DIY LEDs - The write-up

Status
Not open for further replies.
Getting ready to do my first LED build

Getting ready to do my first LED build

What a thread!! Took me a couple of days to go through it. At one point I felt fairly comfortable about what I wanted to do, but now I have doubts, so I would like to have you guy review my plans.

I plan on LPS, SPS and some softies in the tank.

The tank that I am setting up is a 65 gal tank which is 36 X 18 X 24. Total area is 648 sq inches. I am planning on using 48 LEDs with the breakout being 28 RB's (XP-E XPEROY-L1-D40-16-0-01) and 20 white (XP-G XPGWHT-L1-1BO-R5-0-01) Is this correct? Do I need Neutral LEDs?

I am not sure on optics. I plan on having the lights 8-10" above the water but I am flexible. Do I even need to have optics?

LEDS will be wired in series, with Meanwell drivers and I would like to be able to dim the LEDS. Whats the latest and greatest to use for this?

Thanks for any suggestions. Mike/KD9RG
 
Neutrals are personal choice. They will make the tank slightly redder. Some think you need them some don't. If you have the money order half again NW and switch them till you get what you like (means you are stuck with extra LEDs). If not guess. You also might want to consider more blues 32/16 at least that is what people have been writing.

You don't need optics. But 60 (might need to raise it a little more) to 80 would put more light in the tank.

If you don't have a controller then IMHO order the P series and find one of the Arduino threads to follow.
 
I have the same tank. Going with 26 RB & 2 B on one driver and 18 CW and 2 NW on the other. I'm using Thomas Research drivers so mine will be maxing out at 850mA on the blue and 1250mA on the white.

Because the tank is 24 inches deep I am using the 55~65 optics. I checked the angles and (depending on your spacing) if you can mount so that the lense is 10 inches off the water, you should get perfect coverage.

Based on what I have seen posted, this is overkill and I probably wont be running them wide open any time soon but this is going to be an SPS tank so I want the ability to go pretty strong just in case I need it.

Good luck with your build.
 
For Thomas Research I am looking at:
TRC-100S175DT for blue at 34-57V and 1750mA. (2x14 strings at 875mA)
TRC-100S175DT for white at 25-41V and 2450 mA. (2x10 strings at 1225mA)

Still looking to be sure but these are the front runners.
 
Last edited:
For Meanwell I am looking at:
HLG-100H-54B for blue at 27-54V and 1770mA (2 x 14 strings at 885mA)
HLG-100H-42B for white at 21-42V and 2230mA (2x10 strings at 1115mA)

Think I can get these a bit cheaper.
 
Last edited:
On almost all parallel string builds I see these 1 ohm resistors for current measuring. I am thinking about building a setup with 2x HLG 185-42B drivers (if I can find them anywhere). So lots of parallel strings (7 or 8 for RB) and 6 for XPG.
If I get a current measuring tool that works when placed around a current carrying wire would I still need these resistors? See for example this link: http://www.velleman.eu/distributor/products/view/?id=375168. But other brands also make this kind of devices. Anybody any experience with these, the specs look very promising.

Thanks for any replies!
Paul.
 
On almost all parallel string builds I see these 1 ohm resistors for current measuring. I am thinking about building a setup with 2x HLG 185-42B drivers (if I can find them anywhere). So lots of parallel strings (7 or 8 for RB) and 6 for XPG.
If I get a current measuring tool that works when placed around a current carrying wire would I still need these resistors? See for example this link: http://www.velleman.eu/distributor/products/view/?id=375168. But other brands also make this kind of devices. Anybody any experience with these, the specs look very promising.

Thanks for any replies!
Paul.

That link only takes you to their homepage and I can't see what you're looking at but the mini clamp meter looks too big to me.

Instead of resistors I use fast blow glass fuses with holders and when I want to check current I pop the fuse and clip in my meter. since i feel fuses are mandatory this is a good compromise for me.

As to the Meanwell HLG. Go to wattsupply.com and send them an email or call them. They list the entire line but do not stock them all. however, they got me mine in a week or so. Will have to check shipping rate to Netherlands of course.
 
When balancing multiple strings, especially when you are talking 6 to 8, the goal is to get them within 50 mA of each other. I dont have any personal experience with the type of tool you are considering, but I would question its accuracy and weather it would be sufficient to measure such small differences in current.
 
When balancing multiple strings, especially when you are talking 6 to 8, the goal is to get them within 50 mA of each other. I dont have any personal experience with the type of tool you are considering, but I would question its accuracy and weather it would be sufficient to measure such small differences in current.

Ok, if it has to be within 50 mA then this might not be the right tool. The accuracy is specified at 3% of the full scale (2A) this is already 60 mA... However this does mean that the reproducibility must be much better so to compare strings it is probably still ok. I am not so much interested in the exact currents. I will think about it a little more.
 
That link only takes you to their homepage and I can't see what you're looking at but the mini clamp meter looks too big to me.

Instead of resistors I use fast blow glass fuses with holders and when I want to check current I pop the fuse and clip in my meter. since i feel fuses are mandatory this is a good compromise for me.

As to the Meanwell HLG. Go to wattsupply.com and send them an email or call them. They list the entire line but do not stock them all. however, they got me mine in a week or so. Will have to check shipping rate to Netherlands of course.

Thanks for the suggestion of wattsupply.com. I will definitely send them a mail. Swapping the fuse for the current meter is a good idea, will keep it in mind.

A more general question, does anybody know the range I can expect of the forward voltages of XPE and XPG leds? What is the spread that you observed as min/max at for example 700 mA? I ask because I made a little tool (matlab) to calculate the currents going through the different strings when the leds are a bit different in forward voltage. It seems to work reasonably well, but for now I assumed 0.2V per led. Is this reasonable or should it be more or less? Does anybody know?
 
Swapping the fuse for an ammeter is not a good solution to me, for balancing currents.

The current a string uses is directly dependent on the forward voltages, Vf, of all the individual LEDs. The Vfs change greatly with temperature. Turning off your rig to swap out the fuse for an ammeter completely disturbs everything thermally, changing Vfs all over the fixture. It's the same as studying an ant colony by digging it out with a shovel... "Oh, look how the ants run scurrying everywhere! Huh? I don't see how they can get anything done." :lmao:

With the resistors you can measure anytime you want without altering the thermal aspects. You also don't run the risk of interrupting the string with the driver running, nor accidentally trying to measuring a voltage with the meter in current mode - which shorts the voltage source you're measuring blowing fuses, and blowing your meter. (I've done it many times).
 
Swapping the fuse for an ammeter is not a good solution to me, for balancing currents.

The current a string uses is directly dependent on the forward voltages, Vf, of all the individual LEDs. The Vfs change greatly with temperature. Turning off your rig to swap out the fuse for an ammeter completely disturbs everything thermally, changing Vfs all over the fixture. It's the same as studying an ant colony by digging it out with a shovel... "Oh, look how the ants run scurrying everywhere! Huh? I don't see how they can get anything done." :lmao:

With the resistors you can measure anytime you want without altering the thermal aspects. You also don't run the risk of interrupting the string with the driver running, nor accidentally trying to measuring a voltage with the meter in current mode - which shorts the voltage source you're measuring blowing fuses, and blowing your meter. (I've done it many times).

Of course I will always defer to you on such topics and I hear ya but practice has shown the readings to be repeatable and if thermal differences were an issue it wouldn't be repeatable. After doing multiple rigs with resistors and fuses and checking the readings across the resistor and through the fuse holder and always getting the same answer I chose to just go with the fuse on my recent rig.

Agree on the risks and you need to be diligent not to pop stuff with the rig on and make sure the meter is set correctly. When I'm running through the strings I start with voltage measurements on each string and record them, then I start the fuse popping and go one after the other with only time elapsed is to write down the values. Last I start the LED by LED voltage measurements while trying not to blind myself and i write these values right on the heatsink. Rinse and repeat...

Kcress you always have our best interests in mind and your recommendations are always surefire and fool proof. I deviated recently and do not suggest its a better way to go but just that it worked for me.
 
Thanks brassmonkyballs.

You're right. With a 2 stringer or a single stringer, the meter really is fine and resistors can be avoided if you know what you're doing.

For me they're so easy and cheap I use them even on 2-stringers. On larger builds you can occasionally run down all those resistors in about 15 seconds to see if any strings are starting to drift unevenly with age. This allows detection and an opportunity to correct things before a derail occurs. If you have to use an ammeter, just the hassle is enough to make most people pass on a bi-annual test check.
 
Looking for some advice on led layout please!

Here is what I have

60 RB XP-E
30 CW XP-G
6 NW XP-G

I am using 4 60 inch long 20mm wide aluminum t-slot framing members with 24 leds per member.

I am planning to use 5 of the 48 60D meanwell drivers

How would you mix the leds to get even spread and what optics to use on the RB and the CW, for the NW I will not be using optics.

Any help is appreciated.
 
I was going to run the 5 60 48D because I wanted to be able to control the CW, NW, and RB separately.

The only thing I am stumped on is how to space the NW and the CW.

The outer two bars I was going to do a RB-CW-RB-CW-RB pattern that uses 24 RB and 24 CW.

The two inner bars is where I am not sure of pattern I have 6 NW, 6 CW, and 36 RB to space evenly?
 
Status
Not open for further replies.
Back
Top