I can't find any scans

[umartechboy]

First of all, fantastic effort, in putting your work on GitHub. The level of workmanship that you have displayed in this STM is exemplary.

We are attempting an STM here in my university and your mechanical design appears to be a promising start for this.
However, browsing the git, I couldn't find any real scans of the microscope. Were you able to complete this project? Also, have you published this work in HardwareX or a similar journal? I would love to read more about how to put all of this together.

I see these questions have been asked in other issues as well, but here is my list:

1. What software do you use for electronics/mechanical design, I don't recognize the file format
2. All of your designs and PCBs look too mature for a prototype. I find it hard to believe that you haven't produced results with it yet. Have you already run system-level simulations. What parts do you think must take more iterations?
3. How confident are you with the tunneling amp and isolation circuit scheme of the PCB? It would take significant resources at my end to get a pre-fab PCB. Do you recommend doing it on PCB-router first?
4. If you are comfortable, kindly share your email address, mine is [email protected].

[jherkenhoff]

Dear Umar,

thanks a lot for your comment and interest.
Unfortunately I was never able to get a proper scan. However, I have to admit that this project was on standby during the last two years. But thanks to your comment I feel motivated to take the last steps towards getting the scan going :) I might continue in the next days.

Regarding your questions:

1. I use Kicad for electronics and Freecad for mechanical design. Both are open source.
2. Yes, I directly ordered the PCBs without any prior prototyping. That also meant that there were some hardware bugs, but to my surprise there were only minor problems (like a screwed up connection of two op-amps on the motherboard) that could be easily fixed. I did not run any system level simulations. I am actually quite happy with the design - but at the same time I was not able to get a proper scan, so I might be mistaken here :D I believe the main problem is the mount of the SEM tip, which is very loose and flimsy. I was able to get tunneling going, but after a few seconds it became unstable. Tuning the PID loop did not help, which is why I strongly suspect the tip mount.
3. I am quite happy with the tunneling amp. I can do some characterization measurements regarding leakage current if you are interested. This type of "leaked current guarding" is quite common in high sensitivity frontend design, so I figured that it should be good enough. Even though the amplifier is a ver y important part of the STM electronics, its circuit is one of the most simplest, which is why I would not have trouble going straight into ordering "real" PCBs.

Good luck with you project and I'm very interested in how your STM will look like and perform :) Also I'm very happy to further discuss 👍

Cheers
Jost

[umartechboy]

[umartechboy]

Hello Jost, I'm really glad to see your reply. Sure, lemme set up this software, view the source as it is, and work towards replicating here what you have in your lab. It should not take more than a couple of weeks. I see you have performed some characterization on the op-amp (even frequency response). This means that we should not suspect the op-amp, at least for now. We worked on our own optical breadboard-based STM two years ago here ( physlab.org). Unfortunately, I couldn't get a confident tunneling current going. Although, we did have a piezo-nano-translation stage complemented with pico-motors and manual movement screws. I am sharing some photographs here though. However, the main reason I'm trying to replicate your STM is the mechanical structure. There is a relatively smaller mechanical distance between the tip and the sample, making it less prone to vibrations and thermal strains. Some more questions. 1. Have you developed a custom firmware based on the STM32 controller you mentioned? 2. How do you prepare the probe? is it a simple tungsten wire? We sought help from this not for ours: "Note: Advancement in tip etching for preparation of tunable size scanning tunneling microscopy tips", REVIEW OF SCIENTIFIC INSTRUMENTS 86, 026104 (2015). 3. I see the piezo actuator for nano-scale movement, and I see the screws (probably pre-stressed with springs) for course manual movement. How do you achieve micro-scale movement?

…

On Tue, 13 Feb 2024 at 12:48, jherkenhoff ***@***.***> wrote: Dear Umar, thanks a lot for your comment and interest. Unfortunately I was never able to get a proper scan. However, I have to admit that this project was on standby during the last two years. But thanks to your comment I feel motivated to take the last steps towards getting the scan going :) I might continue in the next days. Regarding your questions: 1. I use Kicad for electronics and Freecad for mechanical design. Both are open source. 2. Yes, I directly ordered the PCBs without any prior prototyping. That also meant that there were some hardware bugs, but to my surprise there were only minor problems (like a screwed up connection of two op-amps on the motherboard) that could be easily fixed. I did not run any system level simulations. I am actually quite happy with the design - but at the same time I was not able to get a proper scan, so I might be mistaken here :D I believe the main problem is the mount of the SEM tip, which is very loose and flimsy. I was able to get tunneling going, but after a few seconds it became unstable. Tuning the PID loop did not help, which is why I strongly suspect the tip mount. 3. I am quite happy with the tunneling amp. I can do some characterization measurements regarding leakage current if you are interested. This type of "leaked current guarding" is quite common in high sensitivity frontend design, so I figured that it should be good enough. Even though the amplifier is a ver y important part of the STM electronics, its circuit is one of the most simplest, which is why I would not have trouble going straight into ordering "real" PCBs. Good luck with you project and I'm very interested in how your STM will look like and perform :) Also I'm very happy to further discuss 👍 Cheers Jost — Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACBS4DVRVCBXIHRXEJCU7L3YTMLDZAVCNFSM6AAAAABDEL65XWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNBQGYZDMOJQGA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[danielreda]

Dear Umar,

thanks a lot for your comment and interest. Unfortunately I was never able to get a proper scan. However, I have to admit that this project was on standby during the last two years. But thanks to your comment I feel motivated to take the last steps towards getting the scan going :) I might continue in the next days.

Regarding your questions:

1. I use Kicad for electronics and Freecad for mechanical design. Both are open source.
2. Yes, I directly ordered the PCBs without any prior prototyping. That also meant that there were some hardware bugs, but to my surprise there were only minor problems (like a screwed up connection of two op-amps on the motherboard) that could be easily fixed. I did not run any system level simulations. I am actually quite happy with the design - but at the same time I was not able to get a proper scan, so I might be mistaken here :D I believe the main problem is the mount of the SEM tip, which is very loose and flimsy. I was able to get tunneling going, but after a few seconds it became unstable. Tuning the PID loop did not help, which is why I strongly suspect the tip mount.
3. I am quite happy with the tunneling amp. I can do some characterization measurements regarding leakage current if you are interested. This type of "leaked current guarding" is quite common in high sensitivity frontend design, so I figured that it should be good enough. Even though the amplifier is a ver y important part of the STM electronics, its circuit is one of the most simplest, which is why I would not have trouble going straight into ordering "real" PCBs.

Good luck with you project and I'm very interested in how your STM will look like and perform :) Also I'm very happy to further discuss 👍

Cheers Jost

I would also be interested to see if you can get a scan. How long did this take you to build? Approximately how much would it cost to reproduce this instrument?

Daniel