What do you think to my DIY lamp?
- Thread starter fishlibrary
- Start date
Concrete,
1.How much did all of that cost? US$ please
Totally at around US$250 to US$300.
2.How much time did it take??
Not include buying materials, planning & build prototype for test before, it took me around 40 hours on contructing the 4 PCB.
3.What kind of freakkin phone is that??
Sorry! Can't get your meaning!!
4.Does Each PCB have 900 LED's
4 PCB, each PCB consist of 100x50,000mcd white LED, 100x13,000mcd blue LED & 25xUV LED.
5.Is there any Noticable Heat Output?
No heat detected.
1.How much did all of that cost? US$ please
Totally at around US$250 to US$300.
2.How much time did it take??
Not include buying materials, planning & build prototype for test before, it took me around 40 hours on contructing the 4 PCB.
3.What kind of freakkin phone is that??
Sorry! Can't get your meaning!!
4.Does Each PCB have 900 LED's
4 PCB, each PCB consist of 100x50,000mcd white LED, 100x13,000mcd blue LED & 25xUV LED.
5.Is there any Noticable Heat Output?
No heat detected.
I am from HK too. Has not been back since 1980's.
Anyway, where is the resistor for each LED? It sounds like you may get into issue if you wire the LED in series without the current-limiting resistor. Just guessing, but that's maybe why you burnt your last setup... :?
Anyway, where is the resistor for each LED? It sounds like you may get into issue if you wire the LED in series without the current-limiting resistor. Just guessing, but that's maybe why you burnt your last setup... :?
Hi dupaboy1992,dupaboy1992":xs0x0nfs said:
You are right! There are no current-limiting resistor on the four PCB. Each PCB consist of 45 parallel x 5 series LED that drive by a 15VDC power supply. Each LED consume 3VDC at a low forward current or you can say I am under-drive those LED.
Total consumption of the 4 PCB is listed below and measured by FLUKE 79 Series II Multimeter:
DC: 15V 1.1A
AC: 223V 0.11A (~24W)
From calculation, 4 PCB x 45 parallel array in 15VDC, forward current on each array consume ~6mA.
The reason of previous burn was due to driving LED by 3.2VDC with 20mA forward current (Factory typical specification is 3.4VDC with 20mA). The brightness reduce quite obvious after running it after 30days each day 12hours.
Well, unless the newer LED works differently, I think you should make sure it is alright not to have any current limiting resistor on each and every LED. As far as I know, it is bad practice to wire them up like that. But keep in mind that I only work with electronic engineers, not one professionally or academically. But if you know otherwise, I am all ear to know how to do it right. 
Hi!
Sorry to interrupt your discussion, I own only tropical tanks, so forgive, if i understood something wrong...
Anyway, I was also interested, how the LEDs of this aquarium are driven. Talks about burning some of them might suggest using simple non-controlled power supply with ballast resistors for bigger grops of LEDs only.
Haven't you considered the benefits of pulse-width modulation? One could immitate sunrise and -set with this trick. Moreover, you could double check your LEDs current with Hall-effect (like HX10NP-rated up to 10A or analogous) )current sensors (since the overall current could be too much for a standard resistor sensor).
PWM let's you use bigger cuurents through LEDs due to short "active" period (see datasheets).
Btw, what's the type of LEDs you used? I could record their spectra for you, if they are available at electronics stores in my country.
regards,
/Felch
EDIT: It appears, that i did not read the thread up to the end
Looks like you do not use any current limiting resistors at all. That's exactly how I burned some LEDs since they are not all identical. Some electronic current limiting circuit (per bigger groups perhaps) could be useful. Could give bigger intensity also.
Sorry to interrupt your discussion, I own only tropical tanks, so forgive, if i understood something wrong...
Anyway, I was also interested, how the LEDs of this aquarium are driven. Talks about burning some of them might suggest using simple non-controlled power supply with ballast resistors for bigger grops of LEDs only.
Haven't you considered the benefits of pulse-width modulation? One could immitate sunrise and -set with this trick. Moreover, you could double check your LEDs current with Hall-effect (like HX10NP-rated up to 10A or analogous) )current sensors (since the overall current could be too much for a standard resistor sensor).
PWM let's you use bigger cuurents through LEDs due to short "active" period (see datasheets).
Btw, what's the type of LEDs you used? I could record their spectra for you, if they are available at electronics stores in my country.
regards,
/Felch
EDIT: It appears, that i did not read the thread up to the end
Looks like you do not use any current limiting resistors at all. That's exactly how I burned some LEDs since they are not all identical. Some electronic current limiting circuit (per bigger groups perhaps) could be useful. Could give bigger intensity also.
dupaboy1992, felch,
Would you mind to provide some practical circuitry to me so as to improve my existing LED board. Currently all LED board powered by a 15VDC ballast to 180 parallel in 5 series LED array, 3V per LED.
I just want to improve the intensity & color temp. base on existing connection, any recommendation?
Many thanks!
Would you mind to provide some practical circuitry to me so as to improve my existing LED board. Currently all LED board powered by a 15VDC ballast to 180 parallel in 5 series LED array, 3V per LED.
I just want to improve the intensity & color temp. base on existing connection, any recommendation?
Many thanks!
Easiest way is using a resistor per LED, like mentioned some posts higher. Of course it means lots of rearrangements.
Then I guess that an LM317 connected as current source could be one solution. Suppose using one such stabilizer per one "column" of LEDs?
Possibly you could then drive LEDs at the edge of max. current without being afraid of power fluctuations in mains. Increasing current means increased light output....and heat also, i am afraid.
Another solution could be using chips with current stabilized outputs designed especially for LED displays (Allegro, Toshiba). This leads to an idea, i have long been wanting to realize.
Suppose, you connect all your LEDs into matrix. In this way you could even draw a picture on them. Naturally this is not needed but one could immitate sunrise and -set and also movement of clouds. Naturally you'd need a microcontroller to keep it under controll.
I made a LED display http://parsek.yf.ttu.ee/~felc/P1010035.JPG, using mentioned idea but this was a display only.
This crazy idea is surely overkill for my tropical tank but perhaps not so foolish for reef aquarium...?
/Felch
Then I guess that an LM317 connected as current source could be one solution. Suppose using one such stabilizer per one "column" of LEDs?
Possibly you could then drive LEDs at the edge of max. current without being afraid of power fluctuations in mains. Increasing current means increased light output....and heat also, i am afraid.
Another solution could be using chips with current stabilized outputs designed especially for LED displays (Allegro, Toshiba). This leads to an idea, i have long been wanting to realize.
Suppose, you connect all your LEDs into matrix. In this way you could even draw a picture on them. Naturally this is not needed but one could immitate sunrise and -set and also movement of clouds. Naturally you'd need a microcontroller to keep it under controll.
I made a LED display http://parsek.yf.ttu.ee/~felc/P1010035.JPG, using mentioned idea but this was a display only.
This crazy idea is surely overkill for my tropical tank but perhaps not so foolish for reef aquarium...?
/Felch
Sponsor Reefs
We're a FREE website, and we exist because of hobbyists like YOU who help us run this community.
Click here to sponsor $10:
