Jump to content
Top Shelf Aquatics

Two Lumina 5.1 on 5 700ma drivers?


cocojakes

Recommended Posts

I've been searching for a while now for how people have done this, but most of us on nano reef don't have tanks wide enough for two luminas :P I'm working on a 40B AIO and have ordered two lumina 5.1 units to use as lighting. I currently have two pucks, each with 6RB, 6TV, 4NW, 1OCW, so the longest string is 12 LEDs. I'm using a storm controller, and 5up board of 700mA LDD drivers with 180W, 48V power supply. My question is if I run the two multi chips in parallel, will it just not work, or be dimmer than their max. I feel like because the voltage drop across two channels of lumina would be more than the 48V that the power supply can offer, it would be as though I'm voltage dimming, in addition to the PWM dimming that i'm controlling. Correct me if I'm wrong. If its not doable, what does everyone else do, two 5up boards?

 

as far as two 5up boards goes, does the signal from my storm controller go to them each in parallel, or series? Also can I have a common ground between matching pairs of drivers to reduce the number of wires going up to the pucks?

Link to comment

so whats the best way to power two 5.1s then?



I was looking at changing to the makers 5up PRO + Controller, but if I need two of them then its suddenly doubled in price

Link to comment

okay, thats just another 35 dollars, thats fine, so how do I control 5 pairs of them with my storm controller? Do i connect the PWM signals in series or parallel?

Also do you see any issue with the matching pairs (both channel 1s, both channel 2s) sharing a common ground coming back from the LEDs to the drivers?



also, since its just covering 36"x 13"x 14" of water, with 120 degree lenses, would you suggest 700 or 500 mA drivers?

Link to comment
jedimasterben

Connect the dimming circuits in parallel. The Storm can drive hundred of LDD per pin.

 

Run at least four wires to the boards from the power supply, one ground and one positive to each. I personally run all four (so two ground and two load) and then also tie the two board grounds together, then the ground from my controller.

 

Either driver is up to you. You can always dim them.

Link to comment

I was going to run the 4 wires to the boards, but I'm talking from the boards to the LEDs, can I have only 15 wires going from the boards to the LEDs, 2 pairs of positives, and 1 group of negatives, which just splits off at either end, that way I can use a single 15 pin VGA cord to connect the LEDS to the boards.

 

as far as drivers, is 500mA going to cause the same temperature rise as 700mA when they are both at the same compatible brightness? I know that I currently never run past 25% brightness with my 700mA drivers, so if 500mA will be cooler at the same brightness, it seems like a better idea?

Link to comment
jedimasterben

The output sides of the drivers must always be separate. You will need 10 wires for each Lumia.

 

Current is current, no matter how it is applied. If you run 700mA drivers at 25%, you are giving them 175mA. On a 500mA driver, you would need to run it at 35% to deliver the same current. The LEDs would use the same amount of power and have the same output.

Link to comment
jedimasterben

They are and they do, however, while the 'actual' current is 100% of the rated output pulsed for shorter or longer periods of time, the difference is zilch between that and an analog driver's output which lowers current. Both will measure the same to a multimeter, both will use the same amount of power from the wall, both will have the same light output on the LEDs. It's just arguing semantics at that point :)

Link to comment

P=IxE

 

Or

 

Watts = Amps x Voltage

 

LDDs complicate that a little because they pulse the current, just factor in the percentage like Ben explained above, and youll get your comparison.

 

(Amps (Y%)) x Voltage = Watts

Link to comment

there is a significant difference as far as LEDs are concerned, as they see it in pulses, of 100% The reason for doing this is that the spectral output of an LED will change with the amount of voltage and current being given to it. I understand the concept of averages, and that makes sense, but its more than just semantics between PWM dimming, and analog dimming.

 

Okay, so this is annoying then, I guess I need to rethink my choice of cable since I need 20 pins + 2 for fans. I thought I had PLENTY of pins with my 15pin VGA lol damn

Link to comment
jedimasterben

P=IxE

 

Or

 

Watts = Amps x Voltage

 

LDDs complicate that a little because they pulse the current, just factor in the percentage like Ben explained above, and youll get your comparison.

 

(Amps (Y%)) x Voltage = Watts

there is a significant difference as far as LEDs are concerned, as they see it in pulses, of 100% The reason for doing this is that the spectral output of an LED will change with the amount of voltage and current being given to it. I understand the concept of averages, and that makes sense, but its more than just semantics between PWM dimming, and analog dimming.

What I was really getting at was that power from the wall is identical when comparing PWM and analog, but also remember LEDs will shift their spectral output very slightly no matter how they are run, as they are only rated to output their spectra at their test current and temperature, so running a 350mA 25C chip at 1A 85C, even if you're pulsing the current to hit the output of 350mA, will have a slight shift, but to the majority of the population there will be no visible difference (hell, I'm good at spotting the difference and I can't even tell).

 

 

Okay, so this is annoying then, I guess I need to rethink my choice of cable since I need 20 pins + 2 for fans. I thought I had PLENTY of pins with my 15pin VGA lol damn

I feel ya, man. You know how hard it was to find proper 20AWG solid 8-conductor cables, of which I needed TWO for running my LEDs, and then a 7-conductor 600v 18AWG cable for my T5s? Damned if that T5 cable didn't cost $20 for 8' lol

Link to comment

so why can't I have two positive leads, going to two different strings of LEDs, and the negative leads from the two strings, then combine into a single lead, which then splits back into two leads to go back to the neg end of the drivers?



and just out of interest, what WOULD happen if I tried just using a single 700mA driver to power a string of 20 LEDs? They're only 2W each which is less than the "3W" ones we normally use, but wouldn't they just react the same way they would when using variable voltage dimming?

Link to comment
jedimasterben

so why can't I have two positive leads, going to two different strings of LEDs, and the negative leads from the two strings, then combine into a single lead, which then splits back into two leads to go back to the neg end of the drivers?

It just doesn't work that way. Driver outputs must be separate.

 

and just out of interest, what WOULD happen if I tried just using a single 700mA driver to power a string of 20 LEDs? They're only 2W each which is less than the "3W" ones we normally use, but wouldn't they just react the same way they would when using variable voltage dimming?

You'd be attempting to give 45v to 72v of LEDs. At that point I doubt they'd even strike at all, and you'll probably damage the LEDs or the driver.

 

Don't think of LEDs as voltage devices.

Link to comment

the outputs of the drivers would be separate, just the return line would be combined? I'm just trying to wrap my head around the physics of it, because I can take two 9v batteries, and have them each powering a small fan, and then have the output of the fans combine, and then split again to go back to the batteries, why can't I do the same with LEDs? I'm trying to work out, it must be something to do with the drivers?

 

and yeah, I've accepted that I need to buy more drivers, was just interested if my thoughts were on the right track



could I run the two luminas in parallel from a single 1000mA driver, resulting in two 500mA channels, that each have the full 48V?

Link to comment
jedimasterben

the outputs of the drivers would be separate, just the return line would be combined? I'm just trying to wrap my head around the physics of it, because I can take two 9v batteries, and have them each powering a small fan, and then have the output of the fans combine, and then split again to go back to the batteries, why can't I do the same with LEDs? I'm trying to work out, it must be something to do with the drivers?

Constant current circuits work differently than constant voltage.

 

could I run the two luminas in parallel from a single 1000mA driver, resulting in two 500mA channels, that each have the full 48V?

Yes, but you're introducing points of failure.

Link to comment

Constant current circuits work differently than constant voltage.

at any given point in time, the current and voltage is constant, regardless of how the driver works. Once its connected to a circuit the voltage is constant (the voltage drop of the LEDs doesn't randomly change). Shouldn't it react the same as any other DC circuit? I had to take a bunch of electrical engineering courses back in second year, but haven't used it much since then so I'm kind of rusty on all this, but I'm not understanding the physics behind why it wouldn't work? other than just, "thats not how we do it"

 

 

Yes, but you're introducing points of failure.

yeah i know, its a matter of do I trust my wiring and soldering enough to make sure I won't fry the LEDs if one of them gets disconnected and all the current goes through one string instead of two. Since I never use more than 20% anyway, which works out to something like 150mA, and never would really use more than 350mA, so running them both in parallel off 700mA drivers wouldn't be terrible either, and would have built in security that worst case scenario, they can handle the full power if something fails?

Link to comment

How many conductors is hdmi cable?

 

Thought it was like 32. Probably 32awg too.

 

If you meter for continuity off your LDD board, youll see that the positives are common. But they arent really.

 

You need a pair for each channel out, ie 10 pairs. There is no common on that side of the board.

 

Negative is common on the PWM side though.

 

ince I never use more than 20% anyway, which works out to something like 150mA

A 700LDD will out put 700mA at 10% and 100%. It pulses the output, but the current doesnt change. It doesnt output 150mA.

 

LEDs will turn off at around 200mA anyway.

Link to comment

so running a 350mA 25C chip at 1A 85C, even if you're pulsing the current to hit the output of 350mA, will have a slight shift,

I dont understand.

 

If you have a LDD1000, the LED is getting 1000mA, just pulsed, as long as the LDD is putting out anything, not off.[werent there reports that a maker was getting 540mA or something a while back?]

 

If the LDD1000 is at 35%, ie the same brightness as an LDD350 at full on, youll be using the same wattage approximately, minus any inefficiency, real world resistance, etc, as the LDD350 full on. Theoretically, it is the same power usage.

 

If your eyes can see the color shift between either, youre a wizard and magic is real.

 

 

If you had a slave that stood at your lightswitch in your room hooked to a 100 watt bulb, and flicked the switch on and off real fast, so the light was on for half the time and off for half the time, the 100watt bulb would use 50 watts of energy at the meter. You still have to feed and clothe the slave, so theres your inefficiency there.

Link to comment

How many conductors is hdmi cable?

 

Thought it was like 32. Probably 32awg too.

I feel like HDMI might be a bit of a small gauge for our uses no? Also I don't think I'd be able to solder that small of pins :P At least VGA I can see each individual pin, with enough space to solder (since its damn near impossible to find a break out box for it) HDMI would be hell

 

 

 

If you meter for continuity off your LDD board, youll see that the positives are common. But they arent really.

 

You need a pair for each channel out, ie 10 pairs. There is no common on that side of the board.

 

Negative is common on the PWM side though.

I was just trying to get a common between the negative side of two drivers that are getting the same PWM signal, and are going to identical channels on the lumina, so they should theoreticaly be doing the exact same thing, and would use 3 wires out instead of 4 (2 out 1 in) but have decided to just run the two channels in parallel off a single 700mA driver instead, since I don't need the brightness.

 

 

A 700LDD will out put 700mA at 10% and 100%. It pulses the output, but the current doesnt change. It doesnt output 150mA.

 

LEDs will turn off at around 200mA anyway.

 

Yes I understand that, but it averages out to a lower current (is the same brightness as a lower current driver at 100%) and therefore could be replaced by any driver larger than that. My logic being if I never use more than 20% brightness with 700mA, thats approximately equivalent to 100% brightness of a 150mA driver, and therefore I would be fine splitting the 700mA into two 350mA circuits without fear of not having enough power.

 

 

In conclusion I think thats my plan, to stick with a single 5up board of 700mA drivers, and just run the two luminas in parallel. Which creates a new problem, any pros and cons between using the new makers pro + controller vs storm + 5up board? (both use the same 700ma LDD drivers)

Link to comment
jedimasterben

A 700LDD will out put 700mA at 10% and 100%. It pulses the output, but the current doesnt change. It doesnt output 150mA.

The effective output current is ~150mA, though, meaning it uses the same amount of power as an analog driver outputting a constant 150mA.

 

LEDs will turn off at around 200mA anyway.

That's a limitation of analog dimmed drivers (and it's usually ~10% of rated maximum current, not a set current), not the LEDs themselves.

 

I dont understand.

 

If you have a LDD1000, the LED is getting 1000mA, just pulsed, as long as the LDD is putting out anything, not off.[werent there reports that a maker was getting 540mA or something a while back?]

 

If the LDD1000 is at 35%, ie the same brightness as an LDD350 at full on, youll be using the same wattage approximately, minus any inefficiency, real world resistance, etc, as the LDD350 full on. Theoretically, it is the same power usage.

Correct. Theoretically :)

 

If your eyes can see the color shift between either, youre a wizard and magic is real.

It's extremely slight, and I can't see it, and I can notice just a few nm difference. But, like I said, if you're not running the LEDs at their test current and temperature then they'll have a slight shift. Had brought that up on the big RC LDD thread where people were arguing the shift that analog dimming would produce (which is also basically nonexistant in real-world use). :)

 

If you had a slave that stood at your lightswitch in your room hooked to a 100 watt bulb, and flicked the switch on and off real fast, so the light was on for half the time and off for half the time, the 100watt bulb would use 50 watts of energy at the meter. You still have to feed and clothe the slave, so theres your inefficiency there.

Can you explain why my Kill-a-Watt meter still says that my LEDs (which all use PWM and never peak at 100% on any channel) still use the same amount of power (in kWh) as they would if the LEDs were not being pulsed to get the 'effective' current? Power supply is an HLG.

 

I was just trying to get a common between the negative side of two drivers that are getting the same PWM signal, and are going to identical channels on the lumina, so they should theoreticaly be doing the exact same thing, and would use 3 wires out instead of 4 (2 out 1 in) but have decided to just run the two channels in parallel off a single 700mA driver instead, since I don't need the brightness.

I'm still going to VERY highly recommend to stay away from parallel driving. Spend the $50 and get another board and five LDD - it's a very small price to pay to eliminate future issues you may have, which can cost you far more.

Link to comment

I'm still going to VERY highly recommend to stay away from parallel driving. Spend the $50 and get another board and five LDD - it's a very small price to pay to eliminate future issues you may have, which can cost you far more.

but if I'm using a 700mA driver then even if the wiring failed, the LEDs could still easily handle the power they're getting, so it wouldn't do any damage? It would just mean one or more of the strings of LEDs would be off until I noticed and fixed the issue? Doing it this way allows me to use the new makers 5up Pro instead, which I think looks much neater, and allows upgrade to 0-10v dimming if I ever decide to get a reef controller.

Link to comment

Archived

This topic is now archived and is closed to further replies.

  • Recommended Discussions

×
×
  • Create New...