What's the modern way to dim AC lights?

About 20 years ago I developed a product that, among other things, dimmed a 120V light via micro control. The way I did it then was to generate a zero-crossing interrupt which loaded the dimming delay into a timer. When the timer timed down it generated another interrupt which then turned on a triac for the remainder of that half-cycle.

My question is, what would be the modern way to do this via micro? Is any of this functionality now combined in a simpler package than the separate bits that I have described? Thanks.

Reply to
Mike Silva
Loading thread data ...

Take a modern 1 apart, notice how few " parts are in it, useing a micro would be overkill in the extreem.

Reply to
cbarn24050

Are you thinking of a wall-mounted slide dimmer? I know how those work. What I want is to convert an externally-supplied or internally-calculated digital value to a dimming value, so a micro is going to be in there somewhere.

Reply to
Mike Silva

The key elements will still be just about the same, unless you want to add complexity in order to improve performance (eg. reduce filament 'singing'). You may be able to improve upon the accuracy using a micro with a more modern timer peripheral that does not introduce interrupt latency into the timing error.

Reply to
Spehro Pefhany

I recently interviews with a company that makes many light dimming products. The engineer I spoke with said that they use mosfets to switch off and on the AC line in relation to the zero crossing, much like how you did it except for the usage of fets. The company was Lutron, if you are interested.

-Mike

Reply to
M. Noone

The Atmel LP2052 has a gated PWM timer, that can interrupt on the PWM end. So, you can setup the Zero cross to be a Triac-Time-Off sense, and that enables the timer, which will spin until the PWM trip point, and then fire the triac. That means no SW is in the Triac timing loop, and your SW can either reload at the end (monostable), or run a phase-tracking scheme - tho that could get interesting if the mains ever has rapid frequecy flutter... Or, you could do the same operation in a CPLD... -jg

Reply to
Jim Granville

So take a normal wall-mounted dimmer design; replace its mechanical potentiometer by a digital potentiometer. Done.

--
Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.
Reply to
Hans-Bernhard Broeker

Are digital potentiometers which are designed to go across the AC line common and reasonably cheap?

Reply to
Mike Silva

If you want to control a dimmer by CPU, then your original design is preferred. If you already have a microcontroller with a timer, why build an external timing circuit out of a digital potentiometer and a capacitor ?

Reply to
Artenz

Yes, that was my feeling as well. I would be surprised if suitable digital potentiometers were common and reasonably cheap (but I could be wrong), and on top of that there would be another layer of mapping between the digital value representing brightness and the digital value used to control the potentiometer. And then there are those 25% tolerances in the various parts...

Reply to
Mike Silva

While working with an amateur theatre group I designed a complete lighting system that dealt with the final drive to the triacs in a digital fashion. I had an 8-bit value to give 256 discrete lighting levels. The lamp control node picked up on the zero crossing point to reset the logic. Then, the turn-on point was when the demand matched the count of pulses from an oscillator locked at a rate that would produce a clock pulse at 1/260th portions of the mains half-cycle. The additional four pulses were to give a minimal time band at the end of the half-cycle. This period was useful to keep a little heat in the elements.

By taking care of galvanic isolations in the lamp node I could run the lighting off a three phase bussed supply and mix and match low level control signals in the control desk quite easily.

--
********************************************************************
Paul E. Bennett ....................
 Click to see the full signature
Reply to
Paul E. Bennett

M. Noone wrote: : I recently interviews with a company that makes many light dimming : products. The engineer I spoke with said that they use mosfets to : switch off and on the AC line in relation to the zero crossing, much : like how you did it except for the usage of fets. The company was : Lutron, if you are interested.

Typically for dimming electronic transformers and other devices that don't like the noise generated by turning the triac on at its high current setting in the waveform. Using MOSFETs allows you to turn on at ZC and off at an arbitrary point, just the opposite of triacs. But, at what cost heat and power loss-wise?

DLC

: -Mike

--
============================================================================
* Dennis Clark         dlc@frii.com                www.techtoystoday.com   * 
 Click to see the full signature
Reply to
Dennis Clark

Why would a manually operated dimmer contain a capacitor? All you need is a DIAC, a TRIAC, and a couple of resistors.

--
Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.
Reply to
Hans-Bernhard Broeker

Umm.. how do you think you could get below 50% power like that?

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

I've seen SSRs with dimming capability --- that is, instead of an on/off input, they take (for example) a 1-10VDC control input and do phase-control dimming of the output. Internally, AFAIK, they still use SCRs.

--
   Wim Lewis , Seattle, WA, USA. PGP keyID 27F772C1
Reply to
Wim Lewis

"Mike Silva" skrev i meddelandet news: snipped-for-privacy@p10g2000cwp.googlegroups.com...

Don't know if this is what you are looking for but people building Light ballast applications are using either

  • On-Off - No Dimming...
  • an low voltage analog signal (typically 0-10 V)
  • a digital interface (DALI) which is basically a Manchester coded 16 bit UART to communicate the dimming value and then a microcontroller interpreting this value.

The dimming is handled by MOSFETs controlled by high speed timers. Typically the frequency is around 80 kHz. but can vary between 40 - 150 kHz.

The AT90PWM2/3 chips were built to handle this functionality. It has a 64 MHz internal clock source and resolution enhancement on the MOSFET timer giving an equivalent internal clock of 1.3 GHz. There is extra circuitry to measuring current in the MOSFETs which will disconnect the MOSFET on overcurrent.

May be useable in your application as well.

--
Best Regards
Ulf Samuelsson
Reply to
Ulf Samuelsson

Search on the internet for "DMX 512" :-) Cheers, Randy Hyde

Reply to
randyhyde

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.