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

[megadeth72]

Seeing as you will doing some testing can you also see what current the LDD dimming circuit draws.

I tested one today and I hope I have done something wrong as I got the following results:

100% PWM - 8mA
1% PWM - 52mA :eek2:

If these results are correct, and I hope they are not, then they will draw too much current for an Arduino pin and 8 will draw too much for the Arduino in total.

Please tell me I have done something wrong.

:beer:

I'm running 5 of them right now without any trouble, along with a lcd screen, relays and 3 temp probes

 

[zachts]

I'll parallel drive 3 700ma LDD's into a 100 watt led and see what happens. I figure if that works- driving a couple of 1000's together into the same load should work too. I'll give it a try and report back tomorrow. My 20 LDD 1000-H will arrive Tuesday and I'll re-run the same experiment with them just to be sure.

when your checking make sure to check the voltage and current being output by each of the 3 LDDs to see if they are sharing the load equally or if one is taking the weight of it.

Thanks!!

 

[megadeth72]

the leds draw all of the power available, so that's not an issue

 

[O2Surplus]

I did some testing this morning and came up with a few answers-

1.) You cannot parallel drive the LDD's. The drivers will not be damaged (whew!) if you try, but you're not going to get any light from an led. (Sorry TomServo ):sad1:
My best guess is- This is typical behavior for a buck regulator style of Led driver.
If you really want more than 1000ma per channel, go with a design using multiple CAT4101's, an LM3409, or a larger commercial driver rated to what you want.

2.) The current draw for the PWM pin is 5ma @ 100% on average. Because of this, I would not recommend driving more than 6 LDD's per arduino pin. 40ma is the maximum per I/O pin according to the folks at Arduino CC.

:reading:
Summary
Microcontroller ATmega328
Operating Voltage 5V
Input Voltage (recommended) 7-12V
Input Voltage (limits) 6-20V
Digital I/O Pins 14 (of which 6 provide PWM output)
Analog Input Pins 6
DC Current per I/O Pin 40 mA
DC Current for 3.3V Pin 50 mA
Flash Memory 32 KB (ATmega328) of which 0.5 KB used by bootloader
SRAM 2 KB (ATmega328)
EEPROM 1 KB (ATmega328)
Clock Speed 16 MHz

 

[007Bond]

O2, Thank you for the information. Looks like MeanWell needs a recommendation for the next gen. LDD's 12-14-16-18 and 2000 series

 

[zachts]

Darn, thanks O2.

Back to the drawing board. I may have to try the sure electronics drivers for over 1 amp. It appears they sum them to get a 1400mA version http://www.parts-express.com/pe/showdetl.cfm?partnumber=073-054
so in theory you could go higher, but maybe not using the stand offs, I doubt the traces could handle more than 1.4 amps.
I havent used this model but they seem to just be two of the 800mA versions stacked together. If that's the case you can buy the cheap 1/2 watt version and add reistors to get them to output more. (done this and it works) all the rest of the components seem the same on the board i've used they just add more resistors.......

 

[007Bond]

Zachts, I just looked at the link. They say output v o 7.3. What leds are you using

 

[O2Surplus]

Sorry for the bad news. I tried running 3 LDD-700H, with all their connections wired in parallel into a single BridgeLux BXRA-C4500 (50 watt) LED with no luck. It seems that, however the LDD is wired internally, such drive conditions cause the LDD to go into protection mode. This situation sucks! I was hoping that this little driver would be more flexible, and work in such situations, but I guess we should be happy that can already do so much to begin with.

 

[zachts]

Looking at thier info I believe that is the "typical" output voltage with one 10watt multichip hooked up. I don't know if hooking the two boards together makes a difference in minimum voltage out but all the other versions can run 1 led the data sheet lists different minimums but since they are the same boards and components I think they measured with an LED hooked up

I doubt they new what XMLs were when they wrote the specs on those (3.6v at 3000mA)

they work just like you'd expect.

 

[tomservo]

I did some testing this morning and came up with a few answers-

1.) You cannot parallel drive the LDD's. The drivers will not be damaged (whew!) if you try, but you're not going to get any light from an led. (Sorry TomServo ):sad1:
My best guess is- This is typical behavior for a buck regulator style of Led driver.
If you really want more than 1000ma per channel, go with a design using multiple CAT4101's, an LM3409, or a larger commercial driver rated to what you want.

O2, what if we put diodes on each LDD output? Although I realize it may need a capacitor as well to stabilize the voltage.

 

[O2Surplus]

O2, what if we put diodes on each LDD output? Although I realize it may need a capacitor as well to stabilize the voltage.

We're already putting diodes across the output. LOL I think the design of the driver is what's thwarting us. The Vout- seems to be important, as it's isolated from VCC-, but without an actual schematic for the innards of an LDD, we're all just stumbling around in the dark.

 

[tomservo]

What I'm thinking is that the outputs + are causing the short circuit protection to kick in, and if we isolate the outputs from each other, it may work. I guess I'll give it a go when I've got some to hand. I only really need the extra output on the blue channels, so it would be pretty reasonable to put some diodes on two of the channels.

 

[chema]

2.) The current draw for the PWM pin is 5ma @ 100% on average. Because of this, I would not recommend driving more than 6 LDD's per arduino pin. 40ma is the maximum per I/O pin according to the folks at Arduino CC.

Are you sure about those measurements? I tried to run a LED fixture with 11 LDDs controlled by Arduino UNO and it didn't work (actually I fried the Arduino) I had up to 4 LDDs connected to the same pin.

 

[O2Surplus]

What I'm thinking is that the outputs + are causing the short circuit protection to kick in, and if we isolate the outputs from each other, it may work. I guess I'll give it a go when I've got some to hand. I only really need the extra output on the blue channels, so it would be pretty reasonable to put some diodes on two of the channels.

Hold the phone folks! I just tried what TomServo suggested and it works! I just tried the same experiment again, but this time with forward biased (1 amp) diodes on the positive and negative outputs. The way I see it- the LDD will merely raise it's output voltage to cover the extra Vf caused by the additional diodes. As long as the diodes are rated for the drive current produced by the LDD, there shouldn't be a problem. I'll continue experimenting with the diodes and post more info later on.

 

[O2Surplus]

Are you sure about those measurements? I tried to run a LED fixture with 11 LDDs controlled by Arduino UNO and it didn't work (actually I fried the Arduino) I had up to 4 LDDs connected to the same pin.

I tested all 9 of the LDD's that I have on hand. A 5ma draw was the average between all of the units. I find it interesting that the LDD has such a low input impedance, as I've driven up to 20 CAT4101's from a single arduino pin. My suggestion still stands- when in doubt- use an NPN transistor to do the heavy lifting and give the arduino a break.

 

[O2Surplus]

Here's a schematic that shows how I got parallel output from a pair of LDD's. The strange thing is- Both LDD's operate as a single driver with respect to PWM signals. I was expecting to be able to raise and lower the output level of each LDD independently by using two separate PWM signals, but I found that whichever of the pair gets a signal first, it acts as a master to the other LDD. The slave LDD contributes to an equal share of the current load, regardless of it's own PWM value, Go figure?

 

[MagicJ]

I did some testing this morning and came up with a few answers-


2.) The current draw for the PWM pin is 5ma @ 100% on average. Because of this, I would not recommend driving more than 6 LDD's per arduino pin. 40ma is the maximum per I/O pin according to the folks at Arduino CC.

That is generally in line with my measurement, but what did you measure at low PWM levels e.g. 10% - I don't know enough about this stuff to understand why, but I found that the current draw increased as the PWM level reduced.

Looking forward your results .

:beer:

 

[tomservo]

Here's a schematic that shows how I got parallel output from a pair of LDD's. The strange thing is- Both LDD's operate as a single driver with respect to PWM signals. I was expecting to be able to raise and lower the output level of each LDD independently by using two separate PWM signals, but I found that whichever of the pair gets a signal first, it acts as a master to the other LDD. The slave LDD contributes to an equal share of the current load, regardless of it's own PWM value, Go figure?

Sweet! I was hoping this would isolate them sufficiently. I was just thinking that if the voltage couldn't get from one to the other, it should work.
That's weird, but it's probably better that they work together than otherwise. Did you by chance try it with diodes only on the + or - side?

Even if I can't modify the board sufficiently to stick some SMD diodes on, I bet I can finagle something. The white channels can just be single, 1A drivers since the white is overpowering anyways, and the actinic shouldn't really be run past 1A, so only two channels really need to be dual driver. Going to update the schematic soon anyways so it's done if I ever need to make more boards.

Thanks for trying!

 

[Ian108]

Hi guys very interesting thread, I am, currently building my rig using multiple ldd drivers for a 180g controlled by an audino mega, and I have heard that as I will be using multiple ldd drivers about 5 to 6 controlled from a single pwm output, I should put a resistor or something on the pwm output to protect the ardunio, not the greatest with electronics what would be the best way for me to do this? Cheers

 

[gilles0181]

Hello guys,

Here is a short question from a Dutch Profilux user.

I have 2x Profilux LedControl4 Active. More info on this forum in this link.

LEDControl4 active converts 4 analog input voltages into 4 pulse width modulated (PWM) signals (Open-Drain) and includes drivers for connecting LEDs (max. 3A in sum). In order to protect a connected ProfiLux against possible interferences the input circuits and the output circuits are galvanic insulated through optocouplers. In the case of 0% PWM, the transistors are insulating (output is highly resistive), in case of 100% PWM, the output is permanently pulled to minus potential.

LEDControl4 active can produce "žflashes"œ during a thunderstorm simulation in ProfiLux. During a flash, the PWM-signal is temporarily set to 50%, 75% or 100%, independently from the corresponding analog input. Flashes can be output at channel 1 and channel 2, the flashing behavior can be set

This means i should have a total of 8 PWM channels. Now i would like to use the LDD-700H drivers for each channel but i am not even sure this is possible.

Question 1) Can anybody tell me if this is possible at all?

If i am correct, the PSU is not switched off by default if we send a 0% dimming signal. Is this a simple fix to implement?

Question 2) I do see a "REMOTE ON/OFF" feature on the driver, but how would i implement this?