Lighting A Fluorescent Bulb With DC Voltage Not With Inverter Concept

While you had some good answer votes, I have to tell you that a fluorescent lamp will work on DC indefinitely but all of the mercury that changes electricity into photons will migrate to one end of the lamp over time. I'm not aware of any universities that teach anything substantial about fluorescent lighting. It is an obsolete art like other vacuum tubes.

The railroad used to do this, and they found that after some time they had to reverse the polarity to keep the mercury evenly distributed. This physical property is used to determine the mercury dose in a lamp but I really don't know why anybody cares about any of this anyway.

LED's are well on their way to eliminating all other forms of devices that generate light. By the way, the typical voltage drop across a 4 foot T12 lamp turns out to be anywhere from about 65 Volts to 125 Volts depending on the buffer gas.

If you keep on asking the same old question, you'll keep getting the same old answers.

This↓ was posted on 5-14-2011. It is taking you a long time to make no progress.

Checking the Practicality

Sometimes one does get better answers the second time around....

(Advantage of better spelling?)

Maybe, but does it resurrect the dead horse?

If that concept was true I'd be married

Does the high voltage created initial stage for ignition only or it last for whole span to make this FL glow?

For ignition only. Then the voltage (and, concomitantly, the current) can be reduced to a value that sustains the arc, or possibly a little higher to maximize light output efficiency.

I think some of the detractors may be missing the point.

A Fluro will run on DC after the gas inside has been ionised (although I don't know if it'll work at 6V). So it's (in principle) possible.

The role of a teacher (sir) is to push the students so they develop new knowledge, not getting them to develop a real product. Being asked to design something odd may be a great way to teach a particular concept.

A dc powered Fluro that starts with a HV pulse is an interesting idea.

So can I proceed with the circuit as shown.

There is another doubt in this circuit. The transformer in the circuit is a step down transformer 230V to 9V-0-9V 10A, but the primary and secondary windings are exchanged.

Can that actually be done i.e. reversing the primary and secondary terminals? If that is possible can it drive the FL bulb.

I got the circuit from website Electronicsforu.com

Or just open this in a new tab

http://www.electronicsforu.com/electronicsforu/lab/ad.asp?url=/efylinux/circuit/cir49.htm&title= Low Cost Transistorised Inverter

For getting the whole circuit click on the diagram which enlarges itself in a new page

Will this give output at the least

The transformer in this drawing suggests that it is an inverter.

(As has been said already.)

Yes I am doing an INVERTER. I'll face the sir the same way

Hi!

Have you heard of the "Joule Thief" concept? The web is full of circuits that pulse a high value and High frequency AC voltage across CFL (Compact Fluorescent) bulbs. I have replaced the CFL with a 20W Fluorescent bulb and no migration of Mercury is noticed even after one year of operation.

The Driver circuit of the CFL first converts the 230Volt ac to a low DC voltage using a Diode Bridge and Capacitor set up to smooth the DC output. Next the DC is pulsed using a push-pull transistor pair. I believe the small quarter inch diameter Ferrite toroid with three sets of a few turns of magnetic wire drive the bases of the BR13003 transistors (low power).

Dismantle any burnt out CFL bulb carefully (Be warned about not damaging the blub) and you can see the circuit. Consult Internet web sites that have YouTube videos of how to dismantle CFL bulbs.

Said

Actually I have heard a few things about this. First of all, the fundamental rules of physics always apply to these ideas. By that, I mean that you can never get something from nothing. In the factory, we used to light up lamps with RF transmitters because it did not require installing them into a fixture to check for a vacuum.

Note also, that you can walk a lamp under power lines of sufficient voltage and they will also light up. There is no magic here. But just because you can get one to glow does not mean that it is in any way efficient. The power company does not take kindly to people stealing power from them but it happens anyway.

Back to your circuit, the only possible use of the low voltage DC voltage is to keep the coil warm. This allows the chemical compound to emit electrons more efficiently but it is not really matched to the load factor of the plasma. And in electronic ballasts, they frequently don't bother converting that little bit of power to DC.

Powering a lamp from high voltage spikes (called induction) does have an effect on the source. You might be able to obtain some small percentage of light from the product, but it is far from efficient. LED's perform so much longer (and without all the mess) that the fluorescent business will be gone within a decade or two.

A gimmick, like what you see all over the internet, is nothing but a waste of your time unless you are studying to be a magician. If there was any opportunity to make any serious money on it, you would be the last to know.

In response to "Shreevatasv" 'I have a doubt. I have to do a project on lighting a 40W fluorescent bulb with a dc voltage of only 6V.'

I am also researching the possibility of driving 40W fluorescent lamp with a circuit to be driven off a 6 Volt battery.

The Chinese manufacturers have supplied me with an emergency light that charges its 6 volt sealed lead acid battery off the 230 Volt 60Hz. AC mains and drives 2x20watt Fluorescent Tubes delivering 6 hours of emergency lighting, during any blackout. That is 40Watts of lighting power.

I wish someone could decipher the circuit? If you could I can send you JPEG photographs of all the innards of the circuit boards!

Said I Dias

For whatever it's worth, I've been successful with lighting small fluorescent lamps with DC without any inverters. In fact, the only component was a resistor. The best suited lamp for DC is an F4T5, a small 6" tube. The power supply is 48 V DC. I use 4 small very common and inexpensive 7 AH 12V Gel Cell batteries in series for 48 V. I've also used a "wall wart" power supply that's 48 V DC at 700 mA from a router or network gear as well. The resistor is 100 Ohms 10 Watt ceramic. It is in series with the + supply and the lamp. The - supply goes to the other end of the lamp. To start you must preheat by connecting the unused pins for a few seconds until the ends glow with thermionic emission not just the orange glow of the cathodes. Most lamps will light without a choke. The best lamps for this application are old GE lamps and the 2700K Sunlite lamps. The arc voltage is around 35 V, 126 mA current, 4.4 W delivered to lamp, 1.6 W lost in Resistor. Reliable starting..100% of the time. The worst lamp for this circuit are the Sylvania F4T5. Arc voltage of 25 V, current of up to 200 mA 5 Watts delivered to lamp 4 Watts wasted in Resistor. With the right light bulb, the resistor loss is not bad for a simple circuit. Also there is no noticeable mercury migration in small lamps. I've run one for over a year 24/7 without any problems on a 48 V DC power supply.

As far as running an F40T12 on DC, you'll need about 200 V DC and a choke to help start the lamp. The current limiting resistor has to be fine Nichrome or other thin resistive wire wrapped around the - side of the lamp. The added heat from the resistor creates a thermal gradient that counteracts the effect of mercury migration and keeps the entire tube bright. You have to experiment with resistive wire so that the lamps get 425 mA current. I've successfully run 20 Watt tubes on DC with the "heated cathode" resistor wire with no noticeable mercury migration.

DC operation may not be the most efficient way to run a fluorescent lamp but it is useful for a few reasons. The first being no flicker. The second being is that you can see Crookes dark space, negative glow space, The Faraday dark space and the positive column glow in some lamps and they make great negative resistance oscillator. F4T5 oscillates at around 2.3 KHZ and F20 T12 at 1 KHz. You will see a few V AC sine wave at the anode when the lamp is lit on DC

Hope that helps

Continued:

I was able to efficiently power an 8 Watt F8T5 tube from 12 V without an inverter. I modified a 20 Watt pulsed inductor power supply that I got on eBay. I changed component values and wound my own toroidal inductor so that the circuit produced 450 V DC open circuit to instant start the lamp but limited current to 170 mA once the lamp lit. I had to experiment with core sizes and windings to get cool operation efficient operation and was able to get 85 to 90% efficiency. With your 6 V power supply voltage you need to find FETs that can switch at least 10 Amps of current with very low on resistance and a beefy inductor that won't get too hot or saturated during operation. That's the only way I can think of generating high DC voltage without an inverter topography. You will still have to deal with Mercury migration in a long tube such as an F40T12. You will be much better off with an inverter design..

40 Watts from 6 Volts (and no losses) will require at least 6.7 Amps. My guess is that you will want to aim at operating at just a few Watts or switch to a T8 lamp (32 Watts). The problem is that at normal operating currents, the voltage drop across the plasma amounts to about 10 volts per inch (at rated power). The plasma is basically resistive in nature up to about 400 Hz and then it starts acting differently.

DC power by itself will not start a linear fluorescent lamp. The lamp acts like a capacitor with high resistance until the arc (plasma) is initiated. To get past the capacitor stage, a series of sudden pulses of at least 600 VDC is required and once the arc starts, it attempts to conduct current in a runaway fashion that will blow the end off of the negative end of the lamp. So, inductors are used because they create a spike as current is interrupted and then they limit the operating current in run mode. You can reduce the strike voltage by warming up the coils at 3.6 Volts minimum to 6.3 Volts maximum. Once the lamp is running, the impedance of a 40W T12 lamp is just under 200 ohms. Before it starts it is in the megaohm range, depending on your meter.

Lastly, using a 6 Volt source at 6.7 amps is really quite unreasonable even in a perfect application. The higher current will cause energy losses and probably circuit failures if you manage to build one.

Electronic ballasts converts AC power at 50 or 60 Hz to AC power at 20,000 Hz and higher. Those higher frequencies make the incoming power look like slow varying DC power, but not at 6 volts.

Joe, could you tell me why fluorescent lamps oscillate when powered by DC? I see 2300 Hz AC at the anode...

I would say that is a natural frequency associated with the emitter. Do you observe the attachment point (a small red glow) moving across the coil or does it remain stationary?

Joe I don't fully understand your question. The lamp that I am referring to is a common F4T5/CW running on 48 V DC ballasted by a 100 Ohm resistor. About 1V of 2.3 KHz AC is present on the Anode along with + 30 V DC. You can lower the oscillating frequency of the lamp from 2.3 KHz to 1 KHz by placing a 1 uF capacitor from the Anode to ground. The power source is a pack of 4 12V Gel Cell batteries in series. There is no power supply ripple. The lamp is the source of AC oscillation..

Could be something new. I'm not sure what is causing the oscillation. I was expecting it to be a thermionic emission from the emitter but something else must be going on. Sorry I can't be of more help.

Joe, why would thermionic emission cause an oscillation? Is the lamp acting as a triode and somehow gets positive feedback? I'm guessing that negative resistance of a gaseous discharge coupled with the resistance and capacitance of the lamp is responsible for an RC type of oscillation.