, as mentioned.
apples
Active member
LOL my head is spinning with wiring sickness. I don't know a thing about electronics and I appreciate all the help from everyone but I still don't get it.
I wish I could understand what you mean Aero.
Should I:
Wire all positives together and all negatives together as mentioned before? Which would make things cleaner. Or will that not solve my problem of the power supply turning off?
Get another 4pin molex cable and use a second slot on the PSU?
Something else beyond my grasp?
Edit: ok, so a little of column a little of column b. wire half and half dividing them up on each rail.....now I just need to find out wtf a rail is lol
I wish I could understand what you mean Aero.
Should I:
Wire all positives together and all negatives together as mentioned before? Which would make things cleaner. Or will that not solve my problem of the power supply turning off?
Get another 4pin molex cable and use a second slot on the PSU?
Something else beyond my grasp?
Edit: ok, so a little of column a little of column b. wire half and half dividing them up on each rail.....now I just need to find out wtf a rail is lol
I have found that Molex connectors work fine for small loads but if you try to draw too many amps through them, they eventually fail. I had one fail trying to run my 1.65amp blower fan, turned the pins black and killed the fan. I was lucky to catch it soon because the cab temp was 106F when I found it. As you know the grow went on to be a great success but that crappy connector could have cost me the entire crop. If you can, hardwire the lights because you will be drawing much more current than my exhaust fan with those lights.
Apples, your idea to keep all the pos and negs together is correct. The only other thing to do now is to try to get half of the lights on a different 12v chain of molex connectors. I'll start, how many chains of molex connectors does that PSU have?
Hey well listen to Aero, but here is the illustration I just made to show the wiring...just keep connecting the inverters to the same colours. And make one of these setups for each rail...which aero is probably gonna help you with now.
Edited pic to remove the molex.
Edited pic to remove the molex.
Last edited:
Sorry, I know you are trying to concentrate on wiring, but Apples about your intake slot, it depends what size fan you are using. If it's a 120mm fan then you need 35.4 sq inches to equal 2x intake to exhaust. You have 6. So your slot needs to be 6 times as high, if you have a 120mm fan.
does this help anything?
So I just need to know where exactally I should hard wire these once they are all wired together.
Yes, that tells me you have a modular power supply, and quite a nice one too. How many cables that plug into that thing have molex connectors on them?
apples
Active member
I only have 1 molex cable from the original power supply. (got the pc from a friend b4 I upgraded it)
I have an extra cable from my new power supply but the connecter is 4pin+2. I say that because it has 6 holes but 2 are empty. What if I broke those 2 empty ones off? Or should I just buy a new 4pin cable.
I have an extra cable from my new power supply but the connecter is 4pin+2. I say that because it has 6 holes but 2 are empty. What if I broke those 2 empty ones off? Or should I just buy a new 4pin cable.
those power supplies are slightly different than the regular ones.
although it still has basic rules that apply to it.
all power supplies are split into 3v 5v and 12v connectors on their molex.
cables plug into those holes and the other end of those cables can vary.
The PCI-6P port is for a high power PCI-Express x 16 graphics card, most of them require additional power other than the power provided by the PCI x16 slot its self. This should be a molex connector at the end of this one.
The SATA port one is for sata hard drives. The other end of this cable will be multiple sata power connectors. Not the easiest to work with.
The multiple use holes will be your best bet. Not sure how this particular PSU works but it probably is running different circiuts for each port as was mentioned above you should make use of all those ports and share the power out by connecting in parallel.
If you dont have all the cables that fit into these multiple use ports and have molex on the other end, then im sure you can get them from the same place you bought this PSU from.
best of luck
although it still has basic rules that apply to it.
all power supplies are split into 3v 5v and 12v connectors on their molex.
cables plug into those holes and the other end of those cables can vary.
The PCI-6P port is for a high power PCI-Express x 16 graphics card, most of them require additional power other than the power provided by the PCI x16 slot its self. This should be a molex connector at the end of this one.
The SATA port one is for sata hard drives. The other end of this cable will be multiple sata power connectors. Not the easiest to work with.
The multiple use holes will be your best bet. Not sure how this particular PSU works but it probably is running different circiuts for each port as was mentioned above you should make use of all those ports and share the power out by connecting in parallel.
If you dont have all the cables that fit into these multiple use ports and have molex on the other end, then im sure you can get them from the same place you bought this PSU from.
best of luck
Could someone please advise me I'm good to go with this? Thanks.
12 bulbs, 12 inverters. Each side of the bucket will have a bunch of 6 inverters wired as above. I was going to use one of the sets of molexes to power the fan controller and about 2.5 amps of 12v fans. The other set of molexes will be removed and hardwired to power all the lights/inverters. (Aero, the Vantec Nexus fan controller is rated to handle these fans, so I'm a little skeptical to remove the molexes here. I'd have to remove the molex connector on the controller and solder wires on, blah. So I will have to risk it unless you have any ideas?)
Besides the two molex sets, it has the 20 pin thing, and these two plugs:
Is there a better way I should wire things? (and do I have enough amps with the fans as well?)
I read up on Yaha power supplies and they seem pretty crap. Think I should just upgrade?
Again, thanks. Here is some blurb from the specs
12 bulbs, 12 inverters. Each side of the bucket will have a bunch of 6 inverters wired as above. I was going to use one of the sets of molexes to power the fan controller and about 2.5 amps of 12v fans. The other set of molexes will be removed and hardwired to power all the lights/inverters. (Aero, the Vantec Nexus fan controller is rated to handle these fans, so I'm a little skeptical to remove the molexes here. I'd have to remove the molex connector on the controller and solder wires on, blah. So I will have to risk it unless you have any ideas?)
Besides the two molex sets, it has the 20 pin thing, and these two plugs:
Is there a better way I should wire things? (and do I have enough amps with the fans as well?)
I read up on Yaha power supplies and they seem pretty crap. Think I should just upgrade?
Again, thanks. Here is some blurb from the specs
LadyLargely
Member
You kids are crazy! All this buggering around with DC power, is there a reason you don't just run 18 inch T8 floros? This does all seem to be for curiosity's sake.
ScrubNinja Your power supply should have enough amperage, but the bummer is it only has one 12 volt power rail. That means it makes no difference which plugs you use, they're all taxing the same line. I would just cut all that plug shit off and wire your CCT power inverters up in parallel to the 12 volt main. Cut out the middlemen.
Most decent PC power supplies have at least two 12 volt rails. Lots of the SLI multi-graphics-card rated ones have even 3 or 4.
You've got plenty of amps, its just the nature of the load you are presenting to the PSU. Cold Cathodes run at HUGE voltages, those power inverters you have are like mini HID ballasts. They have to spank your 12 volts up to many thousands of volts in order to push juice through the gas-filled tube. This kind of load isn't very nice to the power supply.
Which brings up a serious issue with your little experiment here: Electrical Efficiency
I don't think your CC grow, or any CC grow, could ever generate a very good Grams Per Watt number. You just aren't making very good use of your wattage.
Electrical power comes to your house as Alternating Current. Somewhere around 110-240 volts at 50-60 Hz depending on what bit of the world you call home. The moment you convert that energy to DC power you throw a good hunk of it away.
*Dielectric withstand: Lnput to frame ground 1800VAC for 1 second
*Humidity: 5-95%RH
*Efficiency: Under full load circumstance, the lowest efficiency is 65%, normal is 70%
*Overload Protection: Max.150%
So right there, before your energy even makes it to the lamp, you've thrown almost 1/3 of it away. Now, part of this is due to your piece-of-shit PSU. Good ones these days are pushing 85% efficiency. But you will pay for that.
And then the converted 12 volt DC power gets to your cold cathode power inverter where it is converted back into a form of alternating current, throwing even more energy away, and that's before it gets to the transformer which steps the hell out of the voltage, tossing out even more.
At the end of the run, I would be astonished if any more than 50% of the energy that came out of your wall actually made it any where near those pretty little tubes. Your lighting wattage total might only come to 35 watts, but if you hooked a Kill-A-Volt or similar energy-metering device to your power supply you would find the final draw to be a lot more than that.
So, not to piss in your Wheaties or anything but any quests for good yield efficiency with Cold Cathodes are surely doomed from the outset. Now, in terms of quality, potency, stealth potential and general intrigue they still have a lot going for them. I'm certainly glad that somebody is trying them out, and given the mostly DC nature of the small Cold Cathode units like you guys are playing with they could make a really nice team along with some LEDs.
So good luck with your runs! Sorry I've not been so helpful with your wiring situations. Just use decent quality gear and do as much direct-drive soldered-in waste-free wiring as you can.
ScrubNinja Your power supply should have enough amperage, but the bummer is it only has one 12 volt power rail. That means it makes no difference which plugs you use, they're all taxing the same line. I would just cut all that plug shit off and wire your CCT power inverters up in parallel to the 12 volt main. Cut out the middlemen.
Most decent PC power supplies have at least two 12 volt rails. Lots of the SLI multi-graphics-card rated ones have even 3 or 4.
You've got plenty of amps, its just the nature of the load you are presenting to the PSU. Cold Cathodes run at HUGE voltages, those power inverters you have are like mini HID ballasts. They have to spank your 12 volts up to many thousands of volts in order to push juice through the gas-filled tube. This kind of load isn't very nice to the power supply.
Which brings up a serious issue with your little experiment here: Electrical Efficiency
I don't think your CC grow, or any CC grow, could ever generate a very good Grams Per Watt number. You just aren't making very good use of your wattage.
Electrical power comes to your house as Alternating Current. Somewhere around 110-240 volts at 50-60 Hz depending on what bit of the world you call home. The moment you convert that energy to DC power you throw a good hunk of it away.
*Dielectric withstand: Lnput to frame ground 1800VAC for 1 second
*Humidity: 5-95%RH
*Efficiency: Under full load circumstance, the lowest efficiency is 65%, normal is 70%
*Overload Protection: Max.150%
So right there, before your energy even makes it to the lamp, you've thrown almost 1/3 of it away. Now, part of this is due to your piece-of-shit PSU. Good ones these days are pushing 85% efficiency. But you will pay for that.
And then the converted 12 volt DC power gets to your cold cathode power inverter where it is converted back into a form of alternating current, throwing even more energy away, and that's before it gets to the transformer which steps the hell out of the voltage, tossing out even more.
At the end of the run, I would be astonished if any more than 50% of the energy that came out of your wall actually made it any where near those pretty little tubes. Your lighting wattage total might only come to 35 watts, but if you hooked a Kill-A-Volt or similar energy-metering device to your power supply you would find the final draw to be a lot more than that.
So, not to piss in your Wheaties or anything but any quests for good yield efficiency with Cold Cathodes are surely doomed from the outset. Now, in terms of quality, potency, stealth potential and general intrigue they still have a lot going for them. I'm certainly glad that somebody is trying them out, and given the mostly DC nature of the small Cold Cathode units like you guys are playing with they could make a really nice team along with some LEDs.
So good luck with your runs! Sorry I've not been so helpful with your wiring situations. Just use decent quality gear and do as much direct-drive soldered-in waste-free wiring as you can.
LadyL! Thank you. I really appreciate the blunt answer. I'm a realist, and I can see that I was wrong about getting a genuinely high yield from these, and I did try to clarify that as it became clear. One of the first and best links I linked to very early has clamp meter tests and discusses the terrible inefficiencies and low lumens per watt. But even with it clearly just being a nerdy fun thing though, AlienBait got pretty good results I thought? I mean, I hope I'm not clutching at straws here but if he got .2gpw in a room where he easily could have fitted 5 plants instead of 2, and that *theoretically* could have given him .5gpw right? (again, only talking about the wattage that the bulb is rated, as we've been looking at it, divided by the dried grams, I don't care about the time it takes personally). I know that's totally cheating on my part but it's all I care about. This is just the micro forum and I'm simply not that hard core to figure the real wattages in. Nobody does here except for the LED nurdz.
Anyway, half a gram per watt, if that is a reasonable ballpark figure (Y/N?) is actually pretty good yield over here on the wee 'lil micro forum. Shit, even if I did take into account all the inefficiencies, I still think it could potentially cane a lot of cfl grows featured here, basing this only on AlienBait's grow. I don't obsessively work out grams per watts but I suspect some guys here struggle to get .2gpw with regular fluoros. And hey, there is the occasional bit of efficiency going on, such as my unique bulb placement.
Well, I hope it doesn't seem like I'm arguing the point cos I'm not. You are right, it's inefficient to hell. Just different ways of looking at it, I suppose. It's not like I was weighing up a HPS, CMH, and CCFLs and thought that I'd get a higher gpw with ccfl's.
A 55w pll bulb (just to give you an idea of the high cost of lighting here - fluoro shoplights etc are equally overpriced)) costs $45aud at the local lighting store, and I have to wait weeks for them to order it in. And I'm pretty sure you would be aware I'm not the most clued up when it comes to fluoro tubes. They may seem pretty simple to you but I struggle to understand the details and I'm highly unfamiliar with them. So basically the effort and cost of using traditional fluoros is too expensive for me, and would be inefficient fun:effort ratio.
Thanks a lot though. I really am glad you put a knowledgeable slant on proceedings. I just wish you had've discussed light spectrums!!!! For me at least, that is much more of a problem/confusion-causer than the inefficiencies.
Again, thanks, and I can tell you ahead of time that it won't let me rep you I'll pick up a more efficient power supply as the first priority and brace myself for the outcome.
Anyway, half a gram per watt, if that is a reasonable ballpark figure (Y/N?) is actually pretty good yield over here on the wee 'lil micro forum. Shit, even if I did take into account all the inefficiencies, I still think it could potentially cane a lot of cfl grows featured here, basing this only on AlienBait's grow. I don't obsessively work out grams per watts but I suspect some guys here struggle to get .2gpw with regular fluoros. And hey, there is the occasional bit of efficiency going on, such as my unique bulb placement.
Well, I hope it doesn't seem like I'm arguing the point cos I'm not. You are right, it's inefficient to hell. Just different ways of looking at it, I suppose. It's not like I was weighing up a HPS, CMH, and CCFLs and thought that I'd get a higher gpw with ccfl's.
A 55w pll bulb (just to give you an idea of the high cost of lighting here - fluoro shoplights etc are equally overpriced)) costs $45aud at the local lighting store, and I have to wait weeks for them to order it in. And I'm pretty sure you would be aware I'm not the most clued up when it comes to fluoro tubes. They may seem pretty simple to you but I struggle to understand the details and I'm highly unfamiliar with them. So basically the effort and cost of using traditional fluoros is too expensive for me, and would be inefficient fun:effort ratio.
Thanks a lot though. I really am glad you put a knowledgeable slant on proceedings. I just wish you had've discussed light spectrums!!!! For me at least, that is much more of a problem/confusion-causer than the inefficiencies.
Again, thanks, and I can tell you ahead of time that it won't let me rep you
I'll pick up a more efficient power supply as the first priority and brace myself for the outcome.
Last edited:
LadyLargely
Member
Oh yea, forgot you where down in Oz. Fucking all lights are expensive for you poor cats, even shitty shop fixtures like I use.
GPW numbers can be quite murky. Usually you just take the total wattage that is says on the packages for your bulbs and divide out total grams of bud yielded. Problem with that is, wattage of the bulb does not equal actual power draw. For instance, a 400 watt HPS that yields 300 grams has managed .75 GPW yes? Not neccessarily. Depending on what kind of ballast is used (magnetic/digital) and the age of the bulb that lamp can be pulling anywhere from 450-600 watts! And even more during peak load times like when it ignites the bulb at the start of the day.
This means using the simple GPW method to compare one type of a grow to another is pretty apples/oranges. A .5 GPW LED grow is actually getting a much better yield efficiency than a .5 GPW HID grow IRL. Same number, but the LEDs made their weed using much less real-world energy usage.
And here we run into yet another light measurement issue. Low lumens-per-watt is generally accepted as a way to point out crappy lights. This isn't true!
Lumens are a measure of illuminating power, relatend to the candle-power measurement. The problem is this measure of illumination is based on illumination fit for the human eye!
Fucking PLANTS! have rather different priorities. What works well for the human eye is not neccessarily what will work well for growing the ganjas.
Perfect example:
I run 2 types of floros. For veg, I run GE daylight tubes, 6500K color temperature, 40 watts, 3050 lumens of light output. When I switch to flower I use GE Plant & Aquarium tubes. They are also 40 watts, but generate only 1900 lumens! Thats 76.25 lumens/watt vs 47.5 lumens/watt.
By any measure, the P&As should be infirior to the daylights right? WRONG!
If you where to measure the light output of the tubes differently you would get a different impression. Where you to guage those two lamps in terms of PAR (Photosynthetically Active Radiation) the P&A would come out as the clear winner. Sadly, very few manufacturers measure their lights in terms of PAR, even the ones who are selling grow lamps specifically, like the P&As. The best we can generally do is to use this tool:
To help us tune good grow lamps. As you can see there is a HUGE dip between 500 and 625 nm. This is around the 'green' part of the visible light EMR. Plants don't use the green bits to make their food, that part of the spectrum is reflected back rather than absorbed, which is why plants look green!
So the green bits of the light spectrum given off a bulb that can help us to see are useless to plants. That whole chunk of a bulb's Lumen output is meaningless for our needs. That's why lumens are a crap measurement for judging a good grow-lamp.
The best we can do is take that graph above and match it to this:
The spectral graph output of your bulbs. We want there to be big peaks of this graph right where there are big peaks on the PAR graph. See the big spikes at around 440 and 610 nm? Those match up perfectly with the chlorophyl A and chlorophyl B spikes. Now, the big jump around 540 nm is unfortunate. Thats in the green bit and hardly any of it will get used. You'll also notice the big lump down in the violet bits. This is actually the ultraviolet part of the spectrum and would not normally be covered on a graph like this. Its there because this particular graph describes my UV tubes.
And thats really all there is to understanding color temparatures and their significance to cannabis. Other than knowing that redder is better for flower and blue is best for veg there isn't a lot else to know.
If the spectral output graph makes your head hurt just hold on a second. You and StealthDragon where actually sortof making your own crude spectral graphs. The trick is to break up the light using a prisim and then look at the relative amounts of resulting color. It should come out looking like a rainbow, but certain strips of the color should be 'thicker' than others.
Honestly mate, this cold cathode business is much trickier than floros! With all the specific color output of them I'd say they are most comparable to LEDs as far as grow style. Only difference is the massive efficiency gap. Floros have much wider output bands, you can use them in more generic terms of 'red' and 'blue' as opposed to hunting out specific spectrum outputs like you kids are doing.
But cheers anyway! I don't think these cold cathodes could hold a candle to conventional floros (especially not High Output and/or overdriven models) in terms of real-world GPW numbers. But given your unique Oz situation it certainly wasn't a completely stupid idea. In America your little rig would be costing like 10x as much as an equivilent floro solution, but down where you are it sounds like its closer to 2x, or perhaps even less.
Its a very fun curiosity anyway. I have high hopes for your potential quality. Deep-red specific-output lighting like you have seems to generate the most amazing little candy-coated nuggets. You're such a goddamn tinkerer SN, I can't help but love this kind of shit even if it makes my inner penny-pincher a bit sad. Full steam ahead I say, let's see what your crazy spiral tunnel-of-light can do! I've already seen the tunnel-of-light concept well-proven with CFLs and such.
Oh! One thing though, I think you could give yourself a tiny bit more help reflectivity-wise. I know you used mylar tape on the inside of the tunnel but something like this...
...could help you claw back a few percent of efficiency. Apply clear-adhesive foil-tape directly onto the light tubes.I do it without trouble on overdriven T12s, so it should be no risk at all for you. You'd prolly have to use a razor blade or something to trim down thin little strips of tape; a tedious job. But if you're into tinkering and wringing out every last ounce of possible performance (seems like you are) then this might be a plus.
Before you pass judgement try this: Just tape up one of your tubes, power it up, and compare it to an un-taped tube. I think the difference will really astonish you, it certainly did me the first time I tried it on floros.
GPW numbers can be quite murky. Usually you just take the total wattage that is says on the packages for your bulbs and divide out total grams of bud yielded. Problem with that is, wattage of the bulb does not equal actual power draw. For instance, a 400 watt HPS that yields 300 grams has managed .75 GPW yes? Not neccessarily. Depending on what kind of ballast is used (magnetic/digital) and the age of the bulb that lamp can be pulling anywhere from 450-600 watts! And even more during peak load times like when it ignites the bulb at the start of the day.
This means using the simple GPW method to compare one type of a grow to another is pretty apples/oranges. A .5 GPW LED grow is actually getting a much better yield efficiency than a .5 GPW HID grow IRL. Same number, but the LEDs made their weed using much less real-world energy usage.
And here we run into yet another light measurement issue. Low lumens-per-watt is generally accepted as a way to point out crappy lights. This isn't true!
Lumens are a measure of illuminating power, relatend to the candle-power measurement. The problem is this measure of illumination is based on illumination fit for the human eye!
Fucking PLANTS! have rather different priorities. What works well for the human eye is not neccessarily what will work well for growing the ganjas.
Perfect example:
I run 2 types of floros. For veg, I run GE daylight tubes, 6500K color temperature, 40 watts, 3050 lumens of light output. When I switch to flower I use GE Plant & Aquarium tubes. They are also 40 watts, but generate only 1900 lumens! Thats 76.25 lumens/watt vs 47.5 lumens/watt.
By any measure, the P&As should be infirior to the daylights right? WRONG!
If you where to measure the light output of the tubes differently you would get a different impression. Where you to guage those two lamps in terms of PAR (Photosynthetically Active Radiation) the P&A would come out as the clear winner. Sadly, very few manufacturers measure their lights in terms of PAR, even the ones who are selling grow lamps specifically, like the P&As. The best we can generally do is to use this tool:
To help us tune good grow lamps. As you can see there is a HUGE dip between 500 and 625 nm. This is around the 'green' part of the visible light EMR. Plants don't use the green bits to make their food, that part of the spectrum is reflected back rather than absorbed, which is why plants look green!
So the green bits of the light spectrum given off a bulb that can help us to see are useless to plants. That whole chunk of a bulb's Lumen output is meaningless for our needs. That's why lumens are a crap measurement for judging a good grow-lamp.
The best we can do is take that graph above and match it to this:
The spectral graph output of your bulbs. We want there to be big peaks of this graph right where there are big peaks on the PAR graph. See the big spikes at around 440 and 610 nm? Those match up perfectly with the chlorophyl A and chlorophyl B spikes. Now, the big jump around 540 nm is unfortunate. Thats in the green bit and hardly any of it will get used. You'll also notice the big lump down in the violet bits. This is actually the ultraviolet part of the spectrum and would not normally be covered on a graph like this. Its there because this particular graph describes my UV tubes.
And thats really all there is to understanding color temparatures and their significance to cannabis. Other than knowing that redder is better for flower and blue is best for veg there isn't a lot else to know.
If the spectral output graph makes your head hurt just hold on a second. You and StealthDragon where actually sortof making your own crude spectral graphs. The trick is to break up the light using a prisim and then look at the relative amounts of resulting color. It should come out looking like a rainbow, but certain strips of the color should be 'thicker' than others.
Honestly mate, this cold cathode business is much trickier than floros! With all the specific color output of them I'd say they are most comparable to LEDs as far as grow style. Only difference is the massive efficiency gap. Floros have much wider output bands, you can use them in more generic terms of 'red' and 'blue' as opposed to hunting out specific spectrum outputs like you kids are doing.
But cheers anyway! I don't think these cold cathodes could hold a candle to conventional floros (especially not High Output and/or overdriven models) in terms of real-world GPW numbers. But given your unique Oz situation it certainly wasn't a completely stupid idea. In America your little rig would be costing like 10x as much as an equivilent floro solution, but down where you are it sounds like its closer to 2x, or perhaps even less.
Its a very fun curiosity anyway. I have high hopes for your potential quality. Deep-red specific-output lighting like you have seems to generate the most amazing little candy-coated nuggets. You're such a goddamn tinkerer SN, I can't help but love this kind of shit even if it makes my inner penny-pincher a bit sad. Full steam ahead I say, let's see what your crazy spiral tunnel-of-light can do! I've already seen the tunnel-of-light concept well-proven with CFLs and such.
Oh! One thing though, I think you could give yourself a tiny bit more help reflectivity-wise. I know you used mylar tape on the inside of the tunnel but something like this...
...could help you claw back a few percent of efficiency. Apply clear-adhesive foil-tape directly onto the light tubes.I do it without trouble on overdriven T12s, so it should be no risk at all for you. You'd prolly have to use a razor blade or something to trim down thin little strips of tape; a tedious job. But if you're into tinkering and wringing out every last ounce of possible performance (seems like you are) then this might be a plus.
Before you pass judgement try this: Just tape up one of your tubes, power it up, and compare it to an un-taped tube. I think the difference will really astonish you, it certainly did me the first time I tried it on floros.
