View Full Version : Problems dimming lower than 7%
I use two channels to dim my leds using L1L2 but when they are at 7% or lower they starts to twinkle irregularly. Anyone have an idea what could be wrong?
Are running dimmable drivers with integral 0-10 control? if so, I would be looking at your drivers first. I suspect this is where your problem is!
Matthias
20.01.2013, 10:45
from what LED lamps are we talking? DIY? please more details
Diy from stevesleds using quaddriver and their pwm controller card
Med vänlig hälsning, Björn
http://stevesleds.com/uploads/Steves_LEDs_Aquarium_Controller_Interface.pdf
http://stevesleds.com/uploads/Quad_LED_Driver_Guide_11.27.11.pdf
Using profilux 3.1a 5.14
Anything else you need?
Thanks for helping me!!
Med vänlig hälsning, Björn
Matthias
20.01.2013, 12:14
than this driver is maybe not able to dim down so much
-> increase the lower voltage of L1 until it works stable at 1%
Thanks Mathias. The driver should be able to dim to zero. This is what stevesleds says: The wire connecting the controller to the ACI is picking up stray voltage/interference, or the wire is too thin gauge.
I think it could be the wire. I use the ghl wire that should be used with tunze pumps and that wire is very thin. Is the wire to the ghl evg thicker and is it shielded? Is it possible to buy just the rj12 cable?
Med vänlig hälsning, Björn
Anyone know if the rj12 cable that uses to the ghl evg is thicker than the rj12 cable that controls tunze pumps?
I suggest it´s not the wire itself. Diming signal is a voltage signal with very low current, so the thickness of the wiring should not matter if you don´t have a very long cable.
And even it would have an influence, the effect would be a lower signal, not a higher one. So I think you have an induction problem from another cable nearby or a leak current from another circuit (poor mass connection)
I suggest it´s not the wire itself. Diming signal is a voltage signal with very low current, so the thickness of the wiring should not matter if you don´t have a very long cable.
And even it would have an influence, the effect would be a lower signal, not a higher one. So I think you have an induction problem from another cable nearby or a leak current from another circuit (poor mass connection)
Thanks for helping me! The wire from the drivers/Controller is a not shielded "22 gauge" (pretty thick in my eyes) cable and It's about 1,5m long it is soldered to the ghl tunze rj12 cable that in my eyes is very thin and not shielded.
Maybe I should change the cable from the profilux to the drivers to a shielded one?!
Thanks
Med vänlig hälsning, Björn
Maybe I should change the cable from the profilux to the drivers to a shielded one?!
This is not so easy due you need to have such a cable first.
For test purposes I would try it first:
- have a s much distance as possible between power cables and the signal cable
- use a tube or an old screen from a screened cable to protect signal cable temporarily. You should ground it on one side.
- don´t know where you located, but some countries have plugs you can turn by 180° - if yes, try this also (changing power and neutral wire)
You also should check if the ground of the drivers is available and o.k.
To my pwm card from stevesleds i use 2 channels and I need on ch1 (+) positive and (-) negative ; ch2 (+) positive and (-) negative
3534
What is the correct connection from the profilux (pinout on the picture)
3533
Is it ok to use the same (-) negative on both my channels? As I see it on the pinout there is only one negative from the profilux (pin 3)
I really want to solve this problem
Anyone have an idea?
Regards Björn
Med vänlig hälsning, Björn
This is not so easy due you need to have such a cable first.
For test purposes I would try it first:
- have a s much distance as possible between power cables and the signal cable
- use a tube or an old screen from a screened cable to protect signal cable temporarily. You should ground it on one side.
- don´t know where you located, but some countries have plugs you can turn by 180° - if yes, try this also (changing power and neutral wire)
You also should check if the ground of the drivers is available and o.k.
Thanks for helping me!
Stevesleds says:
Quote:The wire connecting the controller to the ACI is picking up stray voltage/interference, or the wire is too thin gauge.
The driver won't blink unless it is over heating. End quote
One more think is when i unplug the rj12 from the profilux my ch1 is on about 8% and my ch2 is 0%
Thanks
Med vänlig hälsning, Björn
Another quote from stevesleds
You may want to contact the manufacturer. I think Proflux can output PWM, which plugs directly into the driver with nothing else required.
Is that correct?
Regards Björn
Med vänlig hälsning, Björn
I think Proflux can output PWM,
Not directly but with the help of the PL-0776 LEDControl4Passive
Not directly but with the help of the PL-0776 LEDControl4Passive
Ok, thanks! That was also my thougt. I have this one3542
instead of the ghl PL-0776 LEDControl4Passive - maybe that is my problem?!
Regards Björn
Med vänlig hälsning, Björn
Now I bought the pl-0776
I'm gonna replace my stevesleds controller with that one and I really hope this is gonna work for me to dim down to 0% (it should work with the drivers)
Any thougts about that?
Regards
Björn
Björn please keep us posted. I believe that's the module you need to connect to StevesLED. Please take pictures of what you do and how you do it so others who use StevesLED can follow your lead.
Thank you in advance,
Allen
Björn please keep us posted. I believe that's the module you need to connect to StevesLED. Please take pictures of what you do and how you do it so others who use StevesLED can follow your lead.
Thank you in advance,
Allen
Hi Allen,
Hopefully I can do the installation tomorrow and I can also take some pictures and post here.
/Björn
3550
Now I tried to connect the profilux LEDControl4Passive to my drivers.
3551
The left RJ12 to profilux L1L2 port
LED-driver connector:
No1 to my drivers ground that is connected to my powersupply(dc-) stevesleds
No2 to Led driver channel 1 PWM input (stevesleds)
No3 to Led driver channel 2 PWM input (stevesleds)
When I turn everything on no lights at all! The profilux says it should be lights and I've tried different voltages and percents.
Connected the stevesleds controller back in again and the lights went on but not able to dim down to 0% as I want. Works pretty Well down to 7-8% then it starts to twinkle irregularly :(
Anyone have an idea??
I´m not sure which of the SteveLEDs board you try to connect. I suggest it´s the http://www.stevesleds.com/uploads/Single_LED_Driver_Guide_11.27.11.pdf (or a similar multi-output board)
In this case the wiring should be as follows (taken from documentation):
- The left RJ12 to profilux L1L2 port
- LEDControl4Passive #1 to driver connector No4 (DC 5V+)
- LEDControl4Passive #2 to driver connector No2 (PWM input)
Works pretty Well down to 7-8% then it starts to twinkle irregularly
I´m not sure at this but I think this is a problem of the LED driver itself
I´m not sure which of the SteveLEDs board you try to connect. I suggest it´s the http://www.stevesleds.com/uploads/Single_LED_Driver_Guide_11.27.11.pdf (or a similar multi-output board)
In this case the wiring should be as follows (taken from documentation):
- The left RJ12 to profilux L1L2 port
- LEDControl4Passive #1 to driver connector No4 (DC 5V+)
- LEDControl4Passive #2 to driver connector No2 (PWM input)
I´m not sure at this but I think this is a problem of the LED driver itself
Hi Gunther
I use two of these:
http://stevesleds.com/uploads/Quad_LED_Driver_Guide_11.27.11.pdf
And I am a bit confused of what you said about:
LEDControl4Passive #1 to driver connector No4 (DC 5V+)
- LEDControl4Passive #2 to driver connector No2 (PWM input)
In my manual to Ledcontrol4passive it says #1 led driver minus all channel #2-#5 led driver channel 1-4
3552
3553
Regards
Björn
And I am a bit confused of what you said about:
...
In my manual to Ledcontrol4passive it says #1 led driver minus all channel #2-#5 led driver channel 1-4
Ups ... you are right, my fault.
Unfortunately the LEDControl4Passive is working with common ground, while the LED driver is working with PWM2ground, so it´s not so easy to connect. I think about it ....
Ups ... you are right, my fault.
Unfortunately the LEDControl4Passive is working with common ground, while the LED driver is working with PWM2ground, so it´s not so easy to connect. I think about it ....
Thank you so much for helping!!
3554
This should do the trick. Unfortunately I have no glue about the needed resistiors.
I would try it with an 1 KOhm ressistor or a 0-10k potentiometer. If the light stays on, it´s to low, if it stays off, it´s to high.
The Rx is only reccomended if you use two power supplies to prevent leckage current via the LEDControl
If I want to change drivers to run my leds. Any suggestions to use with ghl LEDcontrol4passive. I use two channels running 28x3w leds on each channel.
3555
Do you know why ghl4passive leds makes led blinking when we set lighting above 5% ?
I use typical driver with uC MBI6651 and 68uH. With other controllers , which I've tested , there was not simillar problems. It's looks like P4L has wrong algoritm or PWM signal.
Sent from my iPhone 5 using Tapatalk
Do you know why ghl4passive leds makes led blinking when we set lighting above 5% ?
Does this also occur with the pullups shown in the manual?
Not yet.
From where should I get V+ ?
From inside Passive4leds? +5 or Vin of driver ( +24V ) ?
In application note for this driver, there is no any information about pullup for DIM channel.
From inside Passive4leds? +5 or Vin of driver ( +24V ) ?
From your driver. It´s similar to my picture in post #25
In application note for this driver, there is no any information about pullup for DIM channel.
See Manual (http://support.aquariumcomputer.com/german/download.php?did=181), image on second page
Hints about dinmension of the resistor:
http://forum.aquariumcomputer.com/showthread.php?6124-LEDControl4-Passive-and-Meanwell-LDD-drivers&p=41019#post41019
http://forum.aquariumcomputer.com/showthread.php?4113-Has-Anyone-attempted-to-connect-Meanwell-ELN-60-48P-drivers-to-an-LED-Control4&p=28715#post28715
According to #2 the resistors should not be lower that 2k
it's not impossible.
Inside my drivers I don't have any additional V+.
Look on driver schema ( it's only 1 channel, I've made a 4 channels drivers where all PWM pin are connected together ):
3559
it's not impossible.
:confused:
Inside my drivers I don't have any additional V+.
There is a +24V
If I want to change drivers to run my leds. Any suggestions to use with ghl LEDcontrol4passive. I use two channels running 28x3w leds on each channel.
3555
My day off working with the aquarium leds/profilux/LEDcontrol4passive ended with a destroyed driver. I ordered a new one which I expect will arrive this week. I am then going to try again. I really want my leds to be able to dim down to 0% so that I can use my Simu L with sunrise/sunset/thunderstorms.
Gunther - You put in a nice pic in post 25 about using a resistor between pwm+ and 5V+. Do you really think that could do it? When I tried to connect the LEDcontrol4passive without the resistor I got 0% light only (I couldn´t get the lights to work at all...)
Regards Björn
Gunther - You put in a nice pic in post 25 about using a resistor between pwm+ and 5V+. Do you really think that could do it?
It should, but find out the right resistor value may be not so easy.
When I tried to connect the LEDcontrol4passive without the resistor I got 0% light only (I couldn´t get the lights to work at all...)
That was a very bad idea ......
If you try it without resistor and no connection between 2+4 nothing will happen for sure and the LEDs will stay dark (that´s what I take from Steve´s manual).
If you connect 2+4 without resistor and witout LEDControl4Passive the LEDs should have 100% - if not, the driver is propably broken - anyway, disconnect it fast in that case!
Now you can set LEDControl4Passive to 100% and connect it to 2/4 wire, the LEDs now should also have 100% - if not, the LEDControl4Active is propably broken - anyway, disconnect it fast in that case!
The resistor only is used to protect the driver´s power supply and the LEDControl4Passive against the +5VDC power source. If you connect 2+4 without resistor to the LEDControl, the driver and/or the LEDControl4Passive may be destroyed due it drops the +5V to ground for dimming. The resulting current without resistor may be much to hight for both devices.
3582
This is the connection I used when the lights didn't work at all (before the driver died)
Could it be something with the ground (mixing ground 24v- and ground from profilux?)
Regards Björn
This is the connection I used when the lights didn't work at all
This cant work (never ever). Connection 4 of the drivers is open, so there is no power for the PWM
Could it be something with the ground (mixing ground 24v- and ground from profilux?)
No. The LCD4P has no connected ground on the PWM side. The LCD4P is like an array of switches as shown in the pic below. It has no connection to the 1-10V channel inputs.
The pic is simplified, but it shows that if you have not connected any power to the circuit, so the input on pin 2 of the driver gets no signal and the LEDs stay off.
3583
What I have done with the resistor in the pic from post #25 is as follows:
If the switch of the LCD4P is open, pin 4 and 2 are connected via the resistor and Pin #2 has +5V. In this case the driver will output 100%
When the switch of the LCD4P closes, it connects the +5V to ground and the voltage will drop down. The resistor limits the current to protect the 5V power supply and the switch in the LCD4P. If the voltage is 0V, the the driver will output 0% aka it is off.
Doing that on/off thing a hundred times a second with various on/off times is called PWM and will dim the LEDs (in fact, it´s on/off switching and not dimming in real, but LEDs are fast enough to have it look like it´s dimmed).
Powered by vBulletin® Version 4.2.5 Copyright © 2024 vBulletin Solutions, Inc. All rights reserved.