Hot to reduce 120 vac to 82 vac, 300 watt load

M

mobus1

Newbie level 4
Joined
Dec 27, 2007
Messages
7
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,341
Hello Folks, here's a unique challenge I dealing with.....
Please bear with my lack of knowledge.
I'm trying to find a way to power a 82 vac, 300 watt project lamp from 120 vac.

The lamp/lens assembly is salvaged from an old Kodak slide projector. The lamp power was obtained from a fan motor winding in the original projector.

I am building a custom projection system using multiple lamp/lens assemblies.
The surplus units I acquired did not include the fan motors, so I need to find a way to power the lamps.
To power them from 120vac directly will burn the lamp out prematurely...like in less than a minute.

I considered a simple triac light dimmer circuit (which I had built years ago for another project) on each lamp, but the RF noise finds it way into my audio sound recording studio mixer/system.

I also considered a simple power resistor mounted to a heat sink, but I dont know to use ohms law to figure out resistor ohm value or watt rating. Does the resistor watt rating need to be that of the lamp (300 watt)?

Any creative ideas?
I dont mind building a circuit at all to accomplish this, but not sure where to start. thanks
 

Hi,

To keep it simple I´d use an autotransformer.

If you don´t find a suitable one, you may also use:
A 120V --> 38V (40V), 150W transformer

Klaus
 
autotransformer was my first idea, however, i could not find anything close other than my bench unit which i tested with the lamp, not to mention the cost of the transformer i did see were over $75 each. Cost prohibitive. I need something I can build cheap as there will be aprox 8 lamps I need to run.
They can be run from a single power supply, if I can build or find one rated enough.
 
Last edited:

You need an equivalent series resistance of 10.2 Ohms. You could use a series capacitor but the value has to be about 250uF so finding one rated at >120V and class X2 would be very hard. However a series inductor of 27mH rated at >3A is more practical, smaller and cheaper. I have assumed 60Hz as the voltage is 120V, for 50Hz the values will be different.

Brian.
 
there will be aprox 8 lamps I need to run.
So where will the 20A, 2400W to power the lamps come from?
--- Updated ---

betwixt said:
However a series inductor of 27mH rated at >3A is more practical,
Click to expand...
It would need to be about 45mH to limit the bulb power to 300W, so that may be somewhat expensive also.

Here's my Eureka solution:
If you run the bulbs off a half-wave signal from a diode, the power to the 82V, 300W, 22.4Ω bulb will be 316W from a 120Vac source.
You want to run half the bulbs from the opposite polarity diode to balance the current and minimize any DC component in the AC, which mains transformers do not like.
Note that this will increase the flicker frequency, so don't know if that will be a factor.

 
Last edited:
Hello Brian, i want to be sure i have the correct understanding of your said use of the capacitor....is it used for filter a triac dimmer rf, or does is reduce the voltage when in series with the lamp?
if i use a 10.2 ohm resistor, what wattage rating do i need?
crutschow said:
there will be aprox 8 lamps I need to run.

So where will the 20A, 2400W to power the lamps come from?
--- Updated ---


It would need to be about 45mH to limit the bulb power to 300W, so that may be somewhat expensive also.
Click to expand...
i figured two separate supply circuits, maybe from a split 240 30amp line or two 120 lines.
 

Do you see my edit to my other post using a half-wave diode to power the lamps?
 
Last edited:
crutschow said:
Do you see my edit my post using a half-wave diode to power the lamps?
Click to expand...
yes, just popped up...i like the simplicity...can you suggest a equivalent diode in a package other the surface mount? stud case?

I suppose I would need to actually set it up to test for flicker, but im thinking it may not be apparent as those mid to high wattage projection lamps take a couple seconds to "die down" after power is removed.

The crazy thing is, if I could find the lamp size (package) in 300 watt at 120 volt, I would be all set. I cant even find a listing for these obsolete incandescents. I always do things the hard way.
 

Thanks crutschow, so heat is not an issue, no stud mount or heat sink needed?
 

Hello Brian, i want to be sure i have the correct understanding of your said use of the capacitor....is it used for filter a triac dimmer rf, or does is reduce the voltage when in series with the lamp?
Click to expand...
One capacitor in series with each lamp. The lamp will appear as a resistor of approximately 22 Ohms, the inductor will appear as about 10.2 Ohms in series with it (ignoring its wire resistance) to drop the voltage. The advantage of using an inductor is lower power dissipation than a real resistor. A capacitor would in theory dissipate no heat but the value needed would be almost impossible to find.

The diode trick might work given the thermal inertia of the lamp, it would be worth a try. Any diode rated > 3A and > 200V should work but note it will dissipate about 2W of heat so it would be advisable to mount it on a small heatsink and if the lamp has a cooling fan, place it in the airstream.

Brian.
 
Reactions: mobus1

    FvM

    Points: 2
    Helpful Answer Positive Rating
forgive me if im wrong, but wont the diode still pass 120 vac but at a half wave cycle?
if so, im concerned its still too high a voltage even though half the time?
 

mobus1 said:
but wont the diode still pass 120 vac but at a half wave cycle?
Click to expand...
Yes, but the average power is reduced since it is only seeing 1/2 cycle.
Look at the Average power dissipation in my simulation (small window in trace area).
 

True, the voltage will still reach the same peak but the average power will be halved. It is the power that matters, the mass of the filament makes the temperature follow the average voltage rather than instantaneous peaks. By eliminating one half of the AC cycle, the effective power is also halved. It is a simple test to try out, the danger is flicker so the trick doesn't work on lower wattage lamps where the filament has less mass or on LED lighting where the light level can change so quickly the brightness rises and falls on each half cycle.

The heatsink is needed because a diode drops a small voltage when conducting, typically around 0.7V - 0.9V at low Amps currents. Ohms Law calculates the power dissipated will therefore be as much as 0.9 multiplied by amp current which will be about 3.5A so a little over 3W of heat will be produced when conducting. It will actually be less because conduction only takes place half of the time but it is still enough to need something to conduct the heat away.

Brian.
 
thanks, im starting to learn new things today....im getting the power concept at half cycle.

im going to dig out a few diodes and do a quick test tomorrow
 

Hi,

although there are already interesting solutions...

I just came up with a new idea:
How about a (used) VFD?

They should be inexpensive and the amplitude can be adjusted. (also adjustable in frequency).
You may need additional filters.


They usually have a 3-phase output and a 3-phase input. That should not be a problem.
Soft start could help extend the lifespan.

Klaus
 

A wrong answer worth trying if pleases you : Use 120VAC wiring two lamps in series and evaluate if performs reasonably well.

120V----------------------lamp1----------------lamp2----------------neutral.
120V----------------------lamp3----------------lamp4----------------neutral.
120V----------------------lamp5----------------lamp6----------------neutral.
120V----------------------lamp7----------------lamp8----------------neutral.


--- Updated ---

Or,
240V----------------------lamp1----------------lamp2----------------lamp3---------------neutral.
--- Updated ---

Or,

120Vsplitphase1-------------lamp1----------------lamp2----------------lamp3---------------120V splitphase2
 
Last edited:

There many Lamp Dimmer IC that work smoothly in the market. Place a Common Mode Filter to the grid networks with 2 Y type capacitor.
I think it will wok.
 

Agreeing with post #18 (Externet)...
You might get the correct resistance from a bulb in series. At an earlier time It was easy to obtain a variety of incandescent bulbs: 25, 40, 60, 75, 100W. Not fussy about voltage or voltage drops.

Three of your 82v lamps in series could work with 220 or 230v or 240v. US installations have two 120v lines at opposite phase. Typical appliances are or were: Water heaters, stoves.
 

You must log in or register to reply here.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…