Skip to main content.

Great espresso with a Livia 90

After two years with my livia and a few months of home roasting, I attribute bitterness in the cup to barista technique and not bean freshness. I’ve found that bean freshness mainly affects the strength of the flavor. A good shot with beans a week old will taste a little duller (ie. less flavor) then a good shot with beans a few days old. I’m pretty sure there’s a discernable reduction in crema after a week.

Before home roasting, I would go to a fair amount of effort to make sure the beans I was getting were as fresh as possible. Usually a couple of days but always less than 7 days. I love home roasting mainly because I now have a very convenient supply of very fresh beans.

As I’ve said, I think bitterness comes from barista technique because whether I’m using a home roasted bean or fresh (<7 days) pro-roasted beans like Blue Bottle, Barefoot, Ecco or Zokka, I can produce bitter shots and awesome shots better or at least equivalent to a good La Marzocco-equiped cafe. The main factors I’ve noticed are brew temperature, no channelling, and shot length.

First major factor, and most important, is brew temperature. How do I control brew temperature on the Livia? By varying the duration of the pause after my cooling flush.

Here’s my basic process:

  1. grind
  2. dose
  3. tamp
  4. flush around 8oz of water through the group head
  5. pause between 12 and 45 seconds based on the bean I’m using. Lock in the loaded portafilter during this pause.
  6. brew!

My understanding of how this works is that the higher temperature steaming water is heating the water in the heat exchanger tube (the brew water). The longer you wait, the hotter the water in the heat exchanger tube gets. I think the Livia’s tube is pretty big like 3oz or something like that which is why I flush 8oz to make sure the tube is filled with water of a known temperature (ie. room temperature). BTW, from my reading at the excellent Home-Barista.com, PID’ing a heat exchanger machine is pointless because it just makes the steaming temperature more stable.

When I get a new bean or am trying a new home roast, I need to figure out the grind AND the duration of the pause. I iterate on the grinder setting until I get a shot length in the 30s neighborhood, then I iterate on the duration of pause. For the grind, it’s the standard “go coarser if shot length is too long; go finer if shot length is too short”. Lately, I’ve started testing this additional heuristic: “go finer if it’s overcast; go coarser if it’s sunny”.

For pause duration, I’ve come up with this:

The back vs. middle of the mouth thing came from me not understanding the similar advice I’d read based on sour or bitter which I can’t really distinguish. So I started asking myself if I could find something more reliable. This is the toughest part of this process for me since I still question “where” I’m tasting that undesirable taste but I’ve been able to dial in a pause duration for almost every bean I’ve tried.

This “getting to know the bean” process usually takes 2-5 shots depending on how lucky I get. While I’m iterating on grind, as long as the shot isn’t way off length-wise (<15s or >60s), I can take a sip and can get a read for the pause duration, too. Also, I take a guess based on what I know about the blend or what the roaster can tell you about the blend. I’ll ask if the bean likes it hotter or colder. Peet’s baristas don’t have a clue what I’m asking; Barefoot baristas often can tell you; Blue Bottle gets specific on their site although I only consider these good initial guesses.

I’m pretty sure the pause duration is machine-specific (ie. duration for Livia would be different than for a Quick Mill Anita) but not sure if each machine is different. I’d be interested in hearing if other Livia’s produce similar results with the same pause. If it’s consistent, we can help each other on this list by sharing our results. Here’s some specific numbers to give you an idea and I’ll publish the rest on my blog:

Second major factor is avoiding channelling. Channelling can kill (from bitterness) an otherwise excellent shot. Channelling is when the puck is unevenly tamped or distributed and the water finds the path of least resistence and extracts your espresso from just *some* of the grounds instead of *all* of grounds. It’s pretty hard to tell if you’re getting channelling unless you use a naked portafilter. You can inspect the puck but I’ve found that unreliable.

With a naked portafilter, channelling shows up as tiny jets of espresso at different angles. So step 7 above is “watch the shot if some jets appear, try to position the cup so they don’t spray into the shot.” A few reasons:

Third major factor is shot length. Bitterness of the espresso goes up as the shot gets longer. This is gradual so there isn’t an exact point here but the observation that if you shorten the shot, you can sometimes reduces the bitterness. Say I calibrate my grind and dosage to a 2oz double which blonds in 27s, there are a couple beans (Blue Bottle’s Temescal comes to mind) where I still get bitterness and if I just go for a shorter shot (say 1-1.5 oz), the bitterness goes away. I always order short pulls in cafes because they always seem to think when I ask for a double I must want more liquid.

If you’ve made it this far, thanks for reading! Hope it stimulates some discussion so I can improve some more.