Near field scanning

From MesoscopicWiki

Jump to: navigation, search

Contents

SCAN QUEUE (leave at top of page)

If you are running a scan, or plan to, please make note of it here in advance to minimize our interference.

(04.30.2008) Scott does 2nd PIC scan with powered circuit?

Done

Some related papers

Image:pdf.gif Download Near-field THz microscopy


Multiple Scattering Statistics

(04.28.2008)

Rayleigh vs. Rician Scattering.



Sweep versus Point by Point scan for resolution

(04.28.2008) I did these scans on Friday, (04.25.2008). The setup was a single horizontal scan line across the copper tape wedge Scott made. The point of the triange wedge was pointed downward and the scan went from the straight edge to the angled edge in direction.

The 'sweep' means I set the motion controller to move at a constant velocity of 0.01 mm/sec across the wedge while the MVNA recorded a 1000 point sweep over the 10mm range. The math works out for this. The 'pt 2 pt' means I used our labview software and recorded 1000 steps at 0.01mm between 0mm and 10mm. The following picture is the results.

New horn setup pic.

Going from teflon to the tape edge (for the straight edge) took about 60 steps, or .6mm. Going off the angled edge back to teflon took ~80 steps, or .8mm. I need to go measure the aperture hole side in the aluminum holders and put those numbers here. At 260GHz, our wavelength is 1.15mm. So our resolution of 0.6mm-0.8mm is definitely sub-wavelength.


New setup JPGs

(04.24.2008)

New horn setup pic.


Optical pulse transmission setup.


Latest probe CAD view

(04.22.2008)

Solidworks assembly of parts for sandstone scan, annotated.




Microcontroller Circuit Board

(05.02.2008) Higher resolution scans of the circuit board. The image is of the bottom corner of the integrated microcontroller. When the power was 'on', the microcontroller program converted a voltage to a digital number with the ADC peripheral and then displayed that number on 8 LEDs. The wire directly to the bottom and the side of the scan are the wires that lead to the LEDs. Resolution is .05mm. The resolution increase did not seem to do much.

Circuit OFF- Reflected dB, 260GHz.
Cifcuit ON- Reflected dB, 260GHz.


(04.29.2008) Here is the first scan of the microcontroller from the demonstration board that I had. The small box is the actual chip while the larger box (that is cut off at the bottom) is the enclosure (plastic housing). You can see the wire connectors from the chip to the outside. Also the solder contact points (at the top of the picture) are not really visible; I believe that this is because they are not flat and would not reflect anything too well directly back.

Reflected Phase, 260GHz.



CSM Blocks, 260 GHz

(04.26.2008)

Picture

Reflected Amplitude, 260GHz.




Dime resolution versus offset, 260GHz

(04.20.2008)

0.2mm offset

Reflected Amplitude, 260GHz.
Reflected Phase, 260GHz.


Transmitted Amplitude, 260GHz.



0.6mm offset

Reflected Amplitude, 260GHz.
Reflected Phase, 260GHz.


Transmitted Amplitude, 260GHz.
Transmitted Phase, 260GHz.



1.0mm offset

Reflected Amplitude, 260GHz.
Reflected Phase, 260GHz.


Transmitted Amplitude, 260GHz.



2.0mm offset

Reflected Amplitude, 260GHz.
Reflected Phase, 260GHz.


Transmitted Amplitude, 260GHz.



U.S. Twenty OVI scan NO slides

(4.16.2008)

Colorbar labels are cut off, do not use yet.

Transmitted Amplitude, 260GHz.
Transmitted Phase, 260GHz.


Reflected Amplitude, 260GHz.



U.S. Twenty OVI scan between slides

(4.14.2008)

Axes are crunched, look bad I know... working on it (bz)

Colorbar labels are cut off, do not use yet.

Transmitted Amplitude, 260GHz.
Reflected Amplitude, 260GHz.


Transmitted Phase, 260GHz.
Reflected Phase, 260GHz.


Brass Cylinder without CSM engraving, 260GHz

(04.15.2008)

Reflected Amplitude, 260GHz.
Reflected Phase, 260GHz.


Reflected Amplitude, 260GHz, 3D.
Reflected Phase, 260GHz, 3D.



Brass Cylinder with CSM engraving, 260GHz

(04.15.2008)

Reflected Amplitude, 260GHz.
Reflected Phase, 260GHz.


Reflected Amplitude, 260GHz.
Reflected Phase, 260GHz.



Brass Cylinder with CSM engraving, 170GHz

(04.15.2008)

Reflected Amplitude, 170GHz.
Reflected Phase, 170GHz.


Reflected Amplitude, 170GHz, 3D.
Reflected Phase, 170GHz, 3D.



Dime hidden in sandstone

(04.11.2008)

Solidworks assembly of parts for sandstone scan, annotated.




(04.01.2008)

Here are the transmitted amplitude plots of with and without the dime.

Transmitted amplitude with dime.
Transmitted amplitude without dime.



Difference of both data sets

Transmitted amplitude difference, normalized, linear scale.
Transmitted amplitude difference, normalized, linear scale, color.



weak diffraction/standing wave test

Here is the amplitude scan of a glass slide cover. These are about 190 microns thick. I scanned one corner, right off the edge in both directions. Shown here is the field just up to the edge of the slide. You can see the degenerate standing wave (the slide cover is square).

Standing waves from a microscope slide cover, transmitted amplitude.


Link to the Slidecover_TdB.eps file.

Sandstone thin section raw data


This is a small chunk of a scan of one of the thin sections. It is 114 by 132 pixels. I scanned at 2400 dpi, which corresponds to 95 pixels per mm. Looking at this, I would estimate the typical grain size at being around 20-30 pixels. This translates into 210-320 microns.


Image:114x132section.jpg

Proposed aluminum holders for aperture size dependence scans

This is what I am proposing for the scans involving dependence on aperture diameter. With 3 small hex head screws, it should be easy to pop off one aperture top and screw on another. These shouldn't take me more than a day to make, and the nice part is we (read 'I') can make a suite of aperture sizes to be swapped out at our choosing.

Pros:

  • Easy to change face plate, no removing of aluminum housing from the mounts
  • High accuracy in aperture diameter and position.
  • Consistency in measurements regarding probe position in relation to aperture.

Cons:

  • I have to make them.
  • I have to make them.
Top view
multiple tops, easy to change with 3 small screws


Dime inside Berea Sandstone

(03.31.08) High Resolution scan through 4.06mm mm of Sandstone

Scan parameters:

  • Teflon probes, small aperture.
  • Scan area (X,Z) = (27,30) mm.
  • Scan resolution is 0.1 mm in X-Z.
  • Sandstone parameters:
    • (actual pictures forthcoming)
    • Sandstone thickness, uncut plate: 4.06 +/- 0.05 mm.
    • Sandstone thickness, cut plate: 4.06 +/- 0.05 mm.
    • cutout thickness: 2.60 +/- 0.05 mm thick
    • diameter of cutout: 18.0 mm
  • Dime head/front facing gun.
  • Link to the DimeInSandstone_hires.txt Raw Data.
dime in sandstone - reflected amplitude, gray colormap
dime in sandstone - reflected phase, gray colormap
dime in sandstone - transmitted amplitude, gray colormap
dime in sandstone - transmitted phase, gray colormap



Blaster Card Scan

(03.31.08) Scan of microchip (see overlay of Blaster Card and data below)

Scan parameters:

microchip - reflected amplitude, gray colormap
microchip - reflected phase, gray colormap
microchip - transmitted amplitude, gray colormap
microchip - transmitted phase, gray colormap



(03.31.08) Overlay of Blaster Card with scan data

CSM - transmitted amplitude, overlay with card

NOTE: fixed overlay of card with data, direction was previously flipped in X. It is correct now.

(03.25.08) Microchip and antenna

Scan parameters:

  • Teflon probes, small aperture.
  • Scan area (X,Z) = (97,57) mm.
  • Scan resolution is 1 mm in X-Z.
  • Teflon probe tips were inside the holders, their points were NOT sticking out the hole/apertures. There was

much stronger transmitted signal this way.

Blaster Card - reflected amplitude, gray colormap
Blaster Card - reflected phase, gray colormap
Blaster Card - transmitted amplitude, gray colormap
Blaster Card - transmitted phase, gray colormap



The box-looking squarish area to the upper right is the outline of the glass slide the card was expoxied to.

Scans of Violet Leaf, Dime

(03.24.08) CSM 3mm high, small aperture, gray colormap

Violet leaf - reflected amplitude, gray colormap
Violet leaf - reflected phase, gray colormap

It looks like we are seeing some mode structure here as well.

(03.21.08) Dime scan with sharpened probe, small aperture, bone colormap

dime at 1200 dpi
dime - reflected amplitude, bone colormap
dime - reflected phase, bone colormap


(03.20.08) Post processing results, colormap comparison

These are images of a 24hour old leaf of a violet plant. These leaves are around 200 microns thick.

Violet Leaf - transmitted phase, processed, colormap tweaked
Violet Leaf - transmitted phase, processed, colormap grayscale

The vasculature of the leaf is now much more apparent. The boundary of the leaf is now crisp as well. The grayscale colormap does a much nicer job by default than the 'color' colormap.


100 Yuan - transmitted phase,raw data, grayscale colormap
100 Yuan - transmitted phase,raw data, default colormap

Compare the grayscale colormap to the jet? colormap made on 3/5/2008.

(03.20.08) Post processing results, default colormap 'publish' quality plots

Violet Leaf - transmitted amplitude, processed
Violet Leaf - reflected amplitude, processed
Violet Leaf - transmitted phase, processed
Violet Leaf - reflected phase, processed

Plots made with mplSurfPlot.m (and associated mfiles) found in the Coder's Corner.


(03.14.08) 260.2 GHZ scans - JS and BZ

Violet Leaf - transmitted amplitude
Violet Leaf - transmitted phase

Higher resolution scan of the full violet leaf, 50 micron step size in X and Z. A bit of the vasculature of the leaf is visible in the transmitted phase.

Link to the violet_fullScanRange.txt Raw Data.

'Mesh' plotstyle used.

Violet Leaf - transmitted amplitude
Violet Leaf - transmitted phase

Scan area is clipped on the top because Brian cannot align things properly... a fix for that is coming. Data is plotted with MplSurfPlot.m, found here.

Link to the Violet_260.2GHz.txt Raw Data.

'Surf' plotstyle used.


Scans of Dime, Leaf Stem, Letter C

(03.08.08) 260GHZ scans - with great SNR, courtesy of John:

Dime - reflected phase
Letter C indented at about 0.25mm depth on metal - reflected phase
Stem of dried bay leaf - transmitted phase


H-Band Scans

Static Probe - Probe Setup

The probe - probe scans use two teflon probes with tips that have and angle at 44 degrees to the horizontal. Both probes are held in a static position while the a sample of {insert favorite sample material here}, mounted on a 2-axis stage setup, moves in a plane of maximum 98mm x 98mm area at sub-millimeter step sizes (generally 0.1mm).

The entire system, VNA and stages, is under Labview control.

Here are a couple pictures of the setup:

Angle Shot
Straigt Shot



Scan of 100 Yuan currency 3/5/2008

100 Yuan, magnetic stripe 

 
scan area: X=10mm by Z=35mm
resolution: 0.2mm step in X-Z

260GHz H band with HPF23 filter, gun oscillator
teflon probes with AL sheathe
Scan Area
Transmitted Phase
Transmitted Amplitude


There is some horizontal striping (horizontal cyan stripes) in the data that is a result of phase wrapping that I am unable to supress.

scan area: X=15mm by Z=20mm resolution: 0.2mm step in X-Z

260GHz H band with HPF23 filter, gun oscillator teflon probes with AL sheathe </pre>

Scan Area
Transmitted Phase
Transmitted Amplitude

Ignore the black box in the scan area picture, it serves no purpose. The rectangular cyan area is the 15mm x 20mm scan area covering part of the '100' written in optically variable ink (OVI) in the lower left corner of the bill. Don't mind the fact that the 100 appears written backwards, I am too lazy to flop the data in X. Note also that the serial number of the bill, written in non-magnetic ink is not visible.


Scan of teflon holes with HPF23 filter (high pass at 170 GHz) 2/27/2008 

scan 1: teflon_2mm-spacing
##########################

sample was perforated teflon, 0.6mm holes at 2mm spacing (bottom row of teflon perf sample)

scan area: X=10mm by Z=5mm
resolution: 0.1mm step in X-Z

170GHz H band with HPF23 filter, gun oscillator
teflon probes with AL sheathe
Transmitted Amplitude
Transmitted Phase

Teflon_2mm-spacing_sample.txt Raw Data The scans of the 0.6mm holes in teflon are much smoother due to the higher S/N operating at a lower frequency, 170 GHz versus 260 GHz. It was necessary to change filters on the gun for this, however.


Scans of teflon holes with HPF34 (high pass at 260 GHz) 2/25/2008?

1mm Hole Spacing
.5mm Hole Spacing

Scan quality is not very good, as the data are quite noisy. Data

Copper Tape Scans

The actual position of the tape is about 2mm (normally) from where the data indicates it starts. .1mm resolution. Scan of Copper Tape -> Data

(March 3, 2008) Here is another scan. I measured the dimensions and scan area of the tape this time. .1mm resolution. Picture -> Data

Leaf Scans

Its tough to tell anything here; bad S/N. The large dip on the left is the main stem of the leaf. .1mm resolution. Scan of Part of Leaf -> Data

(March 3, 2008) Better S/N ratio leaf scan at .1mm resolution. Picture -> Data ‎

44 Degree Probe Scans

Probe - Probe Scans

Picture showing gaussian distribution
Raw Data

The probe - probe scans use two teflon probes with tips that have and angle at 44 degrees to the horizontal. One probe is held in a static position while the other, mounted on a 2-axis motor setup, moves in a plane 1-2mm in front of the other probe.

Scan of probe tips
Raw Data

This scan is a .4mm resolution scan of the center of the probe - probe scan. It shows the decrease in magnitude of the millimeter :wave right at the tip of the probe.

Resolution Scan

Unknown Plastic Sample

The resolution scan was originally intended to find the maximum resolution of the new probe tips made with a 44 degree angle. Instead of finding a max resolution, the data showed how the millimeter waves diffracted through the holes (verify). x/y scanning resolution is .1mm.

Resolution Test Scan

Raw Data

Scanned sample is a small plastic plate 2mm thick. Holes are .5 mm diameter (verify) and spaced at different intervals at different rows.

Scan of Sample (DPI ?)
Better Scan (DPI 1200)
Row Spacing (mm)
1 2.5
2 2
3 1.5
4 1
5 .75 (?)

Teflon Sample

Raw Data

.6mm holes at 2mm Spacing


Old Near Field Scans (Raw Data Only)

Old near field scans done by Nathan Greene.

Half Circle Granite
Oil Shale

1D Line Scans of Perfboard

No Focusing Device
Lens
Two Cones


MATLAB Code

Here is my code that I have used to graph most of the data. Suggestions and comments are welcome.

I would like to suggest a separate page for all m-files and code. I will put it here for now. Updated: March 7, 2008 

%this m file plots a surface of a 2d array of data with a third dimension
%being the value in the array.
%plots from the raw data file
%
%Input:
%   filename- filename of the data file
%   step- step size of data in the x and y direction (assumes they are
%   equal)

function plot_trans(filename,step)
    %handy code to ignore the header information if it is there
    data = textread(filename,'%c',1);
    if (ischar(data))
        data = dlmread(filename,'',2,0);
    else
        data = dlmread(filename);
    end
    leng = length(data(:,1));
    %this makes a square for the data to reside in
    %code later will shave off the zeros on the sides
    arr_size = max(data(leng,1),data(leng,2));
    
    [Y,X] = meshgrid(0:step:arr_size);
    trans = zeros(length(X),length(Y));
    
    for i = 1:leng
        %find coordinates for the data point
        % uses the first and second columns to determine where the point
        % goes into the big square array
        x = round(data(i,1)/step)+1;
        y = round(data(i,2)/step)+1;

        trans(x,y) = data(i,3);
    end
    
    %get rid of any zero row/columns
    test_x = 1;
    test_y = 1;
    for i = length(X):-1:1
        %test row = all 0
        if (test_x == 1)
            if (trans(i,1:length(trans(1,:))) == 0)
                trans = trans(1:(i-1),1:length(trans(1,:)));
            else
                test_x = 0;
            end
        end
        %test column = all 0
        if (test_y == 1)
            if (trans(1:length(trans(:,1)),i) == 0)
                trans = trans(1:length(trans(:,1)),1:(i-1));
            else
                test_y = 0;
            end
        end
        if (test_x == 0 && test_y == 0)
            break;
        end
    end
    
    %graph
    surface(trans);
    %surfl(trans);
    %set(gca,'DataAspectRatio',[1,1,.1]);
    figure(gcf);
    %mesh(X,Y,trans);
end
Retrieved from "http://mesoscopic.mines.edu/mediawiki/index.php/Near_field_scanning"