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

[lomo]

Hi all I'm working on PCB board with 8 LDD-H drivers for full spectrum LED fixture. I would like to use RJ-45 connectors and Cat5 cable for dimming. I took RJ-45 connector from Adafruit library and I'm confuesd about pins used.



So EAGLE used pins marked with red on board, what about two pins marked with blue?! I tought that they will be used instead :S

Is this ok?

 

[rrasco]

Hi all I'm working on PCB board with 8 LDD-H drivers for full spectrum LED fixture. I would like to use RJ-45 connectors and Cat5 cable for dimming. I took RJ-45 connector from Adafruit library and I'm confuesd about pins used.



So EAGLE used pins marked with red on board, what about two pins marked with blue?! I tought that they will be used instead :S

Is this ok?

Refer to the datasheet for the part you are using. They are not all spaced the same either, I had to make my own part for the RJ45s I used on a PCB.

It looks like the pins in red are the shielding and the blue ones are not connected to anything. I don't see pins 5 or 6 on the schematic. You don't 'have' to use all the pins, if you have enough signals for what your are trying to do, it will work. I try not to leave disconnected pins on connections though, for possible future use.

 

[lomo]

I would like to use 8 individual channels for dimming (I can go with 6, but why not 8 when I have arduino mega) and now those pins are confusing me :S

 

[007Bond]

Rick, that is my intention, I did want to try some things once I have some LDDs to play with, like using only one diode per LDD to isolate, so I have not actually updated the drawing yet. It should only take a few minutes to do it, though, if you want the files with the diodes I can probably get around to it today.

Tom, if you have time yes, but don't go out of your way. I ordered my boards from ITead, but did not include the gerber files correctly, so Sunny is holding my order until I get back to her. Kind of worked out well because you and O2 worked out the doubling up thing.
I don't expect the my LDD's for a while, but with your solution I'm open to a few different combinations of LDD's to get towards 1400ma. Thanks---Rick

 

[Neoferdina]

Alrighty, I just want to add an interesting data point to this discussion on these drivers work...

I originally set up an NPN transistor to turn on some fans when the Vout+ pin goes to 15V or more (basically fans are on when LEDs are ON and off when LEDs are OFF). Very simple. Except, this doesn't work! Even with the driver dimmed off, the fans reamin on. Why?

So I started digging a bit deeper into the circuit and poking around with my multi and here's what I found:

LDD drivers don't work as you might expect. You would expect the Vout- pin to be equal the ground and the Vout+ pin drives to a positive voltage to drive the circuit. In this case, when the LEDs are dimmed off, Vout+ = 0V = Vout-, and when it's put to 100% Vout+ = +ve voltage and Vout- = 0V (ground). However...

They work the opposite. Vout+ always equals 24V (or whatever you input voltage is) and Vout- changes. At 0% dimmer, Vout- = Vout+ = 24V. At 100% output, Vout+ = 24V and Vout- drops in potential in order to achieve the specified drive current (in my case, 4.7V versus grnd, which gives 19.3V to drive the LEDs).

Interesting.

 

[Eud]

However...

They work the opposite. Vout+ always equals 24V (or whatever you input voltage is) and Vout- changes. At 0% dimmer, Vout- = Vout+ = 24V. At 100% output, Vout+ = 24V and Vout- drops in potential in order to achieve the specified drive current (in my case, 4.7V versus grnd, which gives 19.3V to drive the LEDs).

Interesting.

Good to know. Don't tie the Vout- to ground.

Can you run your transistor between Vout+ and Vout- instead of between Vout+ and ground?

So does this mean that they consume a bit of power at 0% dim since both of the pins have the voltage on them? Seems like maybe they're just passing along the voltage from the supply on both pins, so it should be similar to just having the supply plugged in and nothing connected to it.

 

[tomservo]

Neo: That doesn't bode well for my PCB, as I shared the ground between two channels... Looks like I'm going to have to do lots of testing on these. It sounds like the isolating diodes may only be needed on the vout- though. I expect I'll have the dimming tied together problem with those channels, but it's not a big deal for me as I can just pair those 2 channels with the white emitters.

 

[Eud]

Neo: That doesn't bode well for my PCB, as I shared the ground between two channels... Looks like I'm going to have to do lots of testing on these. It sounds like the isolating diodes may only be needed on the vout- though. I expect I'll have the dimming tied together problem with those channels, but it's not a big deal for me as I can just pair those 2 channels with the white emitters.

Is that why O2Surplus had to put diodes on both the + and the - when running them in parallel? I hadn't been able to picture why that was necessary.

 

[tomservo]

I believe he did that b/c it's the more certain way to produce the desired effect - I have a feeling that putting the diode on only the vout+ side would not work based on the way the ground floats to control voltage/current.

 

[O2Surplus]

I believe he did that b/c it's the more certain way to produce the desired effect - I have a feeling that putting the diode on only the vout+ side would not work based on the way the ground floats to control voltage/current.

Gimme a minute -and I'll see if the diodes can be omitted from either side of the circuit. The way I see it if the diodes do need to be on both sides, we can always wire in an Led ( XR-E,XP-G ect...) in place of the schottky, or other conventional rectifier diode. That way no permanent mods need to be made to the Pcb, just a little "creative" wiring on the led side of things to achieve parallel output.

 

[O2Surplus]

These drivers do not appear to the rules of physics? I removed the diodes from the Vout+ side of the pair of LDDs and they continued to work in a parallel fashion. But some "weird numbers" showed up when I did some voltage and current checks. Most of the voltage measurements were taken in relation to ground (VIN-)

VIN = 27.3V
LED Vf = 24.4

Both LDD-700H wired in parallel with forward biased diodes on Vout- only and both receiving the same 5V pwm signal.
LDD#1
Vout+ = VIN (27.3V)
Vout- = 2.1V (27.3V - 24.4V - .8V)

LDD#2
Vout+ = VIN (27.3V)
Vout- = VIN (27.3V) ??? What the Heck? Shouldn't it match the first?

Current through the led array measured 1.3amps, which is what was expected. Just for fun I removed the Vout- connection for LDD#2 to see if it would halve the current to 650ma. Nope! Current remained at 1.3amps.
This proves one thing. Current is provided and regulated from the Vout+ side of the LDD. It is not "sunk" through the Vout- connection. When wired in this condition, it appears that LDD#1 began to run hotter than LDD#2? I measured a 20 degree difference between the Two. After reconnecting the Vout- to LDD#2, the temperature began to lower and normalize between the two LDD's.
I'm afraid that-more experimenting that needs to be done, before we can be sure that LDD's can be parallel driven reliably for extended periods of time.

 

[gilles0181]

@O2SurPlus;
Does this new found knowledge has any effect on your PCB files for the 5Up LDD you posted earlier? This is 10x10cm right?

@Rrasco:
Your 4Up LDD-H version is this still the latest file and this is 5x5cm, right?

 

[O2Surplus]

@O2SurPlus;
Does this new found knowledge has any effect on your PCB files for the 5Up LDD you posted earlier? This is 10x10cm right?

@Rrasco:
Your 4Up LDD-H version is this still the latest file and this is 5x5cm, right?

Gilles-

This information is really aimed towards Tomservo, as he's trying to produce a PCB for Parallel driving LDD's. My 5 position Pcb (10cm x 10cm) is not affected by this info, as it was never intended to be set up for parallel drive.
I believe RRasco's 4 position Pcb has to be ordered using the 5cm X10cm option at Itead.

 

[gilles0181]

I hope RRasco can confirm this

 

[Eud]

Anyone order using the cheaper shipping at itead? How long did it take for the boards to arrive? I got the email 1/31 that it was shipped, but so far the tracking still says it's still being processed by HK post to leave Hong Kong. Hmm.

 

[O2Surplus]

Anyone order using the cheaper shipping at itead? How long did it take for the boards to arrive? I got the email 1/31 that it was shipped, but so far the tracking still says it's still being processed by HK post to leave Hong Kong. Hmm.

Ugh, you better have a comfortable seat cuz you're gonna be waiting a while. In my experience, shipping via "the slow boat" takes between 2 and 4 weeks to arrive. That's why it's best to just double down on the $$ and go with expedited shipping. LOL

 

[Kawi9_cf]

Don't worry, the chip holders are under there, along with the 1206 package resistors. It all fits fine.


Here's the board with the LDD's removed.

Can somebody give me a link for the 24 pin ic socket adapters. I ordered some and they are too narrow for my ldd-h drivers. I want to make sure I get the right ones this time.

 

[zachts]

These drivers do not appear to the rules of physics? I removed the diodes from the Vout+ side of the pair of LDDs and they continued to work in a parallel fashion. But some "weird numbers" showed up when I did some voltage and current checks. Most of the voltage measurements were taken in relation to ground (VIN-)

VIN = 27.3V
LED Vf = 24.4

Both LDD-700H wired in parallel with forward biased diodes on Vout- only and both receiving the same 5V pwm signal.
LDD#1
Vout+ = VIN (27.3V)
Vout- = 2.1V (27.3V - 24.4V - .8V)

LDD#2
Vout+ = VIN (27.3V)
Vout- = VIN (27.3V) ??? What the Heck? Shouldn't it match the first?

Current through the led array measured 1.3amps, which is what was expected. Just for fun I removed the Vout- connection for LDD#2 to see if it would halve the current to 650ma. Nope! Current remained at 1.3amps.
This proves one thing. Current is provided and regulated from the Vout+ side of the LDD. It is not "sunk" through the Vout- connection. When wired in this condition, it appears that LDD#1 began to run hotter than LDD#2? I measured a 20 degree difference between the Two. After reconnecting the Vout- to LDD#2, the temperature began to lower and normalize between the two LDD's.
I'm afraid that-more experimenting that needs to be done, before we can be sure that LDD's can be parallel driven reliably for extended periods of time.

verry interesting results......... It looks like one of the LDDs is working harder than the other regulating voltage and the other one is doing nothing in this case? (possibly all return current is routhing only thru LDD #1's Vout-)??

When measuing voltage at the LDD between Vout+ and Vout- do you get the same readings?

I wonder if they would work better if the input voltage side was isolated with diodes, I would think this already would be the case but maybe not. (this is the case on the sure electronics drivers I mentioned earlier, their configuration uses 3 diodes in total, protecting both input and output) I think it needs to be determmined how to get the LDDs to more equitably share the Load before using them long term. It may not be a big deal but if they are only rated for 1 amp and one of them is routing more than that through its Vout- then there could be long term issues with them burning out from over current??

 

[O2Surplus]

verry interesting results......... It looks like one of the LDDs is working harder than the other regulating voltage and the other one is doing nothing in this case? (possibly all return current is routhing only thru LDD #1's Vout-)??

When measuing voltage at the LDD between Vout+ and Vout- do you get the same readings?

I wonder if they would work better if the input voltage side was isolated with diodes, I would think this already would be the case but maybe not. (this is the case on the sure electronics drivers I mentioned earlier, their configuration uses 3 diodes in total, protecting both input and output) I think it needs to be determmined how to get the LDDs to more equitably share the Load before using them long term. It may not be a big deal but if they are only rated for 1 amp and one of them is routing more than that through its Vout- then there could be long term issues with them burning out from over current??

See, That's the "weird" thing that I mentioned. The voltage readings taken between Vout+ and Vout- appear normal at LDD #1. The readings show the Vf of the led and the Isolation diode, which is what one could expect. On the other hand, voltage readings between Vout+ and Vout- of LDD #2 show 0V. In my understanding of electricity, this means that there is no current flowing through the Vout- leg of the circuit, and actual current tests proved that.
I think, since we don't know what the exact internal "wiring" of an LDD is, that the second LDD "influences" the first, but it does not actually carry the additional load. At least not with the isolation diodes removed from the + outputs. This would explain the difference in operating temperature between the two LDD's. I know, from looking at XRay images of an LDD, that there's an inductor coil and probably a Schottky diode inside them. I wonder if the addition of the second LDD's inductor into the circuit, causes the current set point of the first LDD to be thrown off, causing the first LDD to double it's current output?

 

[sfsuphysics]

Another suggestion... test the current running from your LED to the Vout- of the one that is showing a dropped voltage, maybe all of the current is going out through one driver? Since they are wired in parallel, this little electrons might not know where to go

LDD#2 might be 27.3V out+ and 27.3V out- but the difference between out+#2 and out-#1 is still in the ballpark of physics to get you 1300mA from two drivers.