Flux's AI agent is now up to 10x faster and self-corrects in real time, delivering cleaner schematics with less waiting and fewer wasted credits.
Copilot new access to Flux’s live pricing and availability tools so that it can do the supply chain and cost analysis for you. Read on to learn about how we’re leveraging AI to give you the power of an entire supply-chain team right at your fingertips.
At Flux, we want to ensure that your designs are functional and manufacturable—and that means making it easier to navigate the supply chain. That’s why we’re excited to announce a new end-of-life (EoL), not recommended for new designs (NRND) warning feature that automatically flags if components are at or near obsolescence. We’ve also given Copilot new access to Flux’s live pricing and availability tools so that it can do the supply chain and cost analyses for you.
When you’re creating a design that will eventually be manufactured, it pays to start off on the right foot. From the very get-go, it’s imperative that you design with components that are sourceable and will remain in production for the distant future.
With these warnings presented before you select a component, you’ll catch issues earlier in the process and avoid future supply-chain challenges before they even happen. And it doesn’t take a team to do it. Let Copilot and the Flux schematic editor do the heavy lifting for you—all under the roof of one centralized design tool.
Now you’ve got a design consisting of sourceable components with long-term viability. When looking toward manufacturing, you also want to cost-optimize your design to make it as affordable as possible.
With Copilot’s newfound access to Flux’s real-time availability and pricing tools, cost optimization becomes a breeze. Simply ask Copilot,
“@copilot, use real-time pricing data to help me find a more affordable alternate for U1”.
Copilot will then scour the internet and the Flux library, searching for the best component replacements that save you money without sacrificing your original design’s functionality. You no longer have to do all the manual work yourself. No more need to open a dozen tabs, clicking between datasheets and suppliers to find the best alternates. Instead, Copilot automates the whole process for you so that you can feel confident you have the most cost-efficient design possible.
Once your product is manufactured and on the market, the Engineering and Operations teams are still responsible for sustaining the project. That means monitoring changes in component availability and pricing to keep your production line up and running.
With Flux’s new features, this is all done for you.
If a component in an existing design is facing obsolescence, the warning indicators will appear directly on the schematic editor for everyone on your team to clearly see. Or, maybe you want to do an occasional check of your design just to be safe. In that case, you can prompt Copilot to do a BoM analysis for you by asking,
“@copilot, please analyze my bill-of-materials to check if any of my components have reach end-of-life, or if they’re at risk to do so”.
Like an entire supply-chain team at your disposal, Copilot will thoroughly compare your design against data from the world’s leading suppliers to ensure that all of your components are viable options for production. If not, Copilot will alert you accordingly, and even help you find alternates with better long-term prospects.
We want to make hardware less hard, and that means we need to give you, the designers, less things to worry about. With new End of Life Warning Indicators and Copilot’s newfound ability to access and analyze real-time Availability & Pricing information, we’re giving designers the power of an entire supply-chain at their fingertips. Want to experience how Copilot can ensure your design is ready for manufacturing? Start a design on Flux today!
We wanted to bring the power of Python directly into our users' hands so that you can augment your workflows with custom scripts that automate your design and analysis tasks. That’s why, today we’re excited to be introducing Copilot’s Code Interpreter.
Flux Copilot is already the most powerful chat-based AI design assistant for PCB design, but what if we told you it just got even smarter?
Code Interpreter is the newest tool in Copilot’s arsenal. With access to a built-in Python code interpreter, Flux Copilot can now generate and run Python scripts directly in conversation with you. That means that you can automate workflows, analyze data on the fly, and create custom visualizations without leaving the chat interface.
The result? Your team can solve problems more effectively, work faster, and reduce the risk of errors.
To work with Code Interpreter, simply ask Copilot to perform an analytical task or solve a problem and, in some instances, specify that you’d like it to use Python in the process.
First, Copilot will meticulously describe the steps it takes and its line of reasoning in solving the problem. Then, it will generate a comprehensive Python script for you accordingly, including everything from library imports to function definitions. Finally, Copilot will use its new Code Interpreter powers to execute the script, exporting the results in whatever format you specify.
With Code Interpreter, Copilot can provide tables, plots, and charts that help you better organize, visualize, and understand your project.
Need some examples of the ways that Code Interpreter is a game-changer for Copilot? Check out some of the most compelling use cases we’ve evaluated so far.
EEs often have to refer to datasheets in the design process to figure out device performance specifications, tolerances and ratings.
For example, when choosing current-limiting resistors, like in LED circuits, it's important to design for a specific current flow and power consumption and then size the resistors accordingly based on information in their datasheets. With Code Interpreter, Copilot can use Python to do this analysis for you and then compare your design to the expected results. For example, if a resistor is undersized for an expected power, Copilot can flag this and help you find a better component for your design.
Check out this example in action here.
Sometimes, EEs refer to datasheets to extract equations to guide their design efforts, like in the case of regulator designs.
The process of voltage regulator design requires designers to appropriately size the peripheral components for a given output voltage and current. These values are often based on equations given by the component manufacturer in the datasheet. Instead of manually calculating the needed component values, you can use Copilot’s Code Interpreter to do it for you. Looking at your programmable regulator IC, the design information in its datasheet, and the context of your circuit and project requirements, Copilot creates a Python script that calculates what passive components such as inductor, input and output capacitors and resistor values are ideal. You’ll even get multiple options to pick from, so you maintain freedom in your design choices.
Check out this example in action here.
Relatively complex mathematical equations govern the behavior of analog filters. Instead of calculating poles and zeros to graph a transfer function manually, ask Copilot to do it for you. Copilot can use Code Interpreter to analyze your circuit, calculate the frequency response, and plot your transfer function. You then have access to a detailed plot to review and Copilot-created design feedback and recommendations based on the results.
Check out this example in action here.
Determining your system’s overall power consumption can be tedious and arduous. Done manually, the process entails calculating each component’s power consumption and then adding these up individually to estimate the total system value.
With Code Interpreter, Copilot does it all for you. Copilot can analyze your circuit to understand each component’s power consumption, reading through datasheets where necessary to get reliable figures for active components. Then, it can determine your system’s total power consumption and create charts to help you visualize the major contributors to your system’s power draw.
Check out this example in action here.
Another tool in Copilot’s belt, Code Interpreter, makes Copilot more powerful than ever. Now, your team can automate otherwise manual workflows with the power of Python, letting you work more effectively and quickly than in the past. Want to see the power of Code Interpreter in person? Start a new project with Flux today!
Increase your PCB design productivity with Flux Copilot's new AI shortcuts. Automate tasks, optimize designs, and reduce errors.
With Copilot Shortcuts, there is no more crafting the perfect prompt or reading through documentation to figure out what Copilot’s strengths and weaknesses are. To access Copilot Shortcuts, simply right click a component or project in Flux and choose the prompt you want to use. In the past year, users have had access to a handful of Copilot Shortcuts, including;
Today we’re excited to announce the addition of 8 new Copilot Shortcuts so that your team can get the most out of Copilot with the least prompting effort possible.
Now, after much feedback from users and rigorous testing on our end, we’ve created a set of new prompts that are commonly useful to engineers and are written in a way to guarantee the best results possible.
The new prompts we’re adding include
As you work with Flux Copilot, please let us know which use cases and workflows are most valuable by joining our Slack Community. Together, we can revolutionize PCB design by automating tedious and error prone workflows so you can focus on what you enjoy doing most!
Streamline component research with Flux Copilot. Copilot links to components for quick part research, offering multiple options tailored to your needs, and find part alternatives effortlessly without switching between tabs and platforms.
At Flux, we’re bridging this gap and bringing all the hardware design process under one roof. Today, we’re giving Copilot the ability to link to parts when it makes suggestions, so that your team can easily research components without having to leave the tool.
Just click the link and instantly access everything you need to know: specs, datasheets, real-time pricing, and even usable models—all within Flux.
When your team needs to find a part for your design, Copilot is your best friend. Simply provide it with your needs, criteria, and project-specific restrictions. Then watch as it offers you multiple component options, each meeting your team’s unique needs.
Now that Copilot can link to components within its responses, your team has a new advantage. You can go from part option to part research immediately. Simply click the link on the Copilot-suggested part, and you’ll be brought to that part’s page within Flux. There, you’ll find everything you ever wanted to know about the part, including performance specifications, datasheets, real-time sourcing and pricing, and even usable schematic symbols, footprints, and 3D models.
No more switching tabs and software. The entire component selection process lives in one tool, meaning your team can work quicker and more efficiently.
Say your team already selected a part, but you’re in need of alternative options. Maybe you need to cost-optimize your design with more affordable components. Or, maybe your part is out of stock and lead-times are months long.
Copilot shines here by leveraging its knowledge and Flux’s vast database of components to find your team the best options for your specific project. Copilot can understand your project architecture and the interplay between components within your schematic. With knowledge of that context, Copilot can find the best part alternatives that meet your needs without risking your designs functionality or performance
And, with parts linked directly in Copilot responses, you can go from searching for alternative options to designing with those alternatives in minutes. No searching DigiKey. No creating footprints or models. Follow Copilot’s link, read up on your new part, and drag and drop it directly into your design.
Imagine a world where hardware design is streamlined, intuitive, and collaborative. That's the future we're building at Flux.
Hardware design doesn’t need to be scattered, confusing, and slow. Want to learn more about how your team can 10x efficiency? Sign up for Flux today.
It can be daunting when you and your team are looking to embrace cloud software for the first time. This blog discusses the major principles for your team to better understand the ins and outs of choosing the right cloud-based software.
But, it can be daunting when you and your team are looking to embrace cloud software for the first time. There are seemingly dozens of options out there, each with its own marketing jargon convincing you that their solution is the best. Let’s cut through the fog to help your team better understand the ins and outs of choosing the right cloud-based software. The following are the 8 Principles of Choosing Cloud-Based Software for PCB Design.
First things first, you need to be clear about what is driving your team to adopt cloud-based software.
Why do you need new software now? What’s pushing the urgency? Maybe your CFO demands a 12% margin increase by year-end. Or perhaps you need to ship a backlog of products by Q3. Define this compelling event, document past attempts to solve the problem, and honestly assess what worked, what didn’t, and why.
Pro-tip: Try to solve your problem internally first—sometimes better processes can work wonders without extra costs.
Your team isn’t buying software; you’re buying a solution to a problem. Therefore, any software purchase must be justified with a clear return on investment (ROI).
By understanding the magnitude of your problem in terms of time and money, your team can better determine which cloud solutions make the most sense for you. This process means gathering all the relevant data—FTE costs, current software expenses, opportunity costs—and using it to create a metric of your needs. A good software provider will help you document this and build a solid business case.
If the ROI isn't clear, you might not be ready to buy.
Technical teams often wish they could advance projects at their own pace, but in practice, everything needs to go through corporate approval first. That means that when your team wants to switch to cloud-based software, the move needs executive backing.
Whether you start from the bottom up or the top down, ensuring alignment with strategic goals is necessary. It’s often good practice to leverage the DACI framework (Driver, Approver, Contributor, Informed) to streamline roles and responsibilities and get all major stakeholders on the same page. By clearly defining who is responsible for each aspect of the project, who has the final say, who provides input, and who needs to be kept informed, the DACI framework helps prevent misunderstandings and ensures efficient decision-making.
As a technical team, it is also advisable to keep the executive sponsor involved at key points to maintain momentum and ensure the project’s priority.
We’ve all experienced the pain and red tape associated with requests for proposals (RFPs). What most people won’t admit, however, is that RFPs are often a waste of time.
Your decision-making and needs are complex and nuanced, but RFPs try to reduce complex needs to binary yes/no answers. The result? Mismatches and frustration.
Instead, after following the initial steps in this guide, your list of potential providers should be narrow enough to handle individually and more intimately. It’s far better to talk to 1-3 companies and focus on in-depth discussions rather than going through the cursory checkbox comparisons.
When you’re choosing a new software, you want to know that it has a proven track record of success with other customers. Most software providers will assuage these fears by inundating you with case studies, trials, and product demos. But is all of this collateral really relevant to your team’s needs?
When going through the software courting process, you should ensure that the case studies you request are relevant. For demos, focus on your top concerns rather than generic capabilities. Trials should be approached with clear goals, success criteria, and structured test plans.
Remember, the UI/UX shouldn’t overshadow actual value.
Too often, we get caught up in arguing dollars and cents when the ROI is so much more valuable than the upfront cost. Our advice: don’t haggle for the sake of it.
If the solution offers a clear ROI, like solving a $10M problem for $500K, embrace it. At the end of the day, the upfront cost of your software solution is just a drop in the bucket compared to the return your company will actualize from it.
Instead of haggling, it’s ideal to collaborate with the provider to articulate your problem's size and expect a reasonable investment proposal. High-value solutions shouldn’t feel like a negotiation battle but a mutual agreement. A good transaction will be one in which both sides walk away feeling like winners.
Flux, like other cloud-based EDA tools, offers real-time collaboration, AI integration, seamless updates, and scalability. These are game-changers for moving hardware teams from the 90s into the modern age, matching the pace of their software counterparts.
These advantages collectively streamline the design process, enhance productivity, and enable hardware development to be innovative and agile.
It often feels like the hardware industry is stuck in the past, fearing new technologies like cloud and AI. Understand that these fears are rooted in the industry's slow evolution and long-established practices. Unlike the software industry, which has rapidly adopted and benefited from cloud computing, AI, and other innovations, the hardware sector has traditionally been more cautious, prioritizing stability and reliability over agility and innovation.
Flux and similar tools are designed to help you overcome these fears and leverage modern capabilities to stay competitive.
Adopting these technologies can transform your workflow from rigid and outdated to agile and innovative. By embracing cloud and AI, you can reduce time-to-market, improve product quality, and respond more swiftly to changing market demands. The shift may seem daunting, but the long-term benefits far outweigh the initial challenges.
Buying cloud-based PCB design software doesn't have to be daunting. By following these steps, you’ll make informed, confident decisions. Evaluate your needs honestly, quantify your problems, secure executive alignment, ditch the outdated RFP process, and focus on relevant case studies and structured trials. Embrace the cloud for its real-time collaboration, AI benefits, and scalability.
Want to learn more about how to select the cloud software for your team’s needs? Contact our sales team today.
Discover how Copilot transforms hardware design from concept to creation through an end-to-end example of designing a webcam, showcasing the power of AI hardware design at every step.
This project will be designing the camera circuit for an open-source laptop. We’ll be starting completely from scratch to design a functional webcam around OmniVision’s OV02740 HD image sensor, and we’ll be covering everything from the image sensor itself to the supporting power management and passive circuitry. At the end of this project, you’ll see exactly how AI can reinvigorate your PCB design process from start to finish.
The major AI-powered steps in this project will include
You can check out the projects at your own pace using the links below
The first challenge when starting a new project is taking your idea for a product and deciding on the design architecture. The architectural design phase is foundational in hardware development, sometimes determining the success or failure of a project. It's complex, requiring the balancing of numerous variables and aligning diverse stakeholders on a unified vision.
Creating architectures with Flux Copilot is easy and straightforward. You can simply have a conversation with Copilot about what you intend to build using as much information as you know. Copilot can use your requirements and constraints to explore many different architectural ideas and variations quickly.
In this project, we started by asking:
Copilot then provides us with multiple architectural options, each containing suggestions for circuit blocks, components, and their interconnections. Later, Copilot helps us create block diagrams to better visualize and intuitively understand our chosen architecture. Leveraging AI to rapidly generate and evaluate a wider range of options against your specific product requirements ensures a more effective selection process that leads to optimal outcomes.
To learn more about AI-powered architecture design, read our blog.
Once our architecture is determined, we need to choose the core components that will turn our idea into a real circuit. Fortunately, selecting components is one place where Copilot really shines.
Copilot is guided by your company’s guidelines, including regulatory requirements, pricing, power consumption, operating conditions, and more. With these parameters defined in a Template, Copilot finds the best components that fit your specific project requirements.
In this project, given our architecture, we ask Copilot:
Copilot then provides specific components that are interoperable and achieve the needs of our design in an organized, tabular manner.
Learn more about how AI powers component research.
Designing electronics is about more than just creating a system that works; it also requires creating a system that is sourceable, compliant, and robust against a volatile supply chain. Once we have decided upon our main components, we can use the power of AI to help find alternative component options in case we need backups for whatever reason.
For space savings in our webcam project, we ask Copilot to help us find leadless package alternatives for our components. Specifically, we can ask Copilot
Copilot then provides us with an alternative component that meets our architectural needs and even describes the differences between the original and alternative components.
However, we run into a problem here: the new component doesn’t have a part in the Flux library yet. Fortunately, we can use Copilot to automatically generate a part (i.e., schematic symbol, footprint, and 3D model). Our AI-powered workflow allows you to create parts by importing PDF datasheets directly. Copilot will automatically extract part information, generate components, and enable easy review, validation, and editing—all within the browser.
This AI-powered workflow for handling datasheets and part creation offers an entirely new way to approach the task, replacing the tedious and time-consuming manual part creation process.
Want to learn more about AI-powered part creation? Read our blog.
Thanks to Copilot, we have a schematic and component alternatives for our webcam project. But next, we want to try to make this design as affordable as possible. It’s time to cost-optimize our design.
First, we can save costs by eliminating any unnecessary components in our bill of materials (BoM).
BoM consolidation involves identifying outlier components that can be merged with existing values, reducing the number of unique components needed. For example, a circuit may require a unique resistance value, such as 31.23kΩ, but such a unique value is costly to source. Instead, Copilot can suggest implementing this resistance with two more standard and affordable solutions, such as a 30kΩ and 200Ω in series.
Then, Copilot can help us save costs by evaluating components for over-specification. This ensures that no component exceeds the necessary performance requirements, which can reduce costs. Learn more about AI-driven PCB Cost optimization.
For example, assessing whether a high-performance microcontroller is necessary or if a lower-cost alternative can meet the project’s needs without compromising performance. Copilot can investigate the components in your schematic against the operating conditions of your circuit — be it power, temperature, or frequency — to ensure that no component is unnecessarily over-specified.
Now that we have a highly optimized design, it is time for a design review. Follow this blog to learn more about AI-powered design reviews.
Reviews can be meticulous and tedious, demanding the near-impossible task of considering hundreds of variables and comparing your design against organization-specific standards and constraints. AI like Flux Copilot can automate these menial tasks so that engineers can save time and instead focus their efforts on more high-leverage tasks.
In this project, we ask Copilot first to review the voltage ratings of our passive components. Copilot can handle this task without further instruction, but in this case, we give Copilot detailed steps on how to complete the review so that we’re confident that the review is up to our standards. Copilot then compares every passive component’s ratings against the maximum applied voltage in our circuit and provides this data, as well as a pass/fail status, in an organized table for our review.
Then, we take things one step further by asking Copilot to conduct a precise design review focused on verifying the pin-out and configuration accuracy of the components in our schematic. Copilot gives us insight into its design review process by first explaining the procedure it will follow during its review. With our permission, Copilot then validates the pin-out of our components against the datasheet and provides the results in an organized table.
The final step in our design process is now the testing and debugging phase. Flux uses AI to help you test and debug your circuit designs, including everything from test plan creation to failure mode and effects analysis (FMEA).
In our webcam project, we first asked Copilot to help us develop an automated FMEA report that identifies critical failure modes, assesses their impact, and recommends mitigation actions based on severity, occurrence probability, and detectability. Copilot took to the challenge, generating a detailed tabular report of each component type and explaining the table. With an understanding of the detailed interplay between all the components in your design and expert-level knowledge of electronics fundamentals, Copilot creates a thorough and comprehensive FMEA report for us.
Next, we asked Copilot to help us create a detailed step-by-step plan table for this project to verify its functionality. In our case, Copilot responded with a test plan outlining steps for each major component and functional group. To learn more about AI-powered testing and debugging, read our blog.
With a comprehensive test plan, your team can ensure that any design errors get caught early in the process and well before you go to production. This means your team can spend less time correcting errors and less money on unnecessary design revisions. Ultimately, that translates to higher quality products and faster time to market.
AI has the power to completely transform the hardware and electronics design workflow from start to finish. Through this end-to-end project example, we hope to have demonstrated how AI can assist your team’s design process and how each unique AI-powered workflow converges to create an innovative product. To learn more about how your company can streamline processes, reduce costs, and save time with AI-powered hardware design, sign up for Flux today.
