Questions about electronic dimmers and tungsten filaments

[Yoshi]

Questions about electronic dimmers and tungsten filaments

Hi all,

First, I read on this board that this kind of dimmer (electronic dimmer) uses pulsed full 120v voltage, and does not really reduce the voltage. I don't get it- If it uses pulses, then the bulb would actually be turned on and off at very high speeds, which means that on the lowest setting the bulb should flicker noticeably because the time gap between pulses becomes very long. I have seen dimmers that do exactly that- and have seen a GE dimmer that does not. How can a dimmer use pulsed voltage without causing flickering? My idea of how electronic dimmers work is that the time gap between each 120v pulse becomes larger on the lowest setting, and the time gap becomes zero at full power (which is why I don't understand...). When my GE dimmer worked, I tested its voltage with my multimeter, and the voltage at the lowest setting was like, 40 volts, and was something around 80 or 90 volts at the highest setting. ()

I would also like to know what makes one type of filament light at its full rated capacity at 120v, and another to light at its full rated capacity only at 220v. I've seen both types of filaments and they look the same to me. If they are both made of tungsten, and are of the same length/thickness, what makes them light at such different voltages? Shouldn't the 220v filament be a lot longer than the 120v one (like in antique cage fil. lightbulbs)? Or is there a way to alter the resistance of a tungsten filament?

I am worried because in the past I used an electronic dimmer to light a few of my antique lightbulbs. Fortunately, none were damaged but it still worries me to know that I ran them at full power without knowing. I also would like to know the effects of running a 115v lightbulb at 130v: Does it suffer damage that it would never get running at 115v, or does it only burn brighter & last less? These are all interesting things that I would love to know.

Thanks!


-Yoshi

[Mónico González]

Hi Yoshi!

About Dimmers, I would say you that these devices, as you know, are basically constituted by series mounted phase-shift triggered Triacs, so, the power regulation are get by wave "chopping", not by a true voltage control. Depending upon the moment the conduction are established at, the RMS output value are higher or lower. Therefore, the average voltage drop through the whole filament will be proportional to the conduction time and the effective voltage value reached during that conduction time.
Also, the abrupt 100-120 alternating pulses per sec. forces the filament to vibrate at the same rate strongly than at senoidal mode, so this tends to suffer of mechanical stress, giving as a result a shorter life of the lamp.
Because the AC waveform becomes highly distorted when dimmers are used, most meters cannot give an accurate measuring, mainly digital ones, whose particular sample and hold characteristics makes very difficult to get  exact readings. Only a true RMS voltmeter could give you a correct measuring of average volts.
If a Triac becomes short-circuited when we are testing a low voltage lamp, the full line voltage will be applied to the filament, giving as a result that it will be "gone with the wind" in a fraction of second. Worrying!
As a rule, transformers are safer (and expensive) than dimmers to lamp testing.

Slightly overruning tungsten lamps during a few seconds, gives not cathastrophic results nor permanent damage in future, but as a rule, lamps always must be powered at nominal voltage (or below it) but never at voltages at 5% or more above it.
A permanent 5% overruning above the rated voltage, can shorten the life of the lamp almost at a half of its nominal life-span.

The sole way for filaments voltage-power rating are the adequate choice of its lenght/thickness ratio, because the specific conductivity of pure Tungsten cannot be varied in any way.
The apparent lenght of tungsten filaments cannot be taken as a sign of its rating, especially when the filament are coiled or coiled coil (as in all modern lamps) but as a rule, the shorter and thicker filaments are for lower voltages and the longer and thinner ones are for higher voltages.

M. Gonz?lez
 http://mis-bombillas.webcindario.com


[This message has been edited by M?nico Gonz?lez (edited October 21, 2003).]

[byronleer]

Hi Yoshi:
At first I thought the use of a dimmer was a good idea until seeing other info like Monico has supplied. I did wind up using a digital meter to confirm that even at the lowest setting on a dimmer, the voltage remained at the full amout, which is around 118 volts in my area.
Since I like to light some of my bulbs, I decided that I needed to reduce voltage to greatly extend the life of them and still allow seeing the unique filaments. I got one variable transformer that is very handy, especially if you get a bulb without a label and are not sure of the correct voltage. But what I have been using on bulbs that are rated close to my current voltage is called a half wave recitfier diode, spliced into an extension cord. This diode will take the voltage wave and cut off the entire bottom half, thus I get around 59 volts, which still allows the bulb to light nicely.
I only found these diodes from an electrician that I met and he gave me a couple. I tried pretty hard to locate them through Radio Shack, etc. and did not find them, but I'm sure someone must have them available. Good luck.

[Yoshi]

Thank you M?nico and byronleer!

Let's say I want to measure the true output voltage of a certain "electrical device" (be it a ballast, transformer, variac, triac, dimmer, etc).; How can I be sure of the voltage reported by my multimeter? It couldn't measure the correct voltage of the dimmer, so now I don't trust it. I have a transformer from an old printer that produces 40 volts, which I think is great for lighting old bulbs; But, how do I tell if this device REALLY has an output of 40v? I have both an electronic and a mechanical multimeter. Where do I get a true RMS voltmeter? (what's that?; please excuse my ignorance...).

Of course, I would love to buy a variac but their cost is prohibitive for my current "income". I'm glad that my dimmer didn't damage my bulbs, thank you M?nico for confirming it, you've been a huge relief! It had worried me for a long time. Now I can look at my bulbs and smile again.

Right now, the voltage here is so high! (138v!) This is because the weather is so good, that virtually no cooling or heating devices are needed. In the summer the voltage usually is approximately 127v (it is supposed to be 125v). This is how our electrical system always has been. Voltage regulators are a must for computer systems in this country. (no not surge protectors; VOLTAGE regulators, which besides keeping the voltage constant also protect against surges). I also like lighting my lightbulbs ocassionally but I have been unable to light them for a long time, due to the previously mentioned issues. All I have now is a 1-18v variable transformer from a toy train, which produces a very dim glow on tungsten bulbs, but will not light carbon fil. bulbs.

Thanks for all the help!

[Alan Franzman]

quote:Originally posted by Yoshi:
All I have now is a 1-18v variable transformer from a toy train, which produces a very dim glow on tungsten bulbs, but will not light carbon fil. bulbs.

Check if the secondary of the train transformer is isolated from the mains input and from the case (it should be for safety, test for no continuity with ohmmeter between one input and one output terminal as well as one output terminal to any metal parts on the outer case.) If it is isolated, you could connect it in an autotransformer configuration to reduce the mains voltage.

This means to put the primary and secondary in series across the mains, and take your output from the primary winding only. Since this is a variable transformer, you can reduce the reconfigured output by 0 to approximately 12%. This should be enough to drop your 138 volts down almost to 120 volts. WARNING: when you do this the secondary will no longer be isolated - treat all connections of the transformer as dangerous when connected this way.

Also, the above estimate assumes that the primary and secondary windings are connected "in phase". If you connect them out-of-phase, you can get even greater voltage reduction. Try it with a cheap modern bulb connected first, measure the voltage range you can get at the output, then reverse the secondary connections and test it again.

Finally, whether in phase or out, you'll be drawing most of the current through the secondary winding. If you can't find the output current rating on the transformer, divide the transformer's power rating by 18 volts to get the approximate output current rating. Multiply the output current rating by the nominal (125V) voltage to get the transformer power rating in this configuration.  For example, if the transformer were rated for the unlikely figure of 9 watts, output current rating would be 0.5 amps. 125 volts into your load at 0.5 amps is 62.5 watts, so you could operate up to a 60 watt bulb this way.

------------------
 
Alan "A.J." Franzman

Email: a.j.franzman at verizon dot net

------------------

[This message has been edited by Alan Franzman (edited October 24, 2003).]

[Mónico González]

Alan's procedure is almost the same I've done with fixed transformers to reduce the 230v to 220 and the 130v to 125 at my recording studio, as I said to you in my last e-mail, Yoshi.
In all Europe (also here in Spain) mains voltages are being raised from 220-380 to 230-400 volts from almost a year, to meet CE new electrical requeriments. Some appliances are quite sensitive to this little over-voltage, so I've done the system above mentioned by Alan to reduce in some volts the input mains.
I have not only 220-230 devices but 125v too, because this was the lower standard voltage until no more than 15-20 years here in Spain. (of course, always at 50Hz as now).
Therefore, when we need to power any old 125v appliance, (I have a couple of them) they becomes slightly overdriven if we does use the simplest reversible auto-transformers available here, wich are rated to 220-125v. As mains voltage are 230-233v most of times, the self-xformers output are as high as 132-134v! instead the nominal 125-127v. So, the simplest solution are the use of little low voltage transformers whose secondaries has to be connected in series at out-of-phase mode to substract some volts from mains.

About voltage measuring, the newest multimeters gives true RMS readings only when the input waveform are sinusoidal or very near to it. So if you want to measure transformer outputs, fluorescent ballast, or another non excesively distortive device, all readings would be quite accurate.
On sinusoidal waveforms, the peak voltage value are equal to the RMS value multiplied by square-root of 2, therefore, digital meters that often makes a sampling from rectified and capacitively integrated peak values, are not accurate (unless some true laboratory instruments) to measuring non-sinusoidal waveform RMS voltages or currents.
But don't forget: Dimmers are a higly distortive devices, therefore the output from these devices only would be measured by means of true RMS voltmeters (special moving coil meters with built-in rectifier or AC moving iron type instruments).

Best regards,
M. Gonz?lez.

P.S. For Byronleer: 1N4007 diodes does work pretty good, allowing to handle up to 1 amp. at 1Kv, and are very cheap and widely availables.


[This message has been edited by M?nico Gonz?lez (edited October 25, 2003).]

[Yoshi]

Hello Alan, I'm not sure what you mean... I really have no need for a device that can turn 138 volts to 125v because I already have one for my computer, it has 4 outlets and I'm currently using only one. My toy transformer is rated as follows:

Brand: Bachmann
Model #: 6607
Input: 120v 50/60 hz
Output: 17 VDC 20 VAC
Max. output: 7 VA total

Since the actual input for the transformer is higher than 120v, it is giving a slightly higher maximum output than it's supposed to... (18v). It does have two terminals, one is for the train (variable) and the other is for the track accessories (fixed output). The transformer is around 8 years old. I have used it extensively for electrical experiments.

Thanks for all the help, it is very much appreciated.

-Yoshi

[Alan Franzman]

Oh, I see - the variable output is DC.  You must NOT use this transformer the way I described, with the variable output in series with the primary.  You could do it with the 20VAC output in series with the primary for a fixed reduction (or two values of reduction, depending on the order of the secondary connections.)  But with a power rating of only 7 VA, this transformer is too small to be very useful anyway.

[Haberman]

Hi All,
I had started the discussion about extending old bulb life and was told I could use electonic buttons under the bulb that are available at http://www.scottenterprises.com/shop/220.html
They say it converts the AC to pulsating DC and was wondering if this was OK for old bulbs. Any info especially in laymans terms would be greatly appreicated.Thanks, Jack Haberman Davenport, IA

[Ross]

Hi Jack Haberman,

I'm guessing that the device is a diode, which only conducts current in one direction.  This means that the a.c. mains only flows through the bulb during one half of its 60Hz (50Hz Europe) cycle.  This is known as half-wave rectification. The effect is that the bulb burns at approximately half power.  It is a trick I have used here in the UK to make Christmas lights last longer; a diode will fit neatly inside a UK plug.  But in peripheral vision the lights do have a slightly irritating flicker; certainly true with 50Hz mains but maybe less noticeable on 60Hz.  I would be very wary of the effect on antique filaments though. The light turns off during every half cycle, so there is likely to be some vibration and thermal shock.  I note that the advert specifically says to use a new bulb, probably because used tungsten filaments are much more brittle.

Cheers,

Ross.

[gutfinski]

Hi,

For my early colored bulbs, I simply use two bulbs in series, one for the display and the other to serve as a resistor.  By changing the wattage (actually the resistance) of the other bulb, I am able to vary the actual voltage across the filament of my display bulb through a fairly wide range.  Presently, for the Christmas season, I am running a couple of these "series circuits" with true ruby colored glass Mazda bulbs in an outdoor display.   Even at half rated voltage, the red bulbs illuminate with a rich, red color.  Also, since they do not run hot, there is no danger of breakage from the occasional raindrops or snowflakes.

[Yoshi]

Hi Gutfinski,

I was wondering, do you use ALL of your lightbulbs for your display? Or do you keep the better ones indoors and turned off, especially those with mint stickers?

-Yoshi

[gutfinski]

Hi Yoshi,

I don't use anything with a paper label outside but do display most of what I have from time to time on low voltage.  Actually, the cobalt blue and ruby red glass bulbs show up very nicely even on 50% rated voltage.  This is surprising with the blue as both carbon and tungsten filaments give out light mostly in the red-yellow area of the spectrum.  The typical "voltage vs. filament life" curve indicates that at less than 75% rated voltage, filaments should have an extremely long life.  Of course, there are other factors involved, but basically, if you give good care to these old bulbs and run them on reduced voltage, there is no reason to fear displaying them "live."

[Yoshi]

I see... However, I wouldn't light any of my bulbs for more than a few minutes, in any voltage (except maybe used, clear regular-size cage fil. mazda bulbs). You are right in that they should last very very long on low voltage, but... what if they don't? What if one of them has an imperfect vacuum, or defective/offset filament? This is why I don't like to light them. Of course, this is just my opinion. I'm not saying you shouldn't light yours...

By the way, have you taken pics of your display yet? I'd still like to see it. I hope you don't mind. Thanks