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

[zachts]

Sounds like somehow you sent 48V power supply current into the control channel between your Arduino and the LDDs. This fried the LDDs and likely the Arduino PWM outputs. I found that LDDs are not very reliable in general. I did not have fuses in my led strings after just hooking up my lights for the first time. They were not on for even 15 minutes. I heard a pop and whites went out. Result is LDD driver fried and took 2 Luxeons in a string of 10 with it. Had to take the light apart and sent the PCB to Steve's LEDs who promised to reflow new ones on. I heard this happening to other people on another forum as well. EVERYONE ON THIS THREAD PUT FUSES IN ASAP.

To make blanket statements like that you need to have a whole lot of them fry for no reason. One LDD going bad here and there is a fluke and should be expected with any electronic device. Are you certain all your connections were solid? A bad solder could cause them to fail. I would recheck all the connections from the psu all through that white string before firing it back up with a new LDD.

Having to use fuses with them kind of defeats the simplicity and one might as well go back to a parallel LED build at that point as the LDD is acting like the fuse and regulating current.

There are probably well over 1000 LDDs in service just from people following this tread, and the only reports of failure that was not due to user error are from your experience. So 999 out of 1000, that have no issues at all, that sounds very reliable to me.

 

[pwreef]

You dare to question me - the topic starter? The Lewis and Clark of LDD? I use my magic and make all your LDD belong to us!

 

[zachts]

I do dare to question you! As without question there could be no progress. I use my magic to take all YOUR LED's and make them mine! Keep your faulty LDDs :lolspin:

Though seriously adding fuses to the boards designed here by O2 and others would not be difficult. Necessary? We'll see, but I'd need more than a few reports of unexplained problems to warrant that effort, that's just me though, assuming the big company Meanwell does a bit of quality control on their end to minimize problems due to component failure and workmanship. :beer:

 

[O2Surplus]

I do dare to question you! As without question there could be no progress. I use my magic to take all YOUR LED's and make them mine! Keep your faulty LDDs :lolspin:

Though seriously adding fuses to the boards designed here by O2 and others would not be difficult. Necessary? We'll see, but I'd need more than a few reports of unexplained problems to warrant that effort, that's just me though, assuming the big company Meanwell does a bit of quality control on their end to minimize problems due to component failure and workmanship. :beer:

LOL _ I'm sure that there's a very, very small percentage of LDD's that fail for reasons other than user error, but the most of the time it is user error. I get PM's from time to time stating that the user's LDD's have blown and the claim is made that it happened due a design defect in one of the PcB's that I've posted here. I always request photos to asses the damage and do a bit of forensic analysis. Every time I've looked at the photos the cause was obvious, either the input wires were crossed, the bottom of the PcB had been shorted to the heat sink, OR the most prevalent- The LDD's were inserted into their holders Backwards! It really should be repeated more often- "check and double check your connections before applying power". It's also a good idea, when working with multiple LDD's, if there are any doubts concerning wiring, proper polarity ect..., only power up one LDD at a time. This way- if something goes wrong, you only smoke one! Not all of them!
Just my .02

 

[NiRa]

I think I found the problem.

Not quite the same situation as in the picture, but as an example.

In my connector is position a1 +36v en position a2 -36v.
When i connect the connector the voltage spike came and to many volt's running over position a3 and a4 next to the pins (A1/A2) +/- 36v.
In my case the connector is a little different, but the ldd''s nearby a1 and a2are damaged, the rest is still working.

So maybe it's not smart to put the + and - of the power supply in the same connector as the PWM to the LDD.
Or first connect the connector with pwm to the LED's and then connect the 36v power.

would that be an logical explanation?

 

[marspeed]

I usually connect things without the power on

 

[zachts]

LOL _ I'm sure that there's a very, very small percentage of LDD's that fail for reasons other than user error, but the most of the time it is user error. I get PM's from time to time stating that the user's LDD's have blown and the claim is made that it happened due a design defect in one of the PcB's that I've posted here. I always request photos to asses the damage and do a bit of forensic analysis. Every time I've looked at the photos the cause was obvious, either the input wires were crossed, the bottom of the PcB had been shorted to the heat sink, OR the most prevalent- The LDD's were inserted into their holders Backwards! It really should be repeated more often- "check and double check your connections before applying power". It's also a good idea, when working with multiple LDD's, if there are any doubts concerning wiring, proper polarity ect..., only power up one LDD at a time. This way- if something goes wrong, you only smoke one! Not all of them!
Just my .02

Advice worth every penny!

...and as marspeed points out never make connections when anything is powered. Power down both LED power supplies and controllers before making or removing a connection.

@ NiRa, it is generally good practice not to run power and signal wire in the same cable or connector for a lot of reasons, one of which you've experienced first hand. Another good reason not to is that you can create interference with the signal wires that can cause all sorts of bad things depending on the application.

 

[NiRa]

Thanks for all the help.

I'm glad to know what the problem has been, and that the problem was my own fault, so I can avoid it next time.

Thanks a lot.

 

[Dmantz1]

Marspeed have you received the 4up + SWC boards?

 

[marspeed]

Chinese holidays are holding shipments up I think they are over today they have been shipped

 

[O2Surplus]

Chinese holidays are holding shipments up I think they are over today they have been shipped

Marspeed- Just an FYI on the SCW PcB.

When I designed that particular board, I hadn't actually held an SCW in my hand. I designed the circuit based on what I'd read in the data sheet. The data sheet referred to the output pins by calling them Vout+ & Vout-. So based on my experience with the LDD-H, that uses the same pin naming convention, I did not tie the Vout- pin to the PcB's common ground plane.

Now that I have 10 SCW's in hand, I can see that the pin labeling on the case is different? The Vout- pin is labeled "GND" for "Ground". My thoughts are this- If your SCW's don't function properly, try connecting the Vout- pin directly to the PcB's ground plane. This mod can be made easily enough, by scratching off some of the protective coating to expose the copper, near the Vout- pin, on the underside of the PcB. Apply a little solder between the exposed copper and the pin to make a good connection, and you should be set.

 

[marspeed]

ok i am going to hook one up to a power supply and test the pins to see if i get 12 volts out the ones i have say GND too GND and Vout - should mean the same thing sort of

 

[marspeed]

ok i hooked it up to a 48 volt 7 amp power supply then tested for voltage on the gnd pin and vout+ pin i got 12.5 volts straight polarity and when i reversed the probes i get - 12.5 should work as designed

marc

 

[Dmantz1]

Chinese holidays are holding shipments up I think they are over today they have been shipped

Well at least they shipped.

 

[jedimasterben]

Good sirs of possibly the best thread ever ,

I just re-re-re-re-re-re-refinished my array and I think it's pretty final now LOL. I've got two Bridgelux BXRA arrays, each surrounded by four Luxeon M RB and four Rebel blues, with 20x clay-boa hyper violet. A total of six LDD populated on two 4up LDD boards. I also have two T5 tubes on a 0-10v dimmable HEP ballast. I'm currently using a BoostLED Typhon to control the LEDs (thinking about a Storm/X), and my Reef Angel to control the dimming of the T5 (and have it on a relay, which is an unfortunate necessity).

I would like to have everything possible loaded up into the hood and have as few wires as possible exiting. Currently, the Typhon is in the hood, so the only wires leaving are for the T5 ballast, the dimming wire going to the RA, the 48v power supply, the 12v power supply for the Typhon and a 12v power supply for the fans.

I am planning on switching to the 4up LDD + SCW boards (or, if possible, can a 3up LDD-H and single SCW board be made? pretty please? ), but would also love for the Typhon to control the T5. Is there an easy way to convert the 5v PWM signal to the analog 0-10v that the HEP ballast wants?

 

[O2Surplus]

Good sirs of possibly the best thread ever ,

I just re-re-re-re-re-re-refinished my array and I think it's pretty final now LOL. I've got two Bridgelux BXRA arrays, each surrounded by four Luxeon M RB and four Rebel blues, with 20x clay-boa hyper violet. A total of six LDD populated on two 4up LDD boards. I also have two T5 tubes on a 0-10v dimmable HEP ballast. I'm currently using a BoostLED Typhon to control the LEDs (thinking about a Storm/X), and my Reef Angel to control the dimming of the T5 (and have it on a relay, which is an unfortunate necessity).

I would like to have everything possible loaded up into the hood and have as few wires as possible exiting. Currently, the Typhon is in the hood, so the only wires leaving are for the T5 ballast, the dimming wire going to the RA, the 48v power supply, the 12v power supply for the Typhon and a 12v power supply for the fans.

I am planning on switching to the 4up LDD + SCW boards (or, if possible, can a 3up LDD-H and single SCW board be made? pretty please? ), but would also love for the Typhon to control the T5. Is there an easy way to convert the 5v PWM signal to the analog 0-10v that the HEP ballast wants?

Jedi-

Gimme a couple of days and I'll draw it up for you. ( 3up LDD-H + SCW)

As far as using the Typhon to control your T5's- It should be possible. It's already capable of producing a 10V PWM signal. So the addition of a simple "tank circuit" should be enough to smooth the signal into something resembling the analog signal that your Ballast is looking for. IIRC "OREO57" posted a few threads that were in pursuit of the same idea. If you do a search, I bet you'll find more specific information.

 

[letoan]

I have a few of O2's 5up boards with screw terminal/socket/resistor already installed if someone needs it, pm me for details.

 

[hawker1]

storm-x

storm-x

HI,
What is the real world difference between the Coralux Storm and Storm-x when it comes to actual usage? Would there be a noticeable difference in ramping the lights up and down? Or is one easier to use than the other? Thanks,
Clif

 

[nemosworld]

Storm x (16 channel dimming)
12-bit resolution to deliver 4096 levels of buttery-smooth dimming

storm is (6 channels)
8-Bit PWM (0-5V)
256 Dimming Levels

highlighted in red is the big difference in dimming, one is more precise in dimming than the other.
as a storm owner, I am ok with the 256 dimming.

as far as usage, they both operate the same I believe.

 

[007Bond]

That's good to know. Then I guess it comes down to,...do you need 16 channels, or will 6 work ok. I don't know how big of a difference there is in cost between the two.