Hello Everyone,

Summary:

Over the past six weeks, we’ve faced some unexpected challenges with a small batch of parts arriving out of spec, which affected our fulfillment rate. While we had successfully worked with these suppliers before, this batch presented some unanticipated issues both visually and functionally, but we’ve been fully committed to resolving them.

To ensure the highest quality standards moving forward, we've taken proactive steps, including multiple trips to China to meet with suppliers in person and better understand the root cause. As a result, we've already implemented the necessary corrections to improve the parts’ quality.

We’re excited to report that once the newly improved parts arrive at our shop, our team will be able to operate at full capacity. Once at full speed, we’ll immediately look into expanding our team by hiring more technicians to further accelerate fulfillment and improve delivery times.

We greatly appreciate your patience and understanding, and we are confident that these steps will help us provide everyone with an even better experience going forward.

NOTE: As of this afternoon, a shipment of powdercoated parts has just arrived and building has officially commenced!!!

Supplier Visits:

While in between batches 1 and 2, William and I decided to take a trip out to China at the end of October to invest some time in meeting our big suppliers, reviewing some design changes and issues we had with new samples. Since we’re not exactly a large customer for some of these companies, these visits are an important part in building a strong relationship and ensuring high quality and consistent parts.

On top of meeting our representatives and their engineers in person, we were not only able to have real time manufacturing reviews, but also able to see how parts like our boilers are now welded and hydrostatically pressure checked by the supplier as well as how our sheet metal parts are laser cut, get their brushed surface finish, bent, then finally inspected. All of this information is extremely valuable for understanding manufacturing constraints to improve the design in future iterations.

New Parts:

We received 45 sets of stainless steel parts by air from an existing supplier and well, whaddya know, manufacturing just doesn’t get any easier. While the machining quality had significantly improved from the original samples this fall, the brushed surface finish was still not up to par compared to our original Batch 1 pieces. The finish was far too coarse and left some unappealing gashes and perpendicular rippled effects on the parts. This isn’t something we can comfortably ship to our customers. We’ve spent the past 3 weeks working with the supplier and digging through our documentation to see what methods they used in the past vs what was recently done.

It’s been 2 years since we ordered our original parts and this company’s staff, location, and tools have all changed which has led to the inconsistency. Unfortunately, these discrepancies were not visible on the new samples that we had just approved. The supplier has been working on correcting this finish with new processes which we’ve just approved and are pleased with the results after seeing a much larger sample size.

Some lingering issues remain from Batch 1. One of our more annoying problems has been with our boiler pressure gauges. These are a slightly customized gauge that use lead free solder to avoid any leeching into the boiler. As many people know, we had issues with gauges fogging up in our first batch which affected ~10% of our stock. While each gauge goes through 2 rounds of QC testing on the machine before shipping, sometimes the internal fogging would present itself after being shipped. While it’s not difficult to swap this part out, it’s still very inconvenient as there shouldn’t be any internal leaks after they’re produced.

The supplier was informed and we sent back 20 bad gauges for them to disassemble, diagnose, and resolve on the next order. Unfortunately, after receiving the new order of gauges 2 weeks ago, even MORE gauges are fogging up.

The second lingering issue has been inconsistencies in the coiling of our heating elements. This can lead to difficulties with installing in our tightly packed boiler as well as struggles with getting them to appropriately seal. I’ve designed a very simple 3D printable go/ no-go gauge to simulate the bottom half of the boiler with our fluid level and thermocouple probes sticking out. The goal here is for the supplier to check each heating element after coiling to ensure fitment without issues before shipping to us. This will allow them to make any necessary adjustments on their end and save us the headaches during install.

China Part 2:

I never thought to visit these suppliers on the original trip last month but due to these recent issues and our recurring production schedules, I decided to make a quick last minute trip back out to China. I’ve just finished working with our machined plate supplier to ensure we get the appropriate finish and making sure their techs are correctly trained on the new procedures. This shop floor time allowed me to really understand where some of our issues were coming from, approve current procedures in real time, review powdercoating, inspection, and packaging all in real time. I’ve got several action items for things to adjust on the drawings and procedural documentation before we order more parts in a week but I’m confident that this time investment will pay off moving forward with more consistent results.

I also brought some of the new bad gauges to disassemble and diagnose on site with our gauge supplier. This gave me important insight into their assembly and testing process. Overall the process should be quite robust but I think there’s a few small changes we can make to reduce our issues. A potential cause for the fogging may be residual moisture inside the cover from assembly which collects on the upper lens when heated. Another possibility is an improperly mixed adhesive having a reaction to the high temperatures of the brass base of the gauge that it’s bonded to. We’re currently testing several of these possibilities with the supplier to narrow down and mitigate this issue in the future.

With new heating elements being coiled I was able to go on site to learn about the process and ensure that they understand our concerns and struggles with the gauge inconsistencies. Since the final bends and element tolerances are all adjusted by hand after a machine does the coiling, the new 3D printed fixtures allow each element to go through a test fit procedure where I was able to confirm perfect fitment which will significantly reduce boiler assembly time.

While it’s not the most convenient to fly all the way back out to China, I’m always excited to learn about how more of our parts are produced as well as guarantee a timely resolution to our lingering issues. These unexpected issues do delay us a bit, but I think that the additional up front time investment will pay off in the long run and not delay our overall batch 3 delivery times which have a 1 month manufacturing buffer for Chinese New Year.

I had the opportunity to meet up with an acquaintance from College who also does product design and owns a machine shop in China. Since he too has to outsource some of his own parts to other Chinese manufacturers, he assured me that this is all a part of the process, especially when going to pre-production to 500+ production scale. Essentially, I am doing the role of a Supplier Quality Engineer (SQE) which is not something I was ever familiar with in my prototyping background. While this role could be outsourced, I find that I do enjoy the on site problem solving and building interpersonal connections with our key suppliers.

Shop Progress:

We spent 1 week purchasing tools and testing methods to recreate the original surface finish ourselves in the shop but this proved to be much more difficult than expected. Now that I’ve understood the new process on site, I’ll attempt to find a similar machine in the US but if this doesn’t pan out, we’ll send the parts back to be fixed by the supplier with new Stainless plates being sent out by air this week.

While it’s unfortunate that we weren’t able to use these stainless pieces immediately, we’ve still been able to build up over 100 sets of sub assemblies in preparation for the powder coated parts to arrive this week then corrected stainless parts soon after.

New Piston Linkages are in and shipping immediately. These will officially roll out with serial no 600. The only significant advantage of this part is higher yield in the direct configuration making the yield closer to 50g.

Our new piston gauge brackets have resolved the original alignment issue and will make installation much easier than before.

Starting mid January, we plan to expand into the shop adjacent to ours. Our goal would be to open up the wall between the 2 units allowing us to move our mini “machine shop” and additional storage to the other side while making more space for assembly and shipping. This will set us up well for continued growth in the new year as we continue to ramp up assembly and fulfill orders.

Software

The latest firmware version at the time of this blog is 1.5.1 which includes a minor bug fix to revert the temp tolerance for entering sleep mode to be what it was in an earlier release. We had made this too tight and as a result some machines did not enter sleep mode after ~10 minutes of being at the set temp. Please update to this latest version when you can.

We’re still working through Acaia scale integration issues and have been in contact with Acaia representatives for help. Sorry for the ongoing inconveniences!

Steve, our part time app developer, has been working hard on a https://visualizer.coffee/ integration which we’re excited to announce has been released in the app! Go to settings tab and click the gear icon on the top right and select visualizer.coffee and follow the prompts to get connected. Then you’ll have a new “share to visualizer” button for each brew and after it's been uploaded you can click “view on visualizer” and it will open up the brew in your browser. We’re also working on our own platform that allows for user accounts and backing up/sharing shots.

Additionally we wanted to remind everyone that the setpoint offset is set by our technicians during qc with a Scace to ensure accurate and consistent brew temps based on the temp set by the dial and it is specific to your machine. Do not change the setpoint offset yourself because you may affect accuracy and consistency of the temperature stability of the Argos. We’ve made an update to the app to make it harder to accidentally change as well as including a notice.

We really appreciate everyone’s support as you all go on this journey with us while we create these awesome little machines!

Sincerely,

Ross & William