So Vcc = 12V. I will use pwm IC TL494 to make it regulated.
What topology to use flyback or push-pull? I can't drive more than 60W with 12V flyback, which is possible with push-pull. Right?
And when I have DTC I don't need balast for lamp?
What topology to use flyback or push-pull? I can't drive more than 60W with 12V flyback, which is possible with push-pull. Right?
And when I have DTC I don't need balast for lamp?
An externally hosted image should be here but it was not working when we last tested it.
Here is schematic reuploaded.
An externally hosted image should be here but it was not working when we last tested it.
Hi Vasko
Exactly what tube are you trying to "charge"? Normal fluorescent tubes do not like to be run on high voltage, low current supplies as they are intended to run with a hot filament. You can successfully operate low power tubes (4-8W) from 2.4-6V batteries or perhaps even from 1.5V using the Royer converter. The switching supply you show gives high voltage spikes that might be useful for running a neon tube (as you originally mentioned) but the transformer needs to be wound to provide 1000-5000V... not impossible, but insulation is important. The output from such a circuit you show is a "square wave" for half the cycle (transistor "on") but a spike from the flyback when turned "off". Even for neon tubes, this seems to me to be undesirable, and may lead to premature lamp blackening (electrode sputtering). If your lamp is a neon tube it may only light during the spike. I'd recommend designing a sinewave driver of the Royer class with the operating voltage required for the secondary, not relying on a pulse which may stress the dielectrics of transformers not necessarily designed for such voltages.
Cheers
John
Exactly what tube are you trying to "charge"? Normal fluorescent tubes do not like to be run on high voltage, low current supplies as they are intended to run with a hot filament. You can successfully operate low power tubes (4-8W) from 2.4-6V batteries or perhaps even from 1.5V using the Royer converter. The switching supply you show gives high voltage spikes that might be useful for running a neon tube (as you originally mentioned) but the transformer needs to be wound to provide 1000-5000V... not impossible, but insulation is important. The output from such a circuit you show is a "square wave" for half the cycle (transistor "on") but a spike from the flyback when turned "off". Even for neon tubes, this seems to me to be undesirable, and may lead to premature lamp blackening (electrode sputtering). If your lamp is a neon tube it may only light during the spike. I'd recommend designing a sinewave driver of the Royer class with the operating voltage required for the secondary, not relying on a pulse which may stress the dielectrics of transformers not necessarily designed for such voltages.
Cheers
John
Here is schematic reuploaded.
An externally hosted image should be here but it was not working when we last tested it.
You need some sort of protection for that output FET. With an open load, it's going to avalanche and be destroyed. Even an avalanche rated FET will be destroyed by the ensuing high dissipation.
Hi Vasko
Exactly what tube are you trying to "charge"? Normal fluorescent tubes do not like to be run on high voltage, low current supplies as they are intended to run with a hot filament. You can successfully operate low power tubes (4-8W) from 2.4-6V batteries or perhaps even from 1.5V using the Royer converter. The switching supply you show gives high voltage spikes that might be useful for running a neon tube (as you originally mentioned) but the transformer needs to be wound to provide 1000-5000V... not impossible, but insulation is important. The output from such a circuit you show is a "square wave" for half the cycle (transistor "on") but a spike from the flyback when turned "off". Even for neon tubes, this seems to me to be undesirable, and may lead to premature lamp blackening (electrode sputtering). If your lamp is a neon tube it may only light during the spike. I'd recommend designing a sinewave driver of the Royer class with the operating voltage required for the secondary, not relying on a pulse which may stress the dielectrics of transformers not necessarily designed for such voltages.
Cheers
John
I'll use common neon lamps with broken heating filaments.
I have secondary winding from TV HV transformer which gives 20kV I guess.
I will not use selfoscillating, I'll use either IR2153 or TL494. I'd rather use 494 for more control. So if that's the situation with flyback, Ill use push pull then.
You need some sort of protection for that output FET. With an open load, it's going to avalanche and be destroyed. Even an avalanche rated FET will be destroyed by the ensuing high dissipation.
Is this protection ok? (I'll stil use FET, not BJT)
An externally hosted image should be here but it was not working when we last tested it.
You're on the right path. But, I can't tell you if the values are right without knowing the inductance of your coupled inductor.Is this protection ok? (I'll stil use FET, not BJT)
An externally hosted image should be here but it was not working when we last tested it.
The other suggestions of a Royer oscillator are much better than the flyback, though...
Hi
You should take note of FoMoCo's concern about the avalanche rating of the switching transistor. The flyback energy needs to be dissipated somewhere and if it isn't in your fluorescent tube it will be in your transistor - and maybe kill it. Leakage inductance means you will need to use an RC snubber (as you show), but if you limit the primary flyback voltage with a Zener (not an ordinary diode) the output will be limited by the turns ratio of the transformer and zener voltage.
If you use a MOSFET a series gate resistor as you originally proposed is probably needed to stop parasitic oscillation in the FET. I have found that the old TTL circuit (npn driver with two NPN outputs) works well for FETS and bipolars, but does invert the drive signal.
You might be able to resonate the primary of the LOPT with a suitable capacitor and drive the FET with short pulses (<10% duty cycle) to force it into Class C and obtain a sinewave while solving the flyback issue.
Cheers
John
You should take note of FoMoCo's concern about the avalanche rating of the switching transistor. The flyback energy needs to be dissipated somewhere and if it isn't in your fluorescent tube it will be in your transistor - and maybe kill it. Leakage inductance means you will need to use an RC snubber (as you show), but if you limit the primary flyback voltage with a Zener (not an ordinary diode) the output will be limited by the turns ratio of the transformer and zener voltage.
If you use a MOSFET a series gate resistor as you originally proposed is probably needed to stop parasitic oscillation in the FET. I have found that the old TTL circuit (npn driver with two NPN outputs) works well for FETS and bipolars, but does invert the drive signal.
You might be able to resonate the primary of the LOPT with a suitable capacitor and drive the FET with short pulses (<10% duty cycle) to force it into Class C and obtain a sinewave while solving the flyback issue.
Cheers
John
Hi
You can't do a push pull without a dual winding transformer. You'll have to wind one yourself, or use Class C with a (tuned) pulse drive. Normally lamps will need a ballast to run or they present a short circuited laod (or low impedance at any rate) which will ruin the resonant circuit, but this would normally be needed for any tube driver.
You can only set the inductance once you know the frequency your converter runs at, and the leakage inductance of your LOPT etc.
Many cheap circuits on the market do just what you suggested originally and run the tube from a pulse transformer, but the lamps don't last long - this is the point.
Cheers
John
You can't do a push pull without a dual winding transformer. You'll have to wind one yourself, or use Class C with a (tuned) pulse drive. Normally lamps will need a ballast to run or they present a short circuited laod (or low impedance at any rate) which will ruin the resonant circuit, but this would normally be needed for any tube driver.
You can only set the inductance once you know the frequency your converter runs at, and the leakage inductance of your LOPT etc.
Many cheap circuits on the market do just what you suggested originally and run the tube from a pulse transformer, but the lamps don't last long - this is the point.
Cheers
John
Hi
You can't do a push pull without a dual winding transformer.
Offcourse I'll make dual primary winding for push pull, that's implies.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Power Supplies
- Neon lamp charger