Activity for Timo Kandra
-
Timo Kandra posted a comment on discussion Help
Hi, I am trying to figure out the most effective way to export a scan. Ther is pdf and image export but both do now include a scale bar (and a legend where the high to color information can be obtained) Additionally the export does not work with: pixel ratio != size in nm ratio What is the best approach to generate a pdf or png from the scan?
-
Hi, I would like to change the approach method to not fully retract (currently the tip is fully retracted while doing the approach steps.) The reason is, that I need very high voltages for a working z axis coarse motion. Adding the scan retract ontop of the high voltage slipstick might exceed the piezo voltage limits. Is it possible to easily change the approach so it does not fully retract but instead goes to a maximum voltage. (so that the scan channel does not output +-10V while doing the slipstick...
-
This is phenomenal and exactly what I was looking for! Thanks!
-
I would like to have a much higher resolution in the scan line (x-axis) but a low amount of scan line. For example 512data points per scan line and 64 lines per image. This can easily be set in the main window. However if i choose a range that is square (300nmx300nm for example) I would like the scan to be displayed as a square image as well. Currently it would show a 512:64 aspect ration image in the scan preview window...
-
I did a tunnel gap simulation: All information can be taken from the images attached (using a 100Mohm gap simulator and an according bias gain - the resulting current can be seen in the HUD) It indeed is inverse/does not behave correctly without any piezo and amplifier involved: setpoint < measured current ------> extends the viz !!! setpoint > measured current -------> retracts the viz !!!
-
If I set the current setpoint to 0 this happens:
-
Since the issue with the visualization is still not fixed I am adding screenshot for those scenarios as well: I am using the auto tab and slipstick Z movements to get out of the tunnel contact and to fully crash the tip. The current shows what I have done for which screenshot. For all of that I am using the correct feedback setting (Z-servo Neg) Therefore the tip is fully retracted where I have the saturated current 100nA and the tip is fully extended where I am out of the tunnel contact (<100pA)...
-
I also did another test: tip is in tunnel contact. In the patch rack I switch von Z-serve Neg to Z-servor -> with Z-serve Neg the tip is held at the tunnel contact with Z-serve the tip is pulled out and the current is read as ~0A. This lets me conclude that my setup is indeed Z-servo Neg. What is weird is that setting it to Z-Servo shows the tip to be fully extended but also shows a tunnel current of 0... That is the same visualization bug.
-
I tired this configuration. It does indeed invert the Z-Offset but not the coarse motion. That makes sense to me since the coarse motion does not use Z-Offset as the output but instaed uses whatever channel is set in Mover Control/Config/wave/direction->DAC(0-6) why is wave/direction(xyz) not also a set of signals that can be configured in the patch rack? It does not feel consistent that the output channel for basically anything is configured in the patch rack but the coarse motion in the Mover Control...
-
I tired this configuration. It does indeed invert the Z-Offset but not the coarse motion. That makes sense to me since the coarse motion does not use Z-Offset as the output but instaed uses whatever channel is set in Mover Control/Config/wave/direction->DAC(0-6) why is wave/direction(xyz) not also a set of signals that can be configured in the patch rack? It does not feel consistent that the output channel for basically anything is configured in the patch rack but the coarse motion in the Mover Control...
-
During the auto approach procedure the controller will do couple of steps using the sawtooth function: my question is: will those steps be executed through the channel configured in the config tab or will those steps use the z-serve (neg) signal?
-
I managed to get everything working after fully reinstalling ubuntu 22. And using the apt package for the kernelmodule instead of compiling it myself
-
I managed to get everything working after fully reinstalling ubuntu 22.
-
For Ubunut 22 those packes were required for me as well: python3-numpy, libpython3-dev, libopencv-dev, libpopt-dev as root (in addition to the ones form the initial post): apt install python3-numpy libpython3-dev libopencv-dev libpopt-dev
-
The question was mostly focusing on ONLY: Is the sawtooth stepping also done on channel 5 when using auto approach. (If I am just using corse motion, the sawtooth is transmitted over channel 2 (the z offset channel)
-
Is that actually what is happening? (auto approach only uses channel 5? I am not at the setup atm to test this myself)
-
this does not yield any errors. It does not print anything. Secure boot is not an option i the old BIOS the pc.
-
this does not yield any errors. It does not print anything
-
this does not yield any errors
-
This is only if I want to use the svn repo? I would rather use the ppa tbh. if I try to run sudo modprobe in the directory of the .ko file: /lib/modules/5.17.15-051715-generic/updates/dkms$ sudo modprobe ./usb-sranger_mk23.ko modprobe: FATAL: Module ./usb-sranger_mk23.ko not found in directory /lib/modules/5.17.15-051715-generic and the same for the one in extra; /lib/modules/5.17.15-051715-generic/extra$ sudo modprobe ./usb-sranger_mk23.ko modprobe: FATAL: Module ./usb-sranger_mk23.ko not found...
-
This is only if I want to use the svn repo? I would rather use the ppa tbh. if I try to run sudo modprobe in the directory of the .ko file: /lib/modules/5.17.15-051715-generic/updates/dkms$ sudo modprobe ./usb-sranger_mk23.ko modprobe: FATAL: Module ./usb-sranger_mk23.ko not found in directory /lib/modules/5.17.15-051715-generic and the same for the one in extra; /lib/modules/5.17.15-051715-generic/extra$ sudo modprobe ./usb-sranger_mk23.ko modprobe: FATAL: Module ./usb-sranger_mk23.ko not found...
-
This is only if I want to use the svn repo? I would rather use the ppa tbh.
-
Even that does not work now. I removed and reinstalled the apt package but ls -l /dev/sranger* not listing anything. Do I have to follow any procedure: Booting with usb device turned off? and then turning on? This is the output from sudo dmesg | grep srang [sudo] password for spmuser: [ 14.414111] usb_sranger_mk23: loading out-of-tree module taints kernel. [ 14.414156] usb_sranger_mk23: module verification failed: signature and/or required key missing - tainting kernel [ 14.414511] usbcore: registered...
-
Even that does not work now. I removed and reinstalled the apt package but ls -l /dev/sranger* not listing anything. Do I have to follow any procedure: Booting with usb device turned off? and then turning on?
-
Even that does not work now. I removed and reinstalled the apt package but ls -l /dev/sranger* not listing anything.
-
I have trouble installing the sr kernel modules. ubuntu updated to 22 lts. I needed to reinstall everything. First the kernel modules made no issues following the tutorial on this website. I had a fully functional environment. After restarting the PC the sranger usb device was not detected anymore.I treid running sudo make sudo make install sudo make load again but that did not solve the issue. I installed another kernel and reinstalled the kernel modules again,. During installation I get the follwoing...
-
I do not do the combined slip-stick Z-scan anymore. We changed the amplifier and the new one now has an offset input and substracts scan-offset. To confirm: the slipstick output is not used for the approach in my case 258 (but could also be 2. I dont see the need to use adding mode with the seperate offset amplifier channel) So what you are saying is, that when the auto appraoch is used, the slipstick steps are also done using the scan channel 5.
-
I do not do the combined slip-stick Z-scan anymore. We changed the amplifier and the new one now has an offset input and substracts scan-offset. To confirm: the slipstick output is not used for the approach in my case 258 (but could also be 2. I dont see the need to use adding mode with the seperate offset amplifier channel)
-
Isnt teh Z offset also used for the apporach? In that case the polarity would invert the slipstick motion?
-
Reposting this question at the end of the thread since it is still very weired to me: I am still very puzzeled: How can the feedback loop work and the approach direction even though the Zs signal is not inverted but the Zoffset is inverted by the amplifier. (I have not toched any settings. The polarity is negative and the output on channel 2 which is used for Zoffset is not inverted in the gxsm SR MK3 patchrack) With the inverted offset channel should not either the appraoch direction be wrong or...
-
Ah okay. Thanks!
-
I was wrong, the cycloide+ moves in the same direction then all the other functions. So all other functions also can be used for the approach. I am still very puzzeled: How can the feedback loop work and the approach direction even though the Zs signal is not inverted but the Zoffset is inverted by the amplifier. (I have not toched any settings. The polarity is negative and the output on channel 2 which is used for Zoffset is not inverted in the gxsm SR MK3 patchrack) With the inverted offset channel...
-
This repo? https://github.com/pyzahl/Gxsm3 The last change seems to be from the 21st of Juli.
-
Also it would be really nice to have the negative/positive value explained in the help context of the setting? Positive for +HV fully extended, negative for +HV fulmy retracted.
-
Thanks for looking into it so far. What confuses me now is that our setup is indeed Z positive +HVmax = fully extended. But then the question is how the feedback loop only works with the negative setting. ALso the slipstick motion seems to work with negative correctly (it is inverted now, so the approach moves the tip further away from the sample but that is what we want, since the amplified offset channel is inverted as well) My current workaround is using the cycloide+ fkn. That moves the tip inverse...
-
To invert the offset OUT2, you can do this via the PatchRack, setup a "-1" scaling via a "UserConstant" Signal. I'll give you details later. Busy now. Those detials would be really helpful. I thibk from what i ubderstand now I would create the signal that is just a cobstant -1 and use it as a scaling/modulation/multiplier CH2_something so it is inverted.
-
You do have the Z-Piezo Polarity = Negative if I recall right? Yes I think that is my configuration: the tip is extended if the output is +MAX HV. I am not sure what settings is what though. To me those two statements sound to be contradictory (negative setting in preferences is default: -Vmax == fully retracted, +Vmax = fully extended) and Then you have a very usual hardware / piezo configuration and you should use the Gxsm default "negative Z polarity". (+Max HV at Z= fully retracted)
-
Hello, it would be really helpful to get some explanation on the z orientation/polarity. The setup: one piezo for z controlled by an amplifier which mixes the offset channel (ch2) and the scan channel (ch5) like this ampFactor * (ch5-ch2) the z-piezo extends (when close to the surface is crashes) when applying a positive voltage and it contracts (leaves tunneling and I measure ~0A) when I apply a negative voltage. In the HUD the tip moves inverse and no setting in spm-inst can change this. Moves...
-
So to sum up: I choose whatever is correct for the offset output: in this case positive (+Vmax means the piezo is extended but my amplifyer inverts the offset channel => -Vmax fully extended =>GXSM needs to be configured with Positive setting) Then I have issues with the feedback loop which can be fixed by choosing Z servo instead of Z servo neg
-
So to sum up: I choose whatever is correct for the offset output: in this case positive (+Vmax means the piezo is extended but my amplifyer inverts the offset channel => -Vmax fully extracted =>GXSM needs to be configured with Positive setting) Then I have issues with the feedback loop which can be fixed by choosing Z servo instead of Z servo neg
-
Yes, this is the exact situation I am seeing. I dont have any negative values for the piezo sensitivity and for the feedback loop gain (I created a new user where I never touched those values just to be double save) The HUD is jsut strictly boudnd to the output voltage of ch5. Should It change when changing the Z-polarity? If the controller outputs +10V should the visualization with positive show the tip as extended and with negative the tip should be shown fully contracted (also for +10V output)...
-
I followed all these steps numerous times. I even went that far to restart the pc and double check the current configuration before starting gxsm via the configurator so that Everything is setup as I want it to be. The issue is, that the hud still does not work. (it is always displayed inverted.) Should I try updating gxsm I have the ubuntu iso is there a simple way to just pull and make install? To invert the offset OUT2, you can do this via the PatchRack, setup a "-1" scaling via a "UserConstant"...
-
The issue is, that changing the Z polarity does not fix that. In the HUD picture the tip is still fully retracted (at the top of the window and almost invisible) if the out voltage reads a positive value and fully extended when the out voltage is negative. I cannot change the relation between Picture and outvoltage which I need to be inverted so that I makes sense. (The only thing that changes with the polarity gets inverted, is the feedback loop. The loop needs to use a negative gain to work when...
-
Just to be triple save that we are on the same site: if I have a crash (saturated tunnel current) and the feedback loop is turned on the vizualisation should show the tip at the top (the coarse slitstick is so close to the surface, that even a fully retracted tip cannot prohibit the crash) But for me the tip is ALWAYS (independent of the setting) drawn as fully extended in that situation!
-
This is exactly what I want. the visualization schowing an extended tip if the feedback controller applies voltage to extend the tip/close the gap. But that is not haopening with whatever configuration i try. The visualization is the other way aroubd. The only thing that changes is that with the positive setting I need to choose a negative gain and with the negative setting the gain has to be choosen positive for a woerking feedback loop (Which lets me conclude that negative is the correct configuration)...
-
I had no success with z orientation yet. I played around with it an noticed, that the positive/negative setting is exactly the same then setting serve or serve neg in the patch rack, right? Whatever I try, I never can have the output be +10V and the HUD displaying the tip to be fully extended. The tip is always shown as fully retracted if the output is +10V. It really seems like the servo sign only has an impact on the feedback loop but not on the visualization. What parameter changes the output...
-
I could not succeed with the z orientation yet. I played around with it an noticed, that the positive/negative setting is exactly the same then setting serve or serve neg in the patch rack, right? Whatever I try, I never can have the output be +10V and the HUD displaying the tip to be fully extended. The tip is always shown as fully retracted if the output is +10V. It really seems like the servo sign only has an impact on the feedback loop but not on the visualization. What parameter changes the...
-
So to sum up: I choose whatever is correct for the offset output: in this case positive (+Vmax means the piezo is extracted but my amplifyer inverts the offset channel => -Vmax fully extracted =>GXSM needs to be configured with Positive setting) Then I have issues with the feedback loop which can be fixed by choosing Z servo instead of Z servo neg
-
I will check for memory usage and also report further details if I can reproduce reliably. (If its not reproducible I highly suspect its my hardware.) I will keep you updated in case it seems to be an actual bug.
-
I often have GXSM crashing during a scan. I think most often this happens, if I activate the palette coloring (but it also happend without that before) Could that be a hardware performance issue or is there a known bug and are there things I should not do during scanning. Best Timo
-
Hello, it would be really helpful to get some explanation on the z orientation/polarity. The setup: one piezo for z controlled by an amplifier which mixes the offset channel (ch2) and the scan channel (ch5) like this ampFactor * (ch5-ch2) the z-piezo extracts (when close to the surface is crashes) when applying a positive voltage and it contracts (leaves tunneling and I measure ~0A) when I apply a negative voltage. In the HUD the tip moves inverse and no setting in spm-inst can change this. Moves...
-
Yes a MK3-A810. The issue also solved itself after another restart of the computer and gxsm (with a different user account) Updating gxsm3 is a good idea as well. I am using the .iso ubuntu version for now. Also the other issue is solved: it was caused by what you warned for before. adding slipstick motion and scan to the same channel. This caused some weiredness (kind of obvious) when I moved out of the range and the feedbackloop already saturated the channel. then the sawtooth did not work and...
-
Thanks for the resetting approach. I have a complete new setup and also loaded the MK2 defaults on the dsp using the Signal+DSP Manager. I still cannot get any waveforms to work except besocke, sine, and stepper motor. My setup to test is as following: the wave table is X:3 Y:4 Z:5 an external oscilloscope is plugged into out5 (zscan (6) at the back of the MK3) I am changing the waveform int he mover control/Config and then doing 50 steps in mover control/auto The result is that with besocke, sine,...
-
Not sure about you wave fore selection issue. Try to restart everything after reconfigurations incl power cycle the DSP. I tired a full restart but the sawtooth wave still does not give any reasonable output. Is there a way to do a full configuration reset? So everything is setup as default? It worked in the beginning and i am worried I messed sth up ?
-
I think I did use confusing wording: clicking noise and puts the tip back into the center of the tunneling with that I mean, that the tip gets moved up/down so that it is still in tunneling. (Not centering on the xy-plane) - In the HUD the blue tip is in the center. When stepping out (z-, i think that means stepping out) the tip moves closer to the bottom (the tube extends) - at some the blue tip is drawn in red and is at the bottom of the hud. - when doing more z- steps a fairly loud clicking noise...
-
Hallo, when using coarse slipstick steps (in the mover/auto tab) while in tunneling, it seems that I cannot leave the tunneling area. I can make steps so the z position needs to be adjusted but as soon as I almost leave or almost crash (using a broken tip for testing and setting up the auto approach) the stm does a louder (than normal slipstick steps) clicking noise and puts the tip back into the center of the tunneling. Is this a safety measure? how can I try the auto approach If I cannot retract...
-
Thanks for the good replies. It took me some time to process them but now I got the first pictures and the stm is setup with gxsm.
-
You should be able to use Z-Scan also as Wave destination. With "wave destination" you are referring to the output from the coarse motion (sawtooth wave) right? This sounds very much like what I want. Since our setup just only uses one Piezo for the z control (both during scanning and coarse approach) Cant I just set the amplitude of the auto app sawtooth to something small enough, so that the feedback loop has enough room to adjust. Or can I limit the range the feedback loop is allowed to use to...
-
I am in the same situation. The reason for that setup is, that the scan-piezo is also the one used for the slipstick. So I want the scan feedback (Z-Servo) and the coarse control on the same channel so it can be routed through the HV amplifier together. (of course I need to then make sure, that the z offset is set to 0 to get the full scan range) I checked out the patch rack. As far as I understand there are multiple sections for each output. one for add one for substract (and modulation ones). The...
-
I am in the same situation. The reason for that setup is, that the scan-piezo is also the one used for the slipstick. So I want the scan feedback (Z-Servo) and the coarse control on the same channel so it can be routed through the HV amplifier together. (of course I need to then make sure, that the z offset is set to 0 to get the full scan range) I checked out the patch rack. As far as I understand there are multiple sections for each output. one for add one for substract (and modulation ones). The...
-
I am now considering using the spm-100 just as an amplifier. Does anyone know if it is possible to disable the feedback loop? For the tunnel current I am planning to use the RHK pre-amplifier attached to the spm-100 and use the current monitor output on the spm-100 as an input for the spm controller. For the piezo control I am planning to use the scan offset analog input so it should be possible to just keep everything connected as it was before and just use the input overrides for bias and scan....
-
Thank you so much for the responses. Ufff, that is really old. We have not been using it for a long time. The RHK SPM-100 we had (it is still in one corner) is a "standalone" electronics that does all the stuff by itself (it is an analog one), the Gxsm plugin was a kind of frontend to it. One problem was that the linux support for the DAQ card we used for the interface was dropped by the manufacturer at kernel 2.X. That sounds like the exact hardware I am planning to use. I need to do spectroscopy,...
-
I want to use gxsm with the RHK spm 100 hardware (see attached image). First of all is that even a feasible upgrade path to the original software by rhk that only runs on XP (I think). I found the rhk_lan plugin in the old manual pdf. But I cannot find the corresponding hardware card setting in gxsm3 (installed with the ubuntu iso). The goal would be to fully control the spm controller with gxsm including the additional pci interface (additional to the ethernet port) Thanks for any input!
1