Re: HIDing behind watt?

[Luis Marques <marques@ti.com>]

> From: "Graydon D. Stuckey" <graydon@apollo.gmi.edu>
>         I spent a few minutes playing with an HID light today.  This 
> one was made by Hella, and appears to be very close to a production 
> unit.  It has a replacable bulb, which is very good.  Maintenance 
> costs should be much reduced as a result. 

Although expensive, these lights don't have filaments to burn/break,
which makes them an ideal candidate for hi-vibration applications.  They
should last at least 10 times as much as a halogen, so they should last
as much as the car, unless you are involved in an ac$ident.

> ... The bulb was made by 
> Phillips, and was marked as a 35watt bulb.  I found that it produced 
> relatively equal light regardless of input voltage.  when the voltage 
> dropped as far down as 8 volts, then you could tell it dimmed 
> slightly, but as long as I was above 10v, it was as bright as at 14v.  
> At 14v, it drew 3.0 amps.  While it was still cold, the current seemed 
> to fluctuate a little between 5 and 3 amps, but it eventually settled 
> at 3.0 amps.

Yes, that's how they are supposed to work.
HID lamps are basically a glass chamber filled with high pressure gas
and two electrodes.  Light is produced when current arcs trough this
gas.  For the initial spark to jump, several 100's of volts are
required, unless there is a starting resistor to warm up the gas really
hot.  Still, there is an initial high voltage requirement.  Once the
current is flowing through the gas, it (the gas) turns into a VERY hot
ionized gas/plasma mixture and its resistance is so small that only a
couple of volts are required to keep the current flowing.  This is why
the light output doesn't change much with the supply voltage - the
ballast is regulating the voltage going to the bulb, so it doesn't
notice the difference.

During the initial turn on period, the ionized gas/plasma is not hot
enough and the resistance of the arc is a little higher, so the ballast
supplies some aditional current to bring everything up to operating
temperature and that's why the current fluctuates a little.  Also, 

>         I removed the cover to the aluminum box, and was absolutely 
> astonished at the sheer numbers of electronic components!  There were 
> hundreds of resistors, microchips, transistors, capacitors, and a few 
> other things that i couldn't name.  It looked like a computer in 
> there!

All this is needed to boost the voltage high enough to start the lamp
and to control the current going to it once it starts, just like stadium
lighting.  As you can imagine, producing several 100's of volts from 12V
and going back to a steady 5-6V in a timely and precise manner is no
easy task.

>         I can see why they are very expensive at this point.

At this point, they are recuperating development costs.  If they were
manufactured in high quantities, the cost would be much lower.

Later,

Luis Marques
'87 4kcsq