12V 15A AC light dimmer?
Primoz Peterlin
Hello everybody --
My problem: I would want to regulate a 12V halogen lamp (like the one in the
slide projector). Using ordinary 230V (or American 110V) light dimmer on mains
is impossible because the load is inductive (transformer). I have thought of
putting the light dimmer _after_ the transformer, but I would have to design it
myself, and I don't know of any suitable diac/triac pair for 15 or 20 amps.
Has anybody run into this problem already? How did you solve it?
Best regards, Primoz
--
Primoz Peterlin (pete...@biofiz.mf.uni-lj.si) Fax:+386-61-131-5127
Institut za biofiziko MF, Lipiceva 2, SLO-1105 Ljubljana, Slovenija
WWW page: http://sizif.mf.uni-lj.si/~peterlin/
DonEPearce
You have a bit of a problem here. A halogen lamp is designed to run at
high temperature for the halogen cycle to work - that is the means by
which evaporated metal is deposited back on the filament rather than the
glass envelope. If you dim a halogen lamp this does not happen, and the
life span is severely limited. If you want to dim, use a non-halogen bulb.
To make a dimmer that will work through a transformer try this.
Take a pair of similar transformers and wire the two primaries in series
with each other. Now wire this pair in series with the supply to the lamp
circuit. It will act as a high value choke, dimming the light to virtually
nothing.
Now take the secondaries of the two transformers, and wire them in series,
but antiphase so there is no output. Join these to a DC variable power
supply. By altering the current through the connected secondaries, you can
saturate the transformer cores as much or as little as you like. The lamp
will dim and brighten accordingly, with no disruption to the shape of the
mains sine wave, which will work through the lamp transformer perfectly.
Don P
Kevin McMurtrie
In article <4q70r6$4...@sweet.kiss.uni-lj.si>, Primoz Peterlin
<primoz....@biofiz.mf.uni-lj.si> wrote:
The transformer won't be very inductive with a bulb on it.
If you still want to do the low voltage side, use a TRIAC with a capacitor
and variable resistor in series between the gate and power. I'm not sure
what capacitor and resistor range to use for 12V. The cap is there to
slightly advance the phase so the TRIAC can be fully turned on. There
will be a range missing between low and off unless you make a fancy phase
shifting circuit or use a DC PWM circuit.
Markus Imhof
In article <4q70r6$4...@sweet.kiss.uni-lj.si>, Primoz Peterlin
<primoz....@biofiz.mf.uni-lj.si> wrote:
> Hello everybody --
>
> My problem: I would want to regulate a 12V halogen lamp (like the one in the
> slide projector). Using ordinary 230V (or American 110V) light dimmer
on mains
> is impossible because the load is inductive (transformer). I have thought of
> putting the light dimmer _after_ the transformer, but I would have to
design it
> myself, and I don't know of any suitable diac/triac pair for 15 or 20 amps.
>
> Has anybody run into this problem already? How did you solve it?
>
First of all, it is possible to dim a transformer on the primary. Circuits
for that should float around in most electronics magazines (got some at
home - in German), or you can buy them from electronics stores like
Conrad. Disadvantage is that some transformers will hum quite loud when
driven from a non-sinusoidal waveform.
Second possibility: AC dimming on the secondary - been there, done that. I
used some dedicated circuitry on the low-power side (not the simple
RC-delay), but can't access my documentation (or the dimmer) at the
moment. The main problem was the _reliable_ conection of the 16^2 mm cable
to the PCB. A bad contact almost burned the PCB once (remember: dont make
jokes with 20 Amps). Triacs: the TIC263M will handle up to 25 Amps (253:
20 A, 243: 15 A) - but be sure your PCB tracks won't act as a fuse.
Third possibility (and what I'd probably do today, looking at prices and
availability of high power Mosfets): use DC current (only rectified, not
necessarily filtered - caps are expensive) on the secondary, and run the
brightness control via a PWM/PFM/PsomethingM circuit driving a few
parallel high-power Mosfets. Just make sure your driving frequency doesn't
produce visible interference products with the 50 Hz mains frequency (a 51
Hz PWM should make for an interesting display :-)
Bye
Markus
Bob Wilson
: My problem: I would want to regulate a 12V halogen lamp (like the one in the
: slide projector). Using ordinary 230V (or American 110V) light dimmer on mains
: is impossible because the load is inductive (transformer). I have thought of
: putting the light dimmer _after_ the transformer, but I would have to design it
: myself, and I don't know of any suitable diac/triac pair for 15 or 20 amps.
: Has anybody run into this problem already? How did you solve it?
You can *NOT* successsfully dim a halogen light! The reason for this
relates to the lamp itself. A halogen lamp contains a small amount of
one of the halogen family of elements (usually either bromine or iodine)
or a mixture of these, in the envelope. As the filament "boils" off
tungsten (which would normally condense on the glass envelope and cause
gradual darkening), the halogen gas (and it is gas, at the operating
temperature) chemically combines with the tungsten preventing it from
condensing on the inside of the envelope. The second part of the cycle
is when the tungsten bromide (or iodide, as the case may be) comes in
contact with the hot filament. The heat decomposes the tungsten halide,
and liberates metallic tungsten, which is deposited back on the filament
again. The halogen element is freed as gas to continue the cycle again.
Clever, Huh?
However, in order to make the whole thing work, the lamp must operate at
much higher temperature than a normal lamp (which is why halogens have
such an intense light). If you try to dim the lamp more than a
relatively small amount (say less than about 75 or 80% of its normal
brilliance), the temperature will fall below that needed to sustain the
halogen cycle. Then the lamp will quickly darken (eliminating the need
to use a dimmer in the first place, I suppose).
Bob.
Chris Matthaei
bwilson@newshost (Bob Wilson) writes:
>You can *NOT* successsfully dim a halogen light! The reason for this
>relates to the lamp itself. A halogen lamp contains a small amount of
>one of the halogen family of elements (usually either bromine or iodine)
>or a mixture of these, in the envelope. As the filament "boils" off
>tungsten (which would normally condense on the glass envelope and cause
>gradual darkening), the halogen gas (and it is gas, at the operating
>temperature) chemically combines with the tungsten preventing it from
>condensing on the inside of the envelope. The second part of the cycle
>is when the tungsten bromide (or iodide, as the case may be) comes in
>contact with the hot filament. The heat decomposes the tungsten halide,
>and liberates metallic tungsten, which is deposited back on the filament
>again. The halogen element is freed as gas to continue the cycle again.
>Clever, Huh?
>However, in order to make the whole thing work, the lamp must operate at
>much higher temperature than a normal lamp (which is why halogens have
>such an intense light). If you try to dim the lamp more than a
>relatively small amount (say less than about 75 or 80% of its normal
>brilliance), the temperature will fall below that needed to sustain the
>halogen cycle. Then the lamp will quickly darken (eliminating the need
>to use a dimmer in the first place, I suppose).
Don't the new daytime running lights in use in the US use the high beam
headlights run at a reduced voltage? Surely these headlights also use
halogen lamps. You can tell that these DRLs are running at much less than
75% of their normal brightness. Why don't these have the same problem?
Chris
Bruce Bostwick
donep...@aol.com (DonEPearce) wrote:
[...]
>Now take the secondaries of the two transformers, and wire them in series,
>but antiphase so there is no output. Join these to a DC variable power
>supply. By altering the current through the connected secondaries, you can
>saturate the transformer cores as much or as little as you like.
Ah! A magnetic amplifier! ;-)
<BGB> http://ccwf.cc.utexas.edu/~lihan/ mailto:li...@ccwf.cc.utexas.edu
Paul Mathews
Bob Wilson wrote:
>
> You can *NOT* successsfully dim a halogen light! The reason for this
> relates to the lamp itself....... If you try to dim the lamp more than a
> relatively small amount (say less than about 75 or 80% of its normal
> brilliance), the temperature will fall below that needed to sustain the
> halogen cycle. Then the lamp will quickly darken......
People say you can't do a lot of things, but dimming a halogen lamp
certainly can be done, and it is done all the time. It's all a matter
of degree. The halogen cycle does, indeed, prolong the life of a
tungsten filament. Fortunately, so does reducing the power to the
filament. So, in spite of all these dire warnings, you will find that
halogen lamps operated below the temperature required for proper halogen
cycling can have a lifetime which is even longer than the same bulbs
operated at full rated power.
I have a laboratory solar simulator which contains banks of 24V halogen
projector lamps wired in groups of 5 in series used in conjunction with
120V mains and lamp dimmers. We have several thousands of hours of good
service without anything remarkable happening in the way of early
failure of the bulbs.
--
Paul Mathews, consulting engineer
AEngineering Co.
opt...@whidbey.com
non-contact sensing and optoelectronics specialists
ls470
In article <mcmurtri-190...@venus48.wco.com>, mcmu...@wco.com (Kevin McMurtrie) says:
>
>In article <4q70r6$4...@sweet.kiss.uni-lj.si>, Primoz Peterlin
>
>>Hello everybody --
>>
>>My problem: I would want to regulate a 12V halogen lamp (like the one in the
>>slide projector). Using ordinary 230V (or American 110V) light dimmer on mains
>>is impossible because the load is inductive (transformer). I have thought of
>>putting the light dimmer _after_ the transformer, but I would have to design it
>>myself, and I don't know of any suitable diac/triac pair for 15 or 20 amps.
>>
>>Has anybody run into this problem already? How did you solve it?
>>
>>
>
>If you still want to do the low voltage side, use a TRIAC with a capacitor
>and variable resistor in series between the gate and power.
Hmm... Hmmm... I don't think this solution is a good one: a triac will
give a voltage drop of 2V at least, and from 12 to 10V makes a BIG difference.
I had a similar problem a year ago(but for only 1Amp). I just tried to
switch on and off my 16VAC electric train electronically. I first tried
with triacs, but the train ran lots slower, an the triac ran very hot.
Finally, I tried with two BUZ11 power MOSFETS, and that gave very good
results. I had a very small voltage drop (unnoticable for 1A). they can
be used upto 30A. For 15A, you should use other FETS, or put multiple FETS
in parallel to reduce resistance. I used this for switching, but thoose
FETS can also be used easily for dimming light.
Hope this helps.
ls470
Warning: I am not very experienced at electronics!
brian whatcott
In article <4q8asp$3...@newsbf02.news.aol.com>, donep...@aol.com says...
>
///
>Take a pair of similar transformers and wire the two primaries in series
>with each other. Now wire this pair in series with the supply to the lamp
>circuit. It will act as a high value choke, dimming the light to virtually
>nothing.
>but antiphase so there is no output. Join these to a DC variable power
>supply. By altering the current through the connected secondaries, you can
>will dim and brighten accordingly, with no disruption to the shape of the
>mains sine wave, which will work through the lamp transformer perfectly.
>Don P
Well, well, well!
I thought the transductor was dead and buried.
One of the few circuit elements that can amplify and control with no tubes,
transistors OR ICs....
brian
Sir Spamalot
On 20 Jun 1996 06:34:28 GMT, in...@intellisys.net (brian whatcott)
wrote:
--->In article <4q8asp$3...@newsbf02.news.aol.com>,
donep...@aol.com says...
--->>
--->///
--->>Take a pair of similar transformers and wire the two
primaries in series
--->>with each other. Now wire this pair in series with the supply
to the lamp
--->>circuit. It will act as a high value choke, dimming the light
to virtually
--->>nothing.
--->>Now take the secondaries of the two transformers, and wire
them in series,
--->>but antiphase so there is no output. Join these to a DC
variable power
--->>supply. By altering the current through the connected
secondaries, you can
--->>saturate the transformer cores as much or as little as you
like. The lamp
--->>will dim and brighten accordingly, with no disruption to the
shape of the
--->>mains sine wave, which will work through the lamp transformer
perfectly.
--->>Don P
--->
--->Well, well, well!
--->I thought the transductor was dead and buried.
--->One of the few circuit elements that can amplify and control
with no tubes,
--->transistors OR ICs....
--->
--->brian
--->
Us power supply types call them "MagAmps".
--
This is my theory, which is mine.
*** Ms. Anne Elk ***
Dan Mills
In article <4qardk$i...@zoom2.telepath.com>
in...@intellisys.net "brian whatcott" writes:
>In article <4q8asp$3...@newsbf02.news.aol.com>, donep...@aol.com says...
>>
>///
>>Take a pair of similar transformers and wire the two primaries in series
>>with each other. Now wire this pair in series with the supply to the lamp
>>circuit. It will act as a high value choke, dimming the light to virtually
>>nothing.
>>Now take the secondaries of the two transformers, and wire them in series,
>>but antiphase so there is no output. Join these to a DC variable power
>>supply. By altering the current through the connected secondaries, you can
>>saturate the transformer cores as much or as little as you like. The lamp
>>will dim and brighten accordingly, with no disruption to the shape of the
>>mains sine wave, which will work through the lamp transformer perfectly.
>>Don P
>
>Well, well, well!
>I thought the transductor was dead and buried.
>One of the few circuit elements that can amplify and control with no tubes,
>transistors OR ICs....
Before the development of the Thyristor, these were used in the theatre
industry for lighting control, We called them saturatable reactors. They
mess with the power factor something dreadful. I have seen a variation on the
design that used a soft iron rod lowered into the inductor by lever!
--
**************************************************************
* And on the first day the lord said....... * *
* ... LX1, GO! and there was light. * Dan Mills. * *
**************************************************************
Pierre Olivier
What is all that rubbish about Halogen lamps not beign dimmable???!
I work in a theatre: halogen lamps are used all over the place here. You
just have to run the lamps full on every once in a while (which always
happens in a theatrical production), and it takes care of the halogen
cycling. No big deal about it....
P.O.
Adrian
In article <4q8asp$3...@newsbf02.news.aol.com>
donep...@aol.com (DonEPearce) writes:
> You have a bit of a problem here. A halogen lamp is designed to run at
> high temperature for the halogen cycle to work - that is the means by
> which evaporated metal is deposited back on the filament rather than the
> glass envelope. If you dim a halogen lamp this does not happen, and the
> life span is severely limited. If you want to dim, use a non-halogen bulb.
This is true, but the halogen cycle can be maintained if you run it at
full intensity once in a while. For 1.5 A on 120 you should be able to
use a large triac for phase control, with little problems because of
the inductive load. The alternative is to use a dimable ’transformer’
for 12V lamps.
Adrian
Frank Bemelman
donep...@aol.com (DonEPearce) wrote:
>You have a bit of a problem here. A halogen lamp is designed to run at
>high temperature for the halogen cycle to work - that is the means by
>which evaporated metal is deposited back on the filament rather than the
>glass envelope. If you dim a halogen lamp this does not happen, and the
>life span is severely limited. If you want to dim, use a non-halogen bulb.
Theoretically right, but in practice it's not such a problem. I have a
halogen reading lamp here, with a high/low switch. I use it
approximately 4 hours a day, always at the low position, and never
changed the (50watt) bulb.
You can dim the transformer/halogenlamp safely, but.... the problem
with the ordinary dimmers combined with transformers/halogenlamps is
that the circuit tries to switch on the triac whilst the current is
around the zerocrossing point; not a good moment to turn on the triac.
This is why the usual triac-dimmers don't work very well.
However, there are special dimmers especially made for this purpose,
which have a small modification around the RC circuit. An extra RC
network is added, to generate repeating ignition pulses for the triac.
A more expensive (and better) approach is a design using GTO's that
cuts off portions at the end of the half-sine-wave (rather than
cutting of a portion at the start). These dimmers are even suitable to
drive 'electronic' halogen transformers (the switched-type
regulators).
Frank.
>To make a dimmer that will work through a transformer try this.
Leendert Prevo
ls470 (ls...@dds.nl) wrote:
: In article <mcmurtri-190...@venus48.wco.com>, mcmu...@wco.com (Kevin McMurtrie) says:
: >
: >In article <4q70r6$4...@sweet.kiss.uni-lj.si>, Primoz Peterlin
: ><primoz....@biofiz.mf.uni-lj.si> wrote:
: >
: >>Hello everybody --
: >>
: >>My problem: I would want to regulate a 12V halogen lamp (like the one in the
: >>slide projector). Using ordinary 230V (or American 110V) light dimmer on mains
: >>is impossible because the load is inductive (transformer). I have thought of
: >>putting the light dimmer _after_ the transformer, but I would have to design it
: >>myself, and I don't know of any suitable diac/triac pair for 15 or 20 amps.
: >>
: >>Has anybody run into this problem already? How did you solve it?
: >>
: >>Best regards, Primoz
: >
The load looks resistive at the primairy side of the transformer as long
as the lamp is ok. Siemens has a mains-side controller ic that checks phase
and protects triac. The ic works fine on manual control.
Any resistive load from 10% of nominal power at secondairy side will make
the primairy look enough resistive to the triac to work ok.
leendert.
brian whatcott
In article <835454...@abcde.demon.co.uk>, Dmi...@abcde.demon.co.uk says...
>
>Before the development of the Thyristor, these were used in the theatre
>industry for lighting control, We called them saturatable reactors. They
>mess with the power factor something dreadful. I have seen a variation on the
>design that used a soft iron rod lowered into the inductor by lever!
///
I saw an aircraft power supply, where a solenoid compressed a carbon disk 'pile'
which varied the DC control current into a transductor/saturable reactor/mag amp
Now that looked like a heroic design!
brian
Primoz Peterlin
First, let me thank everybody for their valuable contribution.
It seems I have opened up a much more fundamental question
whether halogen bulbs can be dimmed at all -- something I have
not been aware of.
Let me try to summarize what I have learned.
1) The halogen cycle cannot be sustained when the bulb is not
run under full current. The audience doesn't seem to have
reached a consensus about the implications of this, though.
Some claim that running a halogen bulb at a sub-threshold
current for the halogen cycle doesn't make any harm (does
that mean that a halogen bulb runs like an ordinary tungsten
bulb?), and some claim that halogen cycle can be restored if
the bulb is run at full current every now and then.
I am not quite whether I understand this last statement. At
sub-threshold current, the filament is not hot enough to
reduce tungsten bromide (or tungsten iodide). The latter,
however, still forms as long as there is enough free bromine
(or iodine) around. Is the trick that in this case there is
*much* less tungsten evaporating?
2) As long as the bulb is OK, the load looks resistive on the
primary side, the current gets however out of phase the very
moment when the bulb burns out, which may in turn burn out
triac as well. Good mains-side dimmers have built-in
protection for this case.
3) People have made secondary-side dimmers, either with triacs
(TIC243M/TIC253M/TIC263M), or they used DC current with pulse
width modulation and high-power MOSFETs connected in parallel
(BUZ11 seems to be good for up to 1A). The problem with
triac on the secondary side is the voltage drop on it, which
can be significant at 12V.
4) One can also use a variac (variable transformer) or two
regular transformers in series.
Once again, many thanks to:
ako...@iserve.net.mx (Adrian)
bd...@rz.uni-karlsruhe.de (Markus Imhof)
Bruce Bostwick <li...@ccwf.cc.utexas.edu>
bwilson@newshost (Bob Wilson)
donep...@aol.com (DonEPearce)
fu...@MCS.COM (Chris Matthaei)
kn...@netland.nl (Frank Bemelman)
lprevo@rulhe1 (Leendert Prevo)
ls...@dds.nl (ls470)
mcmu...@wco.com (Kevin McMurtrie)
Paul Mathews <opt...@whidbey.com>
Pierre Olivier <oli...@aei.ca>
John 015
<primoz....@biofiz.mf.uni-lj.si> said:
>First, let me thank everybody for their valuable contribution.
>It seems I have opened up a much more fundamental question
>whether halogen bulbs can be dimmed at all -- something I have
>not been aware of.
They can. Consult the lighting companies.
As long as the temperature at some later point is brought
up to "nominal" the Tungsten that went astray under low
temp operation will be scrubbed off and pulled back into
the cycle by the halogen gas.
Exactly what impact such use has on the expected life
time of the filament is hard to predict - there are many
parameters at play.
However, the advantage of halogen lamps: high color temp
and high efficiency is lost when run dimmed.
> 1) The halogen cycle cannot be sustained when the bulb is not
> run under full current. The audience doesn't seem to have
> reached a consensus about the implications of this, though.
> Some claim that running a halogen bulb at a sub-threshold
> current for the halogen cycle doesn't make any harm (does
> that mean that a halogen bulb runs like an ordinary tungsten
> bulb?), and some claim that halogen cycle can be restored if
> the bulb is run at full current every now and then.
There are too many parameters involved to say just what will
happen when.
Imagine you favorite make of halogen lamp typically failed due to
leaky glass seals around the electrodes. Running dimmed would
make such a lamp last longer.
> I am not quite whether I understand this last statement. At
> sub-threshold current, the filament is not hot enough to
> reduce tungsten bromide (or tungsten iodide). The latter,
> however, still forms as long as there is enough free bromine
> (or iodine) around. Is the trick that in this case there is
> *much* less tungsten evaporating?
Halogen lamps run below nominal (at 85% or less voltage)
will blacken their glass envelope. I've seen it.
> 2) As long as the bulb is OK, the load looks resistive on the
> primary side, the current gets however out of phase the very
> moment when the bulb burns out, which may in turn burn out
> triac as well. Good mains-side dimmers have built-in
> protection for this case.
>
> 3) People have made secondary-side dimmers, either with triacs
> (TIC243M/TIC253M/TIC263M), or they used DC current with pulse
> width modulation and high-power MOSFETs connected in parallel
> (BUZ11 seems to be good for up to 1A). The problem with
> triac on the secondary side is the voltage drop on it, which
> can be significant at 12V.
Current at turn on typically is 15 times the nominal
current assuming the power source has zero source
impedance. I use the IRFZ series with lamps up to
3 Amps nominal.
John
Christopher Trask
: >In article <4q8asp$3...@newsbf02.news.aol.com>, donep...@aol.com says...
: >>
: >>Take a pair of similar transformers and wire the two primaries in series
: >>with each other. Now wire this pair in series with the supply to the lamp
: >>circuit. It will act as a high value choke, dimming the light to virtually
: >>nothing.
: >>Now take the secondaries of the two transformers, and wire them in series,
: >>but antiphase so there is no output. Join these to a DC variable power
: >>supply. By altering the current through the connected secondaries, you can
: >>saturate the transformer cores as much or as little as you like. The lamp
: >>will dim and brighten accordingly, with no disruption to the shape of the
: >>mains sine wave, which will work through the lamp transformer perfectly.
: >>Don P
: >
: >Well, well, well!
: >I thought the transductor was dead and buried.
: >One of the few circuit elements that can amplify and control with no tubes,
: >transistors OR ICs....
And they make terrific variable inductors for tuned circuits, par-
ticularly in situations where distortion and intermodulation are of great
concern. Before the advent of the varactor (parametric diode), these were
used for remote tuning, and still have a place in highly demanding linear
RF circuitry.
The YIG elements used in wide-range VFOs and variable filters are
actually a variation of the transductor, although the element in this case
is a ceramic rather than a powdered iron or ferrite. It is still tuned
in the same manner with a control current rather than a voltage.
--
Regards,
Chris
------------------------------------------------------------------------------
Chris Trask / N7ZWY Circuit Design for the RF Impaired
ATG Design Services __ __ ____ ___ ___ ____
ctr...@primenet.com _~_ /__)/__) / / / / /_ /\ / /_ /
(@ @) / / \ / / / / /__ / \/ /___ /
----------------------ooO~(_)~Ooo---------------------------------------------
Rex Hebert
I've never heard of a Transductor.
Where can I learn more online? or off QST?
Rex Hebert
HEBERT Audio-Video
Lafayette, Louisiana
Rex Hebert
Rene Zuidema
On 24 Jun 1996 09:57:18 GMT, lprevo@rulhe1 (Leendert Prevo) wrote:
>The load looks resistive at the primairy side of the transformer as long
>as the lamp is ok. Siemens has a mains-side controller ic that checks phase
>and protects triac. The ic works fine on manual control.
>Any resistive load from 10% of nominal power at secondairy side will make
>the primairy look enough resistive to the triac to work ok.
>leendert.
Many 230 - 12 V xformers are switching mode nowadays.. (check DIY
stores)
Best Regards, Rene Zuidema