Trad-Lighting Photo Gallery

UV laboratory source
This is an old UV light I have in my collection. It has a 4-pin (B4) vacuum tube type base potted onto the electrode lead-ins. I know very little about it other than it was from a bit of laboratory equipment.  I also have an old book that has a drawing of a very similar lamp, noting it as a low-pressure UV lamp containing argon with mercury. It’s made of fused quartz with hollow cold electrodes as shown.  The U-tube measures 30cm in this example.  The book states they are designed to work with leak-reactance transformer.  The lamp is rated from between 20mA to 150mA which takes it a bit beyond the ability of neon sign transformers. I can confirm this as the mercury will not evaporate on a sign transformer and the discharge remains in argon. The closest I can get is on an oil burner ignition transformer but it is over loaded and the mercury discharge only migrates to about 50% of the tube length.  The voltage across the electrodes at this point is still a high 725V AC.  A sodium leak ballast could provide the current but can’t strike the lamp properly.  The lamp produces ozone when the mercury emissions start to grow. 
Keywords: Lamps

UV laboratory source

This is an old UV light I have in my collection. It has a 4-pin (B4) vacuum tube type base potted onto the electrode lead-ins. I know very little about it other than it was from a bit of laboratory equipment. I also have an old book that has a drawing of a very similar lamp, noting it as a low-pressure UV lamp containing argon with mercury. It’s made of fused quartz with hollow cold electrodes as shown. The U-tube measures 30cm in this example. The book states they are designed to work with leak-reactance transformer. The lamp is rated from between 20mA to 150mA which takes it a bit beyond the ability of neon sign transformers. I can confirm this as the mercury will not evaporate on a sign transformer and the discharge remains in argon. The closest I can get is on an oil burner ignition transformer but it is over loaded and the mercury discharge only migrates to about 50% of the tube length. The voltage across the electrodes at this point is still a high 725V AC. A sodium leak ballast could provide the current but can’t strike the lamp properly. The lamp produces ozone when the mercury emissions start to grow.

UV_lamp.jpg half_coated_Hg.jpg HPLR400.jpg BL_g_by.jpg BL400.jpg
Lamp/Fixture Information
Model Reference:UV
Lamp
Lamp Type:low pressure Mercury
Filament/Radiator Type:fused quartz
Base:pin B4
Burning Position:universal
Fixture
Ballast Type:leak reactance
File information
Filename:UV_lamp.jpg
Album name:Tuopeek / Mercury Vapour
Keywords:Lamps
Filesize:281 KiB
Date added:20 Nov 2026
Dimensions:2500 x 2060 pixels
Displayed:10 times
DateTime Original:2026:11:20 21:12:12
Exposure Time:1/125 sec
FNumber:f/5.6
File Source:Digital Still Camera
Flash:No Flash
Focal length:55 mm
ISO:800
Model:NIKON D3400
Software:Adobe Photoshop Elements 2.0
White Balance:0
URL:https://trad-lighting.net/gallery/displayimage.php?pid=1467
Favourites:Add to Favourites

Comment 1 to 7 of 7
Page: 1

Max   [Sat 21 Nov 2026 at 07:59]
Interesting lamp. At what current level were you driving it when you took the picture? The cathodic glow extending outside the hollow cathode, as it does here, is the sign of a too high current - the negative glow is confined entirely inside the hollow cathode during normal operation. Another interesting detail visible here is the deep blue glow of the lamp's quartz material. I have seen that in other quartz mercury lamps such as the Philips HUV 5, a small 4 W argon-mercury hollow cathode UV source. Spectral measurements revealed that this is not fluorescence, but caused by Rayleigh scattering from (sub)nanometer-sized particles present in the glass material.
Tuopeek   [Sat 21 Nov 2026 at 10:51]
Thanks Max, the current as shown is 34mA. If reduced to 15mA the outer glow discharge from the hollow cathodes stops. In either case full mercury discharge along the full length is not achieved. The lamp temperature never becomes too hot to touch. This fits with the documentation I have. It's interesting you mention the outer hollow cathode glow is an indication of too high a current. I am sure I have noticed this in neon tubes I have made with too low a gas pressure but thought it related to the much higher voltage required. I must look at the current next time. In my case it quickly sputters the cathodes badly.

The Rayleigh scattering is an interesting observation and may well be the factor here with the age of the lamp and processing of the gas, at the time of manufacture.
Max   [Sat 21 Nov 2026 at 11:47]
So, they clearly used undersized hollow cathodes here. Either they had no idea about how those electrodes operate exactly or/and it doesn't really matter if sputtering damages occur if the service life of those lamps is determined primarily by their UV output and the rate of flux depreciation. Quartz solarization will require their replacement long before the lamps die out of argon cleanup as a result from trapping by the sputtered electrode materials.

As for high-voltage neon lamps and hollow cathode discharge tubes in general, the maximum operating current is not determined by the electrode size alone. As the gas fill pressure determines the properties of the negative glow (UV emission intensity, charge density, sheath E field, etc), it also determines the maximum operating current of a given hollow cathode as well.
Ria   [Sat 21 Nov 2026 at 14:32]
Interesting; would a parallel ignitor help with ignition if you use a leak transformer..? Wonder
Tuopeek   [Sat 21 Nov 2026 at 15:42]
Max - It may well be the electrode size and type were not a big concern, for the lamps use and application. I must say it looks very clean inside, despite being salvaged from old lab equipment, where I suspect it was used for fluorescence investigations. I have some plates of ‘Woods’ glass from its original enclosure. I couldn’t say how often it had been used though. There is no noticeable sputtering deposit around the cathode either. In my neon experimental tube, I had heavy sputtered mirroring around the electrodes in less than 30 minutes. However, I don't think the current was the issue it was the net power, given the almost 20 kV across the tube.

Ria - I did think about trying that trick with an auto-leak and my small trusty, lights anything, Tesla Coil to trigger it. The only ballast I had to hand was for the 180w SOX and it may still struggle to keep the lamp on with it’s over 700V running voltage. I decided not to try it as the current would be significantly higher if it struck, and from Max’s observation, I may already be over stepping that.
Drew   [Sat 21 Nov 2026 at 15:42]
Never even heard of Raleigh scattering but that is a very interesting phenomenon. I was gonna say if it was fluorescence that wouldn't be too optimal for a UV lamp. Very cool construction, I wonder why the mercury doesn't want to evaporate at lower currents, I suppose this is due to a high pressure of argon used?
Tuopeek   [Sat 21 Nov 2026 at 17:05]
Raleigh scattering is just very small particles scattering light. You may know it better from, 'why the sky is blue.' Smile In the case here, I would probably think first that there is fluorescence happening in the quartz, when it's actually just scattering. The mercury needs to have a small rise in temperature to evaporate like in a fluorescent lamp appearing a little dim at switch on, so it needs a little bit of heat to be generated by the current.

Comment 1 to 7 of 7
Page: 1