PDA

View Full Version : LDD-H driver source voltage


HorseoftheSea
09/07/2017, 08:33 PM
I'm putting together a light for my reef and I want to confirm wether or not my assumptions, about which power supply to use, are correct. I'm using LDD-H dc-dc drivers for powering my leds which are in 4 channels, approx 12v required for each channel(varying current for each). The manufacturer recommends a 48v power supply for powering this array, but I believe that's so that multiple arrays can run on a single psu which I only am going to be running a single one. If my thought process is correct, then there should not be any problem running this same setup off of 24v rather than 48v as long as my wattage input is equal. Is that correct? My led array as a max wattage rating of ~80w, but they won't be run that high as I'm shooting for ~65w. In this situation there shouldn't be any difference wether my psu is 24v 6.25a or 48v 3a as they are close enough in total power. Am I totally wrong or am I on the right track to move forward? I want to make sure before I order my power supply.

Legot
09/08/2017, 01:57 AM
I'm using LDD-H dc-dc drivers for powering my leds which are in 4 channels, approx 12v required for each channel(varying current for each). The manufacturer recommends a 48v power supply for powering this array, but I believe that's so that multiple arrays can run on a single psu which I only am going to be running a single one.

You're right in that it will work, but you'll lose a bit more efficiency using 24v rather than 48v, that's just a characteristic of higher currents.

They recommend 48v for a few reasons, one being the efficiency benefit (less wasted power, therefore less heat), but also because it's a solid source voltage for expansion of the array. It doesn't have anything to do with running multiple arrays, as they would be in parallel with eachother.

mcgyvr
09/08/2017, 06:00 AM
You're right in that it will work, but you'll lose a bit more efficiency using 24v rather than 48v, that's just a characteristic of higher currents.
.

Not really true (in this case).. See datasheet

It would be less efficient with a 48v supply

oreo57
09/08/2017, 09:05 AM
May be fairly safe to say that "typically" if your output Amps are about at the 50% point of your ps's output rating you are close to max efficiency..regardless of voltage though voltage differential does factor in a bit..

Each ps is different though and in many cases the effeciency decrease is minimal anyways..

IF such close efficiency is needed one would need to know 1)exact draw of the leds
2)chart of the ps calculated efficiency curve..


It's really not something one can really guess at..



You're going to find lots of variation in power supply efficiency as a function of both input voltage and output loading.

If the manufacturer only specs efficiency at one input/output condition, don't expect that number to hold anywhere else.

There are some newer directives out there such as 80 PLUS which, if a product is compliant, provide a standardized measurement methodology as well as efficiency minimums for various loads. 80 PLUS Gold for server power supplies guarantees 88% efficiency at 20% load, 92% efficiency at 50% load, and 88% efficiency at 100% load at 230V / 60Hz input.

http://www.overclock.net/t/711542/on-efficiency

perkint
09/08/2017, 01:48 PM
As mcgyvr suggests, 48V supply to a 12V LED string is not the most efficient. The LDDs are at their most efficient when given just enough voltage. So if running a 12V string, 15V would be most efficient. But, to be honest, I wouldn't worry about the difference. If I was confident I wasn't going to add more arrays I'd get the lower voltage unless I already had the PSU.

Tim

karimwassef
09/08/2017, 02:10 PM
Remember that there's an efficiency loss in converting AC input to 48V or 12V DC.

The AC-48V would be more efficient than AC-12V.

The same goes for the connective structure. 48V distribution losses (lower current) are much lower compared to 12V.

I've designed many power solutions and the general rule is that if you focus on just one part, the overall will be worse.

Price varies quite a bit too. 12V is cheapest due to abundant supplies, but you can DIY 48V solutions using cheaper 12V sources.

My advice is to lay it all out from AC to your load "LED strings" and work out the efficiency or power loss and decide what works best for your system.

In general... higher DC voltages add the most value where the power level is high and/or the distribution distances are long. So above 1000W, I would use 48V. 500-1000W, I would use 36V, 250-500W ~ 24V and <250W ~ 12V. This is just my yardstick - that's all.

I broke my rules once and designed 500W at 12V to get to the cheapest solution.... yikes. Use solid copper shafts and still very hot.

jayball
09/08/2017, 02:16 PM
In this situation there shouldn't be any difference wether my psu is 24v 6.25a or 48v 3a as they are close enough in total power. Am I totally wrong or am I on the right track to move forward?

You can not trade volts for amps. You have a constant current power supply so treat them as two seprate things.

Voltage must be the combined voltage of all led's plus 3 volts (for the highest rated string)

for current you add up the rated current for the LDD's you are using
(example for 2 1000MA, one 750MA and one 500MA ldd you will have exceeded the current capacity of your 48V PS above but you will squeak through with 4 750 MA LDD's (in this example I would be sure to not run them at more than 80% overall as you want some headroom on your PS))

Your rated wattage of the PS is not relevant, treat the voltage requirements and current requirements as two separate entities.

karimwassef
09/08/2017, 02:17 PM
One more thing... the bigger the "step" in voltage, the less efficient that stage will be

So AC (115VAC) to 48VDC is a smaller step than AC-12V ... that's why it's more efficient.
Same goes for 48-12 being a large step compared to 24-12, so the bigger step supply is less efficient.

If you look at the big picture, you're starting at AC and going down to your string voltage... overall, all steps included, the efficiency will vary most by how much power is getting distributed for what distance. But what usually matters is heat- you want to make sure that you're using the devices around 50% of their rating. That goes for all the blocks.

oreo57
09/08/2017, 04:45 PM
So AC (115VAC) to 48VDC is a smaller step than AC-12V ... that's why it's more efficient.


Well considering the "choices" are 24 vs 48V ...
Anyways w/ a "good" power supply and an accurate power efficiency curve isn't AC "losses" already factored in?

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=10&ved=0ahUKEwjNz437xZbWAhUL4YMKHerdBoAQFghfMAk&url=https%3A%2F%2Fuk.tdk-lambda.com%2FKB%2FHow-to-Accurately-Measure-Power-Supply-Efficiency.pdf&usg=AFQjCNFmP3hnF1_KtFZLEZ7Kzv7Fpf99zA

;)
The takeaway is, purchase power supplies from a reputable power supply company that employs conservative component deratings and states realistic efficiency ratings.

http://power-topics.blogspot.com/2011/06/power-supply-losses-and-impact-of.html

W/ switching ps really don't see voltage differentials as being all that crucial.. From my naive standpoint mind you..

conversly, and possibly out of context but oddly the higher the AC voltage the more efficient the ps
http://www.shindengenelectric.com.cn/img/product/new_product/hsa2.jpg
http://www.shindengenelectric.com.cn/product_e/new_product/top_topics/2012405.html

karimwassef
09/08/2017, 06:12 PM
The point I'm making is that an AC-48V supply is more efficient than an AC-24V which is more efficient than an AC-12V. The bigger the step from input to output, the lower the efficiency will be.

This is because a switching supply basically turns on and off at a duty cycle that converts the input voltage to the output voltage. An intermediate transformer is also used to affect voltage change (at the switching frequency), and an input PFC and/or rectification bridge convert the AC to DC at the input.

The greater the voltage difference, the more a supply has to be "off" vs. "on". When a supply is "off", the energy doesn't pass through the supply, it's basically circulating inside it. This is a gross oversimplification, but it gets to the point that bigger input to output voltage difference = lower duty cycle = lower efficiency.

Also, the easiest way to get to the total efficiency for a straight shot powertrain is to multiply the efficiencies. So...

if the AC-48 is 89%, the 48-12 is 93%, then the total efficiency is 89% x 93% ~ 83%
if the AC-24 is 86%, the 24-12 is 95%, then the total efficiency is 86% x 95% ~ 82%

lingwendil
09/08/2017, 06:54 PM
For these smaller arrays I really like 19 volt laptop power supplies for a DC source, easy to find in 2.1~7A varieties, And a great way to utilise the LDD-L series, at their much lower cost ($3.49 each at LEDsupply for the 700 or lower) it ends up pretty nice. Just cut the end off of the Laptop PSU, and direct wire it to the fixture, or install the appropriate barrel connector for your project (2.1x5.5mm are cheap and plentiful) to make it easy. I also have a pretty plentiful supply of them for practically free (I get them free at work) so it works out well.

The less voltage you need to drop the better, so LED string voltage plus the 3 volt drop of the LDD, plus maybe a few volts drop headroom for future expansion is my way of doing things usually.