Gas discharge bulb

[philo]

Because he heard I was good at fixing things, a guy with a
silk-screening business asked me to look at one of his lights.


It's a backup unit so not crucial, but he said no one in town can fix it
and it would cost a fortune to send it back to the factory.
He was told that it's old and obsolete and even the factory would
probably not be able to repair it anyway.


I told him I'd look at it, but there was a good chance it would be ready
for the scrap heap when I got done.


It had "complex" circuitry and as far as I could tell, it first applied
a "lower" voltage to the bulb (perhaps 230 volts). Then after it warmed
up, would boost the voltage to 600 or so.


I know nothing about gas discharge lights, I was wondering if anyone
here has an info. As far as I could tell, they emit a lot of UV.


He told me that when he first started his business, he had a home made
500watt halogen light that worked fine...it just took 15 minutes to do a
burn and he wanted to do things a bit faster.


After I gave up trying to repair it, I removed the bulb and sockets and
retrofitted halogen fittings and a 230v 1500 watt bulb.


He has not tested it yet, but the thing is too bright to even look at so
it should work.


But still I'm curious about the gas discharge lamp.

It's tubular and has no filament.

[Terry Schwartz]

I don't know anything about your particular bulb, but as far as I do know, most gas-discharge bulbs require an initial high voltage strike to ionize the gas, which can be followed by a lower maintaining voltage. Neon bulbs and nixie tubes work this way -- they become essentially constant voltage zeners once ionized. A series resistor typically provides the current limiting and drops the excess voltage.

I think other forms of discharge lamps work in a similar fashion. However, some sort of electronic ballast is used and is more efficient than using a resistor.

[philo]

On 02/05/2018 12:55 PM, Terry Schwartz wrote:
> I don't know anything about your particular bulb, but as far as I do know, most gas-discharge bulbs require an initial high voltage strike to ionize the gas, which can be followed by a lower maintaining voltage. Neon bulbs and nixie tubes work this way -- they become essentially constant voltage zeners once ionized. A series resistor typically provides the current limiting and drops the excess voltage.
>
> I think other forms of discharge lamps work in a similar fashion. However, some sort of electronic ballast is used and is more efficient than using a resistor.
>

Thank you.


The best I could do with the unit was to get the bulb to light
dimly...maybe equal to a 50 watt incandescent at best.


I had no schematic and the wires were a rat's nest.

[N_Cook]

Difficult things to test other than by substituting , both ways.
Be very aware, the strike voltage is of the order 5,000 volts, repeated
until something like full current passage.

[N_Cook]

Forgot to say, because of the 5KV initially, for lash-up testing with
known good supply / known good lamp, you need extra sleeving around the
hookup wires.

[Jon Elson]

philo wrote:


>
> It had "complex" circuitry and as far as I could tell, it first applied
> a "lower" voltage to the bulb (perhaps 230 volts). Then after it warmed
> up, would boost the voltage to 600 or so.
>
>
> I know nothing about gas discharge lights, I was wondering if anyone
> here has an info. As far as I could tell, they emit a lot of UV.
>

Are you sure about all these details?

From your description, I'm thinking this could be a Mercury short-arc lamp.
Check eBay item 201564167114 and see if your bulb looks something like
that.

If so, it takes a high voltage to strike the arc, and then runs something
like 4 A at 60 V or so, for a 200 W lamp.

Mercury short-arc lamps (as opposed to Xenon lamps) have a couple of really
STRONG UV lines that are good for exposing photosensitive materials used in
the screen printing industry. And, WATCH out for your eyes, the UV out of
these things can be BRUTAL! ONE minute exposure will give you a blistering
sunburn.

There are other exposing lights I've seen that had a zig-zag tube like a
tiny neon sign lamp. I've never seen the power supply for those.

Jon

[philo]

The high voltage was not working...I was getting 600 volts max

[philo]

It is not one of those, thank you, I will post a photo later.
There is nothing inside that I can see.

At any rate, if this halogen lamp works, it will probably be safer

[tabb...@gmail.com]

yeah, we need to know what type of lamp really. Telling us the light colour would be a start, and over how long the brightness increases.


NT

[philo]

Found a part number , it's this



https://www.bulbworks.com/light-bulbs/1406-04


It seems to be a 5000 watt lamp

[tabb...@gmail.com]

On Monday, 5 February 2018 22:48:44 UTC, philo wrote:

HID lamps aren't my thing, but I believe they take a huge voltage spike to start. Halogen will give a small fraction of the output at that wavelength, increasing process times hugely.


NT

[philo]

He said that it could do a burn in a minute or two but he does not need
to go that fast. He is hoping to do it in less than 15 minutes...so I
will have to wait for him to test it to see if it works at all.


Thank you

[tom]

"philo" <ph...@privacy.net> wrote in message
news:p5armk$olu$1...@dont-email.me...

Be very careful with that lamp. They are under very high pressure and can
explode if dropped or scratched.

[John-Del]

If it doesn't work, I'd check all those alternate part numbers in your link and see if any of those lamps could be matched to a ballast. You might be able to source a stand alone board. Or email companies like Ushio and see if they can help.

[philo]

Thanks for the warning, I'm going to wrap it carefully and put it in a
safe place.

[philo]

OK but I hope the halogen lamp I have in there does the job.

[Fox's Mercantile]

On 2/5/18 6:21 PM, tom wrote:
>
> Be very careful with that lamp. They are under very high pressure and can
> explode if dropped or scratched.

Note also numerous warnings about "DO NOT TOUCH"
Finger oils on the bulb can cause catastrophic failures.


--
"I am a river to my people."
Jeff-1.0
WA6FWi
http:foxsmercantile.com

[philo]

On 02/05/2018 08:54 PM, Fox's Mercantile wrote:
> On 2/5/18 6:21 PM, tom wrote:
>>
>> Be very careful with that lamp. They are under very high pressure and can
>> explode if dropped or scratched.
>
> Note also numerous warnings about "DO NOT TOUCH"
> Finger oils on the bulb can cause catastrophic failures.
>
>

Thanks


That much I knew, I wore clean cotton gloves.

[Jon Elson]

philo wrote:


>
>
> Found a part number , it's this
>
>
>
> https://www.bulbworks.com/light-bulbs/1406-04
>
>
> It seems to be a 5000 watt lamp

5000 W?? YIKES! OK, there should be standard ballasts for this type of
lamp. Possibly bulbworks can point you to a source for the ballast.
Unless you area REAL electronics guru, a failed ballast would be tricky (and
dangerous) to repair.

And, be insanely cautious with this thing, a 5 KW UV light source will take
your skin off with just a few seconds exposure. I got a severe sunburn
trying to set up the optics with a 200 W short arc lamp, and I was barely
exposed for a minute before deciding I needed to pop in a grain of wheat
bulb for initial setup.

Jon

[philo]

At this point it would be impossible for me to put the thing back
together again in it's original configuration.

[philo]

The guy finally tested it and it did not do the job.


The good news is he was able to get a good, used unit locally.


So his shop has two working units.

[Jon Elson]

philo wrote:

> The guy finally tested it and it did not do the job.
>
>

Yes, I used a big halogen lamp MANY years ago, and had VERY long exposure
times. A 400 W mercury vapor lamp with the outer glass shield removed did a
little better. Then, I finally switched to a different photoresist that was
more sensitive, and now use filtered black light tubes and have about 1
minute exposures. (This is for PC board resist, not silk screens, so their
emulsions might still need a lot more light.)

Jon

[philo]

Well, at least he now as a 1500 watt floodlight if he needs one.