La Pavoni Europiccola and Fellini/multi pump physics?

A haven dedicated to manual espresso machine aficionados.

#1: Post by Rickpatbrown »

I've been getting g some great shots from my Europiccola. Typically, I'm dosing at 16g and get around 30g out. I've been doing the "Fellini maneuver " to get that extra few mLs for a normale, but after reading a bit find that the Fellini is kind of a joke.

When I think about it, how would that even work? The Europiccola is just a piston. When the piston rises above the hole, the boiler feeds the cylinder with boiling water. If you were to press the lever down a certain amount and then raise it again, the water would just move up and down. There isnt any sort of one way valve or anything, right? The only way to get extra water is to make it come out into your cup (so much more than a small pull) and then raise the lever.

Doesnt this cause water to move back and forth through your puck? This doesnt seem like a good idea.

Seems much better to just drop my dose to 15g and do a single pull, no pumping, no Fellini, just maybe a light pull for 10-15sec preinfusion followed by a good strong pull.


#2: Post by toolate »

my results with this are mixed and usually i do what you suggest: 15 gms in a deep basket with moderate tamp.
30 gms out in 30 secs or so.
this gives me consistent results


#3: Post by ojt »

The problem with multiple pulls and la Pavoni is that, like you said also, you're pulling in very hot water. Not just boiling but rather closer to 115 / 117 celsius.

There's a difference in this regard between the millennium and premill machines. In the newer groupheads you would have the "buffer" of water sitting in the grouphead, ao most of your fellini water would be that water. On premills I think you get water straight from the boiler. The issue on both is that the grouphead tends to heat up during the pull and you'd then get hotter water for the second pull.

Well, at least this is the way I see it :) Don't have accurate way to measure the actual brew temperature, only the grouphead thermometer.

Oh and second problem could be disturbing the coffee puck.

I would rather just "wiggle" the lever at the top, do sort of mini pump(s)

Rickpatbrown (original poster)

#4: Post by Rickpatbrown (original poster) »

Another way of describing what I'm talking about is an analogy with a car engine. The piston require valves at the top to exhaust spent fuel and to pull in air.

The Europiccola has no such valves that I'm aware of. So a pumping motion wouldn't do anything except disturb your puck.


#5: Post by LObin »

I have a digital thermometer on my gen 2 group head.
Here are some of my observations:

- Temperature of the group quickly rises when the water enters the cylinder. By 12-14*C on mine.

-Group temperature LOWERS during the poor. Quite a bit actually. More than on a commercial lever or E61.

- When lifting up the lever after preinfusion or fully wetting the puck, boiler water re-enters the group and temperature rises again, maybe a couple degree higher than the 1st peak.

- Boiler pressure is higher than the pressure inside the group when the lever is raised so the boiler water inevitably travels from the boiler to the group if the inlet hole isn't blocked.

- The main issue that many face is puck degradation or damage since the puck tends to move up against the shower screen when the lever is raised a 2nd time.

- Best way to drastically reduce this negative effect on the puck is to add a puck screen. Bplus, coffee-sensor and Aeropress filters all work.

-Why do it? It's an efficient way to yield more coffee on smaller lever groups.

- I'm not talking about a 2nd pull where one would lower the lever all the way, completing the extraction, before raising again and repeating forna 2nd time. I only raise the lever again after my preinfusion phase is completed.
LMWDP #592

Rickpatbrown (original poster)

#6: Post by Rickpatbrown (original poster) »

LObin wrote: - When lifting up the lever after preinfusion or fully wetting the puck, boiler water re-enters the group and temperature rises again, maybe a couple degree higher than the 1st peak.

- Boiler pressure is higher than the pressure inside the group when the lever is raised so the boiler water inevitably travels from the boiler to the group if the inlet hole isn't blocked.

-Why do it? It's an efficient way to yield more coffee on smaller lever groups.
So this really this is what I'm asking here. Can you explain the physics of how you get more water in the group head/piston area with your multiple pulls?

Boiler water cant enter the group head if its filled with water.

Water never leaves the group, so you are really just pushing/pulling the same water up and down.


#7: Post by LObin replying to Rickpatbrown »

The coffee puck soaks up a lot of water. If you wait till it's fully soaked/wet (easier to see with a naked PF) and let a few drops fall in the cup, the water that's contained in the puck and cup (a few drops) makes room for more water to enter the cylinder.

One way to measure the amount of water absorbed by the coffee puck is to weigh your spent puck after a pull. I don't know if it's 100% accurate since there's eventually air (and more water... and the gibberish extracted at the end of a pull...) that is also pushed through that puck but it gives you an idea.

I wouldn't call it multiple pulls though. A semi Fellini at best.
LMWDP #592

Rickpatbrown (original poster)

#8: Post by Rickpatbrown (original poster) replying to LObin »

Ok ... I might buy this. Not bad.
I still stand by the argument that it can cause more harm than good.

The other consideration is ... what happens to air that gets introduced into the group? This displaces water. I suspect that air is responsible for changes in water volumes that we see.


#9: Post by jackdaddy »

I used to perform the 'Fellini' as I had read much about it in H-B discussions and I was frustrated with the inconsistencies I was experiencing with the final volume I was getting in the cup. I thought about what was actually taking place in the group and then this maneuver no longer made any sense to me. First, when you raise the lever to the top position the boiler pressure forces water from the boiler into the portafilter and puck. Assuming that the grind is not too fine, if you pre-infuse long enough the puck will become saturated. This is evident when the puck can no longer absorb anymore water and begins to drip into the cup. At this point, the puck cannot absorb any (meaningful) additional water. Keep in mind the lever is still up, pulling the lever down a small amount and then bringing it back up again without displacing any liquid will not change the amount of water in the portafilter unless the puck is not fully saturated. If some coffee drips out then there will be more room for water in the portafilter and will thus increase the final volume whether you pump the lever or leave it at the top position. So, my conclusion was that doing a 'Fellini' maneuver only adds to the yield when the puck is not fully saturated, as it is forcing more water into the puck. If espresso has already started dripping into the cup then leaving the lever raised has already replaced water that has passed through the puck. WRT air in the portafilter, I always raise the lever to near the maximum point before water enters the group, so I don't think that has been an issue for me. Now that I tend to do 20-60 second pre-infusion I get consistent volumes in my cup, and any variation (usually +/- 2 grams) seems to be due to whether or not the puck is fully saturated before starting the pull, which is a function of several variables, i.e., dose, time, grind, bean, bean age, boiler pressure, etc... This is evident when you apply 1 to 2 bar of piston pressure during the initial ramp up and espresso hasn't begun to drip, indicating that the puck might not have been saturated.

As it relates to temperature in a 2nd generation La Pavoni, my machine is modified so the temperature profile is different. In a standard 2nd generation La Pavoni group, the upper chamber of the group is filled with steam. When the lever is raised steam is forced out of the upper chamber in the group back into the boiler. When the lever reaches the top of the piston stroke, water from the boiler (under boiler pressure) enters below the piston and has a free path to the portafilter. Steam is much hotter than water, but it doesn't have a pathway to the portafilter, so if you are measuring near the top of the group, you don't have a good indicator of how hot the brew temperature actually is. If you are measuring near the bottom (the bell, if you will) then you are getting a more reliable brew temperature. My machine is water heated, meaning the top of the group is not heated by steam, rather it is heated by water, so my temperature range is more narrow, but the same measurement location principle applies.

Here is a plot of initial temperature on my machine before lifting the lever and the corresponding temperature of the group 30 seconds after water has been introduced to the puck (brew temperature). I wait 30 seconds to record the measurement to ensure the temperature has become stable indicating that the group bell has matched the water temperature inside. The average temperature increase from initial to brew was 3.5 degrees celsius. So, now I know that if I start a pull at 87 degrees, I can expect my brew temperature to be around 90.5 degrees. This is a small sample set, but it seems to be very consistent, so I didn't continue to record the temperatures. It would be interesting to get another set of measurements in the cup, using a preheated cup.

Here is a plot of the initial temperatures compared to their corresponding final temperatures 30 seconds after the shot was completed. The average rise in temperature was 6.2 degrees celsius.

Although this is a bit of a crude experiment, I think it is useful to have a rough idea about what is happening throughout the pull. I am sure there are more rigorous scientific temperature tests on La Pavonis here on H-B.

I may be wrong, but I think the only benefit of doing the 'Fellini' maneuver as the OP described it, is to maximize the volume of a shot if the puck was not fully saturated or there was air trapped between the piston and the portafilter. Doing it once the espresso has already begun to flow is approaching a lungo. I agree with the OP that doing a 'Fellini' when the lever is at the top increases the risk of disturbing the puck and ruining the shot.
LMWDP #698

User avatar

#10: Post by guijan12 »

I think you completed the answer, Jack.
Puck saturation is part of the answer.
The overheated water from the boiler, turning into steam, upon entering the grouphead is the other half (in fact the first half) of it.
This causes the spongy feeling in the lever.

A few half fellini's make that disappear, replacing the steam with water.

Btw, I get aproximately 40 gr. espresso with 15 gram dry grinds.