Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

unusual dimmer switch design

Status
Not open for further replies.

Roger44

Member
Hello
I unearthed from the bottom of the drawer a dimmer switch dating no doubt from the beginning of the 80's. A big rotating button and an on-off switch. What surprises me is the fact that you just insert it into the phase line, ie you don't need phase + neutral.

It has a 500k lin pot, 3 largish caps, a thick wire coil wound onto a piece of ferrite rod, and 2 or 3 res and diodes I can't see clearly. It still works and it is very quick and simple to modify light intensity.

Just for curiosity, does anyone have an idea of the priciple of how it works?

Thanks
 
The phase control dimmer utilizing a Triac, Diac and a few passive components has been around since the mid 1970s, probably earlier.
Google Triac Dimmer Schematic
 
The standard Triac dimmer circuits do not need a connection to neutral.
 
The standard Triac dimmer circuits do not need a connection to neutral.

But it 'has' one :D

The neutral connection is via the load (assuming the switch is in the live), or the neutral connection is direct, and the live is via the load.

Live and Neutral are just relative terms, and meaningless as far as electricity goes, it's just a 'potential difference' between two points.

The dimmer simply needs power to work, it gets this from the voltage drop (potential difference) across itself - even turned fully ON there's still a voltage drop across the dimmer.

Things get a little more complicated when you add more electronics to it, such as remote control facilities - you then need to generate a suitable voltage to power that electronics - where a simple dimmer is VERY crude.
 
You are quite right, I've modified the standard circuit to make it clear for myself.

**broken link removed**

Thanks everbody, sorry to have wasted your time
 
Hi,

It was not a waste of time for me because i never replied.
Oops, i replied too now too <chuckle>

Seriously though, it was not a waste of time as this is a general electronics interest question and you've found an answer, and i think others have this question too as i've seen questions like this come up from time to time.

It's not always done the way the schematics here show, but that's probably the way yours is done because of the age of the thing.

One of the problems of those kind from back then is that the dimmer can not always dim to a very low level, but has some lower limit that does not dim the bulb to a very dim state. Newer models can dim to a lower light level because they use a different way of triggering the triac. You can test yours to see how dim the bulb gets...maybe you got one of the better models.
 
Last edited:
Diacs must have become unpopular, none of my regular suppliers stock them, I have seen a few circuits that use neon bulbs instead.
There was a reote control dimmer in a electronics mag a while back, the 5v electronics supply also came through the load, its only a fewma.
 
Hi MrAl

I've measured the light level with a Hammamatsu near perfect eye-response photodiode. (>100€!!!)
122 fully lit and 0.1 dimmed fully down, so this one seems to dim well. You can still see a orange glow.

The last dimmer I made used an optocoupler + triac in a 6 pin white IC. So easy.
 
Last edited:
Hi MrAl

I've measured the light level with a Hammamatsu near perfect eye-response photodiode. (>100€!!!)
122 fully lit and 0.1 dimmed fully down, so this one seems to dim well. You can still see a orange glow.

The last dimmer I made used an optocoupler + triac in a 6 pin white IC. So easy.


Hi,

That sounds like one of the better ones. I had at least one myself many years ago and it only dimmed down to a somewhat low level but not what we would really call dim as it was still too bright really. (maybe about 33 percent, which isnt that dim).

I had designed a "dimmer" for a light duty speed controller because i could not find a ready made dimmer that would go down low enough. It used just a single LM358 (an old time favorite op amp of mine ha ha) to do all the control, and did not need an isolation stage. I used the design later in a much higher powered speed controller with a 25 amp triac. This kind of design takes it right down to zero, so using a bulb you could see the filament barely glow on some setting just above the lowest. That's the way they should be i think, but maybe we should check current at that setting to make sure it does not draw excessive current peaks at the low settings (bulb resistance can be 10 times lower when the filament is cold).
I had also designed a dimmer using a microcontroller when i first started getting into microcontrollers. It was designed to run offline without any transformer. In theory it works good, but i never got around to actually building this circuit because i did not need another dimmer or speed controller yet. It's very simple though, just one 8 pin IC chip, a few resistors and capacitors, zener or voltage regulator, potentiometer of course, and triac.
 
I had also designed a dimmer using a microcontroller when i first started getting into microcontrollers. It was designed to run offline without any transformer. In theory it works good, but i never got around to actually building this circuit because i did not need another dimmer or speed controller yet. It's very simple though, just one 8 pin IC chip, a few resistors and capacitors, zener or voltage regulator, potentiometer of course, and triac.

Aren't there such examples in the MicroChip application notes?.

Just had a quick look, much of the old stuff seems to have gone now? - but there is one example:

**broken link removed**

It's certainly been a common project over the years.
 
Aren't there such examples in the MicroChip application notes?.

Just had a quick look, much of the old stuff seems to have gone now? - but there is one example:

**broken link removed**

It's certainly been a common project over the years.


Hi,

Yeah "Now", but im not sure if that was there when i first started with the PIC chip design for the triac controller.
That could be a good design, but still, i would trust my own design over anything i read on the web :)
It was also just interesting to think it out in order to figure out what is needed to do such a thing. As you can see, it's quite a simple circuit problem as it is more in the programming itself. Features that i thought were good i built into the code including fail safe zero crossing detection and stuff like that. Not sure what that web design has as i dont really want to take the time to go over the program code right now...maybe another time. Interesting though.

Yeah people always seem to be interested in this kind of circuit, with or without uC chip.
 
I think min brightness on cheeper designs is down to circuit limitations.
 
Hi,

Yeah and it is also interesting that if you get too close to the max time delay using a delay circuit or delay code and the line frequency gets a tiny bit higher, it could make the phase firing angle much larger than the user had set.

One of the careful design points is to make sure that the unit can not accidentally 'skip' a half cycle or else you end up pumping a half wave rectified sine into the load. A light bulb might not mind, but a motor probably wont like this :)
This can happen with a delay that for some reason ends up being longer than one half cycle. It skips the whole half cycle and triggers the triac on the next half cycle, and this happens once per cycle so we end up with half wave rectification. So the design has to give higher priority to the zero crossing detection (for resetting the delay) rather than the delay itself.
 
Diacs were also used in strobe lights. Once the breakover voltage was reached, it would discharge a small cap into a trigger transformer, which would initiate the discharge inside the xenon tube.
 
Diacs were also used in strobe lights. Once the breakover voltage was reached, it would discharge a small cap into a trigger transformer, which would initiate the discharge inside the xenon tube.

Basically the same as in a light dimmer :D (except the capacitor is 'discharged' in the triac's date).

It would also be a very old and crude strobe, anything remotely 'modern' would use more electronics.

But it's yet again a very small market.
 
Indeed it is old. I first saw this technique being used in the late 70s .

But it showed what could be achieved with a discrete-only circuit.
 
Hi again,

After reading the last two or so posts i almost want to call the LM358 a "discrete part" because everything is a microcontroller these days so when i talk about a simple op amp it's almost like it's a discrete part now.

Yes we can do a lot with discrete parts still, but it's amazing what we can do with microcontrollers.
The only downside i can think of is that devices made with flash microcontrollers wont stand the test of time like other things can because their bits will start flipping after some 30 or 40 years, and that's with the more modern ones. Of course we probably wont actually need the device anymore, but that means that if it becomes a museum piece it will just be for show and wont work anymore unless the curator knows how to reprogram it and has the right code for it. Maybe a new job opening for a programmer :)
 
I have been working on a strobe for vehicle use, this can be setup to trigger at a preset shaft angle so engine problems can be traced and viewed, I'm also thinking of adding a slowmo function so you can see something rotating quickly in slomo to pick up problems such as non rotating vlave stems, worn rockers and the like.
I have an altai rack mounted disco strobe from the 80's, I think it uses a diac.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top