General information on doing your thesis at iMEK

Dear Students,

We are offering a wider range of background information and references for students doing any level of thesis at our institute, most of them on this platform. A very conclusive list of documents can only be accessed after you found your topic and you received your access codes. Nevertheless, we decided to give in this public space some hints on typical theses are done and how the related support documents are structured.

Good Scientific Working

The Hamburg University of Technology offers specific rules for a good scientific working. This goes far beyond the practical guidelines given here, but is the basis of all our doings and must be also a basis for your doings. Please read and understand these general rules: https://www.tuhh.de/tuhh/forschung-und-transfer/gute-wissenschaftliche-praxis.html

Orientation in Confluence

There are different spaces related to different projects. You will see douzends as soon as you received access to the Confluence-server.

• Your username is your email address and your password. These things are given to you by your supervisor.
• You can find the PRJ spaces (PRJ = project) list from the space/Bereiche pop-up menu on the upper left side.
• We have different PRJ spaces such as: PRJ- Therapy System, PRJ- Finding scratches on the turbine surface and haptic systems, Robots for Paralyzed Patients Home that you can get some information about your project and other students projects.

Style of Working

The most important thing when doing engineering work is "methodology". The professor always says: "it is your lifeline". When you stick to engineering-methodologies, engineers can work on any unknown topic and if they follow them they will reach a result which is with a very high likeliness at least "good", frequently even better.  Engineering-methodology is about making an abstract formulation of the task (formulating the requirements), and finding a solution according to those requirements in some structured way. The level of solution (concept/simulation/prototype/measurement) depends on your individual task. At Institute for Mechatronics (iMEK) this methodology is KEY and needs to be visible in each thesis and it is - simply because it protects you from doing wrong things if applied correctly - the most relevant indicator for your grade.

Development-methodology is lectured in a couple of courses at TUHH. All these courses are aligned in content (such as all the engineering companies worldwide are). Our course with a specific focus on Mechatronics is called "Appplied Design Methodology in Mechatronics (ADMM)". If you did not attend it yet, please do also take a look at ADMM (Applied-Design Methodology in Mechatronics), it is one of the key lectures for development methodology and is considered a basis for all thesis done at our institute.

Another thing which you need to be aware of as a future engineer: Change is normal! You will grow with your task and while executing it you will learn more. You will never know exactly what you are going to be faced within the beginning. It would be a pity. If you would know, the task is not challenging enough because you have to grow and you have to learn. The important thing is, to nevertheless plan. You may revise that plan, but without a plan, you can not work towards a target. Plan early, revise your plan on a bi-weekly level to adapt to your current findings. This is the only way how you can do both in parallel, grow with your task and stay in the development methodology!

Entrepreneurship for Yourself

You are the owner of your thesis. Your supervisor, your professor, the institute is interested in it and will support it. But in the end, the decisions taken should be yours. Be convinced of each of them. If you feel your supervisors are pushing in a certain direction, make sure to understand why they are doing so. In the end, it is your task and your responsibility and you will get your grade based on your decisions. Your supervisor's hints are important, because of their experience and broader understanding, but they never release you from being responsible. Be aware of this and take ownership of your thesis!

Starting your work

Confluence and Visibility

When starting a thesis and after you got access to confluence, please make sure that you are visible for iMEK

1. Please check that your project is in the related PRJ space and your name is on the student works page (like as previous ones in the below list).
   http://jira.ha.tuhh.de:8090/display/PTS/List+of+Student+Work
   Ideally, your supervisor already created a base-page of the thesis for you. Remind your supervisor, if they did not do so already. But it is also a good opportunity for you to learn about it. For making a page you should work with the blue button on the top of the page (create), blank page. You can also copy the previous pages or use a template for the thesis.
2.  Firstly, in each project, you should define the requirements and time plan of your project. Below you can see two examples
   http://jira.ha.tuhh.de:8090/display/PRJMES/Requirement+specification+1.1
   http://jira.ha.tuhh.de:8090/display/PRJMES/Zeitplan+Arne+Eschweiler

Prüfungsamt

If you do a bachelor-thesis (BA) or master-thesis (MA) please start in parallel the process with the "Prüfungsamt".

Please note, these official documents of the Prüfungsamt are very important. The end-date is binding. The title is binding. Any deviation of title or end-date will result in serious trouble with the Prüfungsamt. Typically, the application to the thesis is done by you as soon as you decided to start with your thesis, your supervisor and the professor will fill out the documents and resubmit them somewhere near the middle of your thesis. Afterwards, no change is possible anymore! 

Concerning timing:

We usually grant 2 weeks of orientation, in some exceptions also 4 weeks of orientation on a new topic. But afterwards, the official timing is triggered. Your supervisor will align an official end-date with you which is binding then. We are very strict on this, the chances must be equal for everyone! 

Concerning Written Report

Besides the general supervision, before your defence, we should read and review your written report multiple times. Do not submit it before finalizing and alignment! Of course, on a formal level, you can do it. But those who did in the past always had much worse reports than those where we had a chance for early review.

Concerning Defense

Defence happens usually between 1 and 3 weeks after the submission of the written report to the Prüfungsamt. Make sure, that you align a preliminary defence date at the middle of your thesis. Remember your supervisor about it, they may forget to align it! The "slots" for defence are frequently taken early in advance, and especially at the end of summer and winter term, there is always problems aligning suitable dates for everyone.

If you have any time limit due to personal reasons,  let us know before! 

Ordering/Purchase of Components

As in 60% of all thesis, we want to show a real physical built of a sample. So the topic of ordering frequently is an issue for students. Ordering in a professional context needs time. Specialized systems (sensors/actuators/material in general) may have a lead-time of one to three months simply because of the suppliers. On top of that, ordering at any entity requires time as the process of releasing money always requires some levels of confirmation. Something you would order at AMAZON which on a private level may reach you within two days will take at least one to two weeks at a university or any other large company.

How to handle:

• Plan enough time between design-finish and start of built.
• Try to identify critical items early. Start the ordering-process for everything you are sure that you are going to need, even if your design-process is not done yet!
• Align with your supervisor on the ordering.
• After placing the order, do not assume that everything will be handled without your attention. Align with the supplier whether the order was received (in case of very special components and expensive ones, the receipt of the order may by itself take a couple of weeks) and for small components align every one or two weeks with your supervisor. They have access to lists, where they can see the orders issued by the institute!
• In the attached ZIP there are some templates that may help!

Templates (at the end of this page)

Attached you can find some templates that you may need during your project.

• One of them is for ordering
• Another one is the websites that we can order from them (you should send the orders to our secretary (Janick Lokocz)
• One of them is a good sample of the requirement analysis which should be done in the first step of each work and should be brought in your report.

Toolshop

You may need parts being manufactured for your thesis. There are a couple of ways to do so. Your supervisor will navigate you through the different approaches. In any case, make sure to get reviews of your mechanical designs from skilled supervisors or technical personnel. Be sure to judge properly on what you are experienced in, and where some learning may improve your designs.

• We have a certain level of standard manufacturing machinery ourselves in direct access to the institute. It is possible to use them (turning, milling, drilling, ...) after an introduction by the technical team of iMEK (especially Thorsten Düring). If you do the things yourself, of course, it is the fasted way to realize/build something.
• If things are half-urgent, the technical-team of iMEK is happy to help. Issue the tasks in alignment with your supervisor at Thorsten Düring. But the technical resources at the institute are limited, so depending on the level of urgency we may have to reprioritize. In other words: This way works, but it has some risks. It has to be mentioned that the technical-team is a german speaking fraction and you may encounter difficulties if you can not express your concern in german.
• If you plan properly manufacturing is planned with a time-buffer of four weeks after the final models/CAD is done. Then we would use our "Zentralwerkstatt" of the University. They have all manufacturing-methods available and a huge capacity. However, their typical lead-time is between 2 to 4 weeks. If you want to use them, go there with your supervisor, align the task early (e.g. 2 weeks before submission) that they can plan for it and then be in-time to submit your drawings/CAD/CAM. Give them a chance for review. Their input is always of high-quality and will improve your design. The workshops in building K and O are led by different "Meistern" as Dirk Manning https://kontakt.tuhh.de/suchergebnis.php?Lang=en&Suchtext=Dirk%20Manning is responsible for the workshop in O.
• Sometimes, 3D printing is a suitable way to execute mechanical soft designs or early samples. The Institute of iMEK has a very strong collaboration with the "WorkINGLab". We can print there on short notice. However, make sure to make a conscious decision on when to use 3D-printing. As we strive for optimum designs, there are only a few reasons which work well. To give you an idea
  • We needed small weight → good reason but may come as a boomerang if stiffness is not sufficient.
  • We were short in time → May be bad reason, as you may had a wrong planning of your time. But depends on the situation.
  • We needed it, as the design was not manufacturable → It May be necessary, but frequently is only a statement that tries to hide that CAD-design was not done according to manufacturing standards
  • It is only a sample → Can be a good reason, if the function of the sample is limited to things 3D printed parts can do. Absolutely applicable for showcase-samples and volume-studies without major mechanical function.
  • We needed electrical insulation → may be a good reason
  • We needed organic/curved surfaces → may be a good reason
  • We needed functional integration up to a level where our tool shop proposed us to use 3D-printing → very good reason

Spotlight

We are doing spotlight-presentations (two for BA, two to three for MA and PA) at our institute. Details on this format you can find here.

You can find some samples for spotlights here:

http://https://imek.atlassian.net/wiki/spaces/MLP/pages/12419288/Spotlight+-+Studentische+5-min+Vortr+ge

Please arrange a suitable time for your first and second spotlight (5minutes presentation) on this page.
Spotlights take place only on Tuesdays at 11:30 in presence. At the 1st Tuesday of each months we are doing an online session for external students not from Hamburg https://zoom.us/j/3348867229, but see the organisational page for changes on timing.

It is definitely possible to attend spotlights without presenting. It is an open event. You may join if you are interested to listen.

Some Details

The first spotlight is only 5 min introduction of your work (just for a general view) so please only introduce your work and a short talk about the things you do till that time. It can have (stared items are the most important items):
What was the inspiration for solving which problems you want to do this work, what is the objective of your work? The ability that your system should have, and bring some of your developments to show that you are eligible enough to do this project. (attached you can find some samples)
The topics and issues you present are determined by the time in your project, where the spotlight is set in. While you are only able to present ideas, plans and background in an early stage, more information is available at a later point in time.

You should think twice about the focus of your spotlight and the topics you are presenting to the audience. Focusing on interesting topics and issues instead of showing everything leads to a better understanding of your audience and typically results in thought-provoking impulses to problems you encountered or will encounter. This may therefore prevent you from common mistakes and give you an idea of how to continue in certain directions. In short: take the spotlights as a chance instead of a mandatory presentation!

• *Please do not go over 5 minutes → the professor will interrupt you if you exceed and will stop the presentation, it is part of the learning process
• introduce yourself
• *Some previous works as a motivational part and your work differences
• * The requirements
• What should you do
• The steps that you have to do
• Different ideas and their comparison 
• The final design (if already clear, but usually only in the last spotlight of your thesis)
• *The steps that you should do with your final design (show your goals by making some diagrams show that in which step are you)
• What did you do and in which step are you (it is better to be as a graph) the simple design then go for complicated
• What kind of simulation and CAD system do you want to use
• which experiments or tests you should do
• What should be included in your software
• *What are the challenges (flexible system, how big, wight, the impact of different parameters)
• What is the overview of your system
• *Future work

The elements signed by star should be on your spotlights

The second spotlight is only 5 minutes presentation to report what did you do and want to do in the remaining weeks.

Supervision

Normally every week we have a regular meeting between supervisor and student!

Prepare the meetings, the meetings are moderated by the student with an agenda and a plan. A typical structure looks like this:

• Student reports on the findings last week and the status of the project from his perspective
• Q&A on the findings, discussion
• The student proposes an update of his planning for the remaining thesis
• The student formulates an outlook for the next week in more detail

It has proven to be very helpful to either create simple confluence-pages for each meeting (not much formality, just as a kind of protocol) in your space or have a PPT which is used as running documentation for copy&paste of findings, discussion points etc. Some formats are more suited to online-meetings where sketches and visual explanations are difficult. It has proven to be helpful to sketch certain things beforehand and use free-hand tools (e.g. paint) to collaborate on the sketch during the meeting rather than doing everything in e.g. paint such that the results are no longer understandable.

While it is always advisable to refine the information exchanged and gathered during the meeting into results and upload these into a confluence directory for meeting notes, a few tools have proven to be helpful for preparing and hosting a meeting:

• Google docs
• Microsoft OneNote
• Notable
• Microsoft PowerPoint
• Windows Whiteboard
• Pen & Paper

However, every supervisor is open to more frequent meetings if required!

Methodology

Literature Research

After defining the requirements, for each project, you need to do a literature review and find previous works (attached there is a good example). It is a very important step because you can find very good ideas. Sometimes you will find good solutions just by combining them. If you can find a review paper it will help you a lot. For this aim, you can go to scholar.google.com then search for a review paper about your topic. It is advisable to take a look into the google search syntax, to find best what you are actually looking for (https://ahrefs.com/blog/google-advanced-search-operators/). But this is only one source. Another very good source is https://ieeexplore.ieee.org/, as it has a very high-quality level and does not create so many collateral findings as google scholar. Similar to IEE, for other research domains, different search machines exist, such as https://pubmed.ncbi.nlm.nih.gov/ for medical publications. Please note, many specific papers are only accessible when you logged into the VPN of the university. You will find a lot of publications unavailable to you, as the TUHH did not subscribe to every magazine available out there. Typical problems are https://www.elsevier.com/de-de and others. Although this information may be found in darker parts of the internet, you can also check out the names of the creators of the paper and contact them via e-mail as they usually send you their paper willingly. Researchers enjoy other people reading their publications.

Some hints on searching:
Write down the search term and engines you have used to prevent searching the same term over and over. Reviewing 5 to 10 pages of a search is usually sufficient, as results on page 10+ are typically far off your search term.

Some hints on reading a paper:
You don't need to read the whole of a paper. First of all, read the abstract, if you find it suitable you can download it. Then go to the conclusion part and most of the time conclusion will complete the information in the abstract. Then, if you find the paper is suitable, just scroll up through the paper and read the titles and see the pictures. Then read the sections that you think can help you. The other important part is the introduction. The last paragraph of the introduction is about the paper. The other paragraphs are describing the previous works in one or two sentences. Then if you find some of them interesting you can find them in the reference section.

The most common tool for managing your literature is Citavi, as it further provides tools for citing in your paper. To make full use of your findings, categorize them properly and make notes about the contents of the paper, to be able to find specific information (for your requirements etc.) later on.

v-cycle

Also, you can find the v-cycle in the attachment that helps you to structure your work. If you did not attend it yet, please do also take a look at ADMM (Applied-Design Methodology in Mechatronics), it is a key lecture for development methodology and is considered a basis for all thesis done at our institute. Your work and your final thesis should be oriented at this V loop. It is very important, it gives you a structure that other people expect (e.g. requirements are done before concept, the concept is done from a general approach to a more specific choice, testing is developed in parallel to a concept, ...)

Reasoning

The decision needs to be reasoned. The reasoning is based on requirements that are based on a task-exploration in the beginning. Requirements may change based on findings but never based on decisions taken (important difference, of course, a change in a requirement will result in a different approach. But the objective measure of a requirement comes first and decisions are based on that, not the other way round: E.g. "oh, here is some space, let's do the connection here, it is much easier to do it in CAD. We can reduce the workspace a bit" NO "oh, here is some space. If I use it, what impact does it have on my requirements and is this an acceptable change. If yes, I can reduce the workspace a bit".

Ideally, there is traceability of all your design features to the requirement it is connected to (okay, we do not expect that in such detail, but it should be in this direction how you work).

In the attachment, you can find a good example for writing the requirement list for a system.

Written Report

Writing for your report may happen at any time. In parallel to your work or in a hot phase at the end. We give you some hints and advice on this.

Moreover, There is a Confluence page for students on how to correctly type formulas/symbols, units as well as indices, because many students are not aware of the correct notation of formulas (especially indices or italic vs. normal font).
http://jira.ha.tuhh.de:8090/display/MLP/Formelsatz+und+Formelschreibweise

You can find templates for the paper and thesis here:
http://jira.ha.tuhh.de:8090/pages/viewpage.action?pageId=8192042

An important thing is the number of your thesis page numbers. It should be between 30 to 40 pages and not more than that. If you want to bring more information, you should bring it as an appendix. It is very important because your report has 30% of your final grade. Write sharp and clear. If you feel you can not do it, write extensive, but afterwards make a summary (which is your report) and create long appendix-areas which are small reports by themselves on some details. Not page-limit on the appendix!

For writing an engineering paper or report you can use this link: http://jira.ha.tuhh.de:8090/display/MLP/How+to+write+a+good+Engineering+Paper

Written Report Infrastructure

For easier communication, you can define your project as a GitLab project. Then make three parallel branches there. 1. Thesis 2. Practical works 3. documents (you can divide it into more branches). For more information on how to make a GitLab project:
https://www.youtube.com/watch?v=Jt4Z1vwtXT0

Of course you can also use Confluence as a basis to exchange on your written project, but then you need to upload the files.

From this link, you can get an overview of how we evaluate your work at the end.

Culture

In our institute, we are friends and colleagues. There is no supervisor who is not highly motivated and put his heart to support you in your work. We all started this work in this university and want to develop our centre. So it is not for me or for a specific person. It is a place for helping students and our community. Whenever you want, you can call or send an email. 

But at the same time, we will also highlight missing things and will formulate professional criticism if required. We are very "German" in our open communication. But even if professional criticism is formulated, we will forget about that if we see people learn from it. Learning (the gradient of change) is the most important thing for engineers. Criticism is to trigger learning and thinking.

Labs

All our labs have special security and safety concerns. If you are doing your thesis at our institute at some point you will be using either kVolts or kAmps of voltage or current, or you may be operating equipment which has a value of >> 100.000€. You will never do so without any initial advice/guidance by your supervisors and everywhere where there is a risk of harm of course safety measures are taken to prevent anyone - student, employee or visitor - from being at risk. However as these are labs, there is always an option to violate rules and overrun safety-switches and damage material. The following area collects some advices for certain lab-areas.

General

The general Laborordnung/Labaratory-Guideline can be found here.

UR Robot

If your project is with using UR robots there is a page regarding the rules for using UR robot. Please read it. (Rules to Use Universal Robot and Schunk Sensor)