7/30/96


Hi All.

I have been having a bit of a problem with some stray EM (?) fields

on our SEM. The problem has been traced to the power feeding the wall

outlets. Anything plugged into these outlets and the microscope will affect

the images produced. Specifically the backscatter detector and the PC we use

to capture images digitally. The results are the same for images captured

photographically if either the backscatter or PC is plugged into the wall.

They are not the cause of the problem but simply feed the noise to the

microscope via any cables connected to it. If I unplug all from the outlets

or unplug cables leading to the microscope the problem is corrected. It

doesn't matter if they are on or not, it still feeds the noise.

We have so far simply run extension cords to other outlets in other

rooms and on other circuits in order to solve this problem but it has

worsened and I am running out of different outlets. I recently purchased an

APC line conditioner which is supposed to continuously condition the power

and give out a fresh signal but needless to say it hasn't worked or I

wouldn't be asking for your help.

If you would like to see an image with the problem go to the web

address at the bottom of this message and look in "What's New". You will

notice the vertical banding pattern produced.

Your help and expertise will be greatly appreciated. Vendors please

jump in with any advise you may have.



Sincerely







Scott D. Whittaker

Research Assistant

University Of Florida

ICBR EM Core Lab

218 Carr Hall

Gainesville, FL 32610

sdw@biotech.ufl.edu

ph 904-392-1295

fax 904-846-0251 http://www.biotech.ufl.edu/~emcl/

The home of " Tips & Tricks "


Your problem is a ground loop. I have seen this many times (I am a service

engineer for a SEM co.). The grounds for these outlets are not a common

point, until they get back to the power panel.

There are three ways to fix it. 1) Use the extension cords like you have

been 2) Get an electrician in and tell him or her that you need a dedicated

line with multiple outlets to feed all the equipment that will be hooked up

to the scope. 3) You could use a "cheater plug" for anything you plug into

the bad outlet . This is a three prong to a two prong adapter. This will

keep the ground currents from looping.

Good Luck

Rich2442@aol.com


Scott - sounds like the wires are acting as transmiters. Try making up the

power cord from shileded wire or twisted pairs and see if that helps. Also.

you may well find that there is a voltage difference between the

ground/neutral of the wall outlets and the microscope which can cause such

problems. Running everything toa common GOOD ground can often help. We had a

dedicated ground for the EM's a FSU tha twe tied everything into.

Bill Miller

Microbill@aol.com


Scott,

We had a similar problem during installation of our SEM a few years ago. It seems the power feeding the 110V outlets in the room adjacent to our SEM went under the floor beneath the SEM in *plastic* conduit, rather than metal, and thus there were radial AC magnetic fields generated anytime something was plugged into these outlets, and current was flowing.<p.

Try to trace the wiring from the outlets back to the power distribution box (circuit breakers), and make sure it is shielded.

F. Scott Miller

Electron Microscope Lab

University of Missouri-Rolla

223 McNutt Hall

Rolla, MO 65401 USA

voice: 573 341 4727

fax: 573 341 6934

smiller@umr.edu


Scott,

Oxford Instruments make an EM field cancelling system for just this type of

problem. Have you investigated this yet?? If you need I can put you in touch

with your nearest rep and can fax you a copy of the brochure.

Best regards, Les Brownlow

oisydney@ozemail.com.au


Scott,

You might want to start by measuring with a voltmeter to see

if there is any difference in the ground potential between the troublesome

outlet and the microscope ground. Generally speaking all microscope

accessories should always run off he same oulet (or at least "circuit") as

the microscope. Otherwise ground loops can result. Sometimes this requires

getting the electricians to install a new outlet on the same circuit as the

microscope to effectively cure the problem. You might also want to elaboate

on when the problem first occurred. eg: Installation of a new accessory?

Paul Thomson

Technical Director

Thomson Scientific Instruments

Australia

Web: http://werple.net.au/~tsi/

Email: tsi@werple.mira.net.au


Hi,

Your problem sounds like a ground fault. If the microscope and the

accessories (wall outlets) aren't referenced to the same ground, your will

get line noise. You should try to find a power source from the sem for

those accessories, or use an isolation xfmr for them and reference their

chassis ground to the sem. Any ground path with a potential difference

will cause current to flow through the grounding circuit and cause ripple.

Your service rep should be able to handle it, or just do it yourself.

Mike Warfield

mdwarfield@vcnet.com


Scott,

it sounds like you have an earth loop. I have run into this problem before.

The simplest solution is to drive a copper spike into the ground outside

your lab and then make this the new earth for your microscope. Now change

earthing for both the wall sockets and the power supply for the microscope

to reference from this point. This should solve your problem.

If it does not then you may be looking at having phases slightly our of

sink. Check that the microscope power supply and the wall sockets are both

getting the same phases. In my case I have had all power taken from one

three phase supply line. The electrican has tried to balance the current

drawn on each phase to be equal. This minimises the production of stray

magnetic fields.

I would also suggest the book "Design of the Electron Microscope Laboratory

by Ronald H. Alderson, Practical Methods in Electron Microscopy"

Hope this is of some help.

Colin MacRae

C.Macrae@minerals.csiro.au


Hi

I suggest that you try to use a adaptor plug on the three prong plug for you

devices to remove the ground from the system. Make sure it puts three

prongs to two, a ground loop can be causing this problem that you describe.

Then you can connect the ground pin of the adaptor to the microscope on the

column or ground bar. I have seen this type of phenonmena before, this

usually fixes the "field" problem that you see.

Another way is to ask the electrician to ground the outlets to the

microscope for its accessories. This will again eliminate the ground loop

that may be causing your problem

Ray Spengler

ray@raven.cybercomm.net


Hi Scott:

Just to be on the safe side I send again my message directly to you. Please

keep me informed if any progress was made!

Here you are:

I am almost absolutely certain that your problem is associated with an

earthing (or grounding) loop. It is very important to have only one single

earth path. If you have separate earth path as supposedly exists with your

additional backscatter detector and/or PC (the microscope is connected to

the earth terminal, the PC is connected too, and both instruments are

connected to each other through their grounds), there are alternative,

physically separated earth paths, which will result in noise currents, and

almost certainly impair the performance of any electron microscope. Some

years ago we had similar problems with our computer and EDS system, but

there is a remedy. It would be too long and sketches are also needed to

explain. The best I can propose is to check a good textbook such as "Design

of the Electron Microscope Laboratory" by Ronald H. Anderson (North Holland,

1985, p61-66) or consult an expert electrician as a final solution. In case

you cannot obtain quickly the book I can fax you these pages.

Kristof KOVACS

Associate Professor

University of Veszprem, Central Laboratory

P.O.Box 158, Veszprem, HUNGARY

H-8201

Phone: +36-(88)-421-684

kris@almos.vein.hu


Hello Scott, You may be suffering from the dreaded "ground fault" difficulty.

How is it you seem so convinced the problem is magnet? Try this: Get some

of the line cord adapters which eliminate the 3rd wire (the prong) and plug

them on line cords of all your extra instrumentation. If there is a ground

fault or eddie current problem this may eliminate it. Ideally, all your

instrumentation would all be grounded at one point. Also, there can be

problems with stray voltages on both the neutral and ground connections of

wall plugs.

Alex Greene

Scientific Instrumentation Services, Inc.

Austin, Texas.

ablue@mail.io.com


Scott,

You are almost certainly setting up ground loops. ALL your equipment must

go back to a common, local ground. You need to go back to the mains power

source for your SEM and run extensions to all auxillary equipment from this

point. Be careful not to make these extensions too long, or you may suffer

pick-up from other sources. There should be some auxillary power outputs in

the back of your SEM and these are the prefered power source. However, if

you have a lot of auxillary equipment, you may risk overloading these.

Ideally, you should have a mains matching trasnformer between your SEM and

your local power supply - this helps to provide protection agianst power

surges, and some isolation from interference. You should tap in at this

point to provide additional power sources.

Larry Stoter

LPS@teknesis.demon.co.uk


Scott,

If you are using a line conditioner and still have things plug into

the wall outlets, you may be still creating a ground loop between the

scope and the outlets giving you this problem.

You can isolate this by buying the 3-prong to 2-prong power cord

connector. This lifts the ground from the wall outlets. If this

still doesn't fix the problem then you may have to go to a separate

ground for the scope. i.e. dedicated to the scope only and not the

building.

Give this a try and if you need any more assistance please feel free

to contact me directly.

Larry Cessna

Hitachi

Cessna@hii.hitachi.com


I agree with Kris Kovacs: you probably have ground loops in your system.

One easy way to correct this is to modify a power strip.

If you have a strip with a metal shell, make sure that the outlets in the

strip are not electrically connected to the shell (use plastic screws if

necessary). Lift the ground wire from the power cable inside the strip and

connect it to the strip's metal shell for safety. Connect the outlets'

grounds together and attach a 14-guage insulated wire (preferably stranded

wire for flexibility) to that ground. You'll have to drill a hole in the

shell, of course, so this ground wire can fit through it. Connect the

14-guage ground wire to the ground connector on the SEM (or to the metal

body of the SEM if you can't find the connector).

I recently did this to four of my TEM's. Our problems were more subtle

than what you described.

Russell E. Cook

Electron Microscopy Center for Materials Research

Argonne National Laboratory

9700 South Cass Avenue

MSD-212

Argonne, IL 60439

(708)252-7194

FAX: (708)252-4798


Scott, you are probably experiencing a ground loop problem, one way to check

this is to get a adaptor plug with no ground and plug your sem accessories, i.e.

bsd, printer into that.

Regards

Mike Webber

102413.2660@CompuServe.COM


The guy I had contact with for the field compensation system at

Linear Reasearch Associates (it's a small company) is Curt Dunnam

(crd4@cornell.edu). He's the chief engineer, and will likley be happy to

talk.

Ben Simkin (simkin@egr.msu.edu)

simkin@egr.msu.edu


Robert and all,

In my experience, the "high mag jaggies" are most often caused by a

mechanical vibration. Often the roughing pump is the problem. Sometimes

a couple of tennis balls under the roughing pump will decrease the effect

substantially. It's not always this simple though. Also check for cables

touching the vacuum line from the RP. They can transmit these vibes.

John Best

jbest@vicon.net


We don't have Scott's problem with our JEOL 5400 SEM, but we do have this

jagginess problem that John Best deicribes. JEOL has been here and we have

looked for electrical fields. Not a problem. A shield was installed in the

consol to eliminate fields emmanating from the CRT's. No effect The company

service people are stumped, and we are left unable to use the SEM a high mags.

This outlet problem intrigues me, however. Please keep this thread going for a

while.

Bob Schmitz

rschmitz@uwspmail.uwsp.edu

or

rschmitz@macsrv1.uwsp.edu

(note its macsrv"one" not "el")

Robert (Bob) J. Schmitz

Department of Biology,

University of Wisc. Stevens Point.

Stevens Point, Wisconsin 54481

ph 715-346-2420


High Frequency Earth Loops.<>

We also had an interesting stray fields problem which showed up in high

noise counts in X-ray spectra - the amount depending on the scan rate in

the SEM! This was traced to a high frequency earth loop (no DC connection)

with capacitive coupling associated with the thin plastic shin isolating

the X-ray detector from the column. This was solved by passing a few loops

of the whole cable going to the X-ray detector through a large toroid to

increase the high frequency impedance of this part of the loop.

Alan Wilson alan.wilson@dsto.defence.gov.au

Senior Research Scientist

Ship Structures and Materials Division

Aeronautical and Maritime Research Laboratory

Defence Science and Technology Organization

506 Lorimer St

Fishermens Bend 3207

Victoria Australia

ph 61 3 9626 7508, fax 61 3 9626 7816 or 61 3 9626 7087


Hi Bob. I will summerize all of the replies in a bit. Meanwhile,

there was another thread a while back posted to the server which I have

archived. Go to the web page listed at the end of this message. Click on the

"Tips & Tricks" button. You will find a link for "SEM Techniques &

Instrumentation" which wil point to another link called "Dealing with Drift,

SEM". It may prove useful. If you do not have web access, let me know and I

will be happy to get the info to you some other way.

Scott D. Whittaker

Research Assistant

University Of Florida

ICBR EM Core Lab

218 Carr Hall

Gainesville, FL 32610

ph 904-392-1295

fax 904-846-0251

sdw@biotech.ufl.edu http://www.biotech.ufl.edu/~emcl/

The home of " Tips & Tricks "


About a year ago we had an intermittent "high mag jaggies" problem on our

6300. Viewed at TV rate at >50,000x the interference seemed to have a

fairly high frequency and underwent rapid variations in amplitude. Some

days it was there and other days it wasn't. It was indeed caused by a

mechanical vibration -- but in our case it was coming from the stage

motors, in particular the Z-axis motor. The effects of this vibration only

started to become visible above about 15-20,000x, and it could only just be

felt on some days by very lightly touching the motor once you already "knew

it had a vibration". Our Jeol engineer was the one who finally figured this

out and then he measured a ~2 volt sawtooth ripple on top of the DC holding

current to the stage motors. The problem was eliminated by changing some

resistors in the (non-Jeol) motor driver hardware, on the recommendation of

its vendors, to reduce the DC holding current and hence presumably the

absolute magnitude of the ripple that came with it. We have not had any

trouble since.

Arthur Day, Electron Microscope Unit

Ansto Materials Division

PMB 1, Menai (Sydney), NSW, 2234

Australia

http: //www.ansto.gov.au/

Phone: 61-2-717-3457

Fax: 61-2-543-7179

Email: ard@atom.ansto.gov.au


Greetings,

We are in the process of tracking down similar problems. One avenue that we

are examining is mechanic vibration in the column due to the location of the

EDS dewar. The dewar is essentially a large undamped liquid mass hung out

to the side of the object that must remain vibration-free. Well the

location of the dewar is approximately equidistant to the two adjoining

sides of the room. By placing a small mirror over the top of the dewar, we

can watch the LN2 shimmer. By moving sound-absorbing foam sheets around the

area between the dewar and the walls, we can see a difference in the

vibration. We have more work to do on this, but it seems like the dewar is

in the "sweet spot" in the room if you were going to set up a stereo system.

We are also going to move the 3-phase transformer that sits on the other

side of the wall (less than 3 feet from column and console) in the surface

analysis lab to get rid of the known electrical vibration.

Harold J. Crossman

OSRAM SYLVANIA INC.

Lighting Research Center

71 Cherry Hill Dr.

Beverly, MA 01915

Phone: (508) 750-1717

E-mail: crossman@rd.sylvania.com

Our web sites: www.sylvania.com

www.osram.de

www.siemens.com


As the operator of a well used Etec, I have encountered worn and

loose gears/bearings in the stage. Enviro vibes, which normally

did not create a problem, caused "jaggies" as low as 5000x.

Etec FYI: My system has been modified by removing the diff pump and

installing a turbo. Beware of ball bearing units. The Etec is far

too susceptible to their vibrations. I have been using a Leybold

maglev pump since they became available and can see no pump induced

vibes below about 30Kx.

Woody White

Babcock & Wilcox Research Ctr.

woody.n.white@mcdermott.com

woody.white@worldnet.att.net


Vibration can be a nasty problem. You can't assume they are coming from

your area. When I was at George Washington Univ. Med School my lab was

on the 5th. floor. Never had problems with vibration until one day I was

looking at cilia at 200,000 times and saw it. Zeiss came in to look at

the scope and the surrounding area and couldn't find what was causing the

problem. They saw vibration only when they put the meter on the desk of

the scope. Put it on the floor, no vibration detectable. We looked at

another TEM on the floor below us and they had no problem even at 300,000x.

After much hunting we found a air handling unit 2 floors below us and

about 100ft. away from the scope with a bad rubber foot pad. We had the

physical plant dept. replace it and our vibration problem disappeared.

Apparently this was causing low level vibration that my scope was picking

up and the other scope (floor below) wasn't. Vibration, especially low

level can be anywhere, construction several blocks away, bad feet on a

centrifuge, new equipment, just about anything. So if you're having a

problem with vibration check everything you can, don't rule anything out

unless you've checked it out, this includes anything touching your

mechanical pumps such as cables, hoses or even boxes you might have set on

the backing hoses.

rutledge phil

prutle1@gl.umbc.edu


I've had similar problem with our microscope, and if it's truly fields,

I'd recommend just living with not using the outlet. AC fields can be activly

supressed if they are from an external source (we have purchased a system

from Linear Research Associates; they run ads in Microscopy Today (I don't

have the address with me right now)), but this sounds much more like a ground

loop, and the solutions to that (so I've been told) include either sinking

your own dedicated common ground (anywhere from easy to nearly impossible), or

disconnecting the ground connection between your asscessories and your SEM,

and running a "floating ground". I've discussed this as one solution with

our SEM service technician, and he says it sometimes works (assuming the

noise it adds to your instrument signal from the differance in "ground"

potentials is acceptably low).

Ben Simkin (simkin@egr.msu.edu)

Dept. Mat. Sci. and Mech.

Michigan State University


[Return to Tips & Tricks Menu]