Betta Wire, Powercore Bug Fixes, and More.

Intro

Hi everyone, Cooper here! It's been a while since we've had official communication. This blog is how I think the gap can be bridged. A lot of you have been in our Discord chat on a day-to-day basis, and probably have a good idea of where things currently are. 

However, there are plenty of you who are not in our Discord. That's totally fine and is something we need to recognize and respect. I'll be making blog posts at least monthly from now on in order to keep everyone up to date on our developments. With that said, let's get into it!

Betta Wire

 The Betta Wire has undergone a significant amount of development rethinking when compared to the original version. To be completely transparent, the Betta Wire that was featured at RMRRF 2024 was brought from concept to reality in just under 6 weeks.

Problems

While it was entirely possible for us (really it was mostly John) to make this functional in 6 weeks, the machine had several glaring issues:

• Racking and slip-stick on XY motion
• Lack of independent control for each XY gantry
• Lack of closed-loop speed control based on spark behavior
• Difficult wire threading procedure
• Lack of hole popping tool
• Too many 3D prints in frame, leading to mechanical issues
• Difficult servicing and dielectric management
• Total lack of EMI mitigation
• Various QoL issues

Solutions

In order to fix these issues, we went back to square 1 with the Betta Wire. We wanted to create something that was capable of cutting thicker material than the Wire Tool Kit with better performance and user experience.

This means a significant redesign to an all-metal frame, consisting of 10mm precision steel rods and post-machined extrusions. For our latest kinematics test rig, we've also used a steel "halo" around the top of the machine for mounting extrusions and motors. We wanted an extremely solid base to work from while also not breaking the bank.

The next significant change, which may upset some people who have already purchased aquariums, is the switch to a customized dielectric vat. The current state of using an off-the-shelf aquarium causes issues with compatibility and servicing. Basically, how can you actually get to your work or machine parts without drying everything first?

The new design will have a standalone motion system that can be simply lifted out of the vat. We hope to have a drain and stopcock integrated into the vat for ease of cleaning. 

The next change we're looking forward to is an update to the toolheads. If you've been reading along on our Discord, you can see some of the ideas we've been developing on the new belt drive. We expect the belt drive will allow a semi-automatic wire threading experience. We will also be trying to modify this system to add a hole-popper to the toolhead. This should significantly improve the user experience and utility of the Betta Wire. This will also allow us to have far more control over the wire drawing process with parameters such as speed, tension, and direction.

The last update we're working on is some form of EMI mitigating enclosure. By customizing our vat, we're going to have a much easier time with this. We don't have any pictures to share right now on the enclosure, but we will do so as soon as we have something.

Powercore Bug Fixes

What is going on?

It's no secret that the Powercore has been taking longer than expected for Batch 2. This is due to a couple things. Shortly after shipping Batch 1, we received sparse reports of spontaneous unit failure. This is after all units passed active EDM cutting QC. 

Obviously, we cannot ship units that may or may not fail for seemingly no reason. We have recovered a few of the units that failed and have inspected them for the cause of the error. It turns out there was a switching problem on the board that caused the L3 inductor to "pop" and remove itself from the board. These units were not-recoverable.

How are we going to fix it?

Looking into fixing this issue, our priority was not shipping as fast as possible or getting back to making money in the shortest period of time, but to ensure that future unit reliability would be second to none. With this path, we had two options:

1. Modify components and circuits on the V2 as well as rewrite the firmware
2. Do something a bit more drastic

If you know us, you'll know that we picked option 2! Given that we expected V2.0 fixes to take at least as long as making a more significant redesign, we decided to make a more significant number of changes to the Powercore. These changes will improve safety, reliability, pulse rate, and efficiency. 

It may sound strange that it would take just as long to make significant changes as it would to fix what we already had, but there are reasons for this. For the Powercore V2 we worked with an out-of-house electrical engineer. Their work is great, but this person was greatly overworked by their day job and our project on top of it. There is no way we could expect them to spend the entirety of their time on this fix.

Their design and firmware, while generally known to us, would take time to get up to the point of being able to confidently make changes to. Each failure in doing so would generally eat a week at least in the form of a dead board. While we appreciate their work and in no way want to disparage the great EE we worked with, we needed something that we made and knew in house.

Progress on the updated Powercore is going well. We have already been receiving small test boards that we have built in a modular fashion. These will allow us to more quickly iterate upon our new design and to more easily bring the Powercore to you and to Opensource as a whole. Here's an example of one of our test boards:

Schedule

Full disclosure, we expect to begin shipping the new Powercores in Q1 of 2025. We are not getting these out the door any time in 2024. If you have a pre-order (thank you!) and want to refund your order, just email us at contact@rackrobo.io and I personally will help you. We would rather give you your money back than continue to frustrate you.

If you are one of those who unfortunately had their V2 unit fail, you will be receiving one of these new units when they pass FCC requirements. Your names have been recorded and you will be the first to get these new Powercores.

So, with all these updates, is this still a V2? Is it a V2.1, or something else like a V3? We'll have to see. I think you all will be quite happy with what we end up sending out.

Closing Thoughts

New development is shaping up well. We're beginning to zero in on what it means to have proper development strategies and are learning more every day about how to bring you useful products that are actually affordable while still not being locked down.

Kinematics testing on the new Betta Wire is extremely promising. Assembly was nice and easy too, so that's good.

Overall, we greatly appreciate our community. The enthusiasm and understanding that we're trying to do something very difficult is wonderful. Who would have thought there would be bumps along the road when trying to lower the price of an industrial process by 100x-1000x?

We do these things not because they are easy, but because we thought they would be easy.

-Cooper