Regarding parallel vs. series, wouldn't putting the bulbs in series decrease the starting voltage available for each? In series, they share the same current, and in parallel, the voltage drop across each is the same. So much for theory.
Actually, it is a bit more complex than that. Here is a rather lengthy discussion by Philips:
http://www.advance.philips.com/documents/uploads/literature/1-01%20to%201-69_Atlas2012.pdfPage I-7 discusses parallel vs. series. Page I-35 begins a section on wiring. Note that for high-frequency devices, there are other considerations. The best answer might be to look up your specific ballast to see how it should be wired for multiple bulbs.
Parallel vs. series may be complicated by the nomenclature used. If the bulbs share a common (e.g., white wire), then I would say they can't be in series.
John
----- Original Message -----
From: Robin Whittle
To: Homebrew_PCBs@yahoogroups.com
Cc: v321uk
Sent: Wednesday, August 01, 2012 3:18 AM
Subject: Re: [Homebrew_PCBs] PCB light box - hf ballast / tube wattage - 400W floodlight instead?
Hi John,
Generally, gas discharge tubes cannot be operated in parallel. They
need a high voltage to start, but settle down to a low voltage ca. 90
volts (I guess) once they are conducting. With ordinary inductor
ballasts I have seen two 20W short tubes, each with their own starter,
wired in series with a 40W ballast. This is for 240V. Maybe that would
work with a high frequency electronic ballast.
I chose to use a 500 watt floodlamp from the hardware store. The short
linear quartz halogen lamps give off a lot of near UV, which is what I
want for exposing Riston negative photoresist.
At a distance of about 45cm I have exposure times of 2 to 3 minutes.
This is a much narrower angle source of light than having a bunch of
fluorescent tubes only 10cm away, so for a given phototool and possible
distance from the phototool to the top and bottom of the Riston, I get a
much sharper image (shadow).
The only potential concern I can think of is heating of the phototool,
PCB and whatever means is used to hold them together. I have a fan
blowing on them to keep them cool.
There's no warm-up time for a QI light. Tubular fluorescent lamps are
likely to have some kind of warm-up time, making it difficult to predict
their light output unless they have just been running and are fully
warmed up. High pressure mercury vapor lamps have much longer warm-up
times, and so do the compact fluorescent lamps which are now widely
used. All these fluorescent lamps - tubular, 400W (or other wattage)
high pressure mercury vapor lamps, and compact fluorescent lamps - come
in "black light" versions which I think put out the ideal wavelength for
exposing Riston.
http://www2.dupont.com/Imaging_Materials/en_US/tech_info/datasheets/index.html
I haven't looked at all the types, but the types I have looked at are
specified to respond best to 350 to 380nm light. This datasheet for a
400W high pressure mercury vapour fluorescent H500-BL:
http://www.eyelighting.com/tb/Mercury/MVR/EQS-N-52-78-69917.pdf
shows most of the output at about 370nm.
However, I think their warm-up times are a problem.
QI lamps will vary their near UV output somewhat according to their
operating voltage, but unless the mains voltage varies widely, I would
be surprised if this variation was enough to cause trouble with Riston.
- Robin http://www.firstpr.com.au/pcb-diy/
[Non-text portions of this message have been removed]