The problem is also known as drift. Software for drift correction is on the market. For ferromagnetic samples (e.g. Fe-based alloys) a drastic reduction of sample size may be helpful. Bruker replaced the diodes and took measures to avoid electric loadings in our system. None of the mentioned trials solved the problem in our case. We are still looking for a solution as you are. Perhaps it is a system-immanent problem, indeed annoying.

The problem is also known as drift. Software for drift correction is on the market. For ferromagnetic samples (e.g. Fe-based alloys) a drastic reduction of sample size may be helpful. Bruker replaced the diodes and took measures to avoid electric loadings in our system. None of the mentioned trials solved the problem in our case. We are still looking for a solution as you are. Perhaps it is a system-immanent problem, indeed annoying.

If you are using C tape to fix your sample on the stub you will get drift in the initial stages of measurement. You could design a screw in sample holder which has limited drift 

Thanks for your suggestion,Dr. Bergner.As my specimen was ferritic stainless steel,reducing specimen size seems really helpful.Hopefully,we could both find the final solution

Thanks for your answer,Dr. Shuro.Actually ,we are using C tape.A screw would denfinitely be helpful.

Dear Chen,

As answered before, the drifting problem is a lot reduced if your sample is properly grounded to the stage. I work with tiny dogbone shaped specimens that I need to EBSD and later DIC. Therefore, I cannot screw them to the holder. To solve that, I use a carbon tape to fix it on the holder, and I use cooper tape in each end connecting the sample to the holder. I fold a little bit of the copper tape at each end to ensure that layers with no glue are touching the sample and the holder. It has solved the drifting problem for my specimens. Good luck.

You can also give a conduction pathway with a little bit of silver paint from your sample to the stub if you are using copper tape.  

If the drift direction reverses when you mag in/out...then it would be more of an issue with the type of sample as mentioned before...but if it doesn't change direction then the sample may be slowly coming off the mount because the copper tape isn't adhering enough.  Just something else to think about, and I have had issues with samples larger than TEM foils.

Good luck, Mike

Thanks a lot,Dr. mello.Up till now,ensuring sample properly grounded to stage would be the most fundamental solution.A screw can not apply on alll occasions.I will try the copper tape.Hopefully,it will work out.

Thanks Mike,you provided lots details,that is really helpful.I will keep an eye on adhering problem.My sample is much bigger than yours,so i guess ,it is more of sample type issue.Wish you to find the solution,good luck 

On large samples that do pull off the sample holder, I just glue the sample to the stub (with crystal bond or superglue) and then cover the glue with silver paint.  Either of those glues will dissolve in acetone.

Mike

As Frank Bergner said, sample size reduction is a good option. However, one should critically check the alignment and set-up in the microscope itself if that doesn't solve the problem. Often, problems causing drift are only obvious at long measuring time. Permanent, slight drift can be corrected with a drift correction (e.g. in the Bruker software). Our mayor problem are jumps, where the sample suddenly seems to displace so much that the drift correction is not able to bypass the difference. Has anybody seen something like that, too? 

Thanks Hilger.The drift does get worse while measuring for too long.Accroding to my experiences,the critical time would be half and an hour .There were no jump records in our system.It is a thermal filed emission SEM equipted with EBSD detector.Hopefully,you can find the solution.Good luck.  

Here is how we solved the drift problem for EBSD scans of heavy samples. Instead of using double-sided carbon tape, we use:

1. sample holders which have screws to fix specimens;

2. conductive glue to stick samples to normal holders.

Hope this will help.