(Another) DIY LED Controller - Simple Arduino Style

Ok so I got the Arduino environment all set up. Does this tell you errors in your code like some other dev environments?
 
I used aquarium silicone, yes. It was a pretty big dab of silicone on the chip and exposed wires and it worked for a few months. I might try the epoxy+silicone and see how that holds up. I just got my 1wire bus boards so it is time to rewire the probes anyway.

I work with the Arduino IDE. It isn't as good as eclipse, but not that bad either.
 
Woohoo I got my first piece of the project!...The DS1307 RTC...yep that's right I now have a clock....LOL
 
Woohoo I got my first piece of the project!...The DS1307 RTC...yep that's right I now have a clock....LOL

If we have the same DS1307... then it needs assembly? its really requires small assembly, but a lot has issues with the battery... so make sure + and - is correct and solder points are good.

Thats all...
 
excellent info here Katchuppoy!

One question, can this controller make LEDs undergo softstart or softstop?

I saw your post on another thread, and I believe the answer is the same.
If you are using a non-dimmable driver then the answer is "no"

this setup is designed for dimmable drivers.

let me ask you this? You dont like the on and off of your setup? Define soft start? like 1 minute ramp up? 30 seconds? 5 seconds?

Tell me...
 
I saw your post on another thread, and I believe the answer is the same.
If you are using a non-dimmable driver then the answer is "no"

this setup is designed for dimmable drivers.

let me ask you this? You dont like the on and off of your setup? Define soft start? like 1 minute ramp up? 30 seconds? 5 seconds?

Tell me...

I have nothing against dimmiable drivers. Since I had a constant current driver, I figured there may be a way out here :). Looks like I may have to go with dimable driver.
My LEDs come on instantly and go off instantly. All of them are run using a single driver. Kind of photoshock if you will!
Soft start could be anywhere between 5 min to 30 min. 30 mins is better.
 
I have nothing against dimmiable drivers. Since I had a constant current driver, I figured there may be a way out here :). Looks like I may have to go with dimable driver.
My LEDs come on instantly and go off instantly. All of them are run using a single driver. Kind of photoshock if you will!
Soft start could be anywhere between 5 min to 30 min. 30 mins is better.

Im so sorry, the non dimmable ones are very limited to what you can do with them. Maybe sell them and buy a dimmable one?
 
Yeh! That's the plan.

that's why I wanted to know if your controller will work with dimmable driver to give soft start or soft stop. Thanks for answering my question.
 
A transistor inline with your Leds connected to the PWM should allow you to dim and control the LEDS, the PWM causes the transistor to OPEN and CLOSE (turn off and on) rapidly so the faster your pulse (higher frequency) the brighter your LEDs should be the slower the pulse (lower frequency) the dimmer they should get. I havent tried it, but in theory this should work, if i'm wrong I hope someone will chime in. Im awaiting the delivery of my LEDs and heatsinks so I can get my tank up and running.

EDIT: PS Subscribed to this cant wait to start up my project and hopefully I can help contribute and we can build this up more.
 
A transistor inline with your Leds connected to the PWM should allow you to dim and control the LEDS, the PWM causes the transistor to OPEN and CLOSE (turn off and on) rapidly so the faster your pulse (higher frequency) the brighter your LEDs should be the slower the pulse (lower frequency) the dimmer they should get. I havent tried it, but in theory this should work, if i'm wrong I hope someone will chime in. Im awaiting the delivery of my LEDs and heatsinks so I can get my tank up and running.

EDIT: PS Subscribed to this cant wait to start up my project and hopefully I can help contribute and we can build this up more.

PWM stands for pulse WIDTH modulation. It's not the frequency, but the pulse width that is varied. The brightness depends on the ratio of time the the LEDs are on, to the time they are off. The wider the "on" pulse width, the greater this ratio, and the brighter the LEDs will be. When the ratio is one, they are on full brightness, when it's zero, they are fully off. :beer:
 
A transistor inline with your Leds connected to the PWM should allow you to dim and control the LEDS, the PWM causes the transistor to OPEN and CLOSE (turn off and on) rapidly so the faster your pulse (higher frequency) the brighter your LEDs should be the slower the pulse (lower frequency) the dimmer they should get. I havent tried it, but in theory this should work, if i'm wrong I hope someone will chime in. Im awaiting the delivery of my LEDs and heatsinks so I can get my tank up and running.

EDIT: PS Subscribed to this cant wait to start up my project and hopefully I can help contribute and we can build this up more.

IMHO a very bad idea. When the transistor opens the lines the constant current driver will output maximum voltage to try and get the current up. Snce the sting is open it can't increase the current and keeps raising the voltage. When the transistor is then turned on maximum voltage will briefly be applied to the string until the driver can get the current back under control. So for a very small time you will possibly be over driving your LEDs.

If you were to do the transistor thing do what someone above (kcress?) posted and shunt the current around the LEDs.
 
PWM is a square wave and can be measured in Hz therefore it is a frequency. But yes we are on the same page.

Sorry, but it's not square. Square waves have a mark/space ratio of 1:1, i.e 50% - that corresponds to half power/brightness. Varying the pulse width makes it non-square. btw: this all happens at a constant frequency.
 
IMHO a very bad idea. When the transistor opens the lines the constant current driver will output maximum voltage to try and get the current up. Snce the sting is open it can't increase the current and keeps raising the voltage. When the transistor is then turned on maximum voltage will briefly be applied to the string until the driver can get the current back under control. So for a very small time you will possibly be over driving your LEDs.

If you were to do the transistor thing do what someone above (kcress?) posted and shunt the current around the LEDs.

Awesome thanks FishMan this was the part I was un sure of, didnt know if the driver would overdrive the LEDs or not...
 
This is my 2 cents.
Even though it is a variable rectangular pulse from the supply, but when it arrives to the LED , partly due to parasitic conductance and capacitance, it could very well be transient waveform that might like Fishman says that it could over shot and overdrive the LED for a brief duration. I could be wrong here because the frequency may not be that high to cause this. It would be nice to have a scope to see what exactly like.
 
Last edited:
IMHO a very bad idea. When the transistor opens the lines the constant current driver will output maximum voltage to try and get the current up. Snce the sting is open it can't increase the current and keeps raising the voltage. When the transistor is then turned on maximum voltage will briefly be applied to the string until the driver can get the current back under control. So for a very small time you will possibly be over driving your LEDs.

If you were to do the transistor thing do what someone above (kcress?) posted and shunt the current around the LEDs.

+10 with Fishman... Its a bad idea... the constant driver will compensate for the loss of current, and when it comes back, it will become too much for the LEDs?

thats how i understand it, maybe experts here can partake...
 
Tagging along, would love to have a solution to the 5v PWM signal converted to 10vdc analog for the meanwell D drivers.

Why convert if you can supply the 10V yourself. I believe this is the case even before arduino, since you are adjusting your driver via a pot and a 10v supply?

Ooops. pls ignore... I understand now the question... if it can convert pwm to analog...
 
Back
Top