stardustsailor
Well-Known Member
Let's hope that Guod will notice the initial post ...
Now...
Although,comparing to Guod,I'm a "novice" to electronics ...
I've thought about it a bit ...And manage to design some very "raw " circuits ...
Actually there are two -at least approaches - to the issue.
Approach #1 : AC side.
In this approach ,no actual dimming is done.
It leaves the dimming DC circuit as is.
But it has some limitations .
It takes as basis ,that each COB is driven by it's own CC driver.
Simply it switches the first COB ,when a timer IC switches on the LED light fixture .
After a certain-adjustable- time it switches the second ON the second COB ,
then after some time it switches the third and so on.
Up to 10 drivers/COBs can be handled withg this circuit.
(More than 10 can also be handled ,but it needs more parts.)
So ,actually no dimming is done .
The irradiance is gradually increased (in so many "steps",as the number of cobs /drivers.).
Quite "rough " approach ,but that was my first -quick- idea...
So here is the schematic :
Basic Parts :
4x mechanical or solid state relays.
4x NPN transistors ( BC547 or 2N3904 )
1x LM555 timer IC
1x HCF4017B Decade Counter
1x 10 MegaOhm potentiometer
1x 47 uF electrolytic Capacitor
1x 1MegaOhm resistor
1x 10K resistor
4x resistors (value still uncalculated ,but probably in the range to 1-10 K will be fine )
3x 10-100 nF capacitors (for bypassing noise )
I 'm not sure if everything is alright .Later I will note down some of my concerns.
How it works : When the timer switches on the LED grow light ,the Fan power supply switches ON also.
The LM 555 timer IC is then powered up,also.
As it is configured in ASTABLE operation ( http://www.555-timer-circuits.com/operating-modes.html ) ,
the timer will output a PWM signal of ~ 50% duty cycle ,from its OUTPUT pin (3) to the INPUT (CLK ,pin14) of the
HCF4017B Decade Counter.Every time a new "high" signal is reaching the CLK pin of the decade counter ,
it's output pins will be HIGH ,starting from OUT 0 (pin 3) towards OUT 9 (pin 11) .
As V series Tetras fixture ,features four drivers (and four COBs ) ,not all OUt pins are connected.
Only the Outputs 1 ,3,6 & 9 are used ( pins 3,7,5 & 11 respectively ).
Every output pin is connected to a NPN transistor ,which powers a relay .Each relay powers up a LED driver.
At the last relay (#4 ),before it's base resistor ( R4) ,a line lifts the CLOCK INHIBIT (pin 13)of the decade counter to HIGH.Thus the decade counter stops alternating it's outputs ,and the output 9 (pin 11) stays HIGH .
The circuit will reset itself ,when the LED grow light is switched OFF and back ON again (hopefully ).
The LM 555 timer's frequency is adjusted by the 10M pot , from ~ 1min up to ~10 minutes .
( http://www.ohmslawcalculator.com/555-astable-calculator )
So ,actually the minimum " rise time" is 9 minutes ( 1st driver ON ,after 3min the second,at 6min the third and after 9 min the fourth driver.).The maximum "rise time is ~90 min ( 1st driver ON ,2nd after 30 min ,3rd after 60 min ,4th after 90 min )
Concerns:
-The 555 timer will continue it's PWm astable operation.Will this affect the operation of the decade counter?
(As it's input (CLK,pin 14 ) will continue receiving signals from the timer)
-A fifth relay (or MOSFET/transistor) has to be used to stop the 555 timer ,instead of clock inhibiting the HCF4017B Decade Counter ?
- Will inrush current of the driver deteriorate the contacts of a mechanical relay (sparking ) ?
-The high threshold resistance (1M ) & pot (10M ) used for LM 555 timing allow only for low currents ,thus any induced noise can be an issue ?
.....
I just hope Guod,sees all that and comes with a far better solution,than this one ....
Next one, comes the -more complicated- Approach #2 : DC ( DIM ) side.
Cheers.
Edit :
I've already spotted a mistake :the transistors will switch OFF ,every time the decade counter will proceed to the next output.Instead of transistors ,MOSFETs probably have to be used .So this schematic is wrong.(WAYYYYY Wrong !!!!)
Now...
Although,comparing to Guod,I'm a "novice" to electronics ...
I've thought about it a bit ...And manage to design some very "raw " circuits ...
Actually there are two -at least approaches - to the issue.
Approach #1 : AC side.
In this approach ,no actual dimming is done.
It leaves the dimming DC circuit as is.
But it has some limitations .
It takes as basis ,that each COB is driven by it's own CC driver.
Simply it switches the first COB ,when a timer IC switches on the LED light fixture .
After a certain-adjustable- time it switches the second ON the second COB ,
then after some time it switches the third and so on.
Up to 10 drivers/COBs can be handled withg this circuit.
(More than 10 can also be handled ,but it needs more parts.)
So ,actually no dimming is done .
The irradiance is gradually increased (in so many "steps",as the number of cobs /drivers.).
Quite "rough " approach ,but that was my first -quick- idea...
So here is the schematic :
Basic Parts :
4x mechanical or solid state relays.
4x NPN transistors ( BC547 or 2N3904 )
1x LM555 timer IC
1x HCF4017B Decade Counter
1x 10 MegaOhm potentiometer
1x 47 uF electrolytic Capacitor
1x 1MegaOhm resistor
1x 10K resistor
4x resistors (value still uncalculated ,but probably in the range to 1-10 K will be fine )
3x 10-100 nF capacitors (for bypassing noise )
I 'm not sure if everything is alright .Later I will note down some of my concerns.
How it works : When the timer switches on the LED grow light ,the Fan power supply switches ON also.
The LM 555 timer IC is then powered up,also.
As it is configured in ASTABLE operation ( http://www.555-timer-circuits.com/operating-modes.html ) ,
the timer will output a PWM signal of ~ 50% duty cycle ,from its OUTPUT pin (3) to the INPUT (CLK ,pin14) of the
HCF4017B Decade Counter.Every time a new "high" signal is reaching the CLK pin of the decade counter ,
it's output pins will be HIGH ,starting from OUT 0 (pin 3) towards OUT 9 (pin 11) .
As V series Tetras fixture ,features four drivers (and four COBs ) ,not all OUt pins are connected.
Only the Outputs 1 ,3,6 & 9 are used ( pins 3,7,5 & 11 respectively ).
Every output pin is connected to a NPN transistor ,which powers a relay .Each relay powers up a LED driver.
At the last relay (#4 ),before it's base resistor ( R4) ,a line lifts the CLOCK INHIBIT (pin 13)of the decade counter to HIGH.Thus the decade counter stops alternating it's outputs ,and the output 9 (pin 11) stays HIGH .
The circuit will reset itself ,when the LED grow light is switched OFF and back ON again (hopefully ).
The LM 555 timer's frequency is adjusted by the 10M pot , from ~ 1min up to ~10 minutes .
( http://www.ohmslawcalculator.com/555-astable-calculator )
So ,actually the minimum " rise time" is 9 minutes ( 1st driver ON ,after 3min the second,at 6min the third and after 9 min the fourth driver.).The maximum "rise time is ~90 min ( 1st driver ON ,2nd after 30 min ,3rd after 60 min ,4th after 90 min )
Concerns:
-The 555 timer will continue it's PWm astable operation.Will this affect the operation of the decade counter?
(As it's input (CLK,pin 14 ) will continue receiving signals from the timer)
-A fifth relay (or MOSFET/transistor) has to be used to stop the 555 timer ,instead of clock inhibiting the HCF4017B Decade Counter ?
- Will inrush current of the driver deteriorate the contacts of a mechanical relay (sparking ) ?
-The high threshold resistance (1M ) & pot (10M ) used for LM 555 timing allow only for low currents ,thus any induced noise can be an issue ?
.....
I just hope Guod,sees all that and comes with a far better solution,than this one ....
Next one, comes the -more complicated- Approach #2 : DC ( DIM ) side.
Cheers.
Edit :
I've already spotted a mistake :the transistors will switch OFF ,every time the decade counter will proceed to the next output.Instead of transistors ,MOSFETs probably have to be used .So this schematic is wrong.(WAYYYYY Wrong !!!!)
Attachments
-
1.7 MB Views: 1
Last edited: