UV Sterilization Light

[HerrS]

I operate a reverse osmosis system for my home. One stage of that system uses a UV light to sterilize the water output from the system. It uses a UV bulb about 10 inches long and about 3/4 inch diameter. These units don't seem to last very long and have to be replaced. My question has to do with how this thing works:

The light is powered by a small transformer plugged into house current and is labeled as a ballast with an output of 44 VAC @ 6 A. I've examined the bulb units that won't light and see that there's what appears to me to be one of the small neon indicator lamps wired in series with the two filaments of the UV bulb. The filaments on the UV bulb are intact and, if I shunt across the little neon lamp with a 100 ohm power resistor, the UV lights. My problem is that I can't find out what this small neon bulb is. When the unit quits, the little neon bulb is blackened on the interior surface and I can see no lettering on it.

Does anyone know what the function of this neon bulb might be? I've read that the neon lamp can be used as a voltage regulator to give about 90 V and, of course, the amount of power flowing through it must be quite small. I'm wondering if the neon bulb is acting as a starter for the UV bulb.

 

[davenn]

I assume these are fluoro UV tubes ?

the neon lamp thing you are talking about is probably the started
show us a photo or 2

Dave

 

[HerrS]

I assume these are fluoro UV tubes ?

the neon lamp thing you are talking about is probably the started
show us a photo or 2

Dave

I'm posting three photos:
The shell that the bulb arrangement slides inside:

The UV bulb assembly that slides into the shell:

Closeup of the end of the light assembly:

Two of these photos are of a light assembly that has only been lit a couple of hours, so there's no blackening of the bulb around the filaments. The other photo is of a UV bulb that has seen several months of service but will still light when I shunt the neon? with a 100 ohm power resistor. Yes, it is a fluorescent UV light.

Harold

 

[Bluejets]

Can you show the whole device please (electrical not the outer tubing)?
It appears, as Dave says, a discharge lamp.
The small neon device would then be the glow switch or starter.
Operation of the whole unit is, power comes in through the balast to one heater terminal, through the heater to the starter, through the other heater and back to the supply.
As the heaters heat the gas at each end, the starter will begin to flick the internal contacts open which induces a large voltage across the tube via the balast which will eventually strike across the full length.
Following this the starter remains out of circuit and heaters are turned off until the next startup.
A faulty balast is usually the cause of heaters blowing, and if the starter is crook, it would cause the heaters to stay on too long or indefinetely flicker causing short tube life.
Balasts usually have a circuit resistance around 40ohm, depending on wattage of the tube(light)

 

[HerrS]

Can you show the whole device please (electrical not the outer tubing)?
It appears, as Dave says, a discharge lamp.
The small neon device would then be the glow switch or starter.
Operation of the whole unit is, power comes in through the balast to one heater terminal, through the heater to the starter, through the other heater and back to the supply.
As the heaters heat the gas at each end, the starter will begin to flick the internal contacts open which induces a large voltage across the tube via the balast which will eventually strike across the full length.
Following this the starter remains out of circuit and heaters are turned off until the next startup.
A faulty balast is usually the cause of heaters blowing, and if the starter is crook, it would cause the heaters to stay on too long or indefinetely flicker causing short tube life.
Balasts usually have a circuit resistance around 40ohm, depending on wattage of the tube(light)

Following are the photos you requested. Something I don't understand is where you say "Following this the starter remains out of circuit and heaters are turned off until the next startup." It seems to me that the starter (the neon bulb) can't really stay out of the circuit since it's in series with both tube heaters. I can see that it would limit the current through the heaters because of the low current that it's able to carry.

The Unit:

The Ballast:

 

[BobK]

The heaters are only used to start the ionization process in the lamp. Once it is started, current flows between the two ends and keeps the gas ionized, and the heaters are turned off.

Bob

 

[HerrS]

Well, I have some NE-2U neon bulbs coming. I'm going to replace the existing neon with one of these and see what the result might be.

 

[BobK]

A neon bulb is not the same thing as a starter. I doubt that replacing that with a neon bulb will help you.

See here:

http://home.howstuffworks.com/question337.htm

Bob

 

[Merlin3189]

Can I suggest this link http://www.osram.co.uk/media/resource/hires/334214/technical-guide---starter.pdf which explains the starting process. (With all due respect to Bluejets, his explanation does not sound quite right to me. It's like the first explanation in How Stuff Works, rather than their second explanation which matches the Osram explanation.)

Whatever. If the neon-looking object in the photos is a normal starter, then the point HerrS needs is that, once the UV lamp is operating, the voltage across it (and across the starter which is in parallel with it) is now too low to operate the starter, so the starter looks like an open circuit and plays no further part - which is what I think Bluejets means by "remains out of circuit".

Now, looking at the photo the starter does look a bit blackened, suggesting that it has operated rather a lot. It could be old or it could be faulty - in either case it would need to be replaced. The UV tube looks ok and if it gets replaced with the "Unit", then it's unlikely that he's had a series of faulty tubes.
If the starters are repeatedly failing, then I'd guess (emphasise guess) that either the unit is being started very frequently causing rapid aging, or the starter is mismatched to the tube or the power supply, so that it continues to operate after the UV tube has struck rather than going quiescent.

Sorry I can't help with suggestions for a replacement starter, but if you can get the right one, it's a simple chop out and solder in job.
If it were me and I had old failed units around, I'd try replacing a starter with one taken from low power domestic fluorescent tube starter. When you open up the plastic capsule, the bit inside is just this "neon", usually with a capacitor in parallel.

 

[Bluejets]

No, didn't get my info off "how stuff works" but rather from the old grey cells where it was put during my apprentiship some 40 years ago.
(which is why it may seem a bit different)

 

[Merlin3189]

No, didn't get my info off "how stuff works" but rather from the old grey cells where it was put during my apprentiship some 40 years ago.
(which is why it may seem a bit different)

Sorry, I hope I haven't offended. The only reason I knew about the "how stuff works" explanations was because I went to look it up myself! I probably learnt about neons about the same time as you! Starters would be a bit more recent - say 30 yrs ago.
I wasn't suggesting that's where you got your info from, just pointing out that other people seemed to bit a bit hazy about the workings of starters as well.
If anyone wants a more visual explanation, there's a pretty little video on youtube actually showing the starter working.

Back to the OP. The UV tube seems to be marked "Creator UVC-6W" which I found and it is rated 6W, 42V, 160mA, 6000hrs life, but no further info about starters or starting conditions.
I see that there are standard encapsulated starters with the GU(10?) type base, which are rated for 6W.

But looking at the pictures, it DOES look remarkably like a neon. I can't find any starters which look just like that (inside, of course), though I haven't got any very low power starters to disassemble. On the other hand, I can't think of any way that a neon would work as a starter. I don't think a neon would pass enough current to heat the filaments - at least, without going pop fairly quickly! If it did, there's no reason why the circuit should break to trigger the inductive starting pulse (except when it did go pop!)

As a real left field thought, the only other bulb I know which looks like that and has the bimetallic circuit breaker in, is the flasher bulb on old incandescent Christmas tree lights! Now that would pass 100mA or so (the only one I've tested) and break the circuit after a second or two as required to start the tube. Now that might be a cheap Chinese solution to a starter. But of course a second or two later it would short the tube again and restart it, repeating the cycle over and over again. Might account for its sputtered appearance in the photos and short life? Though I'd expect the tube to show something similar, if its filament were repeatedly being energised throughout its life.

Whatever it is, I'd replace it with a proper starter.

One more little question, for my own curiosity. If the only function of the starter is to provide current for the tube filaments and to break the circuit, why are starters rated for tube power? What difference does there need to be between a starter for a 5W tube and a 50W tube? When the tube is running, the current passes through the tube and not the starter. During starting only the filament current passes through the starter (and I don't see why the filament should be different depending on the tube power.) I would expect starters to be rated according to filament current and supply voltage, but not tube current nor power.
Anyone know anything about that? (And maybe what the filament requirements are? I can't find any data on that.)

 

[HerrS]

I certainly appreciate all you folks helping with this. I see now that this thing that looks like a neon bulb is not a neon bulb. It's a fluorescent starter (without the capacitor).

An FS-5 fluorescent starter is rated at 4-8 watts. Since these UV lamps are 6 watt, I'll see if I can come up with one of those and give it a try. I suspect that Merlin3189 is correct that these units don't come with a correctly rated starter. Once you put these sterilization units in operation, they are seldom shut down, so repeated starting shouldn't be causing them to fail. I've had to replace a number of these units (the fluorescent tube with the starter and leads) and each time the fluorescent tube itself looks fine - it just won't light.

 

[Bluejets]

Your 4-8w starter should work fine.
Usual range for starters in Aus is 4-20W or 4-80W.
The former preferred for the smaller tube and for other applications.
With too large a starter, it will never stop flickering.
Reason it looks like a neon is because it is very similar .....filled with gas and has two electrodes.
Difference is, that as the gas heats or glows, the two contacts inside bend and make and break contact giving the necessary high voltage kick from the balast.
Starters with a cap across them is for radio frequency suppression, reduces the spark a bit.

Merlin, no problems , I took it in my usual humor state

 

[HerrS]

I found an FS-22 starter (4-22W), hooked it up with jumpers and the UV lamp lit. Then I removed the jumpers and, as expected, the lamp stayed lit. So, I have a solution, thanks to you folks. I plan to remove the starter lamp from the aluminum can and probably leave the capacitor on it and I think it will all fit into the UV unit.

Thanks again to all of you for your help. This solution will save me some cash. The replacement UV lamp assemblies cost about $25.00.

 

[BobK]

Always good to hear that our advice has resulted in success.

Bob