sergiyr wrote:Is there a specific reason why the handful of the recipes posted here lean towards long pre infusion and long shot times? I'm just getting started with espresso but from what I saw on YouTube everyone is looking for around a 30 second shot time. Is there a fundamental difference on extraction time with a lever machine?

I understanding needing more for lighter roasts but if I'm starting with a dark roast shouldn't the target times be similar to machine times? As far as I understand, we're creating the same brewing environment in the Robot as electric machines just with manual force and perhaps a bit more control on parameters relative to entry machine. Am I misunderstanding something?

The key differences here are PI and ramp up time. Most pump driven machines drop the hammer and go to full pressure when the shot is started. With lever machines like the robot, you have control over preinfusion and the time it takes to get to full pressure. Those things add to the total shot time. Of course those steps are optional. You'll have to let your taste determine if playing with these variables improves the shot for each new coffee you dial in.

But in theory we have control over the Robot to replicate other machines, right? I'm going to bring up that same coffee that I asked about earlier in the thread as an example. Their espresso recommendations are as follows:

ESPRESSO - Modbar EP
Brew Temp: 198°F, Line Pressure: ~3.5 bars, Max Pressure: 9 bars
Pressure Profile: 0 sec to 4 sec - line pressure, from 4 sec till done - 9 bars
19g in : ~45g out @ ~26s
https://onyxcoffeelab.com/collections/c ... ia-worka-2

It seems to me that I can control all those parameters with the Robot so I thought it that would be a good way to brew. It came out extremely sour though. I got another espresso roast from a local roaster with the recommendation 18 in 36 out but no further details besides that so I did a 10 s PI targeting 30 s. I'm using water fresh off the boil at seal level. This one came out sour too so I'm wondering if there is any sense in using espresso recommendations from traditional machines. From my experience, so far it seems like no but I don't understand why not if I can technically replicate the same conditions in the Robot.

Another thought is that maybe the beans are too fresh? I saw on a European Coffee Trip interview that the baristas recommend using espresso beans 2-4 weeks after roasting and I started with maybe just 5 days after roasting.

sergiyr wrote:But in theory we have control over the Robot to replicate other machines, right?

Well, not exactly. Just too many variables. In fact, replicating between even different self-heating machines will be approximate at best, since they have different thermal and flow characteristics. Especially between lever and pump machines. And the Robot basket isn't exactly like a VST, EPNW, IMS, etc., although it is similarly high-quality and machined very precisely.

Recipes may be crudely helpful as a point of departure, but the Robot is its own animal, and you've got to let your palate tell you how well you're optimizing the prep and pull for each coffee you're using. The basic principles of dialing in a shot -- well documented on YouTube and the Internet -- still pertain, and understand that there are certain variables that aren't easily controlled, and there may be some coffees -- especially very light-roasted ones, generally speaking == that will be more challenging on a Robot.

Another thought is that maybe the beans are too fresh? I saw on a European Coffee Trip interview that the baristas recommend using espresso beans 2-4 weeks after roasting and I started with maybe just 5 days after roasting.

Beans definitely evolve between when they're freshly roasted and when they begin to stale, and that will vary between different coffees and roast levels. 5 days off roast isn't usually unreasonable, as most beans are getting into their sweet spot at that point, but some need -- or at least can tolerate -- longer stretches of time off roast. A good approach with a new machine is to get about a KG of a coffee you already know and like pretty well, and then experiment with it exclusively over a week or so until you understand how grind, dose, preinfusion/pull process, and temperature interact for optimizing your shots.

sergiyr wrote:Is there a specific reason why the handful of the recipes posted here lean towards long pre infusion and long shot times? I'm just getting started with espresso but from what I saw on YouTube everyone is looking for around a 30 second shot time. Is there a fundamental difference on extraction time with a lever machine?

I understanding needing more for lighter roasts but if I'm starting with a dark roast shouldn't the target times be similar to machine times? As far as I understand, we're creating the same brewing environment in the Robot as electric machines just with manual force and perhaps a bit more control on parameters relative to entry machine. Am I misunderstanding something?

Pump machines ramp up to their 9 bar-ish target and stay there the whole shot--traditionally. Obviously with stuff like the Bianca that can change. At nine bar you are pushing water more quickly through the puck to get to your target output than if you have a declining pressure. Traditional lever machines have a declining pressure as the shot moves along. This helps maintain puck integrity. If you were to ramp up to nine bar and keep it there then a 25 +/- 5 seconds might be reasonable. Also our water temperature--under normal conditions--will decline as the shot moves along, so different brewing environment. Further, with the manometer, pressure-wiseI feel I get a lot more control than a entry level machine.

My targets, light or dark, are similar pressure-wise. Probably I'll pull at a lower pressure. Temperature wise I like the light roast a little hotter and I will take measures to get the water/group warmer. It all depends on your taste. Experiment. Have fun.

In the Robot manual Paul explains that the 9 bar pressure as "the pressure was taken from the water inlet to the coffee machine, the extraction pressure at the actual group head was much less, possibly even 1 bar lower."

On the barista version of the Robot, I started replicating this by pushing to 9 bars and maintaining that pressure throughout - I also extract by weight so I keep 9 bar until I stop. Now I've tried to switch to 7 bar. Also since I extract by weight, I've been filling the potafilter almost to the brim with boiling water to keep the extraction temperature more even. At the end of the shot, it looks like I have almost half the volume above the puck full of water.


What is the goal of a long (10-15 s) preinfusion? Is the goal just to evenly wet the ground or are you targeting some extraction there? Do any professional machines run long preinfusions?

The main goal of pre-infusion is to evenly saturate the puck so the water flows through evenly and reduces channeling. I suspect the longer the pre-infusion the more thoroughly the puck is saturated and the odds of channeling are reduced that much more. I've never seen an explanation that targets extraction--that doesn't mean it doesn't exist. I guess you could taste what comes out of a pre-infusion and do 'salami' shots and see if you like this. Given the temperature gradient of the Robot I don't do long pre-infusions. Using a declining pressure as the extraction proceeds you are all ready reducing the odds of channeling towards the end. I don't specifically know but I'm sure there are some baristas running long pre-infusions on all sorts of machines. FTR my DW runs longer pre-infusions than I do.

The theory that I am aware of is that up to a point, a longer preinfusion allows the grounds to swell and become saturated in a controlled way. The consequence is that rather than collecting at the bottom of the puck where they would form a compact dense layer that resists the water flow, the smallest of particles become trapped among the larger ones. As a consequence of that, one can grind finer overall without choking and achieve higher extraction levels.

I don't time shots on my Robot, whereas I do on my pump machine. With the Robot I eyeball the bottom of the basket until the coffee dripping covers the entire surface, then I apply pressure. With the same grind settng that I use on my pump machine, but with 1-1.5 gram higher doses, the pull is smooth, but max pressure reaches 8 bar at the most. Presumably because of this, the crema tends to be thinner, but flavors are smoother/softer as well, compared to the more punchy, angular shots from my pump machine.

Thank you all for the education on pre infusion. I never really tried more than 5 seconds intentionally, thinking that it would over extract but that was naive. I also didn't look at the bottom of the basket during extraction since the cup I was using is large enough to cover it up. I will switch to a shot glass and start looking for beading across the whole basket before applying pressure. But it Will be a less scientific test now as I'm about to start a new bag of beans tomorrow.

Edit: tried the new beans today with the preinfusion waiting to see beading at the bottom of the portafilter. Takes about 10 seconds with a 18g dose on my grind setting and makes the best shots I've had yet. Too many new variables today though: 1) new coffee, 2) longer-pre infusion and 3) started using third wave classic profile in distilled water. Not sure if any one of these is making the largest difference but together its a nice espresso.