Meanwell LDD driver: for those who want to dim to 0 using Arduino

[perkint]

That is during PWM mode though, correct? That is exactly how it is suppose to behave in PWM mode. I didnt see anywhere in that thread where he tested it in analog. He is just concluding that while using PWM you are not actually dimming the LED as some might hope you would be.

You wouldn't - as I said, Cornell was testing the LDD-H. Looks like they don't have the analogue dimming option...

The data sheet on Meanwells site shows the output in analog mode. It doesn't state whether the output percentage is Current or pulse. It simply shows a percentage of max.

The fact that the output is labelled as "I" would suggest it is in Amps supplied as % age of the rated current of the driver. That coupled with this comment from the previous page...
"During PWM dimming operation, the output current will change to PWM style."



Tim

 

[Krazie4Acans]

The fact that the output is labelled as "I" would suggest it is in Amps supplied as % age of the rated current of the driver. That coupled with this comment from the previous page...
"During PWM dimming operation, the output current will change to PWM style."



Tim

Good catch. I missed that. So using analog in does appear to cause a variable current out of the LDDL and not a PWM pulse out. To bad as analog to PWM would be preferred. Maybe that will be on Gen3 LDD's Krazie :jester:

 

[oreo57]

Good catch. I missed that. So using analog in does appear to cause a variable current out of the LDDL and not a PWM pulse out. To bad as analog to PWM would be preferred. Maybe that will be on Gen3 LDD's Krazie :jester:

I think your both misinterpreting that statement..W/ the LDD-h you only have either full or PWM dimmed.. which is the meaning of either full current (on) or PWM output (dimmed).
As I mentioned above I thought "full" is full.. any dimming is PWM out dimming... You can't "pulse" on or full current..

but I do see the H is different than the L
still w/ the "L" series you have 1V to 2.5V analog.. 1V has to be a "requirement" of the circuit to kick in..

w/out an O trace of the analog out we are just still guessing...but again for either the Ldd-h or LDD-l any dimmed setting has a PWM output wave form.. at least that is my bet partuclarily for any PWM dimming signal..

hmmm re-reading the spec sheets I could be wrong, but... both list I(out) as a percent (either type).. The style may refer to the shape of the graph not the output type. Do I sound confused???
The LDD-h uses the same verbiage but has no analog dimming..AFAICT

 

[Krazie4Acans]

Agreed but as I mentioned in my reply I am talking specifically about the LDD 1000L-1500L models that allow Analog dimming as well as on/off and PWM. The rest of the discussion does not apply to any other LDD models as they do not provide analog dimming option. Krazie :jester:

 

[oreo57]

Agreed but as I mentioned in my reply I am talking specifically about the LDD 1000L-1500L models that allow Analog dimming as well as on/off and PWM. The rest of the discussion does not apply to any other LDD models as they do not provide analog dimming option. Krazie :jester:

Ok and without someone actually applying a 1V to the dim circuit and scoping the output.. we will just be battling hearsay and semantics..
completely unrelated but maybe some info can be gleaned from this:
http://www.ti.com/lit/an/slua701/slua701.pdf

 

[Scuba_Steve]

I sent an inquiry to meanwell. Thank you for that suggestion Krazie.

So you say that dimming analog is not good? Dimming analog is the exact same thing as running them using a lower current driver, which to my understanding almost every LED fixture does. So if I understand you correctly, your saying to run them at full blast.

The only reason I did not choose to run them with a lower current driver such as a LDD-1200 is I really wanted more out of them. I was hoping to run them closer to 1350. I just do not want to be at the very edge of their maximum capabilities, but pretty darn close to it.

 

[Scuba_Steve]

When using the analog dimming function on the LDD 1000-1500s does the driver actually reduce the amperage provided to the LEDs in order to dim them, or does it internally apply PWM? To better explain using made up statistics, if I apply 2.2v to the dimming pin of a LDD 1500, would it then output 1300ma instead of 1500ma?

Thank you

Hi Steven,

When using the analog dimming function on the LDD 1000-1500, the led driver dim the LED by reducing the output current. The output current can be reduced by adjusting the external voltage source.

Yes, you are correct. According to the dimming chart on the datasheet, the output current will be reduced to ~85% of the rated output current when you apply 2.2V to the dimming pin.

Regards,

Mason Tang| Applications Engineer
Mean Well USA, Inc.
44030 Fremont Blvd., Fremont, CA 94538
United States

 

[gus6464]

Well I connected the DCW05C-12 to O2's converter board and while nothing blew up it also didn't work. On to do another search for these damn elusive SCW chips.

 

[O2Surplus]

Well I connected the DCW05C-12 to O2's converter board and while nothing blew up it also didn't work. On to do another search for these damn elusive SCW chips.

Did the power indicator led on the adapter light up? Did you apply a 5V or Pwm signal to the PWM input for the channel that you plugged into? I tested each one before I sent them to you and they worked perfectly for me?:uhoh2:

 

[gus6464]

Did the power indicator led on the adapter light up? Did you apply a 5V or Pwm signal to the PWM input for the channel that you plugged into? I tested each one before I sent them to you and they worked perfectly for me?:uhoh2:

Yeah the led light on the adapter lit up. I connected LDD board to the 48V power supply then a fan to the LED out of the port where the SCW was connected to and no dice. I am pretty sure it's just because the pinout of the DCW is different than the SCW.

 

[O2Surplus]

Yeah the led light on the adapter lit up. I connected LDD board to the 48V power supply then a fan to the LED out of the port where the SCW was connected to and no dice. I am pretty sure it's just because the pinout of the DCW is different than the SCW.

Ah yes, that could be it. I'd forgotten that you had DCW's instead on SCW's. Maybe grab a multimeter and see what you can figure out? Could be something simple like a floating ground connection?

 

[gus6464]

Ah yes, that could be it. I'd forgotten that you had DCW's instead on SCW's. Maybe grab a multimeter and see what you can figure out? Could be something simple like a floating ground connection?

Yeah at this point I figured it's not worth it to try and mess with it so I am just going to send the DCW back. I found a place in the US that stocks SCW08C-12 for $12 so I got some of those.

 

[Krazie4Acans]

So you say that dimming analog is not good? Dimming analog is the exact same thing as running them using a lower current driver, which to my understanding almost every LED fixture does. So if I understand you correctly, your saying to run them at full blast.

The only reason I did not choose to run them with a lower current driver such as a LDD-1200 is I really wanted more out of them. I was hoping to run them closer to 1350. I just do not want to be at the very edge of their maximum capabilities, but pretty darn close to it.

What I said is that running them at a lower current will effect the color spectrum produced. You have to look at the data sheet for the LEDs that you are using and see what the color shift is at the current you are planning to use. Then decide if that color is too far from the wavelength you want for your setup. For me it is easier and more accurate to run them at the rated current and use PWM. That way there is no guessing if the color is what you wanted or not.

Mathematically speaking, you will actually increase the life of our LEDs by using PWM dimming because they are rated for on time hours regardless of the current. With analog dimming if they are on at all they are considered on and the life counter is running. Because PWM is turning the LED on and off at a high rate of speed the life counter is only on during the on pulse. So a PWM dimmed LED running a 85% is saving ~15% on the life everyday. Analog dimmed LED running at 85% current is still burning 100% of the life hours while it is on. Krazie :jester:

 

[jedimasterben]

To anyone that is saying that PWM is the best because it doesn't cause a shift in chromaticity, I hate to break it to you, but you should really look at the datasheet charts, paying attention to chromaticity vs temperature and chromaticity vs current. If you aren't running the LEDs exactly at their test current and test temperature, then you are still shifting chromaticity, and therefore the spectral quality. Analog dimming can produce a slightly higher shift depending on the LED make and model (as they have vastly different tolerances), but again, the shift is still slight.

On page 20 of the datasheet for the Philips Rebel and Rebel ES color (DS68), you'll find the chromaticity vs current and chromaticity vs temperature. LEDs in a well-cooled array typically run around 60C. If you keep them running at 350mA test current, that's a total shift of -0.003x and 0.002y. Or, if you can keep their die temperature at 25C while running at 1A, then you'll have a shift of -0.002x and 0.001y.

In addition, these chromaticity shifts are incredibly slight. If the shift had a decimal point removed, then some people with very good vision would be able to see a very slight shift, but the average person would not be able to perceive the difference, as it would only equate to a 0.5-1nm shift.

 

[Krazie4Acans]

The temperature shift is one we all deal with regardless of dimming method. So why add the additional current shift to the equation?

Our human eyes may not be sensitive enough to see the shift but corals are much more likely to notice.

I'm not saying that PWM is the perfect solution either. However, we are comparing the two methods available on the LDD 1000L-1500L drivers that we are discussing. Of those two options PWM will give the most consistent and predictable result.

I don't see anyone mounting their LEDs on a supercooler to keep them at an absolute constant temperature to solve the temp shift. I do mount my LEDs on an over-sized heatsync (with fans)to try and keep them as cool as possible. That doesn't mean that everyone else will choose to do the same. Ecotech packs their radions onto a pretty small heatsync, compared to the wattage their lights can output, and they are able to produce a quality fixture with that setup. Krazie :jester:

 

[oreo57]

In addition, these chromaticity shifts are incredibly slight.

Light-emitting diodes (LED) demand different dimming techniques than the analog methods used for traditional lighting because the correlated color temperature (CCT) of solid-state lighting changes with voltage. Such color change can be detectable to the consumer and undermine claims made about the light quality of LEDs.

http://www.digikey.com/en/articles/techzone/2014/feb/led-color-shift-under-pwm-dimming

Commercial LED vendors such as Seoul Semiconductor, Cree, and OSRAM do not include information about how PWM duty cycle temperature alters the chromaticity of their white LEDs in datasheets. However, they do provide data on how chromaticity is affected by junction temperature. (See the TechZone article "œThermal Effects on White LED Chromaticity.")

Seems pronounced enough to concern industry...

OF course they BOTH do..

 

[perkint]

Just been looking at a couple of Cree data sheets. The effect (for example) on an XPG2 from 25 to 150°C is fairly similar to the effect between 100 to 1500mA (0.006 to 0.005 respectively).

The difference being, people do vary their dimming from 15% to 100% (maybe), but the temp ranges most people run at is probably within a much smaller range (keepig the actual LED at 25°C would probably need a room well below that or some form of cooling much more effective than a heatsink and fans - and running them at 150°C? Ow!!).

So based on that (far from conclusive bit of reading) most people would see more colour change if they had analogue dimming (from 0 to 100%) than they would due to the temp increase from PWM dimming.

Not sure how much difference 0.006 would make in real terms tho. And not sure if other LEDs have more effect (did say it was only a bit of reading ).

Tim

 

[crn005]

I know this is a little off-topic from y'alls current discussion, but I can't seem to find the answer to my questions. I need a power chord to hook up the power supply, right? I have an extra cable from a burned out MW (ELN-60-48P) of mine. I think it should work, but it only has a positive and neutral wire coming out of it. How do I hook up the ac ground? Also, I can hook up all of the drivers to just one of the positive and negative power terminals, right? I'm looking to buy a 250w 48v 5.2a dc power supply, for the records. I think I pretty much understand how to do the rest. Oh, and I plan on running 5 drivers, if that helps.

 

[jedimasterben]

I know this is a little off-topic from y'alls current discussion, but I can't seem to find the answer to my questions. I need a power chord to hook up the power supply, right? I have an extra cable from a burned out MW (ELN-60-48P) of mine. I think it should work, but it only has a positive and neutral wire coming out of it. How do I hook up the ac ground? Also, I can hook up all of the drivers to just one of the positive and negative power terminals, right? I'm looking to buy the 250w 48v 5.2a dc power supply from clay-boa, for the records. I think I pretty much understand how to do the rest.

Toss the two pin cord and use a 3-pin with ground. They're dirt cheap to buy them.

I would recommend, though, NOT to use the LED groupbuy power supplies. They're Meanwell knockoffs and pretty much just suck. I'd recommend a Meanwell HLG or CLG. They are much higher quality, much more efficient, and IP65 or IP67 depending on the model. Cost is higher, but with power delivery, you get what you pay for.

 

[crn005]

Toss the two pin cord and use a 3-pin with ground. They're dirt cheap to buy them.

I would recommend, though, NOT to use the LED groupbuy power supplies. They're Meanwell knockoffs and pretty much just suck. I'd recommend a Meanwell HLG or CLG. They are much higher quality, much more efficient, and IP65 or IP67 depending on the model. Cost is higher, but with power delivery, you get what you pay for.

Ah, nevermind. I cut the green so short that I missed it. I have a three-prong that should work :headwally:

As far as the power supply is concerned, I'll have to check those out. Was I right about how you connect the drivers to the power supply though?