dedicated uv flash tubes that don't work

[baffe]

Common uv flashes use modified standard flash units. They also use standard flash tubes but remove coatings by etching, polishing or grinding the coatings away. Or in some cases they use uncoated ones. (And of course they all need additional filtering making them expensive!)

 

This are two standard flashtubes, from a Metz and DJ strobo (sqares 5mm => 3/16"):

 

http://up.picr.de/20142322jb.jpg

 

The tubes above are common types. The U-shaped one has a little more power and it's easier to focus it inside a parabolic reflector. The Function is equal, electrical parameters are similar.

I contacted a glass blower that usually manufactures neon signs. I hope this expression is right or at least that you will understand me. The craftsman makes free formed illumination tubes for example like the classic guitar outline at the hard rock cafe sign. First some measurement had to be done and then he would make me some dedicated discharge tubes to build uv flashes!

 

This plot shows the voltage of the test generator without load. Its voltage rises quickly up to 3.8 div at 10V/div at the instrument (Fluke 199C) which is 3800V at the terminals of the testbench.

 

http://up.picr.de/20142323ts.jpg

 

Connected to a flashtube the generator rises the voltage to about 3250 volts, then the tube ignits and daws current wich makes the voltage drop down to about 75 volts. Then the current drops, the lamp goes out and the voltage rises again to ingition level. This is repetitive, the cicuit is oscillating with t of 178ms until generator ist switched off.

 

http://up.picr.de/20142324dj.jpg

 

The glasblower made two tubes. Each with diameter of 12.5mm (1/2") and a length of about 25cm (10"). the tubes have internal pressures of 80mBar and are filled with xenon or argon with traces of Hg to improve ignition. I only had to put on coils of silver plated copper wire as ignition electrodes.

 

http://up.picr.de/20142441gb.jpg

 

The xenon tube showed this on the testbench:

 

http://up.picr.de/20142325yo.jpg

 

Ingnition voltage is about 1500V to 2000V. That is not too far from standard tubes. Oscillatin time is 25ms.

 

And the argon tubes test looked like that:

 

http://up.picr.de/20142326fh.jpg

 

http://up.picr.de/20142327fz.jpg

 

A completely different behavior opposite the xenon tubes. Opposite both Xenon tubes, the standard and the glass blower's. The voltage rises up to 850V, then there is ingition and voltage drops to 60V. The Voltage rises again to 330V and oscillates between 230V and 330V with a t of about 2.5ms.

 

This is the xenon tube

 

http://up.picr.de/20142328yk.jpg

 

and this the argon tube in operation connected to a high voltage test generator:

 

http://up.picr.de/20142329au.jpg

 

Later I did some other tests with the tubes connected to a standard flash unit the results were:

1.None of the tubes has a siginficantly increased uv output conpared to a standard flash tube

2.The argon tube fires but doesn't go out after some flashes. I keeps oscillating until flash is switched off.

3.The xenon tube shows signs of wear after approximately 30 flashes. It doesn't fire anymore.

A test at the high voltage testbench showed a ignition voltage that is above 3000V.

After talking to the glass blower he made two other tubes. He used UGxy filter glass tubes now of a diameter of 25mm (1") and set the xenon pressure to 8mBar and 20mBar.

http://up.picr.de/20142330zf.jpg

 

I was not able to fire these tubes. Even a ingintion coil of a car (Up 35000V-45000V) was not able to generate enough voltage to "pump" the tubes.

 

The price of the tubes was 200 Euros for the 4. 2 tubes did not work at all, the argon tube was a "single flash type" and the xenon 80mBar did not generate more light/uv/fl than polished standard tubes, and did not reach its lifetime.

[rfcurry]

Fascinating research! Thank you for sharing.

[Alaun]

Nice work, on the english wikipedia I found a link to a design paper for flashs from Perkin Elmer (but perhaps you know this):

 

http://laser-caltech.web.cern.ch/laser-caltech/report/Flash%20lamp%20Eg&G.pdf

 

 

 

Werner

[baffe]

Thank you, I do my best...

 

Werner this perkin elmer document is new to me but I know similar documents from hamamatsu and heraeus.Thank you for the link.

 

Page 12 figure M shows my standard test circuit.

 

Unfortunately the glass blower can not mix different gases under defined conditions and for small volumina. Seems he is not experienced in working with uv transparent glasses. And the pot-type electrodes are not good for high discharge currents.

 

...no way!

[colinbm]

Excellent work Stefan

Are you trying to make a flash output like figure G in the above linked article ?

Cheers

Col

[baffe]

Sorry for not answering but I first had to read and understand some parts of the document.

 

Figure G isn't possible with my handcrafted tubes. The reason is very simple: They are too big in diameter!

 

Page 11 "Operational considerations" tells: "The radiation ... is primarily dependent on current density... lesser...gas type and pressure."

 

It was my fault not to take care of that. Comparing figure G and H shows the uv output is about 6 times more if the discharging current density is 4000A/qmm (amperes per square millimeter) instead of 1000A/qmm.

 

My tubes are 12.5mm in diameter outside. Inside diameter maybe something around 10mm. Cross-section is abut 80qmm then. If 1000A/qmm (means 80000 Amperes!) is not enough current density and 4000A/qmm is required then the tubes need a discharging current of 320000 Amperes!

 

If the 1" tubes should work good they needed about 1250000 Amperes. Or 1.25E6 Amperes.

 

Impossible to handle that. The capacitor not, the wiring not and even the tube not. Not even the 320000A.

 

So first thing is to have tubes with narrow discharging channel like standard flash tubes and not these handcrafted "pipelines".