Via sci.techniques.xtallography newsgroup

From: Mike Meier [mlmeier@ucdavis.edu]
Newsgroups: sci.techniques.xtallography
Subject: XRD Tubes
Date: Wed, 15 Sep 1999 18:22:39 -0700
Organization: U.C. Davis
Xref: daresbury sci.techniques.xtallography:5660

I would like to learn more about the modern x-ray tube, especially
things which effect its useful lifetime.  Can anyone recommend a book,
article or web site?

The reason I ask is that I am about to retire a two year old copper tube
due mostly to the increase in the intensity of a tungsten_L_alpha peak.
It is now at about 4% of the intensty of the copper_K_alpha peak and is
creating problems for some of my users.  The tube seems to be good in
all other respects, and while I could use the Ni filter, the drop in
intensity (about 50%) would be a problem for many of my other users.

I realize that two years of useful life is not bad, but maybe there is
something I can do to keep it alive longer.

Also, what are the advantages of the ceramic tubes?  I just received a
quotation for one that costs about 30% more than my current tube.  It's
power rating is the same, 2.2 kW, but its warantee is longer.  But if I
still get the L_alpha peaks after 1.5 years, what would be the point of
spending the extra money?

Mike Meier
mlmeier@ucdavis.edu
www.matsci.ucdavis.edu

From: l.cranswick@dl.ac.uk (Lachlan Cranswick)
Newsgroups: sci.techniques.xtallography
Subject: Re: XRD Tubes
Date: Thu, 16 Sep 1999 10:10:24 LOCAL
Organization: Daresbury Laboratory, UK
Xref: daresbury sci.techniques.xtallography:5662

Have the ceramic tubes been around long enough for people to
be able to draw adequate conclusions on them of there value
and robustness vs reliability vs glass tubes vs price?
The theory is that ceramic tubes should be better but I have
heard horror stories involving ceramic tubes  - but this was 
mainly due to dodgy generators - not the ceramic XRD tube.

----

For trace phase ID work in Australia (amongst other uses, 
Rietveld, etc), the % amount of parastic Cu K beta lines 
and Tungsten L was quite important for us (grain boundary 
type phases that could overlap with the parasitic lines).   
When the amount of Tungsten reached what for us was a 
high level (~0.3%), it used to get donated to the 
single crystal lab next door and a new tube purchased.

This meant going through around 1 glass Long Fine 
Focus Powder XRD tube per diffractometer per year.  But 
that was the price for doing the work on these types of 
phase systems.  
(Appearance of Tungsten over time could vary with tube).

One possibility is to specify the performance you expect
in what you purchase and see what the vendor will go with.
In Australia, we had mainly Philips Powder XRD equipment and 
Philips Australia were very good on this type of support if things 
didn't come up to a spec we had both agreed upon.

Lachlan.

PS: Given the life-times of good XRD tubes, a lot of this 
area can be folk-lore-ish as it is hard to go through a 
statistically valid number of XRD tubes to draw scientific 
conclusions(?)

-----------------------
Lachlan M. D. Cranswick

Collaborative Computational Project No 14 (CCP14)
    for Single Crystal and Powder Diffraction
Daresbury Laboratory, Warrington, WA4 4AD U.K
Tel: +44-1925-603703  Fax: +44-1925-603124
E-mail: l.cranswick@dl.ac.uk  Ext: 3703  Room C14
                           http://www.ccp14.ac.uk

From: Cristal [cristal@cybercable.tm.fr]
Newsgroups: sci.techniques.xtallography
Subject: Re: XRD Tubes
Date: Thu, 16 Sep 1999 19:18:51 +0200
Organization: Private
Xref: daresbury sci.techniques.xtallography:5665


Lachlan Cranswick wrote :

> The theory is that ceramic tubes should be better but I have
> heard horror stories involving ceramic tubes  - but this was
> mainly due to dodgy generators - not the ceramic XRD tube.

We have horrible story here for the ceramic tubes mounted
on the Bruker D8 Advance. The decay is 15-25% per month,
observed by careful measuring on a good standard. This
cannot be considered as normal, so that the manufacturer
has already replaced graciously 2 tubes since December 1998...

If this was due to dodgy generators, we would expect
the manufacturer to explain something, but we obtained no
explanation up to now.

Are we alone in that case ? It is extremely unpleasant to
buy such a new expensive diffractometer, and to be in the
situation of not being able to use it fully.

--
Armel Le Bail - Université du Maine, Laboratoire des Fluorures,
CNRS ESA 6010, Av. O. Messiaen, 72085 Le Mans Cedex 9, France
http://www.cristal.org/

From: diffract@aol.com (Diffract)
Newsgroups: sci.techniques.xtallography
Subject: Re: XRD Tubes
Organization: AOL http://www.aol.com
Xref: daresbury sci.techniques.xtallography:5666


The 'bible' for building and understanding eletron tubes is "Electron Tubes" by
W.H. Kohl, Reinhold Publishing Corp, NY 1960 (out of print). Or you can try
another classic "Handbook of electron tube and vacuum techniques" by F.
Rosebury, AVS Classic series, AIP 1993. These books are from a bygone era but
so is tube building. The tubes on the market these days change randomly in
quality by the year since tube building is a very complex process few people
really understand or try to understand. One manufacturer may be good today and
become unreliable tomorrow. Regarding tube life, it so much depends how you
treat your tube regardless what other people may tell you. It is a dynamic
device and impatience is the killer. That means at least 30min warm-up time.
For daily use, running a tube at 40-50% of its max. power rating should result
in a tube life of at least 3 years on a decently made device. 

L. Keller
CAMET Research, Inc.

From: Armel Le Bail [lebail@aviion.univ-lemans.fr]
Newsgroups: sci.techniques.xtallography
Subject: Re: XRD Tubes
Date: Fri, 17 Sep 1999 10:04:25 +0200
Organization: Universite du Maine - France
Xref: daresbury sci.techniques.xtallography:5668


> We have horrible story here for the ceramic tubes mounted
> on the Bruker D8 Advance. The decay is 15-25% per month,
> observed by careful measuring on a good standard. This
> cannot be considered as normal, so that the manufacturer
> has already replaced graciously 2 tubes since December 1998...

Other hypothesis this morning : one tube could effectively
have problems. But the main problem could have been due
to the photomultiplier window. Displacing slightly the
photomultiplier position led to 400% increase (!!). The story
is still horrible but the fault is displaced.

Any experience/opinion about that ?

Armel Le Bail - Universite du Maine, Laboratoire des Fluorures,
CNRS ESA 6010, Av. O. Messiaen, 72085 Le Mans Cedex 9, France
http://www.cristal.org/

From: "Mark Bowden" [m.bowden@spamblock.irl.cri.nz]
Newsgroups: sci.techniques.xtallography
Subject: Re: XRD Tubes
Date: Tue, 21 Sep 1999 11:39:43 +1200
Xref: daresbury sci.techniques.xtallography:5679



Mike Meier wrote in message [37E0465F.67DB25A1@ucdavis.edu]...
>I would like to learn more about the modern x-ray tube, especially
>things which effect its useful lifetime
>(snip)
>I realize that two years of useful life is not bad, but maybe there is
>something I can do to keep it alive longer.

Mike,

A colleague of mine (Martin Ryan) looked at the life of conventional tubes a
while back, and presented the results at a conference (AXAA - Australasian
X-ray Analysts Assoc. '93 by recollection).  Our system utilises a
post-diffraction monochromator, so we aren't normally affected by tungsten
lines.

The outcome of Martin's work was that the intensity you get out of the tube
is propotional to:
1. (kV-kVc)^1.5  where kV is the tube voltage and kVc is the critical
voltage for excitation (approx. 8 for Cu tubes) and the exponent 1.5 is
approximate.
2. mA
On the other hand, tube life is proportional to kV * mA.

So, to get the the longest life for a given intensity, choose a high kV and
and low mA since you get more intensity from increasing kV and it has the
same effect on tube life.  Martin's guidelines were to go for about 75% of
the rated tube power, selecting a kV to give 3-5 times kVc.

Another factor to check is the filament current setting on the generator.
The normal kV and mA settings refer to the voltage and current between the
filament and target.  Generators frequently have a second control to adjust
the current which flows through the filament to heat it sufficiently so that
it emits electrons which are subsequently accelerated to the target.  The
tube manufacturers normally recommend a filament current in their product
literature.  Too high a filament temperature might result in the
unacceptable tungsten contamination you are observing.

Hope this is of some use,

Regards

Mark Bowden
Industrial Research
New Zealand                 m.bowden@irl.cri.nz

From: Frank May [Frank.L.May@umsl.edu]
Newsgroups: sci.techniques.xtallography
Subject: Re: XRD Tubes
Date: 21 Sep 1999 10:26:00 -0700
Organization: University of MO-St. Louis
Xref: daresbury sci.techniques.xtallography:5682


I presently acquire tubes from Seifert which last for at least 2-3 years without
objectionable W-L(alpha) lines.  The system is 1981-vintage Scintag which was
originally a PAD II.  When I turn the system on, I slowly ramp up the power and
when I turn it off, I ramp down the power and allow the tube to completely cool
before turning off the cooling water.

I suspect that having sufficient cooling water is a major contributor to long
tube life, because my practice is to leave the system on continuously and
running at 45 KV and 40 mA (1800 watts on a 2 KW rated tube).  I had one episode
where the tube life was shortened to about 1 year, but it was during the time
when there were a significantly number of power failures in the lab.  As a
consequence, the tube was turned off from full power and cooling water removed,
so the tube cooked itself to death.  Anyone else have similar experience?
-----------------

Frank May
Department of Chemistry
University of Missouri-St. Louis
St. Louis, Missouri  63121
frank.l.may@umsl.edu