Need help with my LED build… meal well ELN 06-48 setup PLEASE

[fatoldsun]

I have 3 mean well ELN 60-48Ps.
http://i1231.photobucket.com/albums/ee503/dzipin/IMG_2042_zpsc391cd89.jpg
Plan is to run 3 sets of parallel strings as follows (plan includes: 1.0 OHM 1% 5W resistors that I will use for each string along with 1A fast blow fuses to protect things with the parallel setup)
12x2 royal blues
10x2 colors (4 blue, 2 green, 2 red, 2 UV)
10x2 mixed whites (4 natural, 6 cool)
I’m ready to set up my drivers and I’m a little unsure of the setup with those. I want to adjust the mean well’s outputs to safely operate the LEDs. At first look there appears to be two internal pots – a smaller pot under a metal shield and a slightly larger white pot in the upper right hand corner. Questions are what should I be looking for on my multi-tester?
You can see both here
http://i1231.photobucket.com/albums/ee503/dzipin/IMG_2043_zps9f0f8bcf.jpg
Here’s my cheap multi tester – it’s served me well for the better part of 15 years
http://i1231.photobucket.com/albums/ee503/dzipin/IMG_2041_zps075568be.jpg

I have 1.0 OHM 1% 5W resistors that I will use for each string along with 1A fast blow fuses to protect things with the parallel setup

LEDs are all wired/soldered and tested for good solder joints and shorts, etc…
http://i1231.photobucket.com/albums/ee503/dzipin/IMG_2039_zpseb93cc1f.jpg

...and Happy New Year!!

[Gorgok]

Oh man, i would go crazy just knowing those wires were in something i built... But it all looks fine really.

Really you shouldn't need to adjust the pots, the one under the metal is for voltage, which is self adjusting (being constant current drivers) but that can raise/lower the high end. Only reasons to do that really would be if you are just a little over 48V on the string or you know you never will need more than say 43V (and if you do something has gone wrong). Making more voltage will mean more wattage, depending on how much more it may lead to thermal shutdown.

The other is for current, which is capable of adjusting it -25% +3%. But these are dimming drivers, so you can adjust it down with PWM. Adjusting it up will again mean more wattage, and again may cause thermal shutdown. Adjusting it down would make sense if you know the LEDs can't handle 1.3A (or half of that on a parallel setup) ever and want to be able to PWM up to 100%. Taking it down to the min of ~0.975A would mean you can run PWM to any position with a 1A max LED without issues. Not much help for a parallel setup as you will most likely be running more current down the strings than either one can handle alone, or you will be underdriving the strings quite a bit.

[fatoldsun]

Oh man, i would go crazy just knowing those wires were in something i built... But it all looks fine really.

Really you shouldn't need to adjust the pots, the one under the metal is for voltage, which is self adjusting (being constant current drivers) but that can raise/lower the high end. Only reasons to do that really would be if you are just a little over 48V on the string or you know you never will need more than say 43V (and if you do something has gone wrong). Making more voltage will mean more wattage, depending on how much more it may lead to thermal shutdown.

The other is for current, which is capable of adjusting it -25% +3%. But these are dimming drivers, so you can adjust it down with PWM. Adjusting it up will again mean more wattage, and again may cause thermal shutdown. Adjusting it down would make sense if you know the LEDs can't handle 1.3A (or half of that on a parallel setup) ever and want to be able to PWM up to 100%. Taking it down to the min of ~0.975A would mean you can run PWM to any position with a 1A max LED without issues. Not much help for a parallel setup as you will most likely be running more current down the strings than either one can handle alone, or you will be underdriving the strings quite a bit.

Thanks..
I have a fourth mean well so I may run the whites separately. I thought 650 mA would be enough for my blues since the consensus seems to be to limit them below their 1 A max at say 750 mA. I also have read that most mean well run well over 1.3, closer to 1.6-1.7. If I test mine and find 2 stronger ones I'd like to try it. Just not sure how to run those tests...

[Gorgok]

If you wire up both strings to one driver, then test the voltage across both testing resistors, add up the number and you get the total amps the driver is putting out. The reason for those resistors is that you get a 1:1 reading of voltage to amps, which means easy testing.

With the meter attached to the resistor you could run the pot up and down to limit the max point to whatever you wish. Though if you run a single string it would probably be smart to adjust the pot to as low as possible before you turn it on.

In all cases you need to have a PWM signal to the driver for it to operate though.

[fatoldsun]

Ok, so I can't test the raw wires coming off the driver? Guess they need the resistance of the LEDs to get current. I wasn't thinking about ohms law...
I have a rapid PVM controller. Does the entire thing need to be set up to test including the PVM?

[kcress]

A couple of points. I'm not clear on your wire up.

You cannot run parallel strings of different LEDs. All parallel strings need to have the same LEDs in them. They can be different but you have to have the same counts in each parallel string.

Before you ever power this thing up you should have completely tested it so you know it will light up correctly.

How to preflight your fixture is all covered in this thread starting about post 60.
http://www.reefcentral.com/forums/showthread.php?t=1973462

[fatoldsun]

Thanks Kcress:
I followed his thread and your instructions exactly. I've probably read the first five pages two dozen times and I've pointed others to it. I **hope** my stars are ok. I check each star as described and found one short that I corrected. I took your advice and dud all of that work before even beginning the power source stuff. I've actually pointed others to those instructions several times.
My parallel strings are mirror images. Not only stars but order from 1-10. Not sure how that will work with balancing if needed. In other words, the color string has green in spot #1 on both, blue in #2 & #3, red in #4, etc.
My questions refer to the next step.
I think I was ignoring ohms law but I wanted to check the driver output before I hooked it up

[Gorgok]

The driver has a minimum as well as a maximum voltage it can handle, as it seems many forget. Its trying to maintain a certain current, and to do that it alters voltage. But it can't adjust it past 24V on the low side. If you don't have at least 24V of load on the driver it won't be capable of controlling the current properly, and i am not sure what it will do, but nothing good i would imagine.

I figured your strings would be identical copies since each set was x2, but going so far as having the same order is something crazy (i admit mine are like that, even though they are not parallel).

[fatoldsun]

What can I say, I'm REALLY anal. It's a curse... and a gift :)

[fatoldsun]

I set up a single driver to test parallel strings. First test was the royal blues. Went swimmingly. Then on to the mixed colors. Not as well. The "right" string/fixture was spot on. Left not so good. It fired for 3-5 seconds and then went dim. I tried a few times. Right string great, left - dim. I powered off PVM and proceeded to look for issues. Here's where it gets strange. While testing for shorts or other problems i started at 10 and worked back) the second to last light (red, # 9) appeared to have a short. I checked the next on the string (UV, #8) and while short testing it it lit up red. Weird. I moved on and got to blue #4 it lit the whole string (just the multi tester). Now I unhooked from the driver and everything tested fine. No shorts, etc... I'm perplexed!!??!!

[kcress]

Yeah the mix certainly doesn't need to be the same order.

You can't measure much of anything on the output of a current driver without a correct load. Don't bother even trying.

As for your last testing you must not be following the prescribed testing methods as you would absolutely be finding an issue if the correct methodology was followed. [/Sheldon]

[fatoldsun]

Yeah the mix certainly doesn't need to be the same order.

You can't measure much of anything on the output of a current driver without a correct load. Don't bother even trying.

As for your last testing you must not be following the prescribed testing methods as you would absolutely be finding an issue if the correct methodology was followed. [/Sheldon]

I'm really confused and feeling a little lost -:headwalls: I went through the entire fixture, one star at a time doing the diode test as described in JP’s thread, at #62. It showed all my solder joints were ok. I then did the “short” test. Again, as described - the difference was I used the diode setting instead of testing for resistance (as described in post # 74) In all of the stars, I found one short. I had a few stars that I bought from other builds and one (a blue) was a little messy. I did my best to clean off all of the old solder but I missed a small line that stretched from the pad to the raw edge of the star itself, which in turn caused the short. I found this going from pad to screw and found that the screw was grounding the circuit. I removed the star, inspected it to find the issue. Cleaned it up and resoldered the joint. I re-tested it about 9 different ways (both the screws, scratching a small spot on the sink to make a clean contact and all around the edge of the star. Everything looked good. Then I retested the entire fixture for a short just to be safe. As you said, it only took a few minutes and I was not in a rush. I was trying to be extra cautious (especially after finding the one short) so now I am surprised. Now I am at a loss as to where to look. Admittedly, the pads come VERY close to the screws in some cases, but I used nylon washers on every screw and thought that would cover me where a screw head may sit on top of some spill over on a pad. Now, like I said, I’m at a total loss as to what to do next. I’d prefer not to redo the whole string. I guess I may have to if I still can’t come up with anything else. I will retest every which way again tonight.
Does the result I described sound like a short? – with the lights firing and them running dim. Could it be an issue with the resister, fuse or PVM? If it did short, am I on a mission to find a bad star? Could that be why it’s lighting dim and passing current from the multi-tester while still tied to the resistors? (it’s not a capacitor, I didn’t think resistors stored power the way caps can). After I powered the light down, it flashed on once. Also, I tested across the resistors after checking the royal blue strings. I then used the same set up for the mixed color string – the RBs also flashed – is it possible I damaged the resister, fuse or PVM doing that? IS there something I can test on that side of the equation after retesting all of the stars but before I fire it up again?

Oh, as for the order, I set the two up as mirror images so it wasn’t too difficult to maintain the order – I set up a layout and all I had to do to maintain the order was to start wire in a mirror image. I labeled my neg (-) wires so there’d be no mix-up and thought if I had any balancing issues it would be a little easier if it was organized.

http://i1231.photobucket.com/albums/ee503/dzipin/1D37534E-0D65-4C0D-ADEC-E7F82BCD586E-7023-00000C2C9D39A140.jpg

Thanks for the input guys. I’m really anxious to get this up over the tank. My halides are effectively expired and my coral is suffering without good light. I’m hoping to get something up before I lose the whole tank. At this point my SPS is about 80% STN’d

[Gorgok]

The on, bright to dim, then flash bright on off sounds like what an under-volt string on the driver would do. Basically the driver tried to power up but failed so its not really on, but has some residual power enough to make the LEDs glow a bit. Then when it powers off the caps unload into the string.

But i did some math on your colors string, and assuming its all XP-Es you should be good on voltage... Not really sure what to do without a better understanding of the situation.

[fatoldsun]

Thanks for all the guidance. I found a UV star that was behaving oddly. Still tested ok but I got inconsistent results testing from the pad itself. It tested fine from the wire off the next led which was weird but I was looking for anything so I redid that star and both wires leading to it. I also avoided the pad since the led seemed to be lose on the star I re-soldered it to the star and used that as a substitute for the pad.

So now onto my next step, where I'm really lost is how to set up the drivers.
1. What setting do I want to test across the resistors to get my string mA?
2. Should I use the internal pot to adjust or the PVM controller?
3. How do I test individual for balancing? - how important is balancing if things look ok?
4. Drivers - I have 4 drivers so if I do 3 sets parallel (1 pair royal blue, 1 pair mixed colors, 1 pair mixed whites) ill have an extra driver - or should I run the whites on their own drivers and only parallel the colors? In other words, 3 or 4 drivers. I'd prefer 3 for space savings but not at the expense of coral growth/health. I think with a 90g display and 64 LEDs I will probably have to dim the fixture. Any thoughts?
Here are shots of the strings. my phone's camera struggled here and these are with the interal pot at about 20%...
Mixed colors:
http://i1231.photobucket.com/albums/ee503/dzipin/57A9736D-74D2-4335-90ED-615A79AD190E-918-000001496FE1270C.jpg

Royal blues:
http://i1231.photobucket.com/albums/ee503/dzipin/D6B111A3-EFCD-44EF-AB58-392270590EFC-918-000001497874DA42.jpg

Whites:
http://i1231.photobucket.com/albums/ee503/dzipin/7CFD2B83-9F5F-487E-8107-CB86E7782F8A-918-0000014975AE7399.jpg

[Gorgok]

1. What setting do I want to test across the resistors to get my string mA?
2. Should I use the internal pot to adjust or the PVM controller?
3. How do I test individual for balancing? - how important is balancing if things look ok?
4. Drivers - I have 4 drivers so if I do 3 sets parallel (1 pair royal blue, 1 pair mixed colors, 1 pair mixed whites) ill have an extra driver - or should I run the whites on their own drivers and only parallel the colors? In other words, 3 or 4 drivers. I'd prefer 3 for space savings but not at the expense of coral growth/health. I think with a 90g display and 64 LEDs I will probably have to dim the fixture. Any thoughts?
Here are shots of the strings. my phone's camera struggled here and these are with the interal pot at about 20%...

1. Test with a voltage setting that is capable of the range required (0~1V but but has fine resolution. For your meter that would probably be 2000m, or 0-2V DC.

2. If you see the amperage is higher than expected, or higher than you ever plan to run, go ahead and run the pot down. It won't hurt to have it limited to a lower max even if you end up dimming them all the time to say 80% later on. And if you do eventually think more power is necessary its easy enough to adjust it up again.

3. For balancing you already have the starting point with the volts across the resistors. If they are close then let them run a while and test again (heating will change their electrical characteristics). The closer they are to the same the better the output of each will be.

If they are off quite a bit then you have many ways to check for high/low voltage ones (but only of identical LEDs). The simplest, is to check voltage across each LED pair in the two strings in turn. If you have a bunch of higher ones on one string you can swap out a few to the low string. Having measured the total voltage across the whole strings here would also be handy, to know how much the difference actually is.

4. Not sure on this one. I would probably run them as a parallel unit. It won't be a big deal to separate the strings later if you want any of them to have more power. The other way, if you later want to use the driver for something else as you are having to dim both of them so much anyway, would mean balancing the strings and all that stuff.

[fatoldsun]

Thanks Gorgok – for ALL of your help through this. I'll be testing /setting it up tonight. One last multi-tester question. I know the black lead stays put in the “com” common spot but the top hot input labeled 10A unfused is only for the dial setting directly above it, correct? For mA I should leave it in the lower spot? The dial goes to 2000m (roughly 9:30 on the dial) for testing across the resistors. This may be stupid, bit just to confirm, I have the lights powered on when I am testing across individual LEDs, correct? I feel like I’ll need to wear a welding helmet for that task. They are really bright. Kind of like staring at a camera flash…

Next steps:

*I need to come up with a way to house the terminal blocks, and fuses/resistors, etc... I was thinking of using Delphi type plugs that are used for car harnesses (like these (http://www.electronicplus.com/images/products/8333.jpg)) so I can easily unplug the individual modules if I need to service something and can do so without pulling the whole canopy/hood off to do it.

Choice A:
…would be to use a double gang outdoor outlet box with a sealed cover. With the resistors I got I’m hopeful that resistor heat won’t be an issue. The driver for each will be mounted to the lid for the set it’s running. Keeps the electronics off the wood and provides some protection from the evaporating saltwater. I have to have a relatively contained system – I have a stand under the tank to work with and it has a lot of equipment crammed under there. I’d rather keep the drivers and PWM (I wish I knew why my phone KEEPS auto-correcting to PVM!! – then again I don’t know why it changed my title to “meal well” I can’t blame it for the “06-48” vs. “60-48” – that was a typo) out of the stand:
Choice B:
I have tons of extra wire so I could do a remote control panel – I just would need something that my wife could live with in terms of appearance if it’s not under the stand…. So like a small electrical panel like what’s used for a home security system controller. I found a PWM with a remote so I have flexibility with where I mount that…

**I have to set up the fans although I think I went overkill with 4 fans but I guess it can’t be too cool.

***I need something to mount the lights on/in. I started a hood that was part aluminum frame, part removable wood panels but my drill press went kaput and I need that for the aluminum frame part. I’m going for lightweight if possible. I have the drill working again but I didn’t want to abandon the light itself so the hood was put on hold. I figured I could always set up a temp frame to mount the light just to have it running while I finish the hood.

**** I've collected a bunch of different lenses. I'll need to decide what is getting a lens and get them hot-glued in place

[fatoldsun]

The build continues. I found a perfect “box” - a metal enclosure such as what is used to house the “brain” of a home security system so it looks like I’m settled on “choice B”. Thank you eBay. I did a search and someone in NY had one – 16”x16”x3” (old, but NIB) for $15 with 3 hours left and no bidders. Yay. I had gone to Grainger, Harbor Freight and HD at lunch and the only one that had anything was Grainger – yep for $150 I could get a smaller NEMA 1 box - $50 upgrade for knockouts - so $15 looked like quite a bargain. Now I wanted to make a harness and I found a great source for the Delphi style connectors. If anyone wants to try this, PM me for the website. They were FAR cheaper than anywhere else I found. Like the 6-way male weather-pack was ~$0.70. I think for all of the pieces a 6 way (male, female packs, 12 pins and weather plugs) was < $4.00. Of course I ended up get tons of extra because I love gadgets and weird things like neat homemade harnesses excite me (sadly) so I’ll be waiting until I can make the harnesses to set up the drivers (and balance the stars if necessary). Hopefully this weekend I can get the frame made for the hood and set up the light mounts. The plan is to have the two modules hinged so I can independently raise them if I need access to the tank – one side will stay down lighting the tank (and one will shine in my face, basically blinding me – I never said it was a perfect plan. The only way to avoid that would be to run 6 drivers so I could independently shut off one side. I’ll have to pick up some welder’s goggles – back to harbor freight….)

I’ll post more pics as I progress along (and maybe start this over since the typos in the title bug me….)