Powering an HMI Light

[mrPelagic]

I have been charged with integrating appropriate electronics to produce some sub-sea HMI lights. One option we want is to be able to power them with either AC (120 or 240) or DC(240). Various ballast manufacturers have told me it's not possible to power an AC ballast with DC. Makes perfect sense to me...
But in the real world, we have done it without incident. I don't know for how long, but at least several hours continuously, multiple times. I saw a simplified block diagram indicating that the ballast basically rectifies the AC input to a nice smooth DC, then inverts it back to an AC square wave. Why not take out the middle man, and invert a clean DC imput from our DC bus?

Is this more of a lifetime issue? Are we breaking the laws of physics? Are the manufacturers just covering their butts from litigation?

any input would be greatly appreciated!

 

[waross]

If you have a conventional transformer type ballast you had better stick to AC.
If you have an electronic ballast with a bridge rectifier on the front end, you may be able to feed the correct voltage DC in either polarity and the bridge rectifier will correct the polarity.
An electronic ballast with a transformer, autotransformer, or current limiting impedance ahead of the rectifier will probably fail on DC.
respectfully

 

[itsmoked]

I agree you should have no problem with an electronic ballast with a rectifier front end.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[mc5w]

Some of Advance's Centium fluorescent electronic ballasts are UL listed for AC or DC operation. If you have an odd DC voltage such as a 180 volt or 192 volt lead acid battery backup that powers 180 volts motors or Emerson's MaxiMaestro servo drive the ballast will self adjust to the DC voltage. It is suppose to be able to run on both 120 volt DC and 240 volts AC but I cannot get anything out of the manufacturer that says that the ballast will run on the available 90 100 volts DC when the battery is near the end point and there is some amount of voltage drop in the wiring. End point = 105 volts for a 120 volt lead acid battery.

High power factor low harmonic input current electronic ballasts use a DC to DC boost converter to step up from the terminal voltage of the bridge rectifier to the main filter capacitors. The pulse width modulation controller tailors the duty cycle of the boost converter so as to deliberately create ripple on the direct current coming out of the bridge rectifier so as to closely match a sinusoidal waveform. For DC operation the microprocessor just simply has to be programmed to stop tailoring the current and use plain PWM to boost up the voltage. A wide voltage range ballast just simply has a heftier inductor in the boost converter and a slightly more complex PWM program to account for doing both current tailoring and extra voltage boosting.

Since these will work off of a 180 volt or 192 volt battery, 240 volts DC would be a good choice since a DC to DC buck converter can be used to charge and equalize the battery and a simple set of contactors can be used to switch to the battery if the normal source of DC should fail.

There is a Cahier document ect183.pdf over at the Groupe Schneider website that describes how this rectifier works. I have also done an autopsy on a failed General Electric elonic ballast see

http://home.earthlink.net/~mc5w/badnewsballasts.txt

Mike Cole