How I Stopped Worrying and Learned to Love HXs covers the basic principles behind the operation of HX espresso machines, but your technique has to be adapted slightly for each machine, dependent on the particulars of its design. A thermofilter and temperature datalogger are very efficient tools for fine-tuning the requisite flush routine in minutes, not days as one would need if done by observation and taste alone. With this in mind, I measured the Achille's recovery and rebound times to determine the optimal routine. Everything I found was consistent with Dave's recommendations based on his first week with the Achille, which goes to show that skill and intuition can get you there (almost) as quickly as expensive temperature measurement equipment.
Espresso machines like the Cimbali Junior have an abundance of "thermal memory". Their brew temperature is highly dependent on the temperature of the heavy brew group and the boiler pressure. The ever-popular E61 semi-commercial espresso machines (La Valentina, Bricoletta, Andreja Premium, Giotto Premium, Expobar Pulser, etc.) have thermal memory too, but they'll "reset" in five minutes to a predictable state independent of the prior operation over the last twenty minutes. The thermal memory of the proprietary group of the Elektra A3 was in between these two (Cimbali "elephant memory" Junior and semi-commercial E61). The Achille defines the opposite extreme: Practically no thermal memory at all.
Before offering my guess of why that is, consider its temperature profile at two boiler settings:
Pump displaces approximately two ounces; recording stopped when flow slowed at ~17 secondsNotice in both cases the temperature is rising unabated. Unlike the semi-commercial E61s, the Achille's powerful heating element has no trouble recovering during the extraction. In fact, the heating element kicks in mid-shot and the temperature rise only accelerates, especially if it's starting from a higher boiler setting, as shown by the red curve for the boiler pressure of 1.25 bar. Subsequently the flush strategy is undoubtedly flush-n-go, not the "flush... wait for temperature to rebound... go" strategy of many E61s. You can see in this video how quickly I locked in after the flush:
My haste wasn't for video production reasons - it's required to avoid going straight past the target brew temperature. In many respects, this makes for the easiest-to-explain flush routine. Pump the handle n times, lock in the portafilter, pull. End of story. No fuss, no muss.
The profile is fairly flat towards the tail, but not ironing board flat. Keep in mind, however, that the thermofilter shows boiler performance, not actual brew time temperatures because the plastic puck doesn't precisely simulate a coffee puck:
The "over the lip" measurement doesn't capture the whole picture either; that's a story for another dayAlthough I was very satisfied with the espresso performance of the Achille, I wanted to check if a flatter profile would produce an even better one. I ran the same series as above, but this time turning off the machine before starting the extraction. Since it's a manual, who needs electricity once the water is flowing, right? It did indeed produce a flatter to slightly declining tail profile, proving that the heating element is impacting the brew temperature during the extraction. I'm still judging whether it matters to shot quality; my gut feel is that low temperature, darker roasts may benefit, or extractions that go beyond 25 seconds.
From a design perspective, I find the Achille's HX performance surprising. It has nearly instantaneous recovery, a simple flush routine, and manages to avoid overheating on successive shots. Why, I wondered, didn't the grouphead overheat from the tail-end rise over a series of shots? Once I disassembled the group, I think that I understand: The Achille is an HX-centric espresso machine that relies very little on the grouphead for attenuating brew temperature. Unlike the massive Cimbali Junior group / HX or the heavy E61 group, the Achille's is more modestly proportioned. In terms of heat storage, the upper portion serves more of a structural role than temperature moderation. It encloses a nylon piston pump and gets hot from the boiler, but its contact is incidental. On the other hand, the lower group section (beneath the seam) is directly attached to the boiler to keep it warm. It's a big assumption on my part, but I believe the aluminum (yes, not brass, it's aluminum) dispersion block is a very necessary part of this machine's design. Aluminum warms and cools in a flash, which would be an important attribute of an HX-centric machine. After all, if the dispersion block was brass, it would retain far more heat over a series of extractions, ultimately overheating the group. In addition, the HX is saturated in steam, not boiler water, allowing the heating element to be engaged during an extraction without pushing the brew temperature through the roof. Or at least that's my take on it.
Even if my assumptions about the reasons behind their choices are wrong, my hat's off to the engineers who designed the Achille. The pump design is smart, the brew system's recovery is rocket fast and yet predictable, and when it comes to extraction quality, it's one of the more forgiving espresso machines that I've used, including the famed E61s. Not bad for a machine weighing less than 20 pounds.
PS: If anyone has closing questions about the Achille, please speak up. Dave and I are wrapping up the review for an early December publication.
by cannonfodder on Mon Nov 20, 2006 10:48 pm 
Contact Us
