STM control: purely by labview logic and DAQ device?

[swngiam]

Hi all.

I'm  involved in building a rudimentary scanning tunneling microscope and subsequently be expected to apply the rubricks to further optimise a commercial equipment.

As my background is primarily physics (and may I add with very very limited electronic knowledge) I am tempted to employ labview heavily in the control of the (basic/rudimentary) equipment. (eg to control voltage i would simply attached an analague voltage output channel to the piezo haha).

In principle, an STM works by having an X-Y raster scan. A z-value is obtained in terms of voltage output of a piezo being made to follow the curvature of the atomic surface via a feedback loop to keep the "tunneling current" constant, a n-Ampere which is affected by tip-surface distance. (all movement done via piezo crystals which exhibition nm-scale movement upon application of voltage)

Here are my questions (please bear with me)

1. the design i referred to employs much analogue circuitry for the feedback-z function. The logic is if current drops, increase z-distance (tip closer to surface) and vice versa. can this be handled by labview near real time? On the analog circuit it would read as "The difference between a set-point and the real tunneling current is integrated with an integral controller and the integrated signal is put symmeterically to the z-piezo. The velocit, with with the controller reacts to the current change, can be adjusted using integral-gain"
   
   
2. still on the tunnelling current, the circuit employ ICs and/or op-amps to filter out higher frequency noises. Would i need analog circuit to achieve this or again, can it be done in software?
   
   
3. many of the past posts on STM, thread posters mentioned pixel size for the scan outputs. May I know if anyone knows how did they define the distance between each raster?
   
   
4. on the voltage supply for the XY rasters, words such as "decoupling" (the computer signals) and piezoelements having capacitve properties resulted in 2 op-amps before the the voltage reaches the piezos. what problems might i face if i just hook the piezos up to a analog voltage output?
   
   
5. i have requestion for the quotation of an NI USB 6229 for the above task. would it prove sufficient as a DAQ tool?
   

reference electronics from http://sxm4.uni-muenster.de/stm-en/STMElectronics.html

Many thanks in advance!!

[Brian F.]

Hello swngiam. 

Thank you for posting to the NI discussion forums. 

Although I have not worked directly with STM, I think I can provide some valuable insight into your application.  First, I would recommend looking into the LabVIEW PID Control Toolkit for Windows.  This will allow you to use LabVIEW custom VI's to use feedback in control loops fairly easily.  If you combine the PID.vi with the DAQmx read and DAQmx write VI's, you would be well on your way to achieving your end goal. 

To answer your real time question, LabVIEW's execution is non-deterministic.  Thus, if there are a lot of processes running, there is no guarantee of when or how long it will take to complete a specific block of code.  To see more deterministic performance, you can use LabVIEW Real Time.  There is more information about this product here: 

LabVIEW Real-Time Module

https://www.ni.com/en/shop/data-acquisition-and-control/add-ons-for-data-acquisition-and-control/wha...

In terms of filtering out high frequency noise, you can use one of LabVIEW's filtering VI's.  I would recommend searching for filter on the functions palette and the results will depend on what LabVIEW package you have.  However, when I search, I find a Filter Express VI that is very good as well as many other VI's that focus on specific types of filters such as Bessel and Butterworth just to name a couple. 

The USB-6229 device is a very good M-series device.  In this case, we need to ask ourselves if 16 bits of resolution and 250 kS/s are good enough specifications for what we want to do. 

Let us know if this helps in your application or if you have any other questions.

Have a great day!

Brian F

Applications Engineer

National Instruments