LED lighting on a budget!

Hi Keith,

Thanks for the reply. I will definitely start PMing lol. Now the problem I have right now is I don't remember what I actually asked since you deleted the post :headwally: lol.

I think you have it in the following.


You can not use this supply as it only has fixed outputs. You'd want smooth variability.

so if I used the power supply in the link, by changing the voltage, the amp of the wires will change? So dimming is not a good idea for your design? If it isn't, I will do that non-dimming method you have written out.
If you want to use the method I outlined in here you would set it all up just as I describe with the supply turned up. Then you could dim the entire setup by turning down the supply voltage.

I do have another question. The meanwell driver allows you to set a constant current. On your design, we used total of 10 strings so we need a power supply that can produce 7A. You used an 8A power supply. What happened to the last 1.3 amps? Are they just not used or the power adaptor have a way of controlling its current output?

Do not mix up constant current drivers with any discussion in this thread. This thread is about constant voltage driving. You will just confuse yourself further.

As for all constant voltage sources, they will supply a voltage and the load will draw the current it wants. Seems you have that.

Conversely a constant current source will drive the exact current it's set to and will vary the voltage to achieve that. VERY DIFFERENT.

From what I read elsewhere, it says whatever device is hook to the power adaptor, they will only draw amount of current it needs.

Yes, if the adapter is a constant voltage source like 99.9% of them are.

The lowest running current for the LEDs are 350mA (XP-G) at 3V. So I am assuming the systems knows to take .7A instead of .35mA because of the 3.2V?

Wrong. The lowest running current for LEDs is microamps. Mine run down to about 100uA(0.1mA or 0.0001A). As the voltage increases the current will go up. What exact voltage will cause what resulting current is not predictable because LEDs have manufacturing variation that is many percent different between one LED and the next. It will be what it is. When you use a constant current source these variations can be ignored because the driver doesn't care what they are. It's forcing a specific current thru the string regardless.
 
This sounds like a nice 120 Watt Driver. On a large build one could run 40 LEDs off of it at 3 Watts per LED. If your planning a big build Would definatly consider it. Only thing is what is the price for these? Sometimes you can run 3 or even 4 seperate drivers for less money than one big driver. You need to weigh in the cost difference.
These might not cost me anything as we are doing a lighting upgrade at work and need to change the drivers of the current lights to handle a lower temperature. Currently in a freezer, and won't relight @ -30.
We will have tons of new/old ones so I'm hoping to get a couple
The lights we have now that these run have 96 total leds and are super bright and are considered a Hi-Bay fixture but with some ingenuity I think these might work.

I would like to try and make something up for my refugium first but it might be more money than worth the benefits
 
These might not cost me anything as we are doing a lighting upgrade at work and need to change the drivers of the current lights to handle a lower temperature. Currently in a freezer, and won't relight @ -30.
We will have tons of new/old ones so I'm hoping to get a couple
The lights we have now that these run have 96 total leds and are super bright and are considered a Hi-Bay fixture but with some ingenuity I think these might work.

I would like to try and make something up for my refugium first but it might be more money than worth the benefits

If they are running 96 Watts on these drivers with a max rating of 120 watts comercialy it is my bet that they are running 1 Watt LED's. If you run 3 Watt they will be more than three times the brightness of the ones they are using per LED.

For a Refugium these are probably over kill. Since you realy only need about dozen 3 Watt LED's to handle a large refugium. My refugium is 100 gallons and I'm looking in the range of 16 to 24 LED's to power it up.

For price the biggest cost biggest cost itsself is the LEDs. Then the drivers fall number two and if you can get these as freebees your already saving a mint. The mounting and heat sinking can be costly but not unlkess your running finned comercial heat sinks it is only a small part of the total build cost. With these drivers you could build a nice refugium lighting system for about a 40 gallon refugium for under $50.00.
 
Thanks for the reply Keith and the clarification about the constant voltage and constant amp, it straighten alot of things out of my system. lol. I have one final question. I am using the XP-E Royal Blue LED. On the data sheet, it says the forward voltage is 3.2V at 350mA. On different threads people say I should run the LED at 70% of what max they can handle. Is there a way to calculate the forward voltage at 700mA? Or should I just run it at 350mA? I know running it at higher amp will be brighter, but is it needed?

I am running 24 Cree XP-E Royal Blue and 12 Cree XP-G R5 cool white.

I will be running the cool white at 1A.
 
peter,

If you look in the datasheet LEDs typically post a curve that shows the forward voltage (output) as the current increases. Some times this is a percentage so you have to go back to the tables to find the starting number.
 
peter,

If you look in the datasheet LEDs typically post a curve that shows the forward voltage (output) as the current increases. Some times this is a percentage so you have to go back to the tables to find the starting number.

Thanks!

I ignore the graphs because the first few was about flux and frequency, lol, i see it now. so it is about 3.4V.


But should I run it at 350mA or 700mA?
 
Thanks for the reply Keith and the clarification about the constant voltage and constant amp, it straighten alot of things out of my system. lol. I have one final question. I am using the XP-E Royal Blue LED. On the data sheet, it says the forward voltage is 3.2V at 350mA. On different threads people say I should run the LED at 70% of what max they can handle. Is there a way to calculate the forward voltage at 700mA? Or should I just run it at 350mA? I know running it at higher amp will be brighter, but is it needed?

I am running 24 Cree XP-E Royal Blue and 12 Cree XP-G R5 cool white.

I will be running the cool white at 1A.

When you run the LED's at 350 ma your actualy running tham at roughly .350 ma X 3.2 Volts or roughly 1.12 Watts. These LEDs are rated at 3 Watts so you actualy running them at at 1/3 of there potential.

For saftey sake many people run them at 70% of there maximium which comes out to 700 ma and a voltage someplace around 3.5 to 3.6 Volts. When you multiply that out your running at roughly 2.45 to 2.52 Watts which is roughly 84% of there rating. They will produce more than twice the light at 700ma than what they produced at 350 ma. Cree claims some series are safe at a full amp and will draw between 3.5 and 3.7 volts at a full amp meaning they are actualy running at 3.5 to 3.7 Watts.

When your running a voltage controled system it is lot tougher to guesstimate what voltage you need to run at. Each LED is unique and if ran a series of identical LED's on a current driven system and measure the voltage drop across them you will find that it does vary by sometimes up to 150 mv.

With a current supply once you have current set it will automaticly compensate for the internal resistance of the LED and keep it at the proper voltage. With a Voltage regulated supply you simply have to tune the voltage for the hottest LED in the circuit. With the variable ratio of Voltage to Current you can see where the vlotage would have to much finer tuned to the LED's the closer you get to its maximium to prevent a led from burn out.

Yes a voltage regulated supply is less expensive than a current supply. I worked for years in electronics and had a hard time realizing the need for a current supply untill I actualy started working with the newer brighter LED's.

Personaly I think it has a lot to do on bright your system needs to be, and how riskey you want to be to save a few dollars initialy. Unless you have a good outlet for a voltage regulated power suppies however there is no real reason they should be any less expensive than current regulated supplies. So even cost savings should be a debatable issue.
 
Thanks!

I ignore the graphs because the first few was about flux and frequency, lol, i see it now. so it is about 3.4V.


But should I run it at 350mA or 700mA?

keep in mind that the accuracy of there graphs is not guaranteed. It is realy nothing other a fair guess-amator. I'm unning a series on a 700ma current driver and out of 8 LED's the voltage drop across them ranges from 3.45 volts to 3.60 volts. With that in mind I would set a voltage driver to pump out no more than 3.4 Volts across any one LED. If I ran 3.6 volts across a LED that is now using 3.45 Volts chances are that current which is now at 700 ma may walk up to well over 1 Amp meaning I'm running a 3 Watt LED at over 3.5 Watts.
 
They will produce more than twice the light at 700ma than what they produced at 350 ma.
Can you give me a reference. I have never noticed an LED that doubled the output going from 350ma to 700ma. The XP-G is 100% at 250 ma and maybe 185% at 700ma.

I don't run constant voltage so it is hard to give a good answer on the suggested current. It really depends on how good the voltage regulation is. This depends on the supply to the power supply and the power supply it self. If the power supply was perfect than any current (not exceeding the LED) limit is good. Check the specification on the power supply it should give you the tolerance (I think it might be ripple) or link to them and I will try and read them. Then using the chart figure out the maximum voltage + ripple will give the current max you want as a safety (700ma to 1 amp probably). Then check it with a meter since you could have hot or cold LEDs.
 
In terms of determining efficiency by comparing output (intensity) to POWER it will always be a losing battle - increasing power will always decrease efficiency. Doubling the power will result in LESS than twice the intensity.

When you compare output to CURRENT, you're missing on the fact that higher current also means higher voltage - it's not a linear relationship. Doubling current MAY with some LEDs at some currents create more than twice the output, but by doubling current you are more than doubling power dissipated in the LED.

Also, the current/output graphs LED vendors supply are typically temperature corrected, i.e. they assume a constant operating temperature. When you turn up the current on an LED "in real life" without adjusting the cooling mechanisms at all, you're increasing the temperature of the LED, which means efficiency will drop faster than the current/output graph predicts.
 
When you start wrighiong in all fators your goal might be to prodice X amount of light. So let's look at from this angle.

You desire to produce 4,680 Units of Light.

If you were to run your LED's at 350 ma they each produce 100 units of light so you would need 47 LED's to give you this light. They are running at 3.1 Volts each so they are consuming a total of 148 Volts at 350 ma for 52 watts of total power. At S4.00 per LED and 4 drivers at $25.00 each you have $288 invested.

Using the Same LED's but driving them at 700 ma they produce 185 units of light. So now you need only 26 LED's. Each using 3.5 volts for a total of 91 volts at 64 Watts. You now need only three drivers at $25.00 each and total cost dropped to $179.00.

Now if you want to be brave you can runn these LED's on a 1050ma driver where they will give you 210 units of light each and you only need 23 LED's These will run at 3.6 volts so your usuing 82.8 volts and 87 watts of power. The cost now since you only need 3 drivers is $142.00

So looking at these numbers we sumerize all producing the same amount of light

350 ma cost $288 power 52 Watts
700 ma cost $179 power 64 watts
1050 ma cost $142 power 87 Watts

The real question then becomes if you save 12 Watts of power usage over what lenght of time will it take you to recover the extra $109 you spent on the build.

12 Watts of power running 12 hours a day means 144 watt hours a day. or 52 Killowatt Hours per year. At even 20 cents per KWH it would save you $10.00 per year. or about 11 years to cover the savings of the cheaper build. It you electric costs are less than that the time extends even further out.

Now for driving at 1050 ma over the 700 the difference is a little more questionable. Your saving only $37.00 on the build and going to use up 23 more watts per hour. Your annual cost would be $20 higher and after the first year you would loose the amount you saved.

If you realy want to push effeciency there are lower wattage LED's costing pennies for each. These are the ones often used in bulliten board displays. They will allow you to produce as much light with about 468 of them running a total of about 33 Watts for about $350 total build cost as well as many hours of soldering. and then the question is will the array even fit over your tank.
 
WPersonaly I think it has a lot to do on bright your system needs to be, and how riskey you want to be to save a few dollars initialy. Unless you have a good outlet for a voltage regulated power suppies however there is no real reason they should be any less expensive than current regulated supplies. So even cost savings should be a debatable issue.

Thanks for all the responses Trop! If I decide to go with constant current what would you recommend for the current 350mA or 700mA since you are in this hobby longer than me. lol
 
Thanks for all the responses Trop! If I decide to go with constant current what would you recommend for the current 350mA or 700mA since you are in this hobby longer than me. lol

With the colored LED's the best ones out there are there are Cree's XP-E which I would run on either a set 700ma Driver or a dimmable 1050ma Driver. If I used the 1050 ma driver I would set itut more a max of only 900 ma though.

For the white LED's there are so many new ones out there being released almost daily which each release giving more Lumins per watt that proior versions. I would tend to go with the XP-G that can easily handle to 1050 ma drivers.

They do have new 5 watt series but it still very limited in its selections. Once these hit the market in CFI-90 for whites and Royal Blue I think it will be whole new ballgame again. The preliminary I saw on these are also running a 750 ma with 6.5 Volts and others running at 250 ma but over 20 volts. These will change the picture quickly as with the higher voltage voltage regulation will be more important than current regulation.

Keep in mond that if your run 350 MA you will need 2 to 3 LEDs to give you the same light output as you would have gotten with even 700 ma. While LED's no longer cost $17 each still on a big build with 603 Watt leds your talking $240 as opposed to running them a 1 Watt and needing 150 of them costing $600.
 
Just ordered Rapid LED 24 kit!

Just ordered Rapid LED 24 kit!

Hey all, I'm kinda new here, haven't really posted. I've been toying with all-led for a long time. Finally pulled the trigger, should get here Friday:

"Your Order Contains:
1 x Drilled and Tapped 4.25" x 16" Heat Sink for 24 LEDs
1 x DIY 2 color dimming kit
1 x 24 LED Plug-n-Play Retrofit Kit Dimmable
(Driver Type: D (1-10V analog), Lens Color: Black, Lens Width: 60 Degree, Thermal Compound Type: Thermal Grease (5 gram syringe), LED Color Ratio: 14 Royal Blue + 10 Cool White)"

Came out to $296, would be a fair amount cheaper for non plug-n-play and non drilled/tapped heatsink or a diy. I just wanted to get it up and running asap. Added $7 priority shipping too. $296 will be well worth it I think...

Going on my 37 gal bowfront reef. My 130watt PC fixture ballast just went and it was just not enough light anyway.going to split the heat sink into two pieces to put about a 4" gap in the middle (tank has a brace there). I'll have pics posted of the project this weekend, not a lot in the tank right now, I kind of let it go for a bit and was debating selling it....but couldn't! Doing this, an auto topoff, and a carbon/gfo reactor finally. SPS here I come...
 
IMHO do not use fixed current driver, but get a dimmable driver. If you insist I would recommend the 700ma.

I think that depends a lot of your knowledge of what you want and how to get it. The dimmables are nice when you have each color on a seperate dimmable control. You can then blend the blue to white ratio to make you happy. But if you already know what you want then the fixed driver will save you money, and you simply select the LED's that will give you the balance you desire.

On my builds I like to use at least 3 drivers. This gives me a pre-dawn, post dusk look of Royal Blues, Then a Dawn to Dusk where I add in more Royal Blue, Blue and even a couple Aqua LED's. And finaly the mid day with a mixture of Whites to give me the total midday look I want.

Now if I was running the LED's on a PC control it would be a different picture. With the dimmable drivers I could step the LED's power in probably 8 steps from off to full on and still have different sets stepping seperatly. Could give you a variable look over 3 hours from nothing to deep blue to mid day with no noticable jumps.

But we are talking on budget here and PC control with in itself add a few hundred dollars to the build.
 
Hey all, I'm kinda new here, haven't really posted. I've been toying with all-led for a long time. Finally pulled the trigger, should get here Friday:

"Your Order Contains:
1 x Drilled and Tapped 4.25" x 16" Heat Sink for 24 LEDs
1 x DIY 2 color dimming kit
1 x 24 LED Plug-n-Play Retrofit Kit Dimmable
(Driver Type: D (1-10V analog), Lens Color: Black, Lens Width: 60 Degree, Thermal Compound Type: Thermal Grease (5 gram syringe), LED Color Ratio: 14 Royal Blue + 10 Cool White)"

Came out to $296, would be a fair amount cheaper for non plug-n-play and non drilled/tapped heatsink or a diy. I just wanted to get it up and running asap. Added $7 priority shipping too. $296 will be well worth it I think...

Going on my 37 gal bowfront reef. My 130watt PC fixture ballast just went and it was just not enough light anyway.going to split the heat sink into two pieces to put about a 4" gap in the middle (tank has a brace there). I'll have pics posted of the project this weekend, not a lot in the tank right now, I kind of let it go for a bit and was debating selling it....but couldn't! Doing this, an auto topoff, and a carbon/gfo reactor finally. SPS here I come...

Yes you could get a comercial built kit for close to that proce but you not be happy with the output of light compared with what your getting. With the 14 blue to 10 white ratio my suspicion is if you like your color like I do you will only be running the whites at a fraction of the power you will be running your blues. If you decide to upgrade in the future you may consider swapping some of the whites for Blues (not Royal Blues) so you keep a nice color balance when both dimmers are turned all the way up.
 
Back
Top