[Maryland Guppy]

Good Evening All:

Thought to share a new LED build.

Pucks built from 2-7 bead base plates per heat sink.
Outer loop is 3-3watt full spectrum LED's, 3-3watt 10K white, and 1-3watt 2.7K in the middle.
42 watts per heat sink.

Mounting box coated with shellac for each sink.
Fans added to each if needed.

3 units added to an existing canopy.
Holes drilled in end of each mounting for air flow.

LDD drivers used, 1 for each sink.
PS is at 48VDC, all 700mA drivers.
PWM is attached via 9 pin serial din.
Remotely located StormX controller, about 12 meters of cable.

Lighting start up.
One bad 2.7K bead caused a lot of blinking and aggravation.
Resolved with a 6500K, will need to change though.

Tank is in a state of mess due to driftwood, another topic.
2nd use of full spectrum LED's, 1st was a success.
Costs incurred were LDD's(3) and 4 position board $27.
Heat sink and PS were re-purposed from work and shop.
LED's and base plates $25 from China.
StormX was pre-existing and will also be used for additional aquariums.

All in all not bad for a little over $50 USD for my free 75G.
In the end StormX will be running many tanks, 2 @ present.
One downfall to the StormX running multiple tanks is the ramp and photoperiod will be the same for all involved tanks.
PWM with twisted/shielded cable can go some distance though.

Mention of fans, thought they would not be needed.
Heat sink temps were running @ 125F, right on the boarder.
Applied 4 volts to each 12 VDC fan and temps came down to 95F.
12 VDC PS is now running 3 fans in series full time.
At 12 volts each they were extremely noisy, now undetectable.

Comments and suggestions please.
Thoughts for improvement are always welcome!

Thanks.

Will update with some PAR readings in the future.

 

[jeffkrol]

Do you have the 2 pucks on one heat sink in series or parallel?

I assume series so check the voltage across the total string.

Reason is, even at 700mA you may be exceeding the actual available voltage..
Doesn't hurt things (and helps w/ heating) but doesn't mean they are running at 700mA or 3w ..
PS is only capable of about 45V to the string..
14x 3.5(V(f)) =49v At 45v you average only 3.2v per diode..which "if" it is really pulling 700mA is only 2.24W

You may need to tweak the ps output a bit to get the full 700mA..Most have some room for boosting or cutting voltage..

Other than that looks good..

Makes me wonder..

full spectrum

Voltage: 3.0-3.6v
Current: 700ma

 

[Maryland Guppy]

Do you have the 2 pucks on one heat sink in series or parallel?

I assume series so check the voltage across the total string.

Reason is, even at 700mA you may be exceeding the actual available voltage..
Doesn't hurt things (and helps w/ heating) but doesn't mean they are running at 700mA or 3w ..
PS is only capable of about 45V to the string..
14x 3.5(V(f)) =49v At 45v you average only 3.2v per diode..which "if" it is really pulling 700mA is only 2.24W

You may need to tweak the ps output a bit to get the full 700mA..Most have some room for boosting or cutting voltage..

Other than that looks good..

Makes me wonder..

In series @ 48VDC.
I will need to check mA with meter.
It is very bright @ 75% dimming, is it 700mA?, I will need to check.
Epistar LED's, voltage always states 3.0 to 3.6, you know?
The PAR meter will provide the tale of the tape.
Tested all LED's with ohm meter and brought to applied volts with variable power supply. Diode tested too.
I opted for the cheap here, I know it though.
Some LED's were tossed after initial purchase too.
Under $4 for 10 beads, I'm not heart broken @ losing a few.

Combination of LED's turned out as pleasent to the eye though.
As stated 3 full spectrum, 3-10k, and 1-2.7K.
The addition of full spectrum in the high tech has really spiked plant growth.
Assuming it will do well in this low-tech arrangement.

PS output is adjustable and 2 series 24VDC are set @ 28.6VDVC each.
Believe I have covered the voltage requirement but will check the mA's.

 

[jeffkrol]

Just be aware w/ PWM is is just pulsed current either full or zero.. W/ out "max" meter setting or o-scope reading can get averaged..
Test at full on is best..
Can't condone series ps. It is not that simple.. possibly w/ LED loads it will be no problem but ps need to be built to "share" or you can get an imbalance between the 2..
conscience dictates I mention this..
Connecting switch-mode power supplies in series to increase voltage - Electrical Engineering Stack Exchange
The photo color (not always real color) reminds me of the old t8's..

 

[Maryland Guppy]

I haven't tested anything at this point other than VDC output.
Will bring meter from work to test such and also measure AC ripple from drivers.
Not that concerned if mA are where they should be.
It is just LED lighting, Fluke meters will let me know!
Bottom of DCV and negative of PWM are tied together and function well.
Was more worried about length of run for PWM signal.
Seems like no issues, may run PWM to the upstairs tanks but will measure distance to calculate their voltage drop first.

Everything we do seems to turn into a science, electronics, or chemistry project. I'm just sayin!

 

[jeffkrol]

I haven't tested anything at this point other than VDC output.

well?????......................

 

[Maryland Guppy]

I have neglected to perform the testing yet.
PAR meter is coming very soon too.

LED heatsink temps were @ 125F.
Right on the line for this LED.
I turned on the fans @ 4VDC each, undetectable on noise and lowered temps to 95F.

I would be further along with this tank, tannin and surface scum is hurting me.
I boiled this driftwood for hours, I should have not bothered.
The tannin can be gotten used to, daily surface scum is unreal at this point.

 

[Maryland Guppy]

well?????......................

Some testing today.
Power supplies are set @ 26.1 VDC, combined 52.0 VDC
AC ripple @ power supply output is 10.4mV
AC ripple @ output of LDD's is 12.1mV
Measured % duty cycle and confirmed with ratio to StormX, all good.
Frequency of StormX PWM output tested as 536.0Hz all channels

All the issues began with testing mA.
Anything on PWM the Fluke 83 meter is having problems with.
Will bring home my 87 this week from work.

Maybe the reading is averaging on the slower meter?
When it should have been 700mA, reading was 365mA
When testing a 300mA LDD on a different fixture reading was 183mA
Should have logged with MIN/MAX?

 

[jeffkrol]

Some testing today.
Power supplies are set @ 26.1 VDC, combined 52.0 VDC
AC ripple @ power supply output is 10.4mV
AC ripple @ output of LDD's is 12.1mV
Measured % duty cycle and confirmed with ratio to StormX, all good.
Frequency of StormX PWM output tested as 536.0Hz all channels

All the issues began with testing mA.
Anything on PWM the Fluke 83 meter is having problems with.
Will bring home my 87 this week from work.

Maybe the reading is averaging on the slower meter?
When it should have been 700mA, reading was 365mA
When testing a 300mA LDD on a different fixture reading was 183mA
Should have logged with MIN/MAX?

Never had much luck w/ "max" on my meters. It still averages..
Best to run them at 100% and put a shunt resistor in line on the plus side.. Measure the voltage across it and multiply by appropriate factor..
Putting your meter on 20A and in line should give you the correct mA output w/ the PWM at 100% on. If it is not you may not be hitting the voltage..Check the voltage across the strip..

Frequency seem off too.. Most are 490 or 980-ish

The frequency of the PWM signal on most pins is approximately 490 Hz. On the Uno and similar boards, pins 5 and 6 have a frequency of approximately 980 Hz. Pins 3 and 11 on the Leonardo also run at 980 Hz.

The resulting frequency is equal to the base frequency divided by
* the given divisor:
* - Base frequencies:
* o The base frequency for pins 3, 9, 10, and 11 is 31250 Hz.
* o The base frequency for pins 5 and 6 is 62500 Hz.
* - Divisors:
* o The divisors available on pins 5, 6, 9 and 10 are: 1, 8, 64,
* 256, and 1024.
* o The divisors available on pins 3 and 11 are: 1, 8, 32, 64,
* 128, 256, and 1024.

Not really important anyways..

 

[Maryland Guppy]

I am now building 12/24 more of these "Puck" lights.
Some for me and some for others.

I have squeezed max volts out of the PS's in series.
No change in lumens, no PAR meter yet.
PS's are topped out at this time for sure.
28.something volts was more than enough for the test.

 

[Maryland Guppy]

Ordered the Seneye Reef Monitor today.
I am not interested in the NH3 or pH monitoring features.
Only lighting info.

Some lighting updates are in my near future.
Been in the dark for too long with the DIY lighting.

Thinking of the averaging on the meter.
The readings would have been a bit more than 1/2 the desired mA's
Channels were on about 50-60%, maybe spot on???

 

[Maryland Guppy]

@jeffkrol I owe some PAR data, better late than never I guess.
Played with this Seneye meter checking all sorts of house hold lamps.
Just took awhile to get to all the tanks.

I am shocked the PAR is this high, I should add CO2 to the 75.
I turned the lights up about 6 months ago, they seemed not bright enough.
Now it has been 8 months since my last post.

PAR Data 75 Gallon 11/27/16
Each puck arrangement is 4" above water mounted.
3 pucks equidistant over 48" lentgh of tank
42 watts per puck @ 700mA 87% duty cycle
Seneye measurements

Spreadsheet data looks a little blotched up after pasting and spaces don't hold.

Puck #1 Puck #2
2-(520nm) / 6-(FS) / 6-(10K) 6-(FS) / 8-(10K)
4"/0" 375 PAR 674 PAR
Water Level 13000 LUX 24150 LUX
No K Kelvin No K Kelvin

7"/3" 210 PAR 350 PAR
8025 LUX 12000 LUX
No K Kelvin No K Kelvin

10"/6" 100 PAR 180 PAR
3800 LUX 6600 LUX
No K Kelvin No K Kelvin

13"/9" 60 PAR 112 PAR
1900 LUX 4200 LUX
No K Kelvin No K Kelvin

16"/12" 61 PAR 75 PAR
2100 LUX 3200 LUX
No K Kelvin 22.2K Kelvin

21.5"/17.5" 31 PAR 57 PAR
Substrate 1210 LUX 3200 LUX
No K Kelvin 13.8K Kelvin

 

[jeffkrol]

Thanks.. funny how you get a K value deeper...Must be just enough red attenuation at that depth w/ the 10000k/fs puck..

 

[Maryland Guppy]

While deleting subscriptions today I thought to keep this one alive.
All part of my LED evolution!

This is still up and running and performing well now.
Now is related to the 2.7K batch of ten 3 watt LED's that all failed over short time frame after this projects beginning.
Had to be a bad batch, so I lost about $3, learning curve.

Now the center LED's are 520nm & 10K.
Meaning each dual disk has one of each in the center.
Greens cut the PAR down a little bit but it is an enhancement for viewing.
All things planted look a little better, little more real.

Now the future has brought about some changes.
Every heatsink now has two fans in series @ 12VDC.
All fans had to be tested and paired to ensure that both run @ same speed.
If not paired up one will stop and the other runs @ full speed.

I repurpose these fans from work(expired electronic equipment)
They are filthy when I get them.
Dawn and warm water in a bucket to remove grease, dust, oils, etc...
Just blow dry them with air from compressor.
They are a brushless cheap foreign fan but run like the Energizer Bunny.
Some have been in service 8-10 years before I get them.
None have failed even after the cleaning process.

The additional fans were not added due to overheating.
I just wanted the heatsinks to be a little cooler and additional canopy circulation as to not heat the tank during summer.

LED changes for new builds.
Per disk still keeping the 3:3 ratio of 10K to full spectrum.
Center LED's are 1 520nm and 1 Royal Blue.
This even cut the PAR some more but raised the PUR values.

The future when I run out of parts?
This I imagine will turn into a different surface mount LED and a new disk layout.
The LED evolution regarding availability of older beads?

 

[ChrisX]

MarylandGuppy,

I just found this thread and wish I had read it before I built my DIY leds. It seems that I too have used the Epistar leds.

Do you have any tech specs for the epistar leds? Found nothing on their website.

When I built my hoods, I just guessed how many I would need based on wattage of drivers. I have 40x leds over a 48" 50 gallon tank. Height of tank is 19". Lights probably 18" from substrate.

I suppose my PAR will be in the neighborhood of yours. I am having problem deciphering your PAR data. What was PAR at the substrate on your 75?

How well have they been growing plants? Do you wish you had more leds?

 

[Maryland Guppy]

It seems that I too have used the Epistar leds.

Do you have any tech specs for the epistar leds?

I have 40x leds over a 48" 50 gallon tank. Height of tank is 19". Lights probably 18" from substrate.

What was PAR at the substrate on your 75?

How well have they been growing plants?

Do you wish you had more leds?

Epistar not a bad choice for saving a serious buck, but buy extras!

No specs except for colors, could find no "white" data.
Even checked again this evening.

Each "puck" has 42watts with 3 pucks on the 75 in this thread.
Total of 126 watts, is it enough? For this tank yes.
Plants grow clean, no algae, BBA on all driftwood and GSA on glass.
Run a scraper monthly and the Amano shrimp tear it up.
It will not grow some of the more demanding high-light plants.
It will sustain them and keep them in check with no new growth.
This is a no-water-change tank that gets a quarterly vacuum.
Home to 70+ tri-color platties and Amano shrimp.
They love it, it is in dire need of a trim but I just watch it grow.
Substrate is BBA ridden SafeTSorb.
Also it is very demanding of ferts due to plant mass.

This was PAR @ substrate.
17.5" = 57 PAR, 3200 LUX, 13.8K Kelvin.

This tank needs no more wattage for my purposes.
I throw plugs and weighted extras in here along with a small selection of Buce.
It is basically a grow out tank that I tend to neglect on purpose.
Here is a terrible pic of the chaos.

 

[ChrisX]

Hey thanks, that is great information.

I have roughly the same number of leds, so PAR is probably similar. Medium light. I can add another row of 10 to each side, so that would (probably) push me into high light territory. (Edit: OTH, mine are running at 600 mA, so I may have less output if yours are at 700.. plus a million other variables.)

From an economic standpoint, the Epistar leds make sense, but from a time/effort standpoint they dont.

Are you running CO2? That would probably solve all your algae problems.

 

[Maryland Guppy]

mine are running at 600 mA, so I may have less output if yours are at 700.. plus a million other variables.

From an economic standpoint, the Epistar leds make sense, but from a time/effort standpoint they dont.

Are you running CO2? That would probably solve all your algae problems.

700mA on PWM dimming 70% I think duty cycle.
Meaning they are on 70% of pulse wave form.
Kind of equals 70% of 700mA = 490mA as a comparison maybe.

For what China sends me these LED's for I don't mind pulling a puck every 6 months to change an LED or 2.

CO2 is a yes in this tank.
I didn't realize the PAR was so I high till I got the meter.
Too late @ that point for algae and I'm not going to tear down very soon.
CO2 was added about month 8 on this tank.
When I do it will be another soil capped tank.

The plants are clean so I can live with my algae.

 

[Omar EAZi]

cool setup bro, good job with the lights

 

[Maryland Guppy]

Thanks for the compliment.
This has changed a lot though.
A new LED configuration and 300watts on a 18" deep frag tank.

Science experiment is always ongoing!:grin2: