Easy DIY PVC lighting upgrade

[pepetj]

I've been experimenting with different ways to upgrade lighting systems in most of my tanks, specially the ones aimed at being heavily planted.

Finding suitable lamps is very difficult in my City, never mind looking for retrofit kits locally. Even with somewhat limited but still sometimes available T-8 standard fluorescent lamps, shortage in stock is not uncommon (e.g. T-8 at 20W, rated at 6700K).

So after going through my learning curve, which included getting sealed garden lamps with 13W CF rated at 3,000K, I built a lamp assembly using PVC pipes and caps. I began using standard CFs, rated at 6,400 and/or 6,500K (depends on manufacturer), the newest T-2 lamps seem to work better. I am experimenting with different wattage ratings (from 15 to 27W).

The first unit I left unpainted and it worked pretty well. The second and third unit I'm working on are covered inside with the brightest white spray paint I could find and mate black on the outer side.

I am using high pressure 4" diameter PVC. The sides are covered by 4" PVC caps cut by half (caps not shown in drawings)

Below are a couple of drawings to show what I'm into:

Pepe
Santo Domingo

 

[KyWildFish]

Have you thought of using a reflective substance like foil? I know some other members have used it to boost their existing systems.

Also pics later would be awesome to see what in fact you did, in terms of assembly and what not.

 

[pepetj]

Have you thought of using a reflective substance like foil? I know some other members have used it to boost their existing systems.

Also pics later would be awesome to see what in fact you did, in terms of assembly and what not.

I've used reflective foil (not mylar) and also white paint. White paint (the brightest the better) works better in my opinion (judging from my eyes, I don't have a lux meter).

I am going to upload better pics as I'm done with my 2nd and 3rd units.

Pepe
Santo Domingo

 

[Tony G.]

wow it looks really well! I will deff keep this in mind for the future

 

[KyWildFish]

Nice!

I would think a reflective surface would return more light than a simple coat of paint, although that's just a guess based on what I know of light waves.

Quick question on the set up. Do those type of bulbs build up a lot of heat? I have started switching and noticed they can get kind of warm.

 

[mathas]

I would think a reflective surface would return more light than a simple coat of paint, although that's just a guess based on what I know of light waves.

Pure white is a reflective surface.

The main drawback to applying a metallic reflective material like mylar, aluminum foil, etc. has to do with diffusive reflection (there's a good graphic illustrating this difference ). If the material isn't applied as smooth as glass, any wrinkles will diffuse the light all over the place rather than reflect it downward, reducing the efficiency.

If you've ever tried to wrap something in aluminum foil and get a perfectly smooth, non-wrinkled finish with no tears, you realize just how hard that is to actually accomplish. A can of flat ultra-white spraypaint, on the other hand, takes a few minutes to apply.

This topic has come up before, and I posted several links regarding paint vs. aluminum/mylar in DIY reflectors. Rather than post them all again, check these threads:

 

[pepetj]

Quick question on the set up. Do those type of bulbs build up a lot of heat? I have started switching and noticed they can get kind of warm.

Yes they do heat up, likely as a T-10 does. Be careful if using them in the standard plastic lamp assembly that most tanks come with. I have had my share of trial-and-error evidence that the 27WCF can certainly cause damage to a 20W T-8 rated lamp cover. I downgraded to 19W-20W-22W and seems to be working fine (still checking). That is the reason I used high pressure PVC instead of thiner, easier to cut, cheaper ones.

The 27WCF stained but not caused deformities in the non-painted surface of the high pressure PVC lamp unit #1 (the 28" long). I'm still working on the assembly of the latter units (around 54" long) so I'll let you know how they behave.

Pepe
Santo Domingo

 

[KyWildFish]

mathas
I would think that using the method that pepe has (a semi-circle) would eliminate any issues with improper deflections. The curvature would ultimately focus almost everything down and out.
Foil wouldn’t be the first thing I would try, mainly because of the issues you pointed out with the wrinkling and what not. Instead a highly polished piece of sheet metal might serve the same purpose. Aluminum would be best IMO but would require some clear prep to keep the oxidation buildup at bay. It is the right tensile response to cut it and curve it into place. This would also aid in reducing the heat as aluminum has a low heat capacity.

Your link just shows that white paint is about equal to a reflective metal. Shouldn't a very high polish be assumed to be better than flat white paint? Also remember that white light is the spectrum of visible light, and lots of stuff in tanks need wavelengths that exist outside the spectrum of visible light.

A mirror is probably going to be the best bet in terms of total reflectivity (think microscopes and telescopes). Its possible to outfit a similar geometry that you have in your store bought set up. The hardware store can cut mirror for you at any size and it can then be arranged inside pepe's tube to angle all the light into the tank.

In defense of your system of white paint, its probably a lot more costly than 3$ to do this mirror project, so its up to how much light you really want returned I suppose. What is the cost of a PAR meter? It would be cool (at least for a data hungry science geek like myself) to test as many methods as possible, and create a cost vs. output table for people to reference. This would give you something visual to post for DIY types like us to reference and stop some redundant debates/discusions.

 

[mathas]

Instead a highly polished piece of sheet metal might serve the same purpose. Aluminum would be best IMO but would require some clear prep to keep the oxidation buildup at bay. It is the right tensile response to cut it and curve it into place. This would also aid in reducing the heat as aluminum has a low heat capacity.

I certainly don't disagree, but how many hobbyists really have the knowledge, materials, and/or tools to fabricate an optimally-reflective polished aluminum reflector?

Most people who want that (myself included) buy them premade from companies that specialize in engineering them. For a DIY project in the basement/garage/whatever, that sounds like a bit of a chore to reproduce (though I could be wrong!).

Your link just shows that white paint is about equal to a reflective metal. Shouldn't a very high polish be assumed to be better than flat white paint?

My links were to show that white paint often performs better than any flexible reflective product applied to an existing structure. When you originally used the term "a reflective surface," I misinterpreted that. When this topic comes up, most people who talk about reflective surfaces mean aluminum foil/tape or mylar, and I somehow assumed you were talking about the same thing.

If you're talking about a true polished metal reflector, yes, that is substantially more effective, on the order of 2-3 times as effective according to these findings.

Also remember that white light is the spectrum of visible light, and lots of stuff in tanks need wavelengths that exist outside the spectrum of visible light.

I guess it depends on what your goals are. Most people who worry about adding reflectors are doing so to benefit plants or corals... I don't know what corals need, being a freshwater-only guy, but plants need very, very little (if anything) outside of the visible spectrum.

A mirror is probably going to be the best bet in terms of total reflectivity (think microscopes and telescopes). Its possible to outfit a similar geometry that you have in your store bought set up. The hardware store can cut mirror for you at any size and it can then be arranged inside pepe's tube to angle all the light into the tank.

Would you really need a mirror, though? It seems like the glass portion would be unnecessary, unless you're just suggesting mirrors because they're readily available.

What is the cost of a PAR meter? It would be cool (at least for a data hungry science geek like myself) to test as many methods as possible, and create a cost vs. output table for people to reference.

They're around $200 for a good one.

 

[KyWildFish]

mathas
Sounds like we are in general agreement. I am currently working on procuring a borrowed light meter. I have access at my place of employment, but doubt they would be happy to see me using their highly calibrated expensive equipment to further knowledge for a personal hobby. I am working on the logistics of the "experiment" this week/weekend and will probably consult you personally for focus on the desired results. I just love geeking out fish keeping!

To get the thread back on track...ahem..pepe, do you have any pics of the DIY system you have in use? It would be interesting to see what it does to the tank in an aesthetic perspective. The design you have is small enough to put into a canopy or another kind of box so I am considering borrowing the idea if you don't mind?

 

[pepetj]

Update:

I am leaning toward using T-8s instead of CFs. I believe the initial cost (higher) will in the end make sense quality wise. I could place 2, 3 or even 4 T-8s inside the 4" PVC semicircle. I will need to purchase electronic ballasts, some electrical wire, and the sockets.

I do have a question for mathas and the experts: Is it OK to mix different Kelvin ratings if I keep the Wattage (power consumption) per lamp constant as I use the same ballast?

I was thinking of mixing 6,400 or 6,700K with 10,000K of T-8 40W 48"; as well as in another assembly of T-8 30W 36"...

Thanks

Pepetj
Santo Domingo

 

[mathas]

I do have a question for mathas and the experts: Is it OK to mix different Kelvin ratings if I keep the Wattage (power consumption) per lamp constant as I use the same ballast?

Absolutely. In fact, mixing bulbs with differing spectral outputs (regardless of what color temperature number the manufacturer assigned) can allow you to overcome any weaknesses present in a specific bulb.

For example, let's say you have a bulb that's weak in the 430-450nm range. If you use four of those in a four-bulb fixture, you're always going to have a deficiency there. If you mix two of those bulbs with two bulbs that are stronger in that range, your overall lighting will be better for your plants.

 

[pepetj]

Update: Although not necessarily a fire risk I would like to report that one of the trial units that I have been running with four compact fluorescent (different brands ranging from 22 to 27W, 5,500K, 6,400K and 6,500K) ended up "burning" (more like "brown roasted") with mild plastic deformation due to heat. The lamp units running with T8 lamps haven't presented such problems.

Pepetj
Santo Domingo

 

[slipitysmooth]

Look into using plastic gutter for holding lights...I built a light bar for my 29 with a 2x39w ballast and gutter, it fits right over the glass section of my stock perfecto top, no problems yet with 2 24w T-5 HO bulbs.

 

[mjordan]

my prob (not to but in on the topic late but) is that I would like to know how effective CF from a depot store are going to help plant growth in a way that's its cheaper to use them than a plant or full spectrum bulb.

I have some CF specificall made for aquarium use by the brand All Glass which are 10 watt but I assume these other bulbs might be the same thing and the "said aquarium use" make them be able to sell these for 6 bucks a piece. which really isn't all that bad.

 

[pepetj]

my prob (not to but in on the topic late but) is that I would like to know how effective CF from a depot store are going to help plant growth in a way that's its cheaper to use them than a plant or full spectrum bulb.

I have some CF specificall made for aquarium use by the brand All Glass which are 10 watt but I assume these other bulbs might be the same thing and the "said aquarium use" make them be able to sell these for 6 bucks a piece. which really isn't all that bad.

I have replaced most of the hardware store CF with T8 lamps (only two units running with CFs waiting for me to upgrade to T8). The cheap CF lamps available locally ran out of light intensity quite quick (say 2 or 3 months). In the end it is not cost effective since I need both correct light spectrum and correct light intensity to keep my plants comfortable.

Pepetj
Santo Domingo

 

[mjordan]

he t-8s are they the screw sockets. if so can u provide a link to wat your buying

 

[pepetj]

he t-8s are they the screw sockets. if so can u provide a link to wat your buying

The T-8s are the standard straight lamps. I removed the standard bulb sockets. I used tie-wraps to hold the lamps in place, a solder iron to wire the bulbs, and 1" rubber stoppers to seal the connection (works as end caps). I placed a Normal Output electronic ballast as remote unit and covered the loose wires with those heat-shrink covers (without using heat at all). So far I ended up getting 75W per lamp unit in the 30"L and 90W per lamp unit in 36"L.

Pepetj
Santo Domingo

 

[aspects]

I love this build. very innovative using pvc. I'm definitely going to give this a try, but using the hagen T5 HO ballasts and wires instead of CFL. I have a couple pairs of ballasts/wires/endcaps lying around from old cannopy builds. as long as I can secure the endcaps to the PVC somehow, this should work.
thanks for the great idea.

awesome build.

 

[pepetj]

I use what locally is sold as rubber stopper, it's just an end cap made of rubber. The 3/4" works great for the T5HO and the 1" for the T8.

To secure the end caps you could use tie-wraps, easy to replace and adjust. Once the lamp assembly is in place lamps won't move but be careful when cleaning it or working around your tank.

Pepetj
Santo Domingo

 

[crazy4fish_123]

awesome job..never thought of that before

 

[pepetj]

Fire hazard

Since July 4th 2009 I ran one of my prototype units non-stop in a low-light planted tank. In this particular unit I kept the lamps despite the wear down in light quality and quantity for photosynthesis, as I wanted to see how PVC behaved over time; two lamps were replaced (20 & 22 W) due to malfunction during this time period, one of them twice (22W).

This PVC lamp unit was not coated with paint and used 4 CFL T2 or T3 of the "twisted" type from the hardware store (all rated within the 5,500 to 6,500K) one rated at 27W, another at 25W, another at 22W and another at 20W. All but the 20W produced notable "burn" marks in the PVC wall.

It didn't catch fire but I advise taking the safe road and so keep the CFL lamps wattage at or below 20W if you use High Pressure PVC (which is the one I used in this project); if you use the Mid-Pressure PVC consider lowering that ceiling to say 15W.

If short on budget, this type of easy DIY PVC lamps can buy you two (maybe three) months time before needing replacement (or retrofit to T5HO!) for planted tanks (more if non-planted).

 

[ilikefish]

Wow! thanks for the info and update! This is definitely going to be my next fixture...

I appreciate the testing too!

And while I wholely agree that white paint reflects better than foil... you could have wrapped the pvc in foil to prevent the burn marks thus allowing for higher wattage lights could you have not?

Regardless excellent job!! : )

 

[pepetj]

I ended up painting the "burned" prototype PVC lamp assembly. I placed two F31T5HO at 10,000K (allegedly full spectrum from Coralife). I placed the electronic ballast in the rear, outside part, glued with hot silicon.

Not too fancy but at least it's not aesthetically unpleasant.

Pepetj
Santo Domingo

 

[ilikefish]

Wait... in the picture it looks like you just connected those bulbs directly to AC power... don't you need a "ballast" or something in between??

Don't tell me all of my DIY light contraptions were a waste when I could've used those bulbs lol

 

[pepetj]

LOL. It has a Fulham Workhorse5 Electronic Ballast (remote). I followed the wiring instructions and that's it. Note that I use soldering iron to secure the contact of the wires that go (output of the ballast) into the lamps then I just apply hot silicon glue to insulate the whole deal. I used tie wraps to hold the lamps in place. In the pic there's a lot of wiring all around I will cut it later on.

Pepetj
Santo Domingo

 

[ilikefish]

Oh ok... lol just checking... Looks great!

 

[jwokmuba2]

could you maybe put a few vents in the top of the fixture to let the heat escape? like three slots, or maybe a few holes in a line, dunno, thought is might dissipate some heat

 

[Theepum]

you said you would clean up the electrical wiring... I would very much advise that you do so, i'm an electrician and i've had to replace numerous fixtures because of improper ventilation and heat exposure to the wires - insulation dries, cracks and falls off leaving bare exposed live wire... not good!

but other than that, looks good!