Analog to Digital Dimmer

Your customers seems to require infinite control.

I would go for full electronic control.

I am not in electronics but are giving this suggestion anyway.

Required. CPU (pic or), keypad , etc

The keypad and CPU is used to enter the required setting (1 to 100)

The cpu then determine the crossover points and allow the flow of current to the load for a predetermined duration. Example: 30 degrees = 2% , 40deg = 4% , 45deg=6% etc etc 150 deg = 98%, 180 deg=100%.

Being in water raises a question: Do the sudden closure of a switch not cause electron hammer (similar to water hammer).

hi there ;

in regards to your comment at the bottom of page /

electrical water hammer :

yes a condition like that did occour last september /

on a multi building service 200 amp 480 volt 3 phase/

two 'bright' people connect a 208 single phase motor to the

LINE side of a disconnect with jumper cables [ 12 v kind ]

the resulting short curcuit did the following :

blew the 2 - 200 amp fuses , cut the service cable just past the

tap connection on the service going to the next building /

this is underground service 5 ft[ 1.5 m~ ] down /

ALL 4 CABLES WERE CUT WITH NO LEAKAGE TO GROUND /

the dirt in the area was fused into glass /

the term an electronics tech used was ' flyback effect '

yes , the water hammer effect does apply to all fluids /

electrons do flow

A similar effect is indeed known: at the switch moment you have serious spikes that can be heart in the humming of audio systems and that kill sensitive equipment.

One of the reasons that this technique is normally only used for lighting and the biggest cost in the commercial units is the filter that takes the spikes out (like an expansion vase)

Dimming of other power users like heaters is done in the time frame: 10 cycles on, 10 cycles off is 50%, no phaze cutting there.

What is described above is the effect of the way to high currents: a short can go in thousands/millions of amps (althow the fuse was rated 200A) in fractions of seconds. This caused the wires to be attacted to each other. As they are really close to each other and only separated with some mm of rubber/silicone potting they could have cut through the insulation causing a way bigger short, invoking the copper to vaporise.

The energy from this vaporised copper glased the surrounding material.

The same happens in generators: the current which goes in the same direction causes the generator coils to explode.

Hmmm

Digital display of 'volts' is probably not at all related to brightness (or dimness!).

Wouldn't you want the display to show say percentage of full brightness?

(I don't have a decent answer to the actual Q as posed)

Why do I always answer a question with another question....?

Why not?

This is not a hard thing to design. However, if you are going to build any volume of them you would be better off to build an integral unit instead of an add-on to an existing device. There is no reason the digital device can't use a similar power controlling configuration if that part of what you have suits. The real question is whether you want to just do a few of these or if it is something you anticipate selling enough of to merit an actual design. If you are just doing a few, well get a microcontroller and use it to control the phase timing to fire the (SCR?) I think you mean Triac. Mess with it until you get it working. If you want to invest in someone designing it for you then send me a message and we can talk.

Is this any help? http://www.circuithut.com/index.php?/circuit/content/view/full/363 I don't understand any of it being a lowly mechanical engineer but it seems to fit your requirement.

I agree that a microcontroller that controls the gate of a triac seems to be the easiest solution. A word of caution though, I doubt you really want to make it linear. A linear response will not mimic an analog or other electronic dimmer very well. Due to the way the eye functions light is perceived as a squared relation and you would want to mimic that in your control. If not there will be a lot of perceived light change from 0 to 50 (70.7%) and much less 29.3% from 50 to 100.

Shawn

How about this...

Replace the manual POT with a digital POT IC. Add a simple up , down counter

0 to 100, This can drive your 2 & 1/2 digit display. You will need a simple RC oscilator and a up/down rocker switch. The oscilator will clock both the counter and the digital POT.

I remember designing one of these things a long time ago. If I remember correctly, I used a zero-crossing trigger to gate an oscillator to a digitally presettable counter and used the carry signal to trigger a triac through a diac. Something similar could be built using a pulse-generating pot (e.g. a low-res encoder) to preset a counter (in principal). These days, a PIC or something similar could be used to do all but the power-handling part (triac, etc.). In the big scheme of things, there's a bunch of ways to do it. Like someone else said, is the market big enough to invest in a little R & D?

Carlo Gavazzi has nice modules that translate a ratio in a signal to a dimming ratio.

The basis is a 4-20mA signal but you can do the same with 0-10V

In fact this is what most dimmers do: translate the position of the pot which is read out in voltage to a delay in firing the triac.

The price is also acceptable.

I use these modules with a PLC with 4-20mA outputs. It allows a very nice adjustable dimming. (I can dim 2kW heater +/- 1%)

In the past years I made various digital dimmers but not using programmable devices. Star with zero cross detection. Divide by 2 for obtain 60 Hz (or 50 Hz) 50% duty cycle square wave. If you wish, by example, 256 steps, make an PLL with an CD4046 oscillating at 512 * 60 (512 * 25) Hz. (This is 8 bit, but you make to 64 bits if you wish) Using preseteable down counters as CD4029 in cascade you preload digital data (from 0 to 255) and count up using 30720 (25600) Hz PLL clock (Motorola CMOS manual have an example of 4029 and other of 4046) Wher counters reaches carry on, AC power triac must be triggered, then counters maybe presetted again. The preset pulse maybe obtained from zero cross detector. If you load 255, one high speed clock pulse after, triac be triggered. If you load 128, triac must be triggered in the mid of semicycle. If you load zero, next zero pulse preset counters before they reach maximun value. This system guarantizes triac trigger ever in same point of sinusoidal AC line. Those dimmers keep working after ten years.