The model I played with was a pre-produciton, and by the admission of Jason (Slayer's father) , there are a good number of kinks to work out. For instance, he plans to rotate the steam wand assembly upward to improve mobility and better position them over the drip tray. I opted not to nip pick on this stuff and just focus to the machine's coffee producing ability.
The Slayer group head design is the most unique aspect of the machine.
There is no solenoid valve. The flow of water is controlled through a very precisely machined mechanical valve. So precise that the valve on the right group was leaking a bit, due to the fact that Jason machined it himself (Jason assured me that this issue is already resolved). I'm a bit concerned about the way this valve will wear, and its susceptibility to scale, but I trust Jason has considered both.
The claim to fame with this value is that water doesn't leave the group/boiler (that is until it exits to extract the coffee). The valve is designed to provide a pre-infuse pressure by restricting the flow. The pre-infuse is not progressive; it has two discrete stages: pre-infuse and full-on. At full-on (group paddle at the full left position) the valve is open and the only flow restriction is the 0.6 mm giggler. At pre-infuse, the valve is partially closed restricting the pressure to a set value. The machine I tested had a different pre-infuse pressure setup on each group (2, 3, and 4 bars). This allowed us to compare the direct effect of each pressure. The difference between
the pre-infuse on this machine and a Synesso is that the pump activates as soon as the paddle is moved to the pre-infuse position and the pre-infuse is achieved through flow restriction (as opposed to line pressure, with the Synesso). This allows for pre-infuse in one group while another group is at full pressure, but it also drastically changes the water flow/pressure dynamic (more on this below).
To control brewing temperature the three smaller brew boilers are fed with a preheat boiler. The preheat boiler is PID'd and was set to 160F, which according to Jason (and our tests) was probably a bit low. But this temperature is configurable, a nice feature. I don't recall the sizes of the brew boilers and preheat boiler (wish I'd written them down!). Each brew boiler is equipped with a separate Fuji PXR temperature controller. The thermocouples are mounted in a unique position right next to the outlet valve, as opposed to near to the water inlet on the Synesso and Zocco. I'm not sure how this placement will affect element response time (as the thermocouple is reading a more stable part of the boiler), but one great result is that, as indicated by the picture above, WYSIWYG on the controller display (less some ramp-up loss, due to dispersion). This would allow a barista to watch the dispensed temperature during a shot, if desired.
During our testing the temperature was fluctuating significantly (the setpoint was 201, and the actual temperature ranged from 197 to 202). According to Jason, he hasn't had a chance to iron-out the PID parameters yet and that should cure the problem (raising the temp of the preheat boiler should help as well). This slowed-up our shot pulling, but with a bit of patience (to wait for the boiler to recover), I feel that temperature variations weren't significantly affecting the outcome of our tests.
Now to the shots (we brought some of our espresso roasted by Paradise Roasters)
First we dialed the coffee using full pressure (moving the group paddle all the way left, without stopping at the pre-infuse). With a bit of dialing, we managed (waiting between shots for the group temp to recover) to get our espresso tasting like it does on our Synesso. Then we started to play.
Jason's suggestion was to put the paddle to the pre-infuse position, wait for the first drop to fall, then go to the full-on position for 1 second before returning to the pre-infuse. According to Jason, this emulated a lever machine pressure profile. We watched on the pressure gauge as the pressure slowly ramped up, held full, then ramped down slowly. The shots tasted good, with slightly improved sweetness and a softened acidity, but with a muddled flavor. The real improvement came when we fined up the grind (dropping the dose) and lightened up on the tamp; a common approach to pull a lever shot. The result was fantastic sweetness, a silky body and all the same flavor articulation that we experienced with full pressure.
All this begs the question: how does this differ from Synesso? The way I see it is that the major difference is water flow. Because the flow is so restricted, pressure transitions occur more slowly. On both the upswing (from pre-infuse to full-on) and downswing (from full-on back to pre-infuse). On the Synesso, with a much higher flow rate, the pressure drops quite rapidly if you try to return a full-on shot to pre-infuse (btw, my testing shows negative results, if this is attempted).
My initial impression, after only a few hours of tinkering, is that Slayer is not quite a revolution in brewing, but an excellent step in the right direction. It might be difficult to get great repeatable results from a high volume cafe, such as ours (I'm still pretty enamored with automated pressure profiling), but full credit to Jason and his crew for putting together a great machine, and pushing the industry forward.
Parting note: we didn't talk about the details of servicing the machine, but Jason assured me that as a machine tech himself, he'd put thought into making it serviceable.
-Phil