 |
Buying a New Electron Microscope – A Users Guide | |
|
As
part of the consultancy side of Protrain's business we purchase scanning
and transmission electron microscopes for our clients.
This means we are buying instruments regularly and are therefore
probably one of the few “professional” electron microscope
purchasing organisations? In
our team we have staff who have worked for most of the electron
microscope manufacturers, as demonstrators and/or application
specialists. We have seen
the purchase game from both sides and of course try to use this to our
client’s advantage. Unlike
the average microscopist, who probably only ever buys two instruments in
their whole career, we have over the years formulated a purchase plan, a
purchasing criteria that we apply to each purchasing project.
In our position we are unable to buy from the nicest salesman, or
the normal "gut" feeling, we must be absolutely certain that
we are purchasing the best instrument for our client’s applications.
For us the best instrument may only be the one that produces the
best results across the client’s range of materials, this we can
measure, you cannot measure gut feelings! So
you are going to buy a new instrument, either the money is available or
you know it will be available on a set date, what should you do first?
Well, make contact with ALL the manufacturers, forget just
talking to your favorite salesman, forget missing out those of whom you
have had no knowledge, you must contact everyone.
Particularly in the SEM field there are some new names and there
are some interesting new products, missing out those instruments of
which you have not heard is bad practice. Whilst
you are waiting for the data you need to conduct a survey of all those
who will be involved with the new instrument.
Do not forget those who may be involved with the instrument in
the future provided the appropriate accessories are purchased.
If a number of departments are involved it is very important to
view the requirements on a level footing.
In our experience the people placed in charge of a purchase
always have a bias towards their own work and the resulting purchase may
often be too constraining for others.
This is an area where as consultants we have some of our greatest
problems. Everyone feels
that they should be buying the instrument, everyone feels that their
project is the most important, almost everyone becomes blinkered to
their own requirements. In
truth we have found that only a totally independent person will be able
to correctly access all the users applications and essential
requirements across the predicted application spectrum. It
is a good idea to develop a “desirability assessment”.
In the assessment each requirement or feature is listed and given
a mark out of 10, this will be used to determine suitable test specimens
and in the final analysis of each demonstration’s results.
For example, in a SEM high resolution at low kV may be very
important for all of the client’s applications; it therefore has an
assessment value of 10. Whilst
in the same purchase a totally eucentric stage, which is not required by
many of the clients, could have an assessment value of 3.
In the final analysis an instrument that had a brilliant
eucentric stage, 10 out of 10, would only total 30 points (assessment
value 3 multiplied by performance 10).
But the same instrument in an important area like low kV may not
perform well with a 20 points total (assessment value 10 multiplied by
performance 2). To some of
you this may seem over complicated but if you intend to purchase THE
best instrument for all concerned you need a calculation that sets
client requirement against actual performance. One
of the most difficult areas of a purchase is having enough knowledge
about the new instrument features to enable you to understand the
benefits that they may bring. One
client was looking at field emission SEM using sample preparation and
operating techniques that we thought were rather old fashioned even for
the Once
you have collected ALL the brochures and prices form a view of the
facilities that could be made available in your price range.
If you do not understand any features ASK the salesman.
Check your “desirability assessment” and update it if
required after discussion with the clients.
Your “desirability assessment” is now your purchase
specification. Look at the
price ranges offered and target the possible instruments and the
instruments that are just a little “too expensive”, that may be your
bonus in a bargaining situation. Do
not forget to take into account those accessories that will be essential
to fulfilling your “desirability assessment”, some may be pretty
expensive even packaged with a new instrument? Now
for another difficult area, preparing for the demonstration.
The selection of specimens should be made with great care and
based upon the “desirability assessment” using the higher pointed
features. One always must
have in mind that the new purchase when installed may not perform as
well as the demonstration instrument.
The results obtained at a demonstration must be absolutely
reliable, because they should be the most important factor when
determining which instrument is to be purchased.
You must have high quality data that if the new installation does
not perform as well as the demonstration instrument you are able to
display the differences; then the instrument must be fixed or replaced!
That is the stance that we take as consultants, however, we see
too many people pussyfoot around the problem and they end up living with
a duff instrument; to use a few English expressions.
The results that you obtain must truly relate one instrument to
another which means the specimens selected must of a type that will
allow multiple preparations with identical features.
It is incorrect to use the same specimen in each instrument, as
contamination will always have an affect, at best slightly softening an
image at worst totally obliterating it!
Do not produce too many specimens for each demonstration.
We have found that three carefully selected specimens are best
with a fourth, in our case a resolution test specimen, to use as a
tiebreaker. The fourth
specimen is used in the final analysis only if there is a tie.
Here we work on the principle that if all other things are equal
you are better off buying the microscope with the best resolution! It
is important to realise that the specimen chamber of a SEM, the
specimen-detector geometry, determines what we will see.
For this reason you will find that one instrument will carry out
a particular task better than any other in the price range; you need
test specimens that will sort this out for you.
If you have a multi instrument SEM laboratory we always feel that
it is a good idea to have instruments from different manufacturers
opening up the laboratory for optimising a wider range of applications.
There will be those that will argue that common manufacturers
mean common operating procedures and common service engineers resulting
in lower operating costs. Whilst
this is true and very much the direction that many teaching
establishments take, our experiences dictate that you aim at the
instrument that will out perform the others with your application.
It is our experience in industrial applications that the result
is what maters, not the cost of getting to the result!
Of course different chamber designs from the same manufacturer
could overcome this problem? Transmission electron microscopes all follow exactly the same imaging criteria so the only real differences model to model will relate to the contrast levels, operating procedures and the accessories available. In the TEM case the feel of the instrument, its driveability, will have a considerable influence upon the purchase Once
specimens have been selected that will test the instrument over the
range required for the applications envisaged by the client, a
demonstration procedure should be set down.
List each specimen to be used, the accelerating voltage or
voltages, the magnifications desired and any information that would be
useful for the demonstrator to have in order to get the best from your
specimen. Decide how you
intend to handle the demonstration before you set off.
You must make sure that each demonstration follows exactly the
same path or you could confuse the final results.
For example a TEM or SEM with an oil filled tank will take longer
to stabilise than one with a gas filled high voltage tank, therefore
high resolution tests made at the end of a demonstration should be
superior to those made at the beginning of a demonstration.
Write to each manufacturer and tell them the type of specimen
that you will be using and the demonstration procedures that you wish to
follow. As
a demonstrator there are several categories of customer and if you wish
to get the best from the demonstrator and therefore the best from your
demonstration you should not fall into one of the “bad customer”
categories. Just what can
happen? Well the client
arrives with dozens of specimens and expects you to complete their
current project for them. Or
the client gives you the specimen that they clearly know nothing about;
it becomes quite plain that this is the first time he/she has ever
looked at it! Then there is
the clever client who tells you nothing about the specimen expecting the
demonstrator to learn in one day what the client has taken 20 years to
discover, there is a battle to determine if the demonstrator is as
clever as the client; crazy? On
the other hand there is no point the demonstrator taking up a hunk of
the demonstration showing us their favorite specimen.
Or for the demonstrator to go through the complete TEM alignment
procedure because they think it is really exciting; does anyone buy a
particular TEM because it is easy to align, what about its performance
on a specimen? Be very wary
if, when presenting the demonstrator with a specimen, you are suddenly
whisked off to have a cup of coffee; stay with the demonstration ALL the
time. Do not leave specimens
to be photographed whilst you are away at lunch and do not leave
specimens behind to be photographed later.
You must be absolutely certain the results you obtain are from
the instrument that you intend to purchase.
Results from another model, because the desired accessory was on
the other model, are not good enough to make the correct purchasing
decision. On
demo day you must be strong because YOU are the important person and it
is in YOUR interest that the demo follows YOUR schedule.
Be prepared to give the demonstrator about 90 minutes to show you
over the instrument. In that
period you will become more familiar with each others approach and
develop an understanding in readiness for the really important area of
the demonstration; the microscopy! Make
it clear when you hand over a test specimen to the demonstrator that you
wish them to follow the operating procedures that YOU have set down.
Time each operating sequence to its conclusion but only when you
have explained exactly what you want from the test specimen, so that the
demonstrator is in no doubt in relation to the task or tasks that they
must perform. We need to
know how long someone experienced in the operation of the instrument
will take to obtain a good result under particular conditions.
It is no good buying the best instrument in the world, where its
procedure is very slow, if the results could be obtained much more
quickly on another instrument with little loss in performance.
Once your laid down procedure is complete, work with the
demonstrator to see if they have alternative procedures that may bring
even more from your test specimens; give them a chance to use facilities
on the instrument that your test may not have considered.
At all times during the demonstration give the demonstrator the
benefit of your knowledge of the test specimen.
They will be very suspicious of you at first because most clients
do not help at all! Gradually
as you gain their confidence everyone will relax and you will have an
even better demo. Throughout
the demo it is most important that your standard tests are complete,
never stray from your pre determined path for too long, you need that
standard data! This
type of demonstration procedure being so strict in itself will tell you
about an instrument even before you compare results.
We have an example where one client had demonstrations of the
same procedures at three instrument manufacturers.
One demonstration was only completed after seven hours on site.
The second demonstration was complete in six hours whilst the
third we had all our results just after lunch.
Which one would you purchase? The
final steps in the purchase procedure are no easier than the others.
Lay out all the photographs from each manufacturer in lines that
relate to the specific procedures used.
Find the best micrograph in each section and give it 10 points,
then relate the other micrographs of the same subject under the same
conditions with points appropriate to their quality.
Then look at the time taken for each result and give the fastest
time 10 points and the slower times points related to their duration (10
X fastest time divided by slower time).
Add each manufacturers picture points together and multiply by
the points given for time taken, thus having a point score for each
group of micrographs. Take
the scores and divide them by the maximum points for that test,
therefore converting the points into a percentage value.
This action brings each test to the same level not matter how
many micrographs made up the test. Once
this figure has been attained for each test multiply them by your
“desirability assessment”. The
final action is to set out the points for each manufacturer.
Look back at the results again to see if you have made a glaring
mistake and if not the top points scorer should be your selection.
Use the tie break specimen if you have to! Now
there will be areas important to the selection of an instrument that do
not relate to the demonstration. How
close is the nearest service engineer, what is the company reputation in
your area, it may be a question of will your favorite engineer be able
to service this equipment? Set
down how important this is and assign points in a similar fashion to the
“desirability assessment”. As
consultants we are forced to go for the instrument that performs the
best, not the one with the best service engineer.
Contrary to popular myth microscopes are microscopes and if man
built them man may service them; ask the non-manufacturer service
engineers how they survive? Sure
there are the odd complications but if you want a high performing
instrument you want the best, there may have to be some other
sacrifices. If you are just
interested in using a SEM as a super light microscope you may just as
well go for the one that provided you with the best lunch.
If you wish to sort out the true best performer we see our
“desirability assessment” as the only way Finally
you need to make the purchase, haggle, then haggle then haggle.
The manufacturer wants your order, you are in control, go for the
very best deal you can get! Be
tough and you will be pleasantly surprised at just what you can squeeze
out of the manufacturer. However
if you bleed them dry do not be upset if their warranty service is not
up to the standards you may expect, there is a balance!
Happy hunting. ExampleAn
actual middle range SEM purchase Part 1 – the desirability assessment with areas chosen for evaluation with a test specimenLow
kV performance
rating 10
Used for Specimen 1 A
variable bias control
rating 8
Tested by Specimen 1 & 2 40kV
for backscatter
rating 3
Tested by Specimen 2 An
externally variable final aperture
rating 3
Tested by Specimen 3 Backscatter
facilities
rating 10
Used for Specimen 2 Digital
& Photographic Imaging
rating 8
All A
simple operating procedure
rating 5
All – (take the times) Stereo
Imaging eucentric stage
rating 5
Tested by Specimen 3 Large
specimen capability
rating 7
Used for Specimen 3
Part 2 – Test Specimens and Part 3 Results
Instrument A
Instrument B
Instrument C Test
specimen 1(Possible Points 30 x time factor = 300) 5kV
performance 5,000X
7
10
5
15,000X
8
10
7
50,000X
10
8
7
Time
35 min
29 min
32 min Time
Factor 29/? X 10
8.3
10
9 Test
Total Including time factor
208pts
280pts
171pts Percentage
Value
69
93
57 Test
specimen 2 Backscatter (Possible Points 120 x time factor = 1200) At
40, 30, 10, 5kV
500X
0 10 10 8
0 6 8 10
10 8 6 4 1,500X
0 10 10 8
0 8 8 10
10 8 6 4 5,000X
0 7 7 7
0 5 10 10
10 10 6 3 Time
65 min
70 min
80 min Time
Factor 65/? X 10
10
9.2
8.1 Test
Total Including time factor
770pts
690pts
689pts Percentage
Value
64
58
57 Desirability
Areas
Fixed bias
Variable
Fixed Bias
SS BSE
Scintillator
SS BSE Test
specimen 3 Large
Specimen and Ease of Stereo 10kV(Possible Points 30 x time factor = 300) 500X
15mm
8
7
10
500X
20mm
8
8
10
500X
30mm
7
7
10
Time
39 min
50 min
41 min Time
Factor 39/? X 10
10
7.8
9.5 Test
Total Including time factor
230pts
172pts
285pts Percentage
Value
77
57
95 Desirability
Areas
eucentric
not eucentric
eucentric Test
specimen 4 Best
resolution SE (Possible Points 10 x time factor = 100) 100,000X
15kV 5mm WD
8
10
7
Time
21 min
18 min
19 min Time
Factor 18/? X 10
8.6
10
9.5 Test
Total Including time factor and
Percentage Value
69pts
100pts
67pts The CalculationsTake
the total points earned by each instrument with each test specimen
multiply this by the appropriate time factor.
Take the absolute maximum points total for each test specimen and
divide this into the points gained X 100 to obtain a % points figure;
this balances the weight of each test specimen, making each test equal
in importance. Multiply each
of these figures by the appropriate “desirability assessment”
figure. Add together all
these totals plus the points awarded for feature not used within the
four test specimens. The
results for our three instrument example- Instrument
A
Instrument B
Instrument C Test
specimen 1 – testing low kV performance and ease of operation Percentage
Value
69
93
57 Desirability
Rating 10
690
930
570 Test
specimen 2 – testing backscatter performance and ease of kV change Percentage
Value
64
58
57 Desirability
Rating 10
640
580
570 Test
specimen 3 – testing ease of stage movement with a large specimen Percentage
Value
77
57
95 Desirability
Rating 7
539
399
665 Test
specimen 4 – tie break specimen – not required Totals 1869
1909
1850 We purchased instrument B for the client.
|
