Slow fade-in circuit

-------------------------

..... Phil

Guess nobody wants to play. As usual.

One of these might work:

The lower one starts at half brightness, because of the mosfet substrate diode. Upper one probably doesn't need R2 if you use a protected-gate fet.

I don't have time to Spice these, but either should work.

There are some variants, like an RC and a source follower maybe.

I already started - didn't you see my design?

piglet

Inventive fersure but the mosfet is going to dissipate a few watts for a few seconds so will need some heatsinking.

piglet

Maybe.

As I said, I didn't have time to Spice any of those ideas. Take them as starting points. Brainstorm, please. Play.

Sorry, missed that one. I'd delete the EMI filter, except with a standard dimmer as the starting point, it's free already.

What about ramping down the value of Rt over time? Some variation on one of my circuits? NTC thermistor?

Maybe the entire circuit could be one NTC thermistor, one of those inrush limiter things.

How many is "any number" of posts? Have you counted them? What is his ratio of "ugly" posts vs total posts?

And you have not responded to my earlier post asking "How many is many? Have you taken a census?"

Give us some data.

That's on you.

Yes, I think Tabby (or someone) suggested an NTC already.

piglet

Maybe heat sink it to a chunk of aluminum or two to slow down the turn-on?

With 20w of load a biggish one or 2 may be slow enough as is. Heatsinking i t would require great thermal contact to be effective in slowing interior h eat-up, most THs aren't a good shape for that. Laser slice them? :)

You might be able to slow it further with a series cap. It takes the LED cl oser to the point where it dims, ie lower mains v, and Xc + Rth is going to be larger at any time during warm up. And if you really must you could slo w it yet again with a PTC Thermistor across the bulb.

There are so many ways it can be done. And Ricky is getting snipey again.

NT

My wild guess is that a bit of heat sinking would slow it down. One NTC would sure be simple, but dramatic home theatre lighting may have crucial timing requirements.

NTC-PTC voltage divider. Interesting.

Again?

sniptechnology.com:

s, let alone 3, but I am game to look at anything.

with up to 450V input and 300ma output. It only requires 13 external compo nents, all of which can be 1206 SMDs except for a small inductor. This mak es a trailing edge MOSFET based dimmer using a 555 practical. It's $2.24, but I can handle that.

with a cap on the dim input

The data sheet prohibits me from copying or printing it. OK, deal.

They put it on the web, so they intend for it to be copied. Besides, you are just posting the link, so you are innocent.

Yes, it is surely that. The thought of a DC solution never dawned on me. Does anyone have a schematic of a typical LED bulb? If they have any AC co mponents, this might not work, but otherwise, it should. It definitely wil l work with an incandescent bulb and definitely not with a fluoresent, I ex pect.

l need some heatsinking.

Yeah, probbably, but not much.

Well, I do have several standard dimmers that I just replaced with wifi dim mers. Free is always nice. OTOH, I dislike hanging components off in the air. Unless it is a very simple mod, I would rather print my own custom bo ards. I guess I am kind of anal that way.

Hmm. Unless temperature control is the entire purpose of the device, I dis like temperature dependent solutions. Again, maybe I am just anal that way , but environment can cause a lot of variability, especially with low tempe rature solutions, and high temperature solutions bring their own set of pro blems. Please understand I am not trying to be contrary, here. The more I think about it, the more I like the idea of a simple DC solution. It's no t only unicorn free, it is elegant and very reliable. It also completely e liminates the question of triggering and timing. *IF* it works.

And doubt teaches us how to do science.

Uh-uh. This is just overhead lighting in a home movie theater. 'No drama, at all. The idea is simply to be kind to dark-accustomed eyes at the end of a show. Bright ights going up suddenly can actually be a bit painful. Allowing the pupils time to contract is much more comfortable.

it would require great thermal contact to be effective in slowing interior heat-up, most THs aren't a good shape for that. Laser slice them? :)

closer to the point where it dims, ie lower mains v, and Xc + Rth is going to be larger at any time during warm up. And if you really must you could s low it yet again with a PTC Thermistor across the bulb.

Yes, I suggest a method of solving the problem that doesn't need to be spic ed, is guaranteed to work out of the box and is very affordable... that's s nipey.

Larkin doesn't take the time to simulate the circuits to show they really w ork and I'm not going to take the time to find the guy an exact part number .

This is not a project I would want to spend significant time on designing a circuit that can have hidden risk of fire and/or shock when there are off the shelf solutions.

If you don't mind that sort of risk, there are off the shelf solutions on a liexpress for a song, probably cheaper than the parts to build a dimmer and won't limit the circuit to only dimming and at just one rate.

This conversation has dragged out for days already. I expect it will be so me time longer before anyone has a circuit that might actually work. It's not the sort of thing I would be wanting to design a circuit for. I would want to get it done and move on to more useful endeavors. That's why I rec ommended a module.

I think a thermistor might work with some cheapo LED bulbs but the good-quality ones have an integrated buck converter that likely has some undervoltage lock-out on the rectified mains input.

I think those type which are "dimmable" accomplish it through an auxiliary circuit to sense if the input is chopped and hold-off the lockout if so and translate the angle of the zero-crossing to an appropriate set-point for the regulator.