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

[joshlawless]

No, room for the pull-down resistors has been included on every version of the board I posted. If you follow the red trace from the "PWM" pin under each LDD, you'll see it connects first to one side of the pull-down resistor (looks like two small rectangles next to each other) before connecting to the screw terminal at the edge of the board.

The ones I ordered from iTead (v1.7) were just a little different from the one I most recently posted (v1.6)--I tinkered with aligning and centering things, and got a little creative with the decorations. I've attached a copy of the BRD & SCH files in one zip (these are what you can modify in Eagle) and a copy of the Gerber files in another zip (these are what you can use to order the boards from iTead or Seeed). The pictures illustrate the layout of each side (with the ground planes hidden for clarity).

[Sacohen]

Thanks.
The parts are the same as what O2Surplus would have used on his 3UP RGB board?
Sorry if I'm asking too many questions, I have a tendency to overthink things sometimes.

[joshlawless]

The parts used to assemble the boards I posted are:

• 3.5mm pitch screw terminals. I recommend using 11 of the 2-position kind that can slide together to build up larger (e.g., 4-position) blocks. I bought these. You can, of course, purchase 5 of the 2-position type and 3 of the 4-position type if you prefer.
• 1206-style surface mount resistors (high resistance). I purchased these from Mouser.
• 12x 2.54mm pitch header pins. These come in long strips that you can break into smaller blocks. Each jumper consists of a block of two of these pins soldered to the board. Here is a strip with 40 pins on eBay.
• 6x 2.54mm jumper caps. You place these over the header pins when you want to connect the pull-down resistors. Here are some on eBay.
• 24pin wide DIP sockets (6 per board). Here are some for sale on eBay. Note that these have the nicer machined pins.
  

The only parts you _need_ are the screw terminals.

The header pins, caps, and resistors can be left off, and you can still use the board to break out the connections from the LDDs to the screw terminals. If you don't solder resistors to the board, the jumpers and caps will be unnecessary.

The sockets just allow you to removably attach the LDDs to the board (i.e., you solder the sockets to the board, and just slide the LDDs in). If you'd prefer, you can just solder the LDDs directly to the board.

[Sacohen]

Thanks Josh.
Now I have to get up the courage to do it. I haven't do a project like this since high school electronics.

[joshlawless]

If you decide to go forward with it, read back a few pages to find a couple other people looking to buy some 6-up boards. Both iTead and Seeed make you buy in multiples of 10, and shipping from Taiwan/China can get expensive if you're in any kind of hurry. Splitting costs and slow shipping can get your per-board fab cost down to around $4 or $5.

[Sacohen]

Thanks. I will.

[Sacohen]

2 Layer for the PCB is all that is needed, correct?

From What RRasco had on one of hie very early posts, the rest was...
1.6mm Thickness
HASL Surface Finish
100% E-test

[Sacohen]

Does anyone have any extra 6UP boards?
I only need 2.

[joshlawless]

Originally Posted by Sacohen

2 Layer for the PCB is all that is needed, correct?

From What RRasco had on one of hie very early posts, the rest was...
1.6mm Thickness
HASL Surface Finish
100% E-test

Correct on all counts.

[Sacohen]

Thanks. See I'm getting it.

[Meldrath]

joshlawless - Do you have a picture of a completed board? I would love to see the soldering on it if you could. Thanks!

[joshlawless]

Originally Posted by Meldrath

joshlawless - Do you have a picture of a completed board? I would love to see the soldering on it if you could. Thanks!

Not yet. I have a tracking number from iTead, but don't have an ETA on the shipment.

[Meldrath]

Cool could you post it when you get it up and running mate?

[joshlawless]

Will do!

[Eud]

I have a strange problem that I'm having trouble tracing. I have a 48V DC power supply going to the inputs of my Meanwell LDD's and also to a 48V to 12V voltage converter. I use the 12V output to power three fans through a fanspeed adjuster as well as powering two Arduino boards that supply the PWM dimming signal.

I have tied together the DC negative of the Meanwell 48V output, the DC negative of the 12V output, and, by extension, the negatives of the LDD input and Arduino input and fan negatives are tied together.

When I turn on the main power, but keep the Meanwells off (I have the pull-down resistors installed and am supplying 0 duty cycle PWM at that point), I see my red LEDs flickering very dimly. If I turn the fan speed up the flicker goes away.

I've tried measuring continuity around the board and can see that the positive output of the LDD is tied to the positive input, but the negative is not. This is consistent with what others have seen where the current and voltage is adjusted by raising the negative.

Any idea where the red flickering could be coming from? Could it be a consequence of tying the input negatives together instead of the input positives? I know someone has posted on this thread before that it works better if the positives are tied together instead of the negatives.

Any suggestions?

[n2585722]

Does that LED string have fewer LED's than your other strings?

[O2Surplus]

Originally Posted by Eud

I have a strange problem that I'm having trouble tracing. I have a 48V DC power supply going to the inputs of my Meanwell LDD's and also to a 48V to 12V voltage converter. I use the 12V output to power three fans through a fanspeed adjuster as well as powering two Arduino boards that supply the PWM dimming signal.

I have tied together the DC negative of the Meanwell 48V output, the DC negative of the 12V output, and, by extension, the negatives of the LDD input and Arduino input and fan negatives are tied together.

When I turn on the main power, but keep the Meanwells off (I have the pull-down resistors installed and am supplying 0 duty cycle PWM at that point), I see my red LEDs flickering very dimly. If I turn the fan speed up the flicker goes away.

I've tried measuring continuity around the board and can see that the positive output of the LDD is tied to the positive input, but the negative is not. This is consistent with what others have seen where the current and voltage is adjusted by raising the negative.

Any idea where the red flickering could be coming from? Could it be a consequence of tying the input negatives together instead of the input positives? I know someone has posted on this thread before that it works better if the positives are tied together instead of the negatives.

Any suggestions?

I have a suspicion that your seeing the side effect of "Inductive Kickback". It's probably happening because the fans. Try installing a small diode in reverse bias across the fan's power and ground wires. That will probably take care of the problem. If not, we'll have to dig a little deeper for a solution.

[n2585722]

On the Meanwell 48D's I only had 6 LED's and solved the dimly lit issue when off by adding a 10K ohm resistor across the 48D output. I don't know if that will help in this case but if you have fewer LED's on that string it might.

[Eud]

Originally Posted by O2Surplus

I have a suspicion that your seeing the side effect of "Inductive Kickback". It's probably happening because the fans. Try installing a small diode in reverse bias across the fan's power and ground wires. That will probably take care of the problem. If not, we'll have to dig a little deeper for a solution.

So is that two diodes? One on the power and one on the ground? Or a diode between them oriented so that when power is applied to the fan no current flows through the diode? Sorry to be dense about it. What size diode(s) should I put there?

I was surprised it could light up the reds at all because they should take at least 18V to light and I'm only supplying 12V to the fans. If it makes any difference, this is what I'm using to vary the speed on the three fans:

http://www.newegg.com/Product/Produc...82E16835118217

Originally Posted by n2585722

Does that LED string have fewer LED's than your other strings?

Yes, it does. It has 9 reds which are only 2V each rather than the 3.2V drop on the blues/whites/etc.

Originally Posted by n2585722

On the Meanwell 48D's I only had 6 LED's and solved the dimly lit issue when off by adding a 10K ohm resistor across the 48D output. I don't know if that will help in this case but if you have fewer LED's on that string it might.

Hmm, like a pull down resistor, what do you think of that one O2Surplus?

[CrazyZoa]

Cool looking controller you all have going here! I really need to rebuild my 20L's LEDs and ad a controller to them. I have a Typhon but haven't added it to my current light set up yet. Looking over this thread it seems everything is listed in one place or another for all of the components to build one of these myself. But...there is always a but....I haven't done any PCB work in 20 years and am lost at some of it now.
Now on the flip side if there is enough need or interest I can build a custom box to house one of these controllers. I am currently making housings for the Typhon and the DIM4 controllers and could do the same for these as well.

[joshlawless]

I'm very interested in getting some sort of enclosure for the board I designed, but I wouldn't go so far as to call it a controller. It's just a breakout board for the LDD drivers, to clean up the wiring (and add the feature introduced by O2Surplus, where if the control line is disconnected, the LEDs dim to 0%). I've asked Roberto at Reef Angel to put together some custom 16-channel dimming expansions for the Reef Angel controller to run my setup.

W/r/t the breakout board, you shouldn't have to read too far back to find the most recent info for the 6-up boards. Post 1726 includes, as an attachment, the zip file containing everything the PCB house will need to print you some boards. Post 1728 lists the hardware you'll need to mount to the board after printing.

[007Bond]

Hey CrazyZoa, you done anything on Jarduino controllers?

[joshlawless]

Originally Posted by aaronpentz

Ive also got a favor for the people that can develop circuit boards. I can get pretty much anything you want laser cut I work for an OEM CNC Laser Cutting Machine Manufacturer. I would be willing to trade my services and material for your design capabilities. Id like a custom board designed.

I had fun designing my 6up board, and wouldn't mind the challenge of designing another. I would also be interested in getting some sort of enclosures for the boards I designed -- just a project box appropriately sized for the board, with room for connectors in and out. Is that something you can help with?

[O2Surplus]

Originally Posted by Eud

Or a diode between them oriented so that when power is applied to the fan no current flows through the diode?

That is correct - That's what's called a "reversed biased" diode. The diode acts as a path a to ground for any induced voltage that may be remaining in the fans windings after the power is turned off.

Originally Posted by Eud

Sorry to be dense about it. What size diode(s) should I put there?

Any small diode should work. There's not enough current in the circuit to make the exact rating a worry.

[CrazyZoa]

Originally Posted by 007Bond

Hey CrazyZoa, you done anything on Jarduino controllers?

No I haven't worked on one of them yet.