SPRING 2025
WEEK 5 UPDATE – Project Update 4
Week 5 Reflection:
Last week I had originally selected the BQ24072TRGTR IC for our battery management system. After a closer look at the data sheet, I found the BQ24075RGTR IC would be a better fit for our system. The 24075 has a regulated 5.5 volt output, meanwhile the 24072 only has a 3.72 volt output. With the increase to 5 volts, it became easier for us to provide two power options to users. One option would be to use a rechargeable battery and this IC and the other would be to use a 9 volt battery and a TI LM317 voltage regulator. This selection would be accomplished two ways. One would be through the use of solder jumpers during assembly, and the other would involve simply not populating the components for the option you chose NOT to go with. For example, if you chose the 9 volt and LM317 route, you wouldn’t need to populate the BQ24075RGTR or any of its corresponding components.
Unfortunately didn’t get a PCB design done by CDR. However, we now have a GitHub repository for our KiCAD files, allowing each of us to update separate pages of a schematic and have it auto-update the primary schematic. This should hopefully allow us to pretty quickly develop a PCB design by the end of the week and have it sent off for production during spring break!
To-Do List for Week 6:
- Finalize schematic
- Are component values correct?
- Do all components have a PCB footprint assigned?
- Finalize Bill of Materials
- Are there any parts unaccounted for?
- Display?
- Banana Plugs for leads?
- Power switch?
- Are there any parts unaccounted for?
- Complete PCB design
- Have Carson Belyk review our design before sending it off for production.
- Start writing documentation
- Do we want to document the research process or just the final components needed in the design?
- Maybe start writing up how a multimeter actually functions?
Concerns for Week 6:
My overwhelming concern at the moment is being able to complete a PCB design by the end of week 7 in time for spring break. We are already half way through this semester and the deadline for this project is not far away. That being said, if we can complete a design by the end of this week, I can work with Carson to review it in the beginning of spring break with the hope of sending them off for production by the end of spring break.
Posted Date: 02/16/2025
WEEK 4 UPDATE – Project Update 3

Week 4 Reflection:
I made some pretty good progress this week on the battery management side of things. After selecting the Texas Instruments BQ24072TRGTR IC, I started working on creating a schematic and hoping to move onto creating a PCB in the coming week. I also want to work on integrating a way to switch off the battery management system and instead use just a 9 volt battery or two AA batteries instead. The team has also set up some times for the coming week to put a Critical Design Review presentation together. My original hope was to have a PCB design made for CDR, which is still possible, but I’m not completely sure that will happen in time. Instead, I see schematics being complete in time and saving the PCB design for after CDR.
To-Do List for Week 5:
- Look into BMS ICs
- What if we also wanted the ability to use a 9 volt battery, can it bypass the BMS?
- Make a block diagram for our project
- What systems need to be on the block diagram?
- How do I want to organize the diagram? Is there a flow to the system?
- Start writing documentation
- Do we want to document the research process or just the final components needed in the design?
- Maybe start writing up how a multimeter actually functions?
Concerns for Week 5:
My biggest concern is still ensuring that we stay on schedule and don’t fall behind. I don’t think we’d be able to reach my initial goal of having a PCB design created in time for CDR, but hopefully we can at least have a schematic ready to go. My other concern on cost of our meter is slowly becoming less of a concern. While there are still some ICs and other components that could be a little pricy, I think we will still be under the $30-50 range that I am aiming for if someone were to recreate just one of these. The goal is still to keep the cost even lower, preferably under $20 if several were being made at once.
Posted Date: 02/03/2025
WEEK 3 UPDATE – Project Update 2
Week 3 Reflection:
We made some great progress this week! After some discussion, we decided to switch from the ICL7107 IC to the ATMEGA328P IC for the primary brains behind our DMM. The ICL7107 lacked much of the customizability that the 328P supports, was $18-20 more expensive, and there wasn’t a lot of documentation on it. Meanwhile, the ATMEGA328P was super easy to find, being the brains of many Arduinos currently on the market, was much cheaper at just $1.50-2.50 each, has tons of customizability as well as documentation, making this choice a no-brainer. We also have a rough plan for how we plan to split up the project. Dividing up the project into sections like creating the case and mode selection switch, creating the battery management system, and creating the primary circuitry responsible for measuring and displaying voltage, current, and resistance. To wrap up the week, we also settled on what times we were available for CDR in a few weeks.
To-Do List for Week 4:
- Look into BMS ICs
- Do we want to use an 18650 cell?
- How about a smaller one cell LiPo?
- What if we also wanted the ability to use a 9 volt battery, can it bypass the BMS?
- Make a block diagram for our project
- What systems need to be on the block diagram?
- How do I want to organize the diagram? Is there a flow to the system?
- Start writing documentation
- Do we want to document the research process or just the final components needed in the design?
- Maybe start writing up how a multimeter actually functions?
Concerns for Week 4:
My biggest concern is ensuring that we stay on schedule and don’t fall behind. My goal is for us to have a PCB design created in time for CDR, allowing us to receive some feedback on our design and make the necessary changes, prior to ordering boards and components. My second big concern is the overall cost of our meter. Some of the ICs and other components that we are finding can get a little pricy, potentially pushing the cost of recreating this project somewhat higher than we are hoping for.
Posted Date: 01/25/2025
WEEK 2 UPDATE – Project Update 1

To-Do List for Week 3:
- Decide on IC for DMM
- Do we want to stick with the ICL7107?
- What about using the ATMEGA328P from an Arduino, similar to existing Open Source Designs
- Divide up the project into individual or groups assignments
- Maybe have one or two people working on the case?
- Two to three people on integrating the IC with the ADC and other necessary components?
- Two to three people working on the battery system?
- Do we just want to stick with a 9 volt or two AAA batteries?
- Do we want to create a Battery Management System for a rechargeable battery?
- Sign up for CDR
- When is everyone available based on the available time slots?
Concerns for Week 3:
I have two primary concerns for the upcoming week. The first big concern is how we can divvy up the project into individual or small group sections. I have a few ideas based on how we wrote the project spec but it’ll mostly be up to the team. My other big concern for this upcoming week is being able to find a time that we are all free for CDR based on the available CDR times. There are a few possibilities but again this will be up to the team and their availability.
Posted Date: 01/21/2025
WEEK 1 UPDATE
- Begin the Digital Multimeter Project
- Met with the DMM Team
- Completed the DMM project specification
- Began research into ICs and started planning out the DMM project
Posted Date: 01/13/2025
FALL 2024
WEEK 5 UPDATE – Project Update 3

Week 5 Reflection
This past week consisted primarily of house keeping items. I spent time with cleaning up my schematics so they follow proper technique. I made some modifications to the schematics based on some feedback from the team, such as condensing certain sections and combining things that were used by both the 5 volt and 3.3 volt schematics. In addition to that, I also made some updates to my bill of materials and further updates to the schematic. I have put together a pretty solid initial BOM based largely on the Texas Instruments evaluation board as well as updating all of the footprints on each of my components in my schematics, which was great preparation for board layout design this upcoming week.
To-Do List for Week 6
- OVERALL GOAL: Board layout design, and preparation for board manufacturing.
- Board Layout Design Sub Goal – Create a board layout in KiCAD that meets all necessary requirements per the manufacturers datasheet and fits within sizing constraints (once those are determined), to do so I must:
- Create and/or reference some type of list of requirements based off the manufacturers datasheet.
- Reference the provided recommended typical application board layout for an initial design.
- Alter BOM and part selection if any components were switched to a different footprint size, such as switching from 0805 to 0603.
- Prepare For Board Manufacturing Sub Goal – Ensure everything is prepared for boards to be ordered and manufactured, to do so I must:
- Double check all my work with my team members to ensure that all potential issues or mistakes are caught and addressed before board manufacturing.
- Make any necessary or recommended design changes.
- Return to my team members to double check my board again.
- Repeat process until every team member is satisfied with the state of the board and ready for it to be put into production.
Week 6 Concerns:
My week 6 concerns assign similarly to last weeks. I am concerned about how to properly create multi-layered boards with large planes for things like the main grounding plane. As well as, the second being the ability to fit both buck converter designs into the rough dimensions of a note card, allowing the back of the board to contain our modular cable connectors.
Posted Date: 09/29/2024
WEEK 4 UPDATE – Project Update 2
Week 4 Reflection
Last week’s to-do list was ambitious. I set out to do a lot in a week that I quickly discovered became quite busy. Despite this, I still managed to complete half of my very lengthy to-do list.
I started with altering the original 2.5 volt schematic to step 12 volts down to 5 volts and 3.3 volts via the recommendations set my the Texas Instruments Excel calculator. Utilizing this calculator not only gave me the necessary values for resistor 1 and 2, but also provided me with the values for every component in each schematic. Meanwhile, the calculator also checked that the selected and calculated component values would perform correctly to step down 12 volts to 5 volts and 3.3 volts while also maintaining a 15 amp current output. This proved to be harder than initially expected as I ran into several issues with the selected inductor limiting our potential current output. Through discussions with my team, we discovered that it was calculating this expected outcome with a huge input voltage range, instead of using the already refined input voltage range from our other boards. After altering this voltage input range, our selected inductors would adequately perform under our set parameters. With this change in mind, I quickly incorporated the few component value changes and finished the 5 volt and 3.3 volt schematics.
After completion of the schematics, I moved onto component selection. Through the use of the Texas Instruments calculator, most of the components and their values were already selected for me, but there were still a few components that I had to settle on. The values of the inductors were already selected, but I had to find physical components that could actually provide the theoretical numbers. I chose to selected not only the recommended 800nH inductor from Coilcraft for the 5 volt circuit, which also happened to be used in the evaluation board from Texas Instruments, but I also chose a 600nH inductor from Coilcraft for the 3.3 volt circuit as well. As for the rest of the components, I chose a standard 0805 footprint for board layout design and will alter the components footprint and bill of materials once the board layout has been complete, just in case a few components are switched to a 0603 sizing to save space.
To-Do List for Week 5
- OVERALL GOAL: Finish both 5 volt and 3.3 volt schematics, part selection, board layout design, and preparation for board manufacturing.
- Board Layout Design Sub Goal – Create a board layout in KiCAD that meets all necessary requirements per the manufacturers datasheet and fits within sizing constraints (once those are determined), to do so I must:
- Create and/or reference some type of list of requirements based off the manufacturers datasheet.
- Reference the provided recommended typical application board layout for an initial design.
- Alter BOM and part selection if any components were switched to a different footprint size, such as switching from 0805 to 0603.
- Prepare For Board Manufacturing Sub Goal – Ensure everything is prepared for boards to be ordered and manufactured, to do so I must:
- Double check all my work with my team members to ensure that all potential issues or mistakes are caught and addressed before board manufacturing.
- Make any necessary or recommended design changes.
- Return to my team members to double check my board again.
- Repeat process until every team member is satisfied with the state of the board and ready for it to be put into production.
Week 5 Concerns:
I only have two big concerns for week 5. The first is learning how to properly create multi-layered boards with large planes for things like the main grounding plane. Then the second is being able to fit both buck converter designs into a small area on the back of the board with our modular cable connectors.
Posted Date: 09/22/2024
WEEK 3 UPDATE – Project Update 1

Image Above: Rough sketch of expected final project design as well as general schematic/layout for buck converter.
To-Do List for Week 4
- OVERALL GOAL: Finish both 5 volt and 3.3 volt schematics, part selection, board layout design, and preparation for board manufacturing.
- Schematic Sub Goal – Alter my current schematic to step down from 12 volts to 5 volts instead of the step down to 2.5 volts, to do so I must:
- Alter resistor 1 to set the buck converter IC to convert 12 volts down to 5 volts instead of the initial design that follows the datasheets typical application of 2.5 volts.
- Utilize Texas Instruments provided Excel calculator to assign the typical input voltage of 12 volts and the wanted output of 5 volts and 15 amps.
- Incorporate the calculators recommendations into my initial schematic.
- Part Selection Sub Goal – Select parts that align with my newly updated schematic, to do so I must:
- Start with the typical application parts list and alter parts based on my updated schematics.
- Ensure my part selections are no smaller than the 0603 footprint to allow for hand soldering to be possible without major difficulty.
- Board Layout Design Sub Goal – Create a board layout in KiCAD that meets all necessary requirements per the manufacturers datasheet, to do so I must:
- Create and/or reference some type of list of requirements based off the manufacturers datasheet.
- Reference the provided recommended typical application board layout for an initial design.
- Alter the initial typical application board layout to fit the 0603 or larger footprints and ensure that there is adequate room to hand solder.
- Prepare For Board Manufacturing Sub Goal – Ensure everything is prepared for boards to be ordered and manufactured, to do so I must:
- Double check all my work with my team members to ensure that all potential issues or mistakes are caught and addressed before board manufacturing.
- Make any necessary or recommended design changes.
- Return to my team members to double check my board again.
- Repeat process until every team member is satisfied with the state of the board and ready for it to be put into production.
Posted Date: 09/12/2024
WEEK 2 UPDATE
- Selected the TPS548A28 as the buck converter IC of choice for our application.
- Created the initial buck converter schematic in KiCAD based on Texas Instrument’s datasheet for the TPS548A28.
Posted Date: 09/12/2024
WEEK 1 UPDATE
- Started my first week in OSHE!
- Started reviewing the plans for the buck converter selection
Posted Date: 08/30/2024