[vivekiny2k]

then switch to a very old resistive regulator. electronic regulator is probably causing the problems now experienced by others (using resistive regulator, may be).

[greenhorn]

its the regulator that came with an MDL branded plate switch

[optimist]

Originally Posted by meerkat

The problem is, the bulbs keep getting fused after only a few weeks (even only a couple of weeks, -- almost never lasts much longer than a month), -- much less than the expected life of the bulbs. The failure mode is always the same, -- the bulb fuses on turning on the switch, with a flash.

The switch /lamp holder is defective and is causing loose contact (known as bounce / arcing in technical terms). Replacing it will get rid of the issue.

Your regulator has nothing to do with the problem. The life of your bulb will go up exponentially as you reduce the input power (dimmed).

Consider using a CFL in the place of the incandesent lamp. The problem might go away as they are more tolerant. Get a good Indian brand and not chinese.

hth.

[clevermax]

Originally Posted by optimist

The switch /lamp holder is defective and is causing loose contact (known as bounce / arcing in technical terms). Replacing it will get rid of the issue.

I am not sure how a loose contact can reduce the life of a bulb?

I am pretty sure that its the electronic regulator. When you turn on the regulator, there could be an initial spike of high voltage (in some brands may be) which could cause the bulb to fail.

I remember very well when we had such a bulb connected through an electronic regulator, the bulb sometimes used to glow very bright for a split second when it the regulator is turned on - especially when you turn the knob very slowly.

Hook up a normal switch in series with the regulator, and you turn on the regulator first to desired position, and then turn on the switch. turn off the switch first and then regulator - your problem is more or less solved.

[dinar]

Originally Posted by Fillmore

I have the same setup for 2 sets of fancy lights.
Its been over 2yrs now. No problem what so ever.

I connected a fan regulator in place of the normal switch, works fine.

Again it depends when was the regulator bought and which company and type, there where a lot of technologies is this field.
Plus you are using it as the regulator , ie lower voltage levels.
My I ask what ype of fancy light? are they CFLs?

@Clevermax loose contact can reduce the life of any electrical system, there are couple of factors [explaining which will put you/all to sleep, I know I did in collage ]. the main reason is there is a varying voltage at the contact point which the bulb can handle upto certain time than it give in.

We used a 60Watt bulb during our practicals in lab where we designed the regulator, an it used to last for around 50-100 hours and than blow. but htat was due the the great designing.
So there may be a chance than there is a loose contact, I think that[changing the holder] is a cheaper way to find out what the problem is.

[omishra]

How can I miss this thread?

I agree with all those guys who say -

1. Dimmers are different than electronic regulators.
2. Some electronics regulators are stepped dimmers from inside. (I changed all of perticular brand in my house due to humming noise)
3. SCR, TRIAC based regulators chop the sinosidal waveform and it causes more fluctuation for bulb which it can bear till its healthy.
4. Fan is a inductive load and it finds some part of sine wave is missing when controlled by electronic regulator. Here width of missing part is controlled.
5. Bulb is resistive load and it does not expect sudden on/off which creates vibration due to sudden changes in waveform and heating/cooling by this. ( So dimmer is suggested over electronics regulator)
6. More important, bulb does have some oxygen in its belly. It causes some oxydation of heated filament which degrades over time, and one fine day gives in.

If the electronic fan regulator is always on then your house have ideal power supply which is near to higher range (240V) and sudden change in bulb filament from cold to hot can break it after it is degraded sufficiently. This may be at perticular time too because voltages are different at different times in a day. (Last week, I measured it for a week for some other task.)

[meerkat]

Originally Posted by optimist

The switch /lamp holder is defective and is causing loose contact (known as bounce / arcing in technical terms). Replacing it will get rid of the issue.

1. Arcing would make noise. It didn't (neither at the switch, nor at the holder).

2. It would also leave its mark. There was no mark at the terminals at the holder.
I didn't open up the switch to check there.

3. In really bad cases orf arcing, the light intensity would fluctuate. It didn't.

4. As I've already mentioned, the said problem occured with two separate
(identical) set ups. It's unlikely that both would be similarly faulty.

[meerkat]

Originally Posted by omishra

SCR, TRIAC based regulators chop the sinosidal waveform and it causes more fluctuation for bulb which it can bear till its healthy.

The higher harmonics caused by the chopping of the sinusoid seems to be a
plausible cause of the problem. However, I'd expect that at the max setting
of the regulator (i.e. the way I'd used it), there would be no chopping, and
the input sinusoid would just go through, unmodified. In which case there
should be no harmonics created. I don't know about electronic regulators
enough to say if this ideal situation actually occurs in the real world.

[amitk26]

Originally Posted by meerkat

The higher harmonics caused by the chopping of the sinusoid seems to be a
plausible cause of the problem. However, I'd expect that at the max setting
of the regulator (i.e. the way I'd used it), there would be no chopping, and
the input sinusoid would just go through, unmodified. In which case there
should be no harmonics created. I don't know about electronic regulators
enough to say if this ideal situation actually occurs in the real world.

As explained earlier even on max you will not get pure sinusoid. Even if one design the circuit as ideal carefully measuring each value the input voltage varies from 210 ~ 240 and this will vary the gate bias setting ( depending on circuit used to feed gate current) thus changing the shape of output voltage. For precision applications it is fully possible to regulate gate bias but then cost will go up.
Actually no real world circuit is ideal , component of given tolerance are used as per the application and cost goes up as tolerance range is reduced. As fan regulator is designed for inductive load designer will not bother about harmonics. Even for lamp dimmer rather then decreasing the tolerance range and keeping the gate bias constant an easier solution of introducing damping should be used.

[zaks]

Just replace the regulator with a resistor type or replace the incandescent lamp with fluorescent one.

[amitk26]

Originally Posted by zaks

Just replace the regulator with a resistor type or replace the incandescent lamp with fluorescent one.

Resister type regulator means wastage of power . An easier solution exists.

Originally Posted by optimist

Consider using a CFL in the place of the incandesent lamp. Th.

FYI , You can not dim the CFL or fluorescent lamps you can just turn them on / off, An Ionic discharge needs to be created for CFL to work.


As I stated earlier the harmonics created due to chopping of waves are quite high frequency ( Radio Frequency range) these harmonics are referred to as RFI most of the time.

Due to very high frequency the rate of change that is di/dt is quite high and strong magnetic field is generated in tugston filament which is supported on tiny legs inside glass bulb and causes movement of filament. You can actually hear it with feeble krr -krr noise this movement of filament breaks it after a few days.

So solution is quite easy place a 100 ~ 150 m Henery inductance in the series that will take care of RFI.

[meerkat]

Originally Posted by amitk26

Due to very high frequency the rate of change that is di/dt is quite high and strong magnetic field is generated in tugston filament which is supported on tiny legs inside glass bulb and causes movement of filament. You can actually hear it with feeble krr -krr noise this movement of filament breaks it after a few days.

This I've never heard! I'd surely have heard it, had it been there. The only
time the bulb emitted any sound was at the moment it fused with a flash.
Same experience with all the bulbs.

[amitk26]

Originally Posted by meerkat

This I've never heard! I'd surely have heard it, had it been there. The only
time the bulb emitted any sound was at the moment it fused with a flash.
Same experience with all the bulbs.

It is not entirely necessary that hum due to RFI should be entirely audible to you always it depends on lots of factors ranging from what is the freq of RFI ( entirely depends on components in your circuit) , How loose is the filament and how much distance between bulb and your ear and how good are your ears.

But the point is that harmonics are produced in radio frequency range and any RF freq will produce magnetic field many times more then 50 Hz AC as it depends on the rate of change of current ( Lenz law if you remember from high school Physics) The magnetic field will cause a physical motion in loosely attached element and that causes filament to get destroyed.

Another way to detect this RFI is that keep a radio tuned to medium wave near to your circuit and check if you hear any khrr .. kharr noise on radio.


Actually I made a lamp dimmer long back in 1993 using an SCR bridge. I was in first year at that time and these things were not available cheaply in market . After I finished my degree in 1997 I am kind of away from devices and Circuits so may be I have forgotten few things.
But I am almost sure that to me hum in filament bulb used to be audible and I had to add some inductance by winding copper wire on a ferrite rod. May be possible today my ear may not detect that hum at all which I could do in my teenage.

Take a new filament bulb set the knob to around 50% and observe if you notice any vibrations and if the bulb is with in 12 inch from your ear and check if you notice any krr - krr.