Re: TRIAC dimmer control spreadsheet; was Re: Controlling Holiday Lights

[Marc_F_Hult <MFHult@xxxxxxxxxxxxxxxxxxxxx>]

On Fri, 3 Nov 2006 16:20:34 -0700, sylvan butler
<ZsdbUse1+noZs_0611@xxxxxxxxxxxxxxxxxxxxx> wrote in message
<slrneknjm2.p3a.ZsdbUse1+noZs_0611@xxxxxxxxxxxxxxx>:

>On Thu, 02 Nov 2006 15:38:23 -0500, Marc_F_Hult
<MFHult@xxxxxxxxxxxxxxxxxxxxx> wrote:

>>This certainly applies to the end user. But folks that actually design
>>dimmers  might initially choose a mathematical representation of the known
>>physics for design purposes. (I got the equations shown on my spread sheet
>>from Chris
>
>yeah, back in the days when the dim level was controlled by a variable
>resistor (even when controlling a triac) it was handy to know the value
>of that resistor and the floor and ceiling values of the total
>resistance.

And in the case of TRIAC based dimmers, it would be handy for you to see the
underlying equations so that you would know that what you claimed in this
newsgroup in umpteen posts simply does not work and that despite your claim to
the contrary, there is in fact an "easily calculable relationship to dim
level". See my the equations in the url in the line below.


>> From my  at http://www.econtrol.org/dimmers/TRIACDimmerCalcs_MFHult.pdf
>>n
>> 50% dimming with respect to luminous intensity is achieved at ~76 degrees =
>> 3.53 milliseconds delay.
>
>And that will vary somewhat depending on differences in the actual
>incandescent light (hardware).
>
>> But in the case of TRIAC dimmer control (see thread subject) a single byte
>> holds all the information needed to produce 2^8 = 256 levels
>
>An 8-bit byte always does.

Here sylvan snips out the part of my sentence above that makes his quip above
so silly and relates to the critical importance of timing,  namely:

	*IF* one also knows when the zero crossing occurs as is conventional.
	 sylvan doesn't know the z-crossing and so futilely tries/d to
 	overcome that lack of critical data  .."

He fantasized that with complex waveforms and (newsgroup post by newsgroup)
increasing frequencies and complexity, he could make a TRIAC dim evenly
without skipping half-cycles and without a timing reference to the
zero-crossing. But he couldn't despite the fact that he claimed that he had.

>> by creating complex waveforms with large data requirements that by his
>> admission challenges the timely computing power of a PC.
>
>Marc, I neither said nor alluded to any such thing, and I thank you
>kindly to not put words into my mouth.

I don't. You gave up using the 18.2 timer. Then you gave up trying to time
using progressively higher and higher frequencies. Then you finally broke down
and kinda hint that you would have to time it with reference to the
zero-crossing because no matter how what kilo/mega/gigahertz frequency you
tried, it didn't work. "Timely".  Read what I wrote because it is at the core
of why what you have written flat out does not work. I thank you kindly for
getting that part right finally and ending your obfuscation. When you finally
stop dodging and misquoting, it will be easier to know what you actually mean
to say.

>> Applying 'perceptual coding" to that single 8-bit byte, we could usefully
>> 'compress' it to a 4-bit nibble to produce 16 levels if that is all that is
>> needed.

>That is not perceptual coding and is not in any way related to it.

Of course it is.  When one compresses an 8-bit linear curve to a 4-bit
square-law curve, one uses fewer bits and incorporates the non-linear way in
which humans perceive light. This is an excellent example of perceptual coding
(square-law curves) that has been around for decades.

>> can be used to select values for a new, equal intensity, 16-step dimmer
>> curve with Full ON, Full OFF and 14 intermediate steps each causing about
>> 6% reduction in light intensity.
>
>You certainly do not want "equal intensity" and equal "6% reduction in
>light intensity" in your steps.  If you actually implemented a dimmer
>that way, the bottom steps (starting from dark) would appear
>significantly bigger jumps than the top steps (approaching full-bright).

Why don't I ? Do you think that log is intrinsically better than linear? For
what purpose? A linear curve is in fact one of the conventional choices in
quantitatively calibrated commercial dimmers. See the AL4016 dimmer chip at
http://www.artisticlicence.com/cat11_1.htm that I cited previously.


>THAT is human perception.

Huh? WHAT is "human perception"? Having spent many years with densitometers
and calibrated light meters, and different films and developers and toners and
papers, I can assert unequivocally that a strict log curve is *not* what is
always aesthetically pleasing or convincing or realistic or desired in part
because of the time-dependence of the logarithmic human perception of light.

Perhaps if you were more knowledgeable, you would write that a different,
equally valid, also quantitative, also commercially available curve, that
incorporates a different approximation to the way humans perceive light, is
one that obeys a square-law, such as the AL4017 chip -- also available from
http://www.artisticlicence.com/cat11_1.htm

>Perceptual coding takes place at a lower level in the design,

Lower than what? Lower than the basic physics of electron flow that you got
flat out wrong?

If you know the physics, and can write and use the equations, the "perceptual
coding" could in fact be done at the last moment, on the fly, by choosing a
different equation or coefficient immediately before sending the ON
instruction to the TRIAC.

I can transform the curves shown in my spreadsheet into most any curve that is
wanted: log, square-law, linear, s-shaped, and any combination of them. I can
do this despite the fact that you have asserted that there is no ready
mathematical relationship, because you are also wrong when you assert that.

And practical dimmer curves often follow more than one 'law' for practical
reasons. One common need is for a preheat-level that can be incorporated into
the curve. In my case, I want a flat part at the top of the voltage curve to
close the contact of a mechanical relay in order to effectively switch the
TRIAC out of circuit at 100% on and I want that to be an un-missable target
when physically rotating a knob or moving a slider.

Similarly with the bottom of the curve. I do *not^ want a barely visible low
level (i.e, the logarithmic curve you insist I should want) in a real light
source in a real room because with incandescent lights, that is inordinately
wasteful of electricity.

So I do *not* want a log curve at low levels as you insist I should. It is
much more efficient to switch to a lower wattage lamp for low levels and it is
an outright requirement if one wants anything resembling 'white" (not orange)
light using incandescent lamps.

> realizing

I realize, you realize, he, she, or it (Already!) realizes.

>that human vision can detect a very small change from dark (no lumens)
>to some light (lumens), but cannot detect an equal change in lumens from
>bright to brighter.  Thus perceptual coding is used so that whether your
>dimmer has 16 steps or 256 steps or 65536 steps, the perception of
>brightness is uniform at each step even though the increase in light
>intensity between each step grows MUCH larger as the light gets
>brighter

Duh ...  That's what effectively part of what a square-law dimmer curve
addresses. So what you are saying is that you would prefer a AL4017 over a
AL4016.

You pontificate in a way that makes it appear that you don't realize that
folks accomplished and have made commercially available long ago what you are
still grappling with in theoretical terms while using smoke and mirrors to
avoid the reality that your dimmer design flat out doesn't work because you
are ignorant of the basic physics.

Seems to me that you have interesting ideas that are usefully discussed in
this newsgroup. Thank you for that.

But do try to learn something in the process and do help us fix mistakes or
misstatements rather than trying to cover them up.

Take care ... Marc
Marc_F_Hult
www.ECOntrol.org