Slow fade-in circuit...

[rhor...@gmail.com]

On Friday, September 25, 2020 at 5:00:54 PM UTC-5, Jasen Betts wrote:

no. but terminals, strain-releif, and insulation will take up most of the matchbox
size leaving very little room for any active parts. those things scale
with wire-size and voltage, you have not indicated a willingness to
negotiate on those parameters.

Oh, nonsense. The only thing that will carry any significant current is the output stage, which consists of either a Triac or a pair of MOSFETs. Those will be TO-220 cases sitting off the board. They probably don\'t even need heat sinks, given the small current they will be carrying. There is no need for any insulation other than that provided by the board itself and the plastic case. I won\'t use terminals. The wires will solder to the board..

 

[rhor...@gmail.com]

On Friday, September 25, 2020 at 5:00:54 PM UTC-5, Jasen Betts wrote:

size leaving very little room for any active parts. those things scale
with wire-size and voltage, you have not indicated a willingness to

Oh, yeah. Wire size. The fixture is intended to utilize four 5 watt bulbs.. That\'s 20 watts, or 0.167 amperes. Allowing for NEC rules, we apply a factor of 1.2, or 0.2 amperes. Even 32AWG wire can easily handle more than that, but I won\'t go that low. I will use 18AWG stranded wire, which can handle up tp 16 amperes (over 2000 watts) in free air.

 

[Jasen Betts]

On 2020-09-26, rhor...@gmail.com <rhorerles@gmail.com> wrote:

On Friday, September 25, 2020 at 5:00:54 PM UTC-5, Jasen Betts wrote:
size leaving very little room for any active parts. those things scale
with wire-size and voltage, you have not indicated a willingness to

Oh, yeah. Wire size. The fixture is intended to utilize four 5 watt bulbs. That\'s 20 watts, or 0.167 amperes. Allowing for NEC rules, we apply a factor of 1.2, or 0.2 amperes. Even 32AWG wire can easily handle more than that, but I won\'t go that low. I will use 18AWG stranded wire, which can handle up tp 16 amperes (over 2000 watts) in free air.

NEC allows that only if the wire is protected by a suitable fuse or
breaker. do you have a 200mA fuse in your setup that you have not
told us about yet? It\'s only just today I found out that you had
the light switch built into the fan.

--
Jasen.

 

[Phil Allison]

rhor...@gmail.com wrote:

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

** So called \"Dimmable LEDs\" are not all the same nor dimmable
by usual means like triac dimmers or a variac.

Who on Earth would employ a variac these days?

** It was an example of \"usual means\" - and they are used in circumstances where EMI free dimming is needed.


> I used plenty of them back when I was in college,

** I use one every day - so do most service tech.

Most are designed to use a \"trailing edge dimmer\" that chops the AC wave in the reverse sense to a triac. The circuit inside the lamp uses this clue to control the LED current.

See fig 1 for a schematic of such a dimmer.

https://sound-au.com/project157.htm

A trailing edge dimmer can be used with CFL bulbs,

** Cos their operation is closer to tat of a variac.

> It is true leading edge dimmers may not work as well as trailing edge dimmers with LED bulbs, but unless I am mistaken, I believe most consumer dimmers are leading edge dimmers, and they do work with dimmable LED bulbs well enough.

** IME not, makers specifically warn TE dimmers must be used.


...... Phil

 

[server]

On Fri, 25 Sep 2020 17:29:21 -0700 (PDT), \"rhor...@gmail.com\"
<rhorerles@gmail.com> wrote:

On Friday, September 25, 2020 at 4:56:32 PM UTC-5, John Larkin wrote:

Do you feel like building a circuit? It could be done with 5 discrete
parts, or possibly 3.

Yes, of course! I am having a little trouble imagining only 5 components, let alone 3, but I am game to look at anything.

I did find a nice, small buck converter module that is only 7mm x 4.5mm with up to 450V input and 300ma output. It only requires 13 external components, all of which can be 1206 SMDs except for a small inductor. This makes a trailing edge MOSFET based dimmer using a 555 practical. It\'s $2.24, but I can handle that.

Guess nobody wants to play. As usual.

One of these might work:

https://www.dropbox.com/s/616xu5p4z1bw2yb/LED_Soft-Start.JPG?raw=1

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.

https://www.dropbox.com/s/g3kaaaauft986b9/LED_Soft_Follower.JPG?raw=1





--

John Larkin Highland Technology, Inc

Science teaches us to doubt.

Claude Bernard

 

[Piglet]

On 25/09/2020 18:36, John Larkin wrote:

Given that this is an electronic design forum, we should design
something. I could start. Maybe some others can suggest circuits.

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

piglet

 

[Piglet]

On 26/09/2020 03:25, jlarkin@highlandsniptechnology.com wrote:

On Fri, 25 Sep 2020 17:29:21 -0700 (PDT), \"rhor...@gmail.com\"
rhorerles@gmail.com> wrote:

On Friday, September 25, 2020 at 4:56:32 PM UTC-5, John Larkin wrote:

Do you feel like building a circuit? It could be done with 5 discrete
parts, or possibly 3.

Yes, of course! I am having a little trouble imagining only 5 components, let alone 3, but I am game to look at anything.

I did find a nice, small buck converter module that is only 7mm x 4.5mm with up to 450V input and 300ma output. It only requires 13 external components, all of which can be 1206 SMDs except for a small inductor. This makes a trailing edge MOSFET based dimmer using a 555 practical. It\'s $2.24, but I can handle that.

Guess nobody wants to play. As usual.

One of these might work:

https://www.dropbox.com/s/616xu5p4z1bw2yb/LED_Soft-Start.JPG?raw=1

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.

https://www.dropbox.com/s/g3kaaaauft986b9/LED_Soft_Follower.JPG?raw=1

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

piglet

 

[server]

On Sat, 26 Sep 2020 08:56:37 +0100, Piglet <erichpwagner@hotmail.com>
wrote:

On 26/09/2020 03:25, jlarkin@highlandsniptechnology.com wrote:
On Fri, 25 Sep 2020 17:29:21 -0700 (PDT), \"rhor...@gmail.com\"
rhorerles@gmail.com> wrote:

On Friday, September 25, 2020 at 4:56:32 PM UTC-5, John Larkin wrote:

Do you feel like building a circuit? It could be done with 5 discrete
parts, or possibly 3.

Yes, of course! I am having a little trouble imagining only 5 components, let alone 3, but I am game to look at anything.

I did find a nice, small buck converter module that is only 7mm x 4.5mm with up to 450V input and 300ma output. It only requires 13 external components, all of which can be 1206 SMDs except for a small inductor. This makes a trailing edge MOSFET based dimmer using a 555 practical. It\'s $2.24, but I can handle that.

Guess nobody wants to play. As usual.

One of these might work:

https://www.dropbox.com/s/616xu5p4z1bw2yb/LED_Soft-Start.JPG?raw=1

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.

https://www.dropbox.com/s/g3kaaaauft986b9/LED_Soft_Follower.JPG?raw=1






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.





--

John Larkin Highland Technology, Inc

Science teaches us to doubt.

Claude Bernard

 

[server]

On Sat, 26 Sep 2020 08:54:01 +0100, Piglet <erichpwagner@hotmail.com>
wrote:

On 25/09/2020 18:36, John Larkin wrote:
Given that this is an electronic design forum, we should design
something. I could start. Maybe some others can suggest circuits.


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

piglet

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.



--

John Larkin Highland Technology, Inc

Science teaches us to doubt.

Claude Bernard

 

[John S]

On 9/25/2020 6:47 PM, Ricketty C wrote:

On Friday, September 25, 2020 at 4:58:36 PM UTC-4, John S wrote:
On 9/25/2020 1:44 PM, Ricketty C wrote:
On Friday, September 25, 2020 at 1:36:55 PM UTC-4, John Larkin wrote:
On Tue, 22 Sep 2020 21:08:07 -0700 (PDT), \"rhor...@gmail.com\"
rhorerles@gmail.com> wrote:

I need a simple circuit design for a slow fade-in AC dimmer. In a dark theater, suddenly turning on the lights is a somewhat unpleasant experience. I want a simple circuit compatible with 110 volt dimmable LED lights that will slowly ramp the light output over a period of about 5 seconds or so. I have found a number of simple designs that would work for 12VDC LEDs, but none for 110VAC dimmable LED lights.

Given that this is an electronic design forum, we should design
something. I could start. Maybe some others can suggest circuits.

How much power are the LED lights, in total?

A triac dimmer typically won\'t see a low enough impedance when dimming
some LEDs. I have one string of LED spots, with one incandescent left
to make the dimmer work.

I would design using modules. I already posted a link to two. Done and dusted!

Since when is Larkin actually concerned about discussing electronic design other than when he intends to use it as a put down for someone? That is a truly nasty individual.

It\'s no wonder that Larkin is not respected by many here. On a personal level he is the sort of person the world would be better off without. But he has a highly specialized skill in a corner of electronics design that still has some minimal demand, so the world pays him to do that.


You are a good contributor to the group, Ricketty, except when you\'re
not. There are not that many people here that dislike Larkin. He, too,
is a good contributor. He does not post the kind of diatribe as you have
done. You do much better at describing projects than you do at this kind
of attack. Please take my post as constructive criticism.

Fine, but it\'s not accurate. There have been any number of posts by Larkin responding to nothing in particular by others where he talks about his disfavored few being \"nasty\" and other choice terms. That was why I used the term.

It is rather un-evenhanded to call my two, short paragraphs as a \"diatribe\" but ignore the many ugly posts by Larkin.

That\'s on you.

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.

 

[Piglet]

On 26/09/2020 15:56, jlarkin@highlandsniptechnology.com wrote:

On Sat, 26 Sep 2020 08:54:01 +0100, Piglet <erichpwagner@hotmail.com
wrote:

On 25/09/2020 18:36, John Larkin wrote:
Given that this is an electronic design forum, we should design
something. I could start. Maybe some others can suggest circuits.


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

piglet

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.

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

piglet

 

[server]

On Sat, 26 Sep 2020 17:11:54 +0100, Piglet <erichpwagner@hotmail.com>
wrote:

On 26/09/2020 15:56, jlarkin@highlandsniptechnology.com wrote:
On Sat, 26 Sep 2020 08:54:01 +0100, Piglet <erichpwagner@hotmail.com
wrote:

On 25/09/2020 18:36, John Larkin wrote:
Given that this is an electronic design forum, we should design
something. I could start. Maybe some others can suggest circuits.


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

piglet

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.




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?



--

John Larkin Highland Technology, Inc

Science teaches us to doubt.

Claude Bernard

 

[Tabby]

On Saturday, 26 September 2020 17:37:43 UTC+1, jla...@highlandsniptechnology.com wrote:

On Sat, 26 Sep 2020 17:11:54 +0100, Piglet <erichpwagner@hotmail.com
wrote:
On 26/09/2020 15:56, jlarkin@highlandsniptechnology.com wrote:
On Sat, 26 Sep 2020 08:54:01 +0100, Piglet <erichpwagner@hotmail.com
wrote:
On 25/09/2020 18:36, John Larkin wrote:

Given that this is an electronic design forum, we should design
something. I could start. Maybe some others can suggest circuits.


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

piglet

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.




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 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?

You might be able to slow it further with a series cap. It takes the LED 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 slow 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

 

[server]

On Sat, 26 Sep 2020 11:33:10 -0700 (PDT), Tabby <tabbypurr@gmail.com>
wrote:

On Saturday, 26 September 2020 17:37:43 UTC+1, jla...@highlandsniptechnology.com wrote:
On Sat, 26 Sep 2020 17:11:54 +0100, Piglet <erichpwagner@hotmail.com
wrote:
On 26/09/2020 15:56, jlarkin@highlandsniptechnology.com wrote:
On Sat, 26 Sep 2020 08:54:01 +0100, Piglet <erichpwagner@hotmail.com
wrote:
On 25/09/2020 18:36, John Larkin wrote:

Given that this is an electronic design forum, we should design
something. I could start. Maybe some others can suggest circuits.


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

piglet

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.




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 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?

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.

You might be able to slow it further with a series cap. It takes the LED 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 slow it yet again with a PTC Thermistor across the bulb.

NTC-PTC voltage divider. Interesting.

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

Again?

NT

--

John Larkin Highland Technology, Inc

Science teaches us to doubt.

Claude Bernard

 

[Lasse Langwadt Christensen]

lørdag den 26. september 2020 kl. 04.25.53 UTC+2 skrev jla...@highlandsniptechnology.com:

On Fri, 25 Sep 2020 17:29:21 -0700 (PDT), \"rhor...@gmail.com\"
rhorerles@gmail.com> wrote:

On Friday, September 25, 2020 at 4:56:32 PM UTC-5, John Larkin wrote:

Do you feel like building a circuit? It could be done with 5 discrete
parts, or possibly 3.

Yes, of course! I am having a little trouble imagining only 5 components, let alone 3, but I am game to look at anything.

I did find a nice, small buck converter module that is only 7mm x 4.5mm with up to 450V input and 300ma output. It only requires 13 external components, all of which can be 1206 SMDs except for a small inductor. This makes a trailing edge MOSFET based dimmer using a 555 practical. It\'s $2.24, but I can handle that.

Guess nobody wants to play. As usual.

One of these might work:

https://www.dropbox.com/s/616xu5p4z1bw2yb/LED_Soft-Start.JPG?raw=1

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.

https://www.dropbox.com/s/g3kaaaauft986b9/LED_Soft_Follower.JPG?raw=1

https://www.monolithicpower.com/en/documentview/productdocument/index/version/2/document_type/Datasheet/lang/en/sku/MP4001/document_id/2012/

with a cap on the dim input

 

[server]

On Sat, 26 Sep 2020 14:12:30 -0700 (PDT), Lasse Langwadt Christensen
<langwadt@fonz.dk> wrote:

lørdag den 26. september 2020 kl. 04.25.53 UTC+2 skrev jla...@highlandsniptechnology.com:
On Fri, 25 Sep 2020 17:29:21 -0700 (PDT), \"rhor...@gmail.com\"
rhorerles@gmail.com> wrote:

On Friday, September 25, 2020 at 4:56:32 PM UTC-5, John Larkin wrote:

Do you feel like building a circuit? It could be done with 5 discrete
parts, or possibly 3.

Yes, of course! I am having a little trouble imagining only 5 components, let alone 3, but I am game to look at anything.

I did find a nice, small buck converter module that is only 7mm x 4.5mm with up to 450V input and 300ma output. It only requires 13 external components, all of which can be 1206 SMDs except for a small inductor. This makes a trailing edge MOSFET based dimmer using a 555 practical. It\'s $2.24, but I can handle that.

Guess nobody wants to play. As usual.

One of these might work:

https://www.dropbox.com/s/616xu5p4z1bw2yb/LED_Soft-Start.JPG?raw=1

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.

https://www.dropbox.com/s/g3kaaaauft986b9/LED_Soft_Follower.JPG?raw=1


https://www.monolithicpower.com/en/documentview/productdocument/index/version/2/document_type/Datasheet/lang/en/sku/MP4001/document_id/2012/

with a cap on the dim input

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



--

John Larkin Highland Technology, Inc

Science teaches us to doubt.

Claude Bernard

 

[Steve Wilson]

jlarkin@highlandsniptechnology.com wrote:

https://www.monolithicpower.com/en/documentview/productdocument/index/v
ersion/2/document_type/Datasheet/lang/en/sku/MP4001/document_id/2012/

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.

 

[rhor...@gmail.com]

On Saturday, September 26, 2020 at 2:56:43 AM UTC-5, piglet wrote:
> Inventive fersure

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 components, this might not work, but otherwise, it should. It definitely will work with an incandescent bulb and definitely not with a fluoresent, I expect.

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

Yeah, probbably, but not much.

 

[rhor...@gmail.com]

On Saturday, September 26, 2020 at 9:56:55 AM UTC-5, jla...@highlandsniptechnology.com wrote:

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

Well, I do have several standard dimmers that I just replaced with wifi dimmers. 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 boards. I guess I am kind of anal that way.

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.

Hmm. Unless temperature control is the entire purpose of the device, I dislike temperature dependent solutions. Again, maybe I am just anal that way, but environment can cause a lot of variability, especially with low temperature solutions, and high temperature solutions bring their own set of problems. 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 not only unicorn free, it is elegant and very reliable. It also completely eliminates the question of triggering and timing. *IF* it works.

> Science teaches us to doubt.

And doubt teaches us how to do science.

 

[rhor...@gmail.com]

On Saturday, September 26, 2020 at 3:20:20 PM UTC-5, jla...@highlandsniptechnology.com wrote:

NTC would sure be simple, but dramatic home theatre lighting may have
crucial timing requirements.

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.