Searching \ for 'Pic Lamp Dimmer' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/microchip/devices.htm?key=pic
Search entire site for: 'Pic Lamp Dimmer'.

Truncated match.
PICList Thread
'Pic Lamp Dimmer'
1998\05\01@032817 by David Duffy

flavicon
face
I was thinking of using an 'F84 for serial controlled 240Vac lamp dimmer.
Does anyone know how many steps you need for really smooth fade up/down ?
I thought 256 discrete levels (1 byte) would do but I imagine that some
sort of look-up table is required so that 'level 127' for example will
actually be half brightness due to the lamp's non-linearity. Levels 0-255
makes it easy to assign min/max/etc levels as a byte each. I realise that
each type of lamp will have a different power/intensity curve but is there
a general purpose 'average' type of curve available out there ?
Regards...

_________________________________________________________________
Dave Duffy      Audio Visual Devices        E-mail: spam_OUTAVDTakeThisOuTspamuq.net.au
Phone: +61 7 38210362                   Facsimile: +61 7 38210281
Unit 8, 9-11 Trade Street, Cleveland, Queensland  4163  Australia
_________________________________________________________________

1998\05\01@165431 by Stuart Broad

flavicon
face
part 0 1913 bytes
<META content=text/html;charset=iso-8859-1 http-equiv=Content-Type>
<META content='"MSHTML 4.71.2016.0"' name=GENERATOR>
</HEAD>
<BODY>
<DIV>Most&nbsp; lamp dimmers usually follow either a log law or &micro; law
curve. &nbsp;</DIV>
<DIV><FONT face=Arial size=2>{Original Message removed}

1998\05\01@220604 by Harold Hallikainen

picon face
       I found a "standard" curve published somewhere that is supposed
to take into account nonlinearity of lamp output and nonlinearity of
human perception of lamp intensity.  The curve gave an RMS voltage for a
specified percentage intensity (with 100% = 120Vrms).  I took a bunch of
points on the graph and developed a power series to represent this curve
between 0 and 100%.  Here's the power series.  You can drop it into your
favorite graphing calculator, math program (I'm using the shareware CC4)
or computer language to generate a table or whatever.

Harold



Vrms(x)=3.7192063561x-1.9826283199e-1x^2
+8.2505490236e-3x^3-1.8179894417e-4x^4
+2.1421958023e-6x^5-1.2870370627e-8x^6
+3.1084656826e-11x^7

x=intensity in percent





_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]

1998\05\02@054343 by Ron Fial

flavicon
face
People I have known that have designed light faders for light shows have
told me that the brightness change vs phase angle of the AC is very
nonlinear, and that they were never able to get it perceptually smooth with
only 256 steps.  The best designs actually used a look-up table with
12-bit control, but for the PIC, you probably want to do this with an
algorithm to save table space.  Note that changing bulb types and filaments
may make a big difference in linearity/smoothness, and that the eye has a
log response to brightness.

At 05:31 PM 5/1/98 +1000, you wrote:
>I was thinking of using an 'F84 for serial controlled 240Vac lamp dimmer.
>Does anyone know how many steps you need for really smooth fade up/down ?
>

1998\05\03@190729 by Dennis Plunkett

flavicon
face
At 05:31 PM 1/05/98 +1000, you wrote:
>I was thinking of using an 'F84 for serial controlled 240Vac lamp dimmer.
>Does anyone know how many steps you need for really smooth fade up/down ?
>I thought 256 discrete levels (1 byte) would do but I imagine that some
>sort of look-up table is required so that 'level 127' for example will
>actually be half brightness due to the lamp's non-linearity. Levels 0-255
>makes it easy to assign min/max/etc levels as a byte each. I realise that
>each type of lamp will have a different power/intensity curve but is there
>a general purpose 'average' type of curve available out there ?
>Regards...
>
>_________________________________________________________________
>Dave Duffy      Audio Visual Devices        E-mail: .....AVDKILLspamspam@spam@uq.net.au
>Phone: +61 7 38210362                   Facsimile: +61 7 38210281
>Unit 8, 9-11 Trade Street, Cleveland, Queensland  4163  Australia
>_________________________________________________________________
>
>
I have seen some interesting comments on your problem. Yes the basic
response that the output is log is correct, however I do assume that it is
Yellow light that your controlling? If this is so, you will have to look at
the number of lamps being controlled, as this and distance also will change
perception of brightness. But as a simple rule of thumb (Here come the
"critics"), the eye will see only some 16 different brightness levels with
the majority of the sensitivity being in the low light area (Electronic
score boards like that at the MCG and SCG and Waverly park use only some 16
(Waverly park) to 48 (MCG and SCG)). However a smooth increase in brightness
from off to fully on will require some 64 steps in your case (This assumes
that you are not using fixed wattage lights, Note that other colours like
GREEN and BLUE will require a very different curve to that of FULL YELLOW
and RED).


Dennis

1998\05\03@212008 by smg

flavicon
face
Hi:
       I don't know if it is of any value to you, but a quasi PWM technique has
worked quite well for me as a method of lamp intensity control.  If you
count zero crossings, you can easily ratio the counts for time-on to counts
time-off.  I belive the response you received here about the visual
(perception) characteristics are certainly valuable and informative and are
necessary to condiser in your design spec.
       In my own experience I have used 8-bits and I do not recall seeing any
flicker.  As pointed out by others, the visual response is not linear
(i.e., of the Y=mx+b functional relationship), it will be a log function. I
assumed you were concerned about smoothness in terms of no flicker and no
discrete jumps or steps in the perception of changed intensity.

----------
> From: Dennis Plunkett <dennisspamKILLspamRDD.NECA.NEC.COM.AU>
> To: .....PICLISTKILLspamspam.....MITVMA.MIT.EDU
> Subject: Re: Pic Lamp Dimmer
> Date: Sunday, May 03, 1998 6:44 PM
>
> At 05:31 PM 1/05/98 +1000, you wrote:
> >I was thinking of using an 'F84 for serial controlled 240Vac lamp
dimmer.
> >Does anyone know how many steps you need for really smooth fade up/down
?
> >I thought 256 discrete levels (1 byte) would do but I imagine that some
> >sort of look-up table is required so that 'level 127' for example will
> >actually be half brightness due to the lamp's non-linearity. Levels
0-255
> >makes it easy to assign min/max/etc levels as a byte each. I realise
that
> >each type of lamp will have a different power/intensity curve but is
there
{Quote hidden}

is
> Yellow light that your controlling? If this is so, you will have to look
at
> the number of lamps being controlled, as this and distance also will
change
> perception of brightness. But as a simple rule of thumb (Here come the
> "critics"), the eye will see only some 16 different brightness levels
with
> the majority of the sensitivity being in the low light area (Electronic
> score boards like that at the MCG and SCG and Waverly park use only some
16
> (Waverly park) to 48 (MCG and SCG)). However a smooth increase in
brightness
> from off to fully on will require some 64 steps in your case (This
assumes
> that you are not using fixed wattage lights, Note that other colours like
> GREEN and BLUE will require a very different curve to that of FULL YELLOW
> and RED).
>
>
> Dennis

1998\05\04@004704 by David Duffy

flavicon
face
My original questions:

>I was thinking of using an 'F84 for serial controlled 240Vac lamp dimmer.
>Does anyone know how many steps you need for really smooth fade up/down ?
>I thought 256 discrete levels (1 byte) would do but I imagine that some
>sort of look-up table is required so that 'level 127' for example will
>actually be half brightness due to the lamp's non-linearity. Levels 0-255
>makes it easy to assign min/max/etc levels as a byte each. I realise that
>each type of lamp will have a different power/intensity curve but is there
>a general purpose 'average' type of curve available out there ?
>Regards...

Dennis Plunkett wrote:

{Quote hidden}

The plot continues !

All I want to do is control standard household lighting. (home automation)
My aim is to go between 0% and 100% smoothly. (with no perceptible steps)
Harold Hallikainen kindly gave me some information on how to calculate the
curve points and I will follow that up. I've since seen a dimmer project on
the PicPoint page but have not looked closely at it yet. Thanks so far !
Regards...   :-)

_______________________________________________________________
Dave Duffy         Audio Visual Devices           AVDspamspam_OUTuq.net.au
Unit 8, 9-11 Trade Street, Cleveland, Queensland 4163 Australia
Phone: +61 7 38210362                 Facsimile: +61 7 38210281
_______________________________________________________________

1998\05\04@012156 by Rob Symmans

flavicon
picon face
At 02:46 PM 5/4/98 +1000, you wrote:
{Quote hidden}

Using a 100 tap xicor eepot under control of a C84 I have dimmed a 20W 12V
halogen lamp from nil to full on, the changes (obviously) appear smooth,
with no problems at all.
Hope this helps.
rob
-----------------------------------
Rob Symmans

Lions Eye Institute
Lasers and BioEngineering
2 Verdun Street
Nedlands 6009
Western Australia
Tel 08 9381 0754   Fax 08 9381 0700
-----------------------------------

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...