Help me with circuit that enables dimmer capability to a group of white LEDs

[edaedaeda]

Here is a concept of a circuit where my circuit can interpret any type of household dimmer switch (triac) adjusting the output voltage of my lm317.
The intent is to enable dimmer capability of a group of 6 150mA white led's.
Truthfully I am not feeling 100% on this so any input will be outstanding. Already I have had some great feedback on an earlier version so I am optimistic of the level of expertise that can be found on this forum... Particularly in my amigo Pranam. Also, critisism is very welcome.

Eda

 

[rikie_rizza]

Is this plausable?

Oh dude, come on!
If you dimmed the AC power, in some way your loosing many other valuable parameter such as current and AC frequency. your led will blink at very dimmed level.

Remember, to control led intensity you should control the current not the voltage.

 

[Eugen_E]

Re: Is this plausable?

The transformer may act abnormal if feeded from a dimmer, because it will be powered from pulses possible with a DC component.
Why you added the backward transistor for the LM317 ADJ control?

On one hand you want a dimmer and on the other you use a regulator, it's strange.

It's easier to build a dimmer in the DC part of the circuit.

 

[edaedaeda]

Re: Is this plausable?

Thanks for the input guys,
I have much to learn.....

The Transistor was just an idea I was throwing out there... Glad I asked.
I know that I should control dimming with PWM. How can this be done from a household dimmer switch? What I want to do with this circuit is use led lights to replace a recessesed light in my home. Using the household dimmer is essential.

Eda

 

[laktronics]

Re: Is this plausable?

Hi,
May be your idea would work. Use another parallel rectifier bridge without any output capacitance and connect its output to the transistor base resistor R7. Interchange the transistor emitter /collector connections.

See what you get now.

Regards,
Laktronics

 

[edaedaeda]

Re: Is this plausable?

Laktronics,
I appreciate your input on this. Here is what I put together based on your feedback. Tell me what you think.

Eda

 

[laktronics]

Re: Is this plausable?

Hi,
Yes, I was talking about similar arrangement. But you need not have a separate winding and fuse arrangement. You can connect the second bridge directly from the output of the first fuse iteself. Also,
1.You need not put a pair of capacitors like C1/C2 and C3/C4. Use only one like C1 and C3.
2. You may need to increase thevalue of C1 to around 470µF, 30V for reducing ripple at 317 output.
3. You can probabaly go for an unreg. voltage of 16 to 20 V using a secondary voltage of 0-16 V AC.
4. You have not used the standard voltage setting resistors for 317 IC. How much peak voltage drive are you expecting for the LED circuit? In the present case when the transistor goes to cutoff, Vadj. terminal will be floating and in such a case, some regulator output may go to its input volatge and if so, it may be OK in this case. But I do not know how 317 regulator would behave if its Adj. pin is floating. Please check on this aspect. If you do not want it that way, use the voltage setting resistors to give required peak output voltage.

Regards,
Laktronics

 

[edaedaeda]

Re: Is this plausable?

Hey all,
Thanks for all the input you have helped out quite a bit in my understand what needs to be done here. Another idea I am considering is using an ac frequency to voltage converter. What would be a good chip or approach to this?

Thanks again!

Eda