Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Can this circuit be used as the dimmer of the fluorescent?

Status
Not open for further replies.

bhl777

Full Member level 6
Joined
Sep 30, 2008
Messages
363
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
5,140
Hi all, I found one circuit like this: a PWM signal with tunable duty cycle is used to drive a power BJT. The output voltage is connected to the collector terminal of the BJT.
dimmer.png

I think the output voltage will be a square wave voltage source. I am wondering if this one can be used to connect to the fluoresent. Would anyone advise me the following questions?
(1) is my understand correct that this circuit will generate a square wave voltage source with tunable duty cycle?
(2) If it is a square wave voltage source, will the high level around VCC(15V) and low level equals to GND?
(3) I am not sure if the fluoresant lamp can be modeled as the RLC tank. If so, and the square wave voltage source has voltage level from 0 to 15V, will this square wave voltage source good enough to power up the lamp?
(4) since this circuit has tunable PWM duty cycle, will it still be good to have the square wave with non-50% duty cycle and still maintain the operation of the lamp?

Thank you!
 

Everything depends on the 'fluorescent' whatever that is.

1. Only if the driving signal is able to turn off and saturate the transistor, otherwise the waveform will likely have a rounded edge or not meet the fully 'on' or fully 'off' states.
2. Only if the transistor is driven to saturation and completely off and the emitter resistor is zero Ohms. It will never go to GND, the best it can manage is VCEsat of the transistor.
3. There are many kinds of fluorescent lamps and drivers/ballasts so you need to be more specific about the type you are using.
4. Provided the driver/ballast is suitable, it makes no difference what the PWM ratio is. However, if you are thinking of standard lighting tubes, they have an ignition voltage that has to be met and if the ballast is a standard choke, it's impedance at typical PWM frequencies will probably stop the tube working altogether.

Brian.
 
  • Like
Reactions: bhl777

    bhl777

    Points: 2
    Helpful Answer Positive Rating
(1) is my understand correct that this circuit will generate a square wave voltage source with tunable duty cycle?

YES

(2) If it is a square wave voltage source, will the high level around VCC(15V) and low level equals to GND?
NO; When the transistor is on, the voltage will be something between VCC and GND (depends on the resistor values). When the transistor is off, the voltage at Vout will be very close to VCC (depends on the load impedance)

(3) I am not sure if the fluoresant lamp can be modeled as the RLC tank. If so, and the square wave voltage source has voltage level from 0 to 15V, will this square wave voltage source good enough to power up the lamp?
Fluorescent lamps are gas discharge lamps and they need a high voltage pulse for initial ionization. Once the discharge is started, the voltage can be reduced but 15V may not be enough to keep the discharge alive.

(4) since this circuit has tunable PWM duty cycle, will it still be good to have the square wave with non-50% duty cycle and still maintain the operation of the lamp?
That depends on the frequency. If the frequency is low (I am not sure how low is low enough) the tube will go out completely and will not start at the next voltage pulse.
 
  • Like
Reactions: bhl777

    bhl777

    Points: 2
    Helpful Answer Positive Rating
YES


NO; When the transistor is on, the voltage will be something between VCC and GND (depends on the resistor values). When the transistor is off, the voltage at Vout will be very close to VCC (depends on the load impedance)


Fluorescent lamps are gas discharge lamps and they need a high voltage pulse for initial ionization. Once the discharge is started, the voltage can be reduced but 15V may not be enough to keep the discharge alive.


That depends on the frequency. If the frequency is low (I am not sure how low is low enough) the tube will go out completely and will not start at the next voltage pulse.

Hi c_mitra, would you advise what is the typical switching frequency for such a fluoresent dimmer? Do you think is 10kHz good enough? Thank you!
 

Normal fluorescent lamps can survive 10-20ms backout; they work perfectly ok with 50-60Hz. So, 10kHz is certainly ok.
 

Normal fluorescent lamps can survive 10-20ms backout; they work perfectly ok with 50-60Hz. So, 10kHz is certainly ok.

Hi c_mitra, thank you very much! I have an additional question of this circuit.
If we do not care about the settling and just to let the fluoresant lamp to operate in the steady state, is it pssoible to tune the duty cycle (say the PWM frequency is 60Hz) to change the brightness? For example, can I have a 100% duty cycle (a DC voltage at "high" level) to have the full brightness, a 50% duty cycle square wave to have the half brightness, and a 0% duty cycle (a DC voltage at "low" level) to have the minimum brightness?
 

They normally work on 50Hz or 60Hz sine waves so you have to be careful when using square wave ratios to define a proportion of time. If you took say 10% PWM ratio and synchronized to zero crossing, the peak voltage across the tube would only be about 10V and it wouldn't light at all. As the PWM ratio increased it would eventually start to flicker and then stay on all the time. This assumes of course that the tube is warm and the heaters are left running all the time.

You still haven't told us what kind of 'fluorescent' this is, there are many different kinds. If I again use a standard white 'domestic' tube as an example, you will get some dimming but the light output to PWM ratio would also be hard to calculate because the tube has a degree of 'afterglow' which would tend to level the light output during off periods.

I think the only way you might achieve any reasonable dimming would be to drive each heater through an isolating transformer and take the ends of the tube to a 'H-Bridge' driver running on a high DC voltage. That would allow instantaneous ignition, even over brief PWM periods and still allow the polarity to be reversed to simulate AC.

Brian.
 

They normally work on 50Hz or 60Hz sine waves so you have to be careful when using square wave ratios to define a proportion of time. If you took say 10% PWM ratio and synchronized to zero crossing, the peak voltage across the tube would only be about 10V and it wouldn't light at all. As the PWM ratio increased it would eventually start to flicker and then stay on all the time. This assumes of course that the tube is warm and the heaters are left running all the time.

You still haven't told us what kind of 'fluorescent' this is, there are many different kinds. If I again use a standard white 'domestic' tube as an example, you will get some dimming but the light output to PWM ratio would also be hard to calculate because the tube has a degree of 'afterglow' which would tend to level the light output during off periods.

I think the only way you might achieve any reasonable dimming would be to drive each heater through an isolating transformer and take the ends of the tube to a 'H-Bridge' driver running on a high DC voltage. That would allow instantaneous ignition, even over brief PWM periods and still allow the polarity to be reversed to simulate AC.

Brian.

Thank you Brian, I am using standard white 'domestic' tube.

If we can assume the lamp is already lighten up with a default PWM duty cycle=50% and a higher "high" voltage (say, 28V), do you think this circuit will work by increasing PWM duty cycle to 100% or decreasing to 0% in changing the dimming? The relationship does not matter at this momemnt, we just want to see if the concept is OK, and if the duty cycle needs to be limited. Do you think we can go from 0 to 100%, or we need to limit it to a certain level, such as 10% to 90%?

Thank you!
 

With the limitation that you will probably have to keep the heaters running, at least at low power, I would guess you might be able to dim down to about 50% before flicker and flashing become too obvious. It certainly isn't recommended though and it certainly won't work if you 'PWM' the incoming AC to the ballast unit, you would have to do it directly to the tube itself.

You have to understand how tubes work, they are not like conventional filament lamps where the brightness is related to the power you feed them. They are more like an arc lamp, you have to go over ignition voltage to start them and then control the current through them. A standard tube will not light up if you just apply a voltage across it, they need assistance from heated electrodes at each end and a high voltage kick to ignite them. If the voltage drops below sustainable level, you have to repeat the process to make them light again. When they are lit, the tube conducts heavily and becomes almost a short circuit across it end to end, thats why the ballast unit is needed, it limits the current to a safe level. When you use PWM you hit two problems, the first is it is a square wave which is unipolar, you have to convert it to AC, the second is you have to produce the PWM at low frequency so the ballast can work and synchronize it to the mains AC waveform.

Try a simple experiment: A standard triac light dimmer, the kind you fit to a wall instead of a switch, uses phase control which in this case is almost identical to PWM. Wire one in series with your lamp and see what happens. The control adjusts the firing time of the triac from almost zero to almost 100% and it does it at mains AC frequency so the results should be comparable with a real PWM driver.

Brian.
 
  • Like
Reactions: bhl777

    bhl777

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top