I just took my first trip home since I moved into college, and I wanted to try to use my NA-1 to observe the interference fringes between two pieces of glass. Well I was able to do this successfully, but not without struggle.

I flipped the power switch for the lamp, and the cathodes slowly started to glow red (like normal). But even after waiting 15 seconds, the discharge still had not struck. The cathodes were just sitting there, glowing. I waited a little longer, and then broke out my mini tesla coil to help it out. Bringing the tesla coil near the lamp struck the discharge, but I know I shouldn't have to do this just to run the lamp. It warmed up and ran normally after finally striking, and I was able to successfully use it for my experiment, but I am worried about its lifespan.

The lamp did not need this kind of help the first few times I used it. It isn't like I was short cycling it, all of the times I ran it I did it for about an hour. I know argon cleanup is not the problem because when I excite the lamp with the tesla coil it glows pure purple, which means it has plenty of argon. I know there is supposed to be some amount of neon in there as well, but that is not what I see when it glows. Could this be the problem? Where would the neon have gone?

Is there any other cause that you can think of that I am overlooking? The lamp definitely has some hours on it because there is some sodium condensed in the lower envelope which isn't supposed to be there, but it doesn't look that bad in general. The glass is still pretty transparent.

Any help would be greatly appreciated.

What preheats the filaments and strike the lamp?

@dor123
It works a lot like a rapid-start fluorescent ballast, with two filament windings and a main winding. There is no starter switch or anything like that.

Have the cathode glow been of the same brightness as in the past ?

Can you measure the Voc in the state when the cathodes glow and there is no discharge ? Do you know what it should be ?

@Ash
To be honest I haven't really paid attention to the brightness of the cathode glow before it stopped working right, I have only lit it a few times. It gets brighter than cherry red, but not quite yellow hot, just a middle orangey-color. It shouldn't really need to get super hot considering it is supposedly barium oxide coated.

I am not home right now (and probably won't be for a while), so I can't measure the OCV. I remember measuring it before, but I forget what it was, and I don't know what it is supposed to be anyway. The ballast looks pristine from the outside, so I would bet that it is electrically ok on the inside as well, but of course I can't know for sure.

I know that the filament windings are on the shunted side of the transformer, so any load on the filament windings will also reduce the voltage of the main winding even with no main load connected. I wonder if I were to preheat the cathodes and then disconnect the heating circuit temporarily so the OCV goes up and it strikes, and then I can reconnect it while it runs.

What i am trying to get to with my question - Could there be a fault in the ballast ?

The filaments are on the shunted side of the ballast in order to reduce the heating supply when the lamp have struck (and clamped the ballast voltage). Of course it is supposed to be sufficient to start the lamp with the Voc in the state when the filaments are preheating

The other possibility is absorption of argon into the glass or otherwise loss of gases from the lamp, which messes the penning mix ratio. It is possible that starting it (with the Tesla coil) and letting it stay hot for a while can bake the gas back out sufficiently to restore normal operation, but i would like to see somebody confirm this before i recommend you do this

@Ash
There definitely could be, that is possible, but the lamp seems to run normally after it strikes. Still possible.

I am conflicted with the gas absorption theory, I don't really know what to think. Here is what I know:

- Modern SOX lamps have a penning mix, making it orange-ish. They have a problem with argon absorption which leaves behind just the neon, making it hard to strike.
- My NA-1 also has neon and argon, but not in an ideal penning mix. It doesn't glow orange or red, but mostly purple, meaning that there is a lot more argon than there really needs to be, or a lot less neon than there really needs to be.
- It is possible that the borosilicate glazed glass used in the NA-1 absorbs argon, but I don't think that is much of a problem since there is evidently plenty of argon in there.
- It is possible that the borosilicate glazed glass used in the NA-1 absorbs neon, which would throw off the penning mix and explain the purple color.
- I have no idea what ratio of argon to neon was used to fill these lamps, but according to Lamptech there is both argon and neon in there in some capacity. The picture of the lamp just starting up on Lamptech also looks purple, so I think the gas fill in my lamp is generally what it is supposed to be

There is evidence in both directions, I just have no clue. It could also be cathode wear, maybe the BaO coating wore off so much that it has trouble striking.

I imagine that in any mix, the majority gas is not a concern

@Ash
Well the lowest-voltage mix of neon:argon is 99.75% neon with .25% argon, and as you stray further from that then the voltage goes up, so it almost certainly makes some sort of difference. That means that the mix in NA-1s is probably not the lowest-voltage mix, despite it having both argon and neon inside.