Hello All,

As always, both good and bad news comes with the level of transparency that we try to provide. Let’s start off with the bad and work our way up from there.

I’ve Dropped the Ball with Reduced Shot Yield

I’m putting this first as not to hide it as it may be very important to some customers. This has been brought up and discussed on the EAF (Espresso Aficionados) Discord over a month ago but I’m realizing that I have not given a proper update on the topic of output yield. The original claim for the Argos was a maximum shot yield of 55g. This number came from the shot volumes that I was getting on the original prototype machine.

While making many small design updates for production including improved spring performance, reducing lever force required, gauge/ transducer interface updates, and overall usability, enough changed thereby reducing maximum shot output to 45g with a single pull. While I have been pulling shots on a production machine since January, there were still many changes to the spring design and the chaos of preparing for production that I did not think much of the reduced yield and how that might impact customers that pull longer ratio shots.

We recognize that this is a significant change from what was originally ordered which is why I want to make it clear before shipping (the rest of) the machines. The good news is that I’ve identified a simple change that will be made to the piston linkage allowing direct lever pulls to achieve up to ~53g yield. This will be a backwards compatible part made in the next batch that can easily be swapped into any machine from the original batch to achieve these results.

We have no plans to modify the machine to achieve this yield in the spring configuration as that would require significant modifications to the design that would have negative repercussions in other areas. A common thing done for spring (or direct) lever machines is what’s referred to as a “Fellini” from a clip of Fellini’s Orchestra Rehearsal where the barista in the back is seen pulling the lever of a commercial spring machine down once, releasing, then pulling again. (https://youtu.be/xuiXwOOkib4?t=161) The concerns of this maneuver are lack of consistent extraction pressure and unseating the coffee puck during the second pull. Unseating the puck should not be a concern on the Argos with our piston air bleed valve. I personally have performed a Fellini on the Argos for testing high ratio recipes and find the taste and balance to be enjoyable.

Processes and Documentation

This past week has been slow but crucial for assembly. Our head Assembly Technician has taken a deep dive into our assembly process, documenting detailed steps throughout. With our assembly team growing quickly, we needed a way to ensure consistent high quality builds for all of our machines. This document clearly lists each step with photos, tools, steps, and tests required for each step of the way.

As mentioned in the last update, we still had leaks due to improper hydrostatic testing. We’ve found ways to properly test at multiple stages throughout the build process and found that the leaks were a direct result of inconsistent tightening torques. We purchased a larger torque wrench and specific adapters to allow us to torque all heating elements to the appropriate amount as noted in our new documentation.

We’ve started 3D printing assembly jigs to hold large components during assembly which protects both the parts and the technician while performing alignment and tightening procedures.

Wood Bonding inconsistencies

Issues with bonding our wood parts to metal have started to show up. This is specifically a problem with the magnets in our side panels as well as the wood knob on the metal boiler cap. We’ve determined this to be a direct result of poor surface preparation and inconsistent bonding practices. We use a very high quality adhesive that is FDA approved as food safe (even though it does not see any water) and high temperature resistance but we quickly learned that surface preparation is extremely critical.

We have now tested and started using better surface preparation techniques as well as using a more appropriate amount of adhesive for bonding our materials together. We have also come up with strength tests to perform on each part to ensure an adequate bond before using those parts on a machine.

Grouphead Pressure Gauge and Transducer Fitment

Unfortunately we’ve had issues with the alignment and fitment of the grouphead pressure gauge and transducer. This is a part that ships pre-assembled with the machines but requires final assembly by the user. Due to slight misalignment of the shipped parts, it’s been very difficult for customers to install. We are now testing a new style of fitting into the grouphead that is the same size as our original fitting but now performs a compression seal instead of an o-ring seal which will make installation much easier for our customers while also ensuring a more secure fit. We will also be doing a better job of setting these sub assemblies up on a fixture in our shop before shipping to ensure an exact fit on the machine. All current customers will be shipped a new assembly (minus transducer) with the new compression fitting installed to swap onto their machines.

Hiring

We’ve hired 1 new assembly technician this week. This brings our total up to 4 part time technicians! One has been out for the past month and the other out for the past 2 weeks. This has reduced our output but also allowed us the time to properly document to ensure all machines will be going out with consistent quality from here on out. Sometimes you need to slow down before you can speed up.

We are talking with 2 more part time technicians that we would like to bring onto the team. It currently takes 1 hour to build up a complete machine. If we assume 4 technicians working on average 20 hours per week, 10 - 15 hours would be spent building and testing machines, and the other 5 - 10 hours would be spent cleaning parts, organizing, bonding wood, building small sub assemblies, packing orders, and shipping. Even if we bring on the 1 to 2 more technicians mentioned above, I think it’s still safe to assume on the low side where we can build anywhere from 40 - 60 machines per week with 4 technicians once everyone is ramped up.

Lot’s of numbers thrown out here but if we continue to slowly ramp up we should have all machined shipped out within 4.5 months.

CE Certification Updates

We’ve received our formal FCC/ ICES (US/ Canada), VCCI (Japan), KS C (Korea), AS/NZS CISPR (AU/ NZ), and CNS (Taiwan) documentation. UKCA and CE documentation are in progress.

Shipping Updates

We’ve finished assembly on 10 units and have shipped 7 of them out! If you ordered a spare piston assembly, it will be located on the underside of the lower packaging insert in the same way that your standard piston assembly is packaged on the top of the upper insert when opening. Spare springs will be labeled on the top of the spring. We are still figuring out a more sophisticated way to label these moving forward.

Please remember, when accepting and signing for your package, if there are any signs of significant damage please tell the delivery person to note that your package was damaged upon receiving. You may still accept the package and contact us but the receiver must explicitly tell the driver to list the damage when signing for your machine in order for us to file a claim with the shipping company.

We are continuing to update the user manual located on our website: USER MANUAL

For a video on unboxing, setting up, and using your Argos with tips and tricks to get the most out of your machine, watch here: Odyssey Argos Unboxing and Set Up

New Videos will be posted next.

Reviews and Feedback

While machines are continuing to head out the door each week, here’s a reddit post with replies from one of our initial customers! https://www.reddit.com/r/espresso/comments/14i70lc/its_finally_here_odyssey_argos/

Other customers have told us:

“Regarding shots, already better than anything I ever got with my La Pavoni. To be fair, I kind of gave up after modding it with a thermocouple on the GH and the water heater mod. Around that time I ordered the Argos and it's just kind of been sitting there since. Today I had my 9th, 10th, and 11th shot all hours apart and they were pretty much all great and identical.”

“Shots are soft and quite defined. Clarity is excellent.”

Next Batch Pre-Orders

Pre-orders will re-open on Monday, Jul 3, 2023 10am MST for batch 2. Despite the slow start, we’re feeling very confident in our new assembly documentation, processes, and organization. The plan is to accept new pre-orders with the goal of ordering more inventory 2 months from now. It will then take about 3 months for the parts to be manufactured and shipped to our facility in Denver. These parts should arrive as the final Argos units from the first batch are shipped out to customers. By planning ahead we hope to prevent our technicians from running out of work between batches. With this timeline, batch 2 units should start shipping 6 months after ordering.

As mentioned, the price of the Argos will increase to $950 USD. We selected this price as a balance between keeping up with the rise in material and manufacturing costs as well as keeping the price of a base Argos + US shipping less than $1000 USD. Future increases in manufacturing and material costs will be dealt with by further optimizing our design for improved manufacturability to reduce part costs.

Our shop machine showing off the “cup warmer” capability

Argos Performance

We’re continually improving the software but I’ve finally taken some time to record consecutive shot temperature data with the Scace 2 and thermometer. All machines are currently cleaned and tested by pulling 4 shots then steaming. We ensure the peak temperature of 4 shots in a row all lies within 2C of the set temperature. This accuracy is achieved with what we call a “perfect pull” where all of the air is bled out of the air release valve in the piston as explained in our unboxing video.

We’re pleased with this data as comparable dipper (no pump) style lever machines require additional thermal management by the user and tend to see on average 5C peak temperature increases with each consecutive shot.

Consecutive shot temperature data. Argos set at 89C.

Sincerely,

Ross & William