Integration, Assembly, & Test Flashcards
Preliminary Assembly
A dry run before actually assembling the systen
- Case 1: If you’re welding or using epoxy, then do a dry fit before applying to make sure all your parts fit (sucks to start assembly and as you’re adding parts realize that some parts won’t fit)
- Case 2: if you’re buying or using small/medium wooden or metal structures, you should 3D print them based on the manufacturer specs and see if they can be made to fit together the way you intend before you buy (just make sure you use the callipsers to ensure the print matches the part dimensions)
- This is also something you can do with your PCBs. If you have a PCB design ready make sure it can actually be integrated into your built system before you buy/solder/test it. Don’t just trust the CAD, actually 3D print it and try to integrate it into your mock up.
IAT Things you should never ever skip
*it will be tempting to skip each of these at some point in the project
- Calibration (unless the sensor is so good that it can automatically calibrate with no user input -but this is rare)
- V & V
- Trade Studies
Importance of not skipping V&V
-especially the verification testing for your components (verifying that they function the way they are supposed to is a good way to identify problems early enough that you can design around them)
-it is tempting to skip this step, but if you do, it could bite you in the ass
-should also validate that each of the pin in your microcontrollers function as promised (sometimes a PWM pin might not actually function as a PWM pin)
-make SURE you do your Validation testing (like a test drive)
-designs that do not go through Validation testing have a high probability of failure
-if it seems like there is no way to validate your design, just figure it out (get creative and use your imagination)
-dont be afraid to try something a little crazy. If you think it might not work, you could be suprised
-someone somewhere has a piece of equipment you could use
-ask one of your mentors if they have any ideas on how to validate your product
Murphy’s law in IAT
There is an excellent chance that if you built it, it’s going to break right before you demonstrate it; which means that you need to know how to fix it and fix it fast
General “fix it” tips:
-follow the line of power
-do point-tests for your voltages
-don’t ever assume that something “used to work” before
-do continuity tests for your electrical connections (just because it looks like it’s still connected doesn’t mean it really is)
Redundance in procurement
When ordering parts that are small and/or in larger quantities, order more than the quantity you need for your design. Inevitably some of those parts will get lost, broken, or turn out to be defective
-You can save any extras for future projects
-This sometimes allows you to take advantage of a quantity discount.
-You can keep this strategy from driving up your costs by designing your system so that you have as few line-items on your “bill of materials” as possible: Keep the variety of electrical components (capacitors/inductors/FETs/resistors/switches/motors/batteries) and mechanical components (screws/bearings/extrusion/filament/pipe fittings/valves) to a minimum
-This also helps you save on shipping
Multifunctional Test Stands
Look at all of the tests you have to do for that hardware and see if you can make one stand that can help you accomplish two (maybe even three) of those tests
-ideally, your test stand should enable you to do as wide a variety of testing as possible
Importance of a Pre-task note review
1) Reinforces your understanding
2) Improves performance
When you are going to do a process, look up your notes on that process first and refresh your memory
-If you’re using the Internet for more detail, the more specific you can be about what you’re trying to accomplish and how, the better. Include as much detail as possible when you do the search (ie what accessories will you need, how to tell the difference between a good job and a bad job, considerations for the kind of material you are using, PPE, etc)
-AND MAKE SURE YOU TAKE NOTES ON IT
-don’t just assume that you know the right way to do it, see if you can find more information about how this can be done correctly/better
-this is especially true if you have never done the process before
-unless you just did it the day before, literally just pretend that you forgot all the steps and best practices
-rereading the proper procedure before you do the task reinforces your understanding and improves performance. This helps prevent mistakes, rework, and saves time and material (like when you were soldering the proto boards for Terminus)
Fragile material during testing
If your system includes fragile materials (glass, sheet metal, thin plastic) during testing (example: there’s a good chance it could tip over or even just fall straight down on the ground during testing), then try to use a non-brittle material for that test so that you’re not constantly having to reprint or procure between tests
-saves you time and material
-juat adding a cushion underneath it is not enough
3 aspects of Pre-manufacturing research
1) Physical
2) Material
3) Procedural/Settings
If you are going to be doing a process that you haven’t done before and have very little experience with, you need to look up precisely how that process is done
-Physical needs: specific types of hardware, specific types of accessories, specific types of materials, PPE
-you need to know the storage requirements for these as well, otherwise you could end up with another surprise like the carbon fiber
-Actually know how much material you need. Make sure you get enough material for a practice run. If you’ve never done it before you will almost certainly fuck up the material on the first try or two
-Proceedural needs: video tutorial, forums, enough workspace, aftercare/cleaning
-Equipment settings (especially speed and temperature)
Checking and double checking dimensions/values needed for manufacturing assembly
When working on a project with multiple people, always ask for the most up-to-date numbers before you do an assembly task (it’s like the group-work version of measure twice, cut once). Things like size and power constraints change all the time, so you cannot assume that the numbers you have are the most recent. It really sucks when you complete a task only to find out that your work was in vain (like your first attempt at the secondary board for Terminus)
Component manufacturing (in DFM)
For DFM, you need to make sure that your method for making the parts is actually viable for the use case you have in mind. EXAMPLE: If you are 3D printing grears, you need to be extremely careful with your filament and settings or else they will not work
- can those components be modified somehow so that they made be manufactured in the way you intend?
Kinetic testing for small fragile systems
When you are doing kinetic testing with components that are senitive but will need to move/spin very quickly (like PCBs or glass), you should use “mass analogs” so that they are not damaged during testing, but you still get the same effect. The mass analogs should have roughly the same size/shape/mass/COM as the actual component. These can be made using 3D printed parts, ballast, or even legos
Pre-assembly/manufacturing version verification
always ask for the most up-to-date numbers before you do an assembly task (it’s like the group-work version of measure twice, cut once). Things like size and power constraints change all the time, so you cannot assume that the numbers you have are the most recent. It really sucks when you complete a task only to find out that your work was in vain (like your first attempt at the secondary board for Terminus)
Importance of using “dummy loads” and “mass analog”
Integration testing is very very complicated. nstead of testing with the actual component -especially you are doing something electrical/mechanical that could damage that component with current or force
-For example: in electrical, this means you can use “dummy loads” to test your power supplies
Eight steps in test/experiment design
- A Justification: what are your long-term goals for this hardware and why are they important? (Need to first make sure that this test is actually necessary and why)
- A question: What is the most important thing that you need to know involving this hardware? (This should be a lower level version of ‘Is it going to work in situ?’ -don’t be afraid to ask SMEs to help you flesh this out)
- Desired knowledge: what will you need to know before you can answer the question? (Finding equations can help with this)
- Desired data: what measurements could you take in order to obtain that knowledge?
- Uncertainty minimization: what sensors and equations should you use in order to minimize your propagated uncertainty when collecting that data?
- Calibration plan: what will you need to do in order to calibrate the sensors? What equipment/data will you need and who has that equipment/data?
- Data processing plan: what is your plan for eliminating outliers, what equations will you use for data analysis, and how will you implement those equations numerically? Have you talked with a SME about it?
- Test CONOPS: specifically how will your test-stand help you collect this data? What will it do and how will it be designed?
Preparing to meet SMEs during IA&T
-Writen documentation from Design and IA&T. It should be correct but concise and organized. Don’t info-dump.
-Make sure you’re bringing test data (pictures when necessary). You want the SMEs to look it over because it will reveal problems you might miss)
-if you can, bring the hardware with you
Preliminary fit-testing of PCBs
Always 3D print your PCB design to make sure that it actually fits in the structure in a way that you can still install it, wire it up, and access any on-board connectors. This is the best way to VERIFY that the placement of your terminal blocks and connecters do not need to be changed on the PCB before you order
Guidance for substituting electrical components
ALWAYS make a record of the substitution.
When looking at example circuits in data sheets, the values for some of the passive components usually represent the upper/lower boundary for that component, not the recommended value (they should be marked with ≥ or ≤, but are not). With smoothing caps for example, if you use a higher capacitance than what is indicated on the datasheet you get an even smoother signal with no downside
What to ask when discussing your work with a SME/mentor
1) Ask them specifically where you went wrong and how to avoid that in the future. If it can be done better, you should try to do it better.
2) Get your next steps validated and ask about best practices
3) Do I actually understand the problem I am trying to solve?
4) Want their opinion on each of your risks (did you forget anything? are any of these a non-issue?)
5) Ask them for referals and sugested materials you could use to improve
Philosophy on replacing damaged components/subassemblies
Replacing something should be a last resort. Always do a little bit of research and see if you can fix it first
Importance of equipment settings
The fewer things you have to set on your equipment the more important those settings are. For equipment with only one setting (like a soldering iron), you need to understand what the different levels/modes are for before you can use it properly (in this case temperature settings -in what situation is a lower temperature more advisable than a higher temperature)
-impropper use will damage your hardware
-the more expensive the material you are working with, the better you need to understand the equipment settings to avoid damage
Importance of screw trays
Screw trays: segmented tray designed to hold screws for different components in their own isolated compartments.
It’s even better if it has a see-through lid that you can close, that way you don’t risk spilling them all on the floor if the tray gets bumped by something.
-It doesn’t have to be anything fancy
-literally just print out a bunch of reusable 5x5 screw trays and when you need to use one, add a label from a label maker or even just a piece of tape (that’s the fastest way to get a the functionality you need
-when you have enough screw containers that you’ve emptied (or even just put away in those little cabinet drawers, use the leftover cases for this, that way you have a lid and you don’t risk spilling/dropping your screw tray
Documentation of wires during IA&T
Taking pictures of the wiring is extremely helpful. It is also wise to make “wiring diagrams” to help with the inevitable integration/deintegration (if you’re using multiple of the same PCB, they to make sure the wire colors are consistent every time you use it so that you only need one diagram)
Case 1: Whenever you fry a piece of hardware, take pictures of how everything was wired up will so you can show the SMEs. You will need a definitive answer as to what caused the problem before you proceed, otherwise you could fry another piece of hardware
Case 2: Whenever you breadboard something and it finally works, take pictures of all the wiring in case it gets changed later.
55) For some applications, modular reusable
Documentation of wires during IA&T
Taking pictures of the wiring is extremely helpful. It is also wise to make “wiring diagrams” to help with the inevitable integration/deintegration (if you’re using multiple of the same PCB, they to make sure the wire colors are consistent every time you use it so that you only need one diagram)
Case 1: Whenever you fry a piece of hardware, take pictures of how everything was wired up will so you can show the SMEs. You will need a definitive answer as to what caused the problem before you proceed, otherwise you could fry another piece of hardware
Case 2: Whenever you breadboard something and it finally works, take pictures of all the wiring in case it gets changed later.
55) For some applications, modular reusable
Importance of questioning and validating each manufacturing/assembly process
Just because the process you are using is effective doesn’t mean it’s the right way to do it.
-is there a simpler or more modern way to design/manufacture/test something?
-need to ask the experts or read documentation to make sure youre operating efficiently
Practical role of aesthetic improvements
More often than not, aesthetic impromemts/features can be done in a way that improves the design
Example 1. Paint and Surface finishes can improve longevity
Example 2. Decals can be made to include reference markings or QR codes for documentation
Example 3. Making something black can improve it’s longevity and help with heat management
Importance of validating the assumptions you made during the design phase once you have integrated hardware
Making assumptions about the performance/application of your design is totally fine (you do not have to engineer every single aspect) but when you do make an assumption, you should document it AND test it.
-Example: if you assume there will be no problems with antennas/comms you will need to prove that through testing just like you would test to prove that a requirement was satisfied.
-Do not let the assumption you made in your design continue to be assumptions in testing (testing is meant to test your assumptions)
Optimization and troubleshooting (OAT) phase
1) Good enough is a thing (prevents damage and schedule overruns)
2) Relies on the extensibility/scalability of your design
You must factor in time for tweaks in your project. When scheduling, you should budget about a quarter of your time for OAT.
-OAT sometimes results in damage to your hardware and you will have to fix it. That time should be factored in as well
-Need to know when to stop. After a certain point optimization is more trouble than it’s worth.
-This is also part of why you should make the design as extensible as possible
Cabling lengths
Your cabling should be slightly longer than needed anyway to make assembly and disassembly easier (be advised that this may cause communications problems like crosstalk, but do not assume that this will be a problem. Test it and shorten the cables if need be)
Response to incidents that fry or destroy a component
If your system has broken/fried something (batteries, microcontrollers, etc.) do NOT replace the broken component. Inatead, investigate your process and your system to find out if there is a problem and fix it (take pictures, write an ‘incident report’ and if necessrry make diagrams)
-Otherwise you risk breaking/frying more and more hardware
-It doesn’t have to be two things one immediately after the other. Sometimes multi circuitry can fry two batteries days apart
-Just because you are using software control or electronic protection, doesn’t mean that it is actually the right kind of control/protection for your system (example: dynamic systems usually require different protection/control approaches than continuously steady state systems. A DC component with an ideal/target voltage or current could use an NTC thermistor to limit inrush current… but a motor on PID control is getting a varying signal and needs more active protection circuity)
-talk with your SMEs about the problem if you are not sure
Importance of wire type
Choosing solid-core v. stranded/multi-thread depends on the application:
-In general, solid core is electrically reliable whereas stranded is mechanically reliable. For an application like a robotic arm, you MUST use stranded because it can handle motion/stress, whereas solid core cannot.
-Solid core offers lower electrical resistance, lower cost, and better signal integrity (due to its lower capacitance and continuous material).
-Stranded offers more effective heat dissipation and is able to distribute bending/vibrational/thermal stress between multiple strands
-Always test the wires immediately after soldering. Make SURE you are getting the data/signal quality you want
Using “test configurations”
If you have to do lots of repetitive testing, you might want to get some unique configurations for each test to make repetitive tasks easier. (Example: if you have to integrate and deintegrate part of the system between each test, you should have a configuration specifically designed to make that process as efficient as possible )
-Use your best judgement, this often isn’t necessary
Importance of finding opportunities to eliminate needless connectors
In the path taken by a signal, you want as few connectors or solder joints as possible
-Always be on the lookout for opportunities to remove a connector from your system. (Example: If you have a cable with a connector and you need it to fit into a terminal block, don’t just add wires between the connector and the terminal block. Instead, cut off the connector and put the wires into the terminal block directly)
Importance of never skipping function testing
Skipping function testing or function-level troubleshooting is tempting but you must resist.
-If that component/circuit/whatever fails it will be MUCH harder to troubleshoot because you won’t know if it ever even worked in the first place.
-Example 1: All computers/microcontrollers need to be tested to make sure they actually boot up. If you try to use one and it won’t boot up it will be impossible to know whether it was due to ESD damage or just a faulty unit
-Example 2: When you know the circuit works, but you’re having software problems and you’re stuck on the troubleshooting, you might be tempted to just do the soldering and deal with the software later. DON’T do it. Demonstrate that it works, THEN implement it.
Importance of function testing any electronics modules before using them
ESPECIALLY if you got them used/salvaged
-if you integrate something into your system and it doesn’t work, it’ll be very difficult to troubleshoot unless you know whether it ever actually worked in the first place. Test it once when you first get the hardware… and then again IMMEDIATELY before integrating it.
-this also applies if you buy bigger quantities (if you bought four Pis, you need to test all four Pis)
Basic checklist before applying power
-Confirm there’s no shorts
-Confirm all wires are hooked up correctly
-Confirm there’s no loose conductors (including exposed/unsoldered wires -cap or clip the ends so they cannot suddenly short out your board)
-Confirm output setting on the power supply
-When you’re ready, always connect ground THEN power
Three things that you should do for each process in IA&T that you will have to do more than once on the project
1) Read your notes. You want the correct way to do this to be fresh in your mind. Make sure that you are refreshing your memory on the things that you learned by doing this in the past.
2) Test your results after the first time to make sure that it worked. If it didn’t work, trouble shoot it before doing any of the others
3) Actually look at the work you did while you repeat the process (helps avoid mistakes). Example, if you are soldering to identical boards, when you solder the second board you should actually be looking at the first board you did.
Deceptively easy component/structure manufacturing
Just because a component seems easy to manufacture, doesn’t mean you know how to make it well enough for your application
-if there’s any question about how much precision/quality you will need for your application, you need to ask your SMEs/mentors and if they say so, hire someone to make it for you professionally
-Example 1: Fiber Composites -If all you need is a lightweight rigid structure, that’s easy to make using carbon fiber. But for some applications it will need to be professionally made in a vacuum bag with breather cloth in a special oven
-Example 2: Reaction Wheels -Most people do not realize that reaction wheels require a lot of precision manufacturing to function. If you just 3D print some wheels, it won’t meet the precision requirement and will be pretty much useless.
When to fight for features/capabilities v. when to leave them behind (in IA&T)
Integration testing also often leads to sudden de-scopes. If testing reveals that some functionality or behavior you were hoping to use will not be possible with the entire system integrated, it is sometimes best just to ditch that functionality or behavior -especially if you are on a time crunch.
Example: taking the Picosats off the deployable instead of of troubleshooting the mesh network)
Fight for it if you can, but don’t risk schedule overruns or cost overruns for a feature that’s not even part of the requirements
When to fight for features/capabilities v. when to leave them behind (in IA&T)
Integration testing also often leads to sudden de-scopes. If testing reveals that some functionality or behavior you were hoping to use will not be possible with the entire system integrated, it is sometimes best just to ditch that functionality or behavior -especially if you are on a time crunch.
Fight for it if you can, but don’t risk schedule overruns or cost overruns for a feature that’s not even part of the core functionality/requirements
Five care/procedures when uploading code to hardware
- Any code you upload to your system could break it in some unforeseen way. Always keep a log of every time you uploaded code to your system and what that code was (that way you can easily revert if necessary). Maintain your code in a version control system like git
- Ensure the embedded system is adequately powered and stable during the upload process to prevent interruptions that could corrupt the system.
- Before uploading, you may want to test the code in a simulated or emulated environment (emulated is better because it can catch incompatibilities with the system’s architecture) to catch errors and ensure it behaves as expected. Can also use linters or static analysis or unit tests to verify the code will not cause memory-related problems.
- Bootloaders are programs that protect the system by stepping it through an initialization process. (Loading the OS into memory, executing firmware, etc). This is especially if you’re doing system rollback, over the air updates, or using custom hardware.
- Avoid implementing multiple changes at once (if something goes wrong and you only changed one thing, you’ll know exactly what caused it)
Three areas where you MUST function testing everything
- Power
- Comms
- Controls
Skipping the function testing can lead to problems that will result in a de-scope and/or force you down a troubleshooting rabbit hole
Three areas where you MUST function testing everything
- Power
- Comms
- Controls
Skipping the function testing can lead to problems that will result in a de-scope and/or force you down a troubleshooting rabbit hole
Importance of knowing the target temperature to use for a process
Pretty much everything is temperature dependent -but the level of sensitivity varies. Some of those temperature sensitive items are obvious (like the fact that soldering at too high temp for too long can melt the pad right off the board) but not all of them are that obvious. Need to find out (or at least Google/GPT) what temperature you need for best results
Importance of “threshold objectives”
Threshold objectives tell you the minimum possible values or functionality for your system.
Sometimes it’s best to settle for the threshold capability or threshold value (as long as you can get validation that it is in fact good enough). This is because an attempt to optimize further can actually make it worse in ways you cannot foresee.
Also helps keep you motivated, especially if you add a deadline and call it a “gateway objective”
“No-Connect Enclosure checking”
ALWAYS test and make sure that your electrical enclosure does NOT have a voltage should not have conduction to any part of your circuit. Do this when you integrate your PCBs and before you demonstrate the system, especially if there’s lots of wires or a coating or insulator might have worn away and introduced an electrical contact between your PCB and your enclosure
DO NOT assume the enclosure is non-conductive. Test it and make sure
Bringing your project to a competition or large integration
There are some things you need to do if you are taking your project somewhere like competition/launch/etc:
-Drive, do not fly. Chances are you will need a large quantity of tools (maybe even a 3D printer) to fix some technical problem and shipping that toolbox is going to be prohibitively expensive
-Security is paramount. People get robbed at events like these. Double lock. Set up a tile. Maybe even alarms?
-Memorize the details of how your project will interface with the system with which it will be integrated (ie pin assignments)
-Bring as many of your lab tools as possible in an efficient storage container (efficient meaning that you can carry as many tools as possible without straining yourself like a rolling crate, with bungees to attach smaller boxes on top)
-Most relevant tools at the top. Containers should be organized so that the tools you are most likely to use are most accessible.
-Bring some tools that could be useful even if you do not routinely use them (ie scissors)
-Bring games, books, Nintendo, etc. because you may end up with lots and lots of downtime. Board games and card games are ideal since you can still engage with the team
-Snacks too, you don’t know when you’ll get a chance to eat
Avoiding the “28 V enclosure problem”
This can cause electric shocks and may damage the system with which you are integrating
-Either caused by faulty assembly (example: wire pinched in the enclosure, lose wiring left uncapped or untrimmed) or faulty design (example: structural material acting as a conductor, failure to ground mounting holes)
-Possible ways to avoid:
-Be aware of the conductive properties of structural material. Materials like metal or carbon fiber WILL conduct electricity unless coated with some sort of protective surface finish, and even that can run away over time
-Use QUALITY capton tape. If it’s low quality you are better off just using electrical tape (verify)
-Be aware that some electrical insulators have a high dielectric constant and could behave like a capacitor if you are discharging into your stucture
-Unless you KNOW you are using non-conductive material isolate your electrical enclosures so that they do not conduct to the rest of your structures. This involves separating them using insulators like vinyl or polypropelene
-Assemble with care, make sure your enclosures and cable management system do not risk pinching the wires
Avoiding the “28 V enclosure problem”
This can cause electric shocks and may damage the system with which you are integrating
-Either caused by faulty assembly (example: wire pinched in the enclosure, lose wiring left uncapped or untrimmed) or faulty design (example: structural material acting as a conductor, failure to ground mounting holes)
-Possible ways to avoid:
-Be aware of the conductive properties of structural material. Materials like metal or carbon fiber WILL conduct electricity unless coated with some sort of protective surface finish, and even that can run away over time
-Use QUALITY capton tape. If it’s low quality you are better off just using electrical tape (verify)
-Be aware that some electrical insulators have a high dielectric constant and could behave like a capacitor if you are discharging into your stucture
-Unless you KNOW you are using non-conductive material isolate your electrical enclosures so that they do not conduct to the rest of your structures. This involves separating them using insulators like vinyl or polypropelene
-Assemble with care, make sure your enclosures and cable management system do not risk pinching the wires
Importance of grabbing arms, clamps, etc.
For solo projects, get lots of clamps, holders, external arms, vices, etc. A big part of being on a project with a large group of people is asking someone to help you hold something or holding something for someone else. If you’re on a project without that support, you will need to plan for it.
Troubleshooting best practices
1) When troubleshooting a problem, always test between each change. This means that once the problem is solved, you know exactly what change you made to remedy the situation (that’s information that you can use to help you find the root cause)
2) Characterize the circuit by testing “point voltages” relative to ground
3) Never assume that a piece of hardware actually worked before integration unless you FULLY tested it
4) Always “follow the line of power”
Troubleshooting strategies, depending on the situation
Different troubleshooting strategies:
1. Car-in-driveway method: you have an important/high-reliability system but you cannot diagnose the problem. Start replacing/rebuilding parts of the affected system starting with the least “expensive” (the thing you can fix with the least sacrifice/lowest investment of resources)
2. Car-in-road method: you have an important/high-reliability system but you are in a situation where you cannot/should not give the resources necessary to fix/find the root cause. You just need a quick-fix to get you back in motion. You need to understand the problem, but you do not need to root cause. Once you understand the problme, apply the least invasive intervention necessary to get the system back to a working state. Make sure the system can still ‘fail safely’; ensure that any temporary fixes don’t compromise the safety of the system or its users). When the system is working again, watch it closely.
3. Car-in-your-shop method: you have an important/high-reliability system but your top priority is getting it root-caused as quickly as possible. You must get the dignosis done quickly and correctly (to minimize wasted time). This can be done by relying on a list of common problems for that system (or similar systems). Also consider whether the repair is even worth it (sometimes it’s better to reccomend the system be replaced)
Troubleshooting strategies, depending on the situation
Different troubleshooting strategies:
-Car-in-driveway method: you have an important/high-reliability system but you cannot diagnose the problem. Start replacing/rebuilding parts of the affected system starting with the least “expensive” (the thing you can fix with the least sacrifice/lowest investment of resources)
-Car-in-road method: you have an important/high-reliability system but you are in a situation where you cannot/should not give the resources necessary to fix/find the root cause. You just need a quick-fix to get you back in motion. Only go straight for the root-cause if you are forced to (ie a safety issue)
-Car-in-your-shop method: you have an important/high-reliability system but your top priority is getting it root-caused as quickly as possible.
Rate of occurrence of technical problems
For a sufficiently complex project, during the later stages of integration and test, you can expect to see roughly one technical problem per day.
Having/using a sensor/actuator/micro log
Keep a review log of every sensor/actuator you buy/test/implement so you know whether you should recommend/buy based on experience
The relationship between the cost of a material, and how careful you need to be with your equipment settings
In general, the more expensive the material is the more scrutiny you need to have when determining what temperature is and speeds you need for your equipment (the supplies to basically all settings and techniques)
FOD minimization
FOD = FOREIGN OBJECT DEBRIS
Messy processes are risky and time consuming. If you’re going to do something, make sure you are doing it the cleanest way you can (example: moving a fluid is easiest with a syringe. If it’s too thick to pour, do not try to scoop)
