DIY LEDs - The write-up - (split again)

[2Quills]

Thanks, I will keep that in mind and try to isolate a matrix of all vf on each LED. Then I may be able to pick and chose which ones to swap. Hopefully I can then reduce the number of times I need to do this.

I will check the solder on the blues, they do seem to have a pretty big spread, especially when compared to the whites. I do have some regular blue LEDs mixed on the same string as the royal blues, they probably have a different spec and may be part of the reason that the blues are farther off. Although I also have some NW mixed in with the CW also and they seem to be pretty close. I am going to measure all the LEDs tonight or tomorrow and I am sure I can get these balanced over the weekend. Unforunatly I can only balance 1/2 the rig until I get the 2 replacement drivers. I am going to finish installing the other 3 heat sinks and wire them up using the 2 good drivers and I will blalance these out. I don't think it is a good idea to balance the other 1/2 until I get the permanent drivers for them.

thanks

I don't know if you caught this on your other thread or not.

We had my friends rig with the 240's pegged at 5 amps / 6 strings for 833mA per string Max.

What they actually saw on both drivers after running them for a little while and then doing the initial testing was 810-860mA between the highest and lowest string on one driver and 800-860mA on the other driver.

I believe he did swap a few leds around to minimize that just a bit. He did very much what Kcress just described. He swapped out 1 led in two strings on each driver. The led with the highest voltage reading on the string that had the lowest amount of current with a led that has the lowest voltage reading from the string that had the highest amount of current. I think it knocked about 20mA off the difference.

 

[GUILLO1]

One more ? When I connect the resistors do I have to leave part of the resistor wire exposed or can I cover it all. I mean in order to test the strings

Sent from my SPH-D710 using Tapatalk

 

[kcress]

One more ? When I connect the resistors do I have to leave part of the resistor wire exposed or can I cover it all. I mean in order to test the strings

Sent from my SPH-D710 using Crapatalk

The entire point of the resistors is to allow easy measurement of the string current by direct measurement of the voltage across the resistor. So. You need electrical access to the resistors lead voltages. If you have terminal blocks or something similar then I guess you can cover the leads. Keep in mind the resistors can't be completely covered because they need cooling.

 

[TheFishMan65]

reeferdale,

Do the blue strings all have the same number of Royal Blue and regular blue. Say all strings have 10 RB and 2 RB. Or are some 9 and 3. If they are all the same then the variation should not be because of LED type.

 

[reeferdale]

reeferdale,

Do the blue strings all have the same number of Royal Blue and regular blue. Say all strings have 10 RB and 2 RB. Or are some 9 and 3. If they are all the same then the variation should not be because of LED type.

Thanks, they are all balanced strings. I am going work on this today.

 

[reeferdale]

I don't know if you caught this on your other thread or not.
.

2Quills, you were right about the SN# on the meanwells. I have 2 'old' drivers and now I have to hunt down some new ones. Bill at reefledlights agreed to refund them for me, but he does not carry these anymore! :headwally:

I found the threads about this entire SN# issue and the 2 drivers I have are definitly older than the date meanwell states.

Back to waiting again!

 

[reeferdale]

OK, now that I know I need 2 HLG-240H-48B drivers - and can't find them - does anyone want to throw me some names of companies they may have use for drivers? If we can't post names, a PM would be great.
So now I am on the hunt for 2 drivers.

 

[2Quills]

OK, now that I know I need 2 HLG-240H-48B drivers - and can't find them - does anyone want to throw me some names of companies they may have use for drivers? If we can't post names, a PM would be great.
So now I am on the hunt for 2 drivers.

So are you saying that the drivers with the old serial numbers are working but the newer ones arent?

Id contact power supply emporium and ask to speak to a gentleman named Mike. He was extremely helpful to my friends in getting the situation with thier drivers rectified.

Just explain whats going on and what youre looking for. He'll probably remember whats going on. Even if they dont have them in stock they can get them in house usually within a couple of days and ship them out to you. My friends only had to wait for a little less than two weeks before they had replacements in hand.

 

[reeferdale]

So are you saying that the drivers with the old serial numbers are working but the newer ones arent?

Id contact power supply emporium and ask to speak to a gentleman named Mike. He was extremely helpful to my friends in getting the situation with thier drivers rectified.

Just explain whats going on and what youre looking for. He'll probably remember whats going on. Even if they dont have them in stock they can get them in house usually within a couple of days and ship them out to you. My friends only had to wait for a little less than two weeks before they had replacements in hand.

actually just the opposite. the newer drivers are good. Meanwell ack. the issue. Drivers after 7/2011 are OK. I found 2 at WattSupply.com and they checked the SN# for me. They are RB18xxxx. which is August 2011.

I am going to use the 2 good drievrs I have (RB18xxx) and wire up the other half. I figure I can balance the strings on these and then when I get the other 2 drivers, I will put them on the pther sinks and balance them.

Do you think if I balance all the strngs using these 2 drivers that they will be the same if I swap out the drivers later? I just don;t want to do this just to find all the readings are different once I put the replacement drivers on

 

[TheFishMan65]

If the current remains the same the balance should also.

 

[GUILLO1]

is anyone using a typhoon controller with MW HLG-185-42B? I connected one yesterday and I could not get the typhoon to control the LEDs. I am going to trouble shoot it some more tomorrow but would like to hear if anyone is already using this combo. I called the guys from BoostLed and he told me to try a few things. Guy from BoostLed seems pretty legit he told me that the typhoon should work with that driver and if for any reason the controller was faulty he would send me a new one.

 

[2Quills]

is anyone using a typhoon controller with MW HLG-185-42B? I connected one yesterday and I could not get the typhoon to control the LEDs. I am going to trouble shoot it some more tomorrow but would like to hear if anyone is already using this combo. I called the guys from BoostLed and he told me to try a few things. Guy from BoostLed seems pretty legit he told me that the typhoon should work with that driver and if for any reason the controller was faulty he would send me a new one.

I'm actually going through the same thing right now. I don't have HLG's I have LPF's. I'm not sure what the issue is on your side but for me my 10v channels won't control my leds. I just get 100% output on the 10v channels and nothing else. They sent me a replacement that I got within a matter of just a few days but the second one they sent is doing the same thing.

Everything works perfectly normal if I run my drivers on the 5v channels just not the 10v channels. But of corse I only get the obvious 50% max output on the 5v channels.

I'm currious if you'd be willing to check something. Can you test the voltage on your 10v channels? For whatever reason the two controllers that they've already sent me start out at 5 volts on the 10v side and they're not supposed to do that.

 

[GUILLO1]

I'm actually going through the same thing right now. I don't have HLG's I have LPF's. I'm not sure what the issue is on your side but for me my 10v channels won't control my leds. I just get 100% output on the 10v channels and nothing else. They sent me a replacement that I got within a matter of just a few days but the second one they sent is doing the same thing.

Everything works perfectly normal if I run my drivers on the 5v channels just not the 10v channels. But of corse I only get the obvious 50% max output on the 5v channels.

I'm currious if you'd be willing to check something. Can you test the voltage on your 10v channels? For whatever reason the two controllers that they've already sent me start out at 5 volts on the 10v side and they're not supposed to do that.

View attachment 176734

I'll test the voltage on the controller tomorrow. Today I'm wiring up the final set of strings.

I am having a little problem. I've tested every single string by itself and they all work. When I'm wiring the strings in parallel I am daisy chaining the - (neg) so to only run 1 wire from the last string back to the driver. Is there some sequence that has to be followed? I am getting a few strings that are not lighting up. But if I test that string by it self with my small driver that string lights up fine. What could I be doing right or should I just run a wire from the last led to the driver?

 

[2Quills]

I'll test the voltage on the controller tomorrow. Today I'm wiring up the final set of strings.

I am having a little problem. I've tested every single string by itself and they all work. When I'm wiring the strings in parallel I am daisy chaining the - (neg) so to only run 1 wire from the last string back to the driver. Is there some sequence that has to be followed? I am getting a few strings that are not lighting up. But if I test that string by it self with my small driver that string lights up fine. What could I be doing right or should I just run a wire from the last led to the driver?

Do you have any pictures or diagrams of your wiring scheme?

There should be one wire leading from the - side of the last led in every string. They all need to connect and lead back to the driver. Same thing at the head of the strings. Lots of people employ the use of terminal blocks for this purpose but I've seen it done with simple wire nuts. Hopefully you've also employed the use of some fuses and resistors to make things easier and safer for yourself but that's optional.

Have you done any testing to look for shorts between the leds and the heat sinks anywhere? That should always be done first before you ever put power to anything. A thorough inspection of all your solder joints and connections should be done at least a few times to make sure everything looks good.

 

[wardda]

Cooling

Cooling

I'm putting 3-4 48" strips of 12 LED's each on 1.25x1.25x1/8" channel in a canopy. I'm just wondering what number of fans I need to keep this cool. The LED Area is fully enclosed glass below the LED's and wood above, so the airflow will be coming from the fans...

Should I do 1-3 80mm fans blowing in parallel to the strips or should I do 1-3 120mm fans blowing down on the strips?

I just don't know how much outside air is needed to keep them cool and how much of that needs to be directed right on the channel vs just blowing through the LED section of the canopy.

Thanks,
Dan

 

[der_wille_zur_macht]

Is the area completely enclosed on all sides as well as top and bottom?

I would vote for fans in the sides/back blowing parallel to the channel, with vents at the other end.

Larger fans are sometimes quieter for a given cfm, but using multiple smaller fans instead of one larger one gives you a degree of redundancy if a single fan fails.

 

[wardda]

OK, do I need an individual fan blowing on each channel or will I be fine just sticking like 2 fans on one end with vents on the other? I guess I'm wondering how much direct cooling is needed vs just cooling the chamber.

One more question, if I put the fans on one side of the canopy and then put the vents on the top of the other end will that cause problems? Aesthetically that would look better, but if it reduces the cooling effects by a lot then I will cut into the side.

Dan

 

[der_wille_zur_macht]

12 LEDs spread over 48" of channel is not really a ton. I would be comfortable with two fans.

Putting the vent in the top (or even the fans) shouldn't be a big deal as long as they are at opposite ends of the enclosure, to help eliminate the chance of stale air.

 

[wardda]

One other set of options, I can cut into the back pretty easily if there is a venting or fan option that would be beneficial. The left side, back, and top are the easiest to access/ modify.

Last question (for now) how may led's would you comfortably put on a 48" channel 1.25x1.25x1/8".

Thanks,
Dan

 

[der_wille_zur_macht]

It depends on the LEDs and their power levels. With fans to provide strong airflow, I would try to shoot for ~3 square inches of aluminum per watt. On a 1.25" U-channel you've got 3.75 square inches of area for every linear inch of channel. If you're running typical LEDs they are dissipating 2 - 5 watts each, so you're talking about maybe one LED per 2 - 4".

The ~3 square inches per watt figure is a rough guess based on technical papers from Luxeon that I remember reading a year or two ago. If your LED manufacturer publishes any papers on thermal design you might want to read them.