When a fluorescent tube starts to turn pink, either at one end or throughout the whole tube, it could be one of many things:

Pink at one end:
• Rectification/mercury migration
• A slow leak at the end that is turning pink
How do I tell the difference? There has to be a way

Pink throughout:
• Mercury starvation
• A very slow leak somewhere
How do I tell the difference between these two as well?

I have seen both styles of fluorescent tube pink-ness, I always know it is one of two causes, but I never know exactly which one. Not that it really matters, but how do I tell?

Thanks!
 

Mercury starvation, nor any atmosphere contamination by itself won't cause rectification.
Rectification comes mainly from electrode assymetry and that comes most often from one of them failing (end of life) or in case of voltage fed filaments (RS and some programmed stsrt ballasts) fault in the heating circuit of one of the filaments.

I am aware of the causes of starvation, contamination, and rectification, I am just unsure on how to visually tell them apart due to their similar visual characteristics.

I see...
Normally no one care.
Either the Hg starvation is not that severe yet and the tube somewhat wakes up once warming up, or it won't (more severe starvation or the contamination by a leak or so) and the tube is being replaced by a fresh one.

Usually tubes that were contaminated when manufactured swirl a lot during the first few hours of running. Then the swirling stops. Lamps that are rectifying also swirl.

Ahh I see, so nobody really knows or cares because no matter what, the tube is toast and just needs to go.

Worth a shot…

Though not attempted this in person, I bet getting a spectrum from a seriously leaked tube will certainly show wideband radiation of nitrogen. But for the lamp that is just mercury starved, you will see only the clean lines of whatever buffer inert gas/mix is in use.

Ahh yes, that should be true. Not exactly practical if you are just walking by, but if you really need to know, then pulling out the spectroscope (or a CD disc) might be worth it.

When a fluorescent tube starts to turn pink, either at one end or throughout the whole tube, it could be one of many things:

Pink at one end:
• Rectification/mercury migration
• A slow leak at the end that is turning pink
How do I tell the difference? There has to be a way

Pink throughout:
• Mercury starvation
• A very slow leak somewhere
How do I tell the difference between these two as well?

I have seen both styles of fluorescent tube pink-ness, I always know it is one of two causes, but I never know exactly which one. Not that it really matters, but how do I tell?

Thanks!
 

quite simple really,  a tube that is mercury starved for whatever reason will generally be pink throughout the whole length of the tube, (I have heard that sometimes a small air leak can cause a tube to go dim, but as you will see below in my experience when a tube gets contaminated like that, they behave differently)


a tube that is contaminated, will often be a more orangey colour, and also generally very hard to actually run, heres a couple examples of contaminated tube, you often need special control gear setups to light (and hopefully revive) a contaminated tube


https://www.lighting-gallery.net/gallery/displayimage.php?pos=-243904

https://www.lighting-gallery.net/gallery/displayimage.php?pos=-27604


generally pink at one end is just down to mercury migration for one reason or another, usually when the tube goes end of life and is on control gear that lets it rectify, if you search the gallery for "mercury migration" you will see plenty examples, for example https://www.lighting-gallery.net/gallery/displayimage.php?pos=-102024

and you can often tell its mercury migration since the dark end will be the end that has also gone EOL with a black sock on the end so to speak

But...

In the real life I suspect things are getting somewhat complicated and many cases of borderline mercury starvation are probably caused by slow leaks. Leaked oxygen supposedly reacts with mercury and precipitates solid HgO on the tube surface, depriving the discharge of mercury vapor. At that time you sill will not see much of free oxygen and nitrogen in the tube volume as these are consumed by mercury fill and electrodes.

 

And...

For a modern '3 mg' tube this leak even don't have to be a real leak like a hairline bubble at dumet seal, it may be so called phantom leak where reactive gases slowly arise from poorly outgassed glass surface and nearby layers at enough quantity to consume the mercury available.