DIY Icecap...its not what you think

[reefburnaby]

Hi,

VHOs on an electonic ballast (4xF32T8s)

Yes, you can drive one VHO with this setup...although, the lamp will only be as bright as a Icecap (or roughly). Regular VHO ballasts (the real ANSI standard ones that most people can't buy), are brighter. It is possible to hook the VHOs in series and light them both.

How it works is that the 4x configuration converts this ballast in to a 600mA to 800mA lamp current ballast. Lamp current is the current flowing through the actual fluorescent tube. Normal T8s need about 235mA of lamp current. Normal VHOs need 1600mA of lamp current. Icecap delivers around 600mA to 800mA of lamp current to their lamps -- this overdrives T8s/T12s and underdrives VHOs. By underdriving VHOs, they can improve the lifetime of the lamps (from 6 months to more than a year).

The power drawn might be higher than the rated wattage (as stated on the box), but they tend to put safety factors around those numbers. So, if you wish (knowing the risks), you can use them to drive two VHOs at 65W to 80W a piece (I didn't measure the actual power yet). The limiting factor with electronic ballasts is the heat that they produce...higher heat...less power it can handle before blowing up. So, if you can keep the ballast cool (with a fan or in a well ventalated area) than you should be fine. These ballasts are designed to work in crawl spaces with no ventalation at 70 to 85 degrees Celcius (~190 F).

So, if you have an old set of VHO tubes, you can give them a try. Keep in mind that VHO tubes may only last 6 to 10 months...unlike Icecap ones. This is mainly due to the fact that our ballasts do not have the fancy starting mechanism -- even through the reduced lamp current may increase lifetime.

Another alternative (a better one) is to use Workhorse 7 or 6 ballasts.

Finding those elusive T8s ?

Grainger should carry these lamps and they have outlets all around the country. Last time I looked, they sold them for $2.75 USD. 6500K is commonly referred in the lighting community as daylight. Cool = 4000K and warm = 3200K.

There should be other places that carry them....mainly light bulb distributors. Companies that sell lamps from Phillips, GE or Sylvania should sell or be able to get them for you.

Local contractors should be able to tell you where they buy their lighting stuff.

Hope that helps.

- Victor.

 

[waj]

I do not think that 4x overdriving with this ballast with T12 lamps really overdrives any T12 lamp all that much. For some reason my crappy meter cannot measure lamp current directly but I can measure the current the ballast draws and the lamp voltage. By assuming ballast losses are constant I can then guess at the lamp current and wattage.

The more you overdrive a lamp, the more the voltage drops so even if you put 4x current into the lamp the lamp wattage does not increase proportionately. I think overdriving the lamp reduces the lamp resistance meaning that the wattage of the lamp is nowhere near 4x. There is only about a 7% increase in light output between 3x operation and 4x operation using T8 lamps. The difference is not caused by heat losses as the ballast draws only 8% more current. Lamp efficiency therefore drops very little.

There is a limit on how much a ballast like that will drive a lamp and I guess it is would be nowhere near VHO levels even if VHO lamps have higher than normal resistance. There would be no point in doing this with a VHO lamp. They cost too much for what you would get. Might as well use NO or HO lamps.

Overdriving NO lamps is very useful but it would be a lot better if we could quantify how much extra light we are actually getting. I can do this with T8 lamps because I can use the ballast factor of of a 4 x 32 watt ballast as a reference and can therefore calculate the ballast factor for T8 lamps by sampling the light increase when the a lamp is overdriven. Measuring T12s is more of a problem. Since the lamp cross section is different I cannot directly compare my T12 light samples with my T8 light samples. I need an accurate reference ballast to do this.

Wayne

 

[StirCrazy]

I did a test on overdriving light and recorded power consumption vs light output. you can see it on this thread.
http://www.canreef.com/ubb6/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=7;t=000014

Steve

 

[Boomslang]

Hi

I read throught this thread and I see there is a lot of people that worry about the tubes exploding, from what I have read there is a vacuum in the tube, so when you break the tube it does not explode but implode (sp?). When the tube gets hot the chance of it exploding is just about zero, because there is very little air in the tube to expand from the heat.

 

[waj]

Really what I am looking for is the ballast factors for different lamp overdriving combinations. This is critical for sizing lighting systems and predicting light output from fixtures without having to actually measure them. I have them for 2' and 4' T8 lamps but not for anything else. This is something that seems to be completely ignored by most hobbiests or companies that cater to hobbiests. Icecap for instance won't publish this info. Perhaps if they had, people would have realized a lot sooner that they are not really a VHO ballast at all. They are HO ballasts. You can buy HO sign ballasts much like them from any electrical supplier.

One of the problems with your setup is that lamps of different cross sections or lengths cannot be accurately compared because the light is dispersed differently by different lamps. You also cannot compare lamps of different spectra as a lux meter as I am sure you know, takes it's sample mainly from the green area of a spectrum. Obviously, lamps with different spectral power distributions will give different results even if they have the same light output in watts.

What I did was to make a box that enclosed the center portion of the lamp and had a hole cut out for the light sensor. I measured 2' and 4' T8s and T12s in about a dozen configurations. My setup would be much more sensitive to lamp shape than yours but better for comparing lamps of different lengths. I measured the amperage draw on the ballast using the meter in series and the voltage across the lamps as well as a lux sample from each lamp. I then calculated the ballast factor for the T8 lamps using the 4 32 watt lamp configuration. Since I knew the ballast factor was .88 for this setup I used that as a reference to calculate the other ballast factors for other configurations. To make it work for T12 lamps I would have to use a ballst designed for T12 lamps with a known ballast factor to use as a reference. Mine are all pretty old magnetic ballasts and I don't trust them.

What do you do with the filament heating circuits of your rapid start ballast when you overdrive? Do you just leave the spares off or do you use them? It would seem better to not have any filiment heating circuits at all if you are driving a non standard lamp and that way you won't get the wrong filament current or preheat time. Is it better to use a program start ballast even if the preheat current is wrong or is it better to just use an instant start ballast?

Wayne

 

[StirCrazy]

Really what I am looking for is the ballast factors for different lamp overdriving combinations. This is critical for sizing lighting systems and predicting light output from fixtures without having to actually measure them. I have them for 2' and 4' T8 lamps but not for anything else. Icecap for instance won't publish this info. Perhaps if they had, people would have realized a lot sooner that they are not really a VHO ballast at all. They are HO ballasts. You can buy HO sign ballasts much like them from any electrical supplier.

my one Ballast is a 3 bulbs T8 electronic ballast, I will get you the power factor later, about the Icecap being a HO ballast, this is what I have stated sence befor the begining of this thread.. I actualy believe they are two HO ballasts in a commen box wich would account for the pickeyness in how you wire them.

One of the problems with your setup is that lamps of different cross sections or lengths cannot be accurately compared because the light is dispersed differently by different lamps. You also cannot compare lamps of different spectra as a lux meter as I am sure you know, takes it's sample mainly from the green area of a spectrum. Obviously, lamps with different spectral power distributions will give different results even if they have the same light output in watts.

Yup, this is why I stated in the test that this could only be used to compare the out put of the same bulbs and not different bulbs.. sence then I have bought a PAR meter and I am going to be testing different bulbs and ballasts as I can aford to.

Is it better to use a program start ballast even if the preheat current is wrong or is it better to just use an instant start ballast?

Wayne

Laitly I have moved to the WorkHorse line of ballasts by fulham, they are a instant start ballast and so far I have found several advantages over the Icecap. the main three are, 1 the Fulham is a "Type 1 Outdoors" ballast whare the IceCap is a "Type 2 Outdoors" and 2, the IceCap has a lot of electric generated noise that will interfere with X-10 componants whare the Fulham workhorse are a "Class A" noise rating. I curently have 3 of them and they don't screw around with my X-10 in anyway. and 3, they are a lot cheeper. a Ice cap 660 is about 300.00 + the wire harness + tax, the WH7 is about 55.00 + tax.

and just for side info the WH will run different types of light at the same time also. I know a couple guys that are mixing 55 watt PC's and NO actinics on the same ballast.

Steve

 

[waj]

I wrote a long reply but now I don't know what happened to it. I am not used to this type of format. Anyway at the risk of repeating myself I am interested in ballast factors and not power factors. The ballast factor is a term that is used to describe a multiplier that you use to figure out the actual lumens produced by a lamp. Your 3 x 32 watt ballast often has a 1.2 or 1.15 ballast factor. This means that whatever the lumen rating of your lamp, you just multiply the ballast factor by the lamp lumens to get the actual light output. There is no way of knowing the ballast factor when overdriving lamps unless you actually measure it. Without those factors it is very hard to know if it is worth it to do so.

I am quite impressed that you have a PAR meter you will have lots of fun with that, especially when comparing MH lamps to fluorescent lamps. I use PAR/lumen values calculated from lamp spectrums to convert lux measurements to PAR measurements but I think a PAR meter is preferable. It is hard to know if published lamp spectrums are accurate so the conversion factors are suspect. I think you will find in the end though that most fluorescent lamps of the same shape are pretty much the same per watt consumed regardless of the phosphor. In theory lamps with more red would produce more PAR per watt but in practice I don't think this varies all that much. It is the shape of the lamp and length of the lamp and the ballast that makes the most difference in PAR output as well as lumen output.

I also think that a 4 x 32 watt ballast is just about the same animal as a workhorse 5 and can run a pair of 55 watt lamps using two outputs per lamp. Those lamps will probably push the ballast to the maximum, the lamp wattage should then end up at around 54 watts if you get a high enough voltage across the lamp. I believe this would be even cheaper than a workhorse 5 as the 4 x 32 ballasts are only Can. $35 at my Home Depot. Can anyone confirm this?

Wayne

 

[qik64]

hi there,i hope someone can help me.i have a aqua one 620 tank and would like to know if i can overdrive the lights in it,they are 23.5 inch tubes.
brett.

 

[ToeCutter]

They are 17 bucks at home depot.

Advance Transformer Co.
High Frequency Electronic Ballast for (3) or (4) F32T8 Lamps
Model: #REL4P32SC35I
SKU: #261255
17.49

Thanks for the info. I picked up a pair today. I wired one up as a test, looking pretty good.

Now I just need to find the right bulbs. I gambled on GE "aquarium" bulbs which are actually 3050K - back to HD they go.

And maybe score some 18" reflectors... yeah...

 

[Chrismo]

18" tubes OVRNO for Nano

18" tubes OVRNO for Nano

I just hooked up 2 18" (15W each) T8 tubes in series to my 2 bulb 32W electronic ballast. The bulbs look really bright! Maybe I''ll try some SPS in my 10g nano?

I just noticed in small print on the ballast that it says it can ALSO run:
(2) 25w,
(2) 17w,
(1) 32w.
(1) 25w

How do I figure out how it's running the bulb? How does it determine what kind of bulb it should be outputting for? or does it just put out onw wattage, and the ballast factor changes a bit depending on what bulb is in there?

It looks pretty darn bright, so I'm happy, but I wonder if I'm running it at 2x or 4x or what?

Chris

 

[jjharrisx4]

I use it in my 15g tall canopy with 2x36w PC and 1x15w NO. It made the tank much brighter. I only am concerned about the adhesive. I think I'll be Ok because I sealed the canopy, it is open so not too humid, and the adhesive is damn strong stuff. It was 5.00 for a roll at home depot...looks like a roll of shiny duct tape.

 

[waj]

When you hook the lamps in series and overdrive 2x the ballast thinks it's running a 2 3' lamps. I believe the resistance of the lamp plays a part in the fact that the ballast changes it's output when you put in different lamps. Of course the ballast is designed to run high frequency T8 lamps so it is hard to predict exactly how much more light you are getting. One good indication is just to measure how much power the ballast draws. The light output somewhat proportional to the power consumption except the more you overdrive the less efficient the lamp becomes and the less light per watt you get.

Wayne

 

[blulegend]

I use it in my 15g tall canopy with 2x36w PC and 1x15w NO. It made the tank much brighter. I only am concerned about the adhesive. I think I'll be Ok because I sealed the canopy, it is open so not too humid, and the adhesive is damn strong stuff. It was 5.00 for a roll at home depot...looks like a roll of shiny duct tape.

So are you using the NO ballast to power your PCs AND NO?

 

[ronrca]

Overdriven T8's

Overdriven T8's

I currently have a 90G planted tank with 6 3ft fluorescent tubes in my diy tank hood. Of course, 6 32W tubes will not provide the lighting that the plants require (unless only low light plants). I was made aware of this method of overdriving T8 electronic ballast on this site some months ago and purchased 1-2 lamp 1-3 lamp and 2-4 lamp electronic ballasts. They have been doing great (plants included). Just on thursday I bought an additional 2 4 lamp ballast therefore overdriving 4 tubes with 4 lamp ballasts, 1 with a 2 lamp ballast and 1 with a 3 lamp ballast. When I open the hood, I have to watch not to look into the lights because they are very very bright.

I would really like to thank whoever thought of this idea because if I would of had to buy a retrofit VHO kit, I would of had to spend around $400+. By using this method, it costed me $150 and Im calculating around 4W/G (maybe closer to 5W/G). Thank you!

5w/g may sound high but my tank is quite tall, at least 24" and Im curious to see if I can start growing some 'carpet' plants now that I just install the 2 ballasts. I think this is a great cost saver and great diy project.

I have not been able to read all the posts in this topic but I wanted to share my experience and gratidute.

 

[waj]

My main interest is in planted tanks also. I have a 90 gallon tank with 6 32 watt lamps overdriven by a 1.2 ballast factor. This is enough light to grow carpet plants on the bottom of a 90 gallon tank. The light level you have is much greater than this. The ballast factor when overdriveing 4x is over 2.0. Just 4 4' 32 watt lamps overdriven 4x would be as much light as anyone would ever need for a planted tank. Even 4 lamps overdriven 3x would be plenty for all but the most light demanding plants.

The person that thought of this has my gratitude as well. It has put a whole new light so to speak on DIY lighting. I acn now DIY an excellent lighting system that is just as good as any PC system for a fraction of the cost. I can also vary my lighting according to my needs just by rewiring the ballasts and I can use the same ballast for every application.

Wayne

 

[ByTor]

for the T8 bulbs in actinic and daylight Triton & zoo-med makes them from 18"to 48"....
I also found these bulbs

Someone posted about the Voltarc bulbs a few months ago..Thanx!

Voltarc also has some really nifty end caps....

 

[ronrca]

Thank you waj for your insight. Since I only have increased light levels on thurday, it will take awhile to see how the carpets will preform. Before, Hemianthus micranthemoides (I think thats what it is or maybe Micranthemum umbrosum) would not grow out like a carpet plant should but up. Im hoping that will change though now.

One thing Im kicking myself now for is when I built the hood, I did not build it to accomodate 4' fixtures therefore Im only using 3' lamps. You would not believe it but my hood is about 1" to small to fit 4' lamps. Maybe if I get ambitious, I'll re-design and re-build the hood to fit 4' lamps and also better air movement thru the hood. Maybe I dont have though.

You can see my pics at http://photos.yahoo.com/ronr_ca

Please comment on what I can do to improve the aquascaping. Thanks!

 

[ToeCutter]

You would not believe it but my hood is about 1" to small to fit 4' lamps.

They make space-saver bulbs that are about 1 1/2" shy of 4' or so, just for that circumstance.

 

[DerrickT]

For all of you that are overdriving NO, do you see the ripple effects similar to those of MH? If so, can you post pics. I would like to see it very much.

 

[DerrickT]

Hi

I read throught this thread and I see there is a lot of people that worry about the tubes exploding, from what I have read there is a vacuum in the tube, so when you break the tube it does not explode but implode (sp?). When the tube gets hot the chance of it exploding is just about zero, because there is very little air in the tube to expand from the heat.

Oh, for safety can one of those tube sleeves be used on overdriven bulbs? Can they withstand the heat?