the pavoni has two V section seals- the bottom one faces down and the top one up.

Now that I think about it- I am not sure what is happening when the piston is at rest- what stops the space above the piston filling with water...

I don't think the water passes thru the gaskets at all. Water surrounds both ends of the piston.
Here's a pic of the my Pavoni Pro piston, at it's highest point inside the cylinder wall.

You can see the water inlet ( square hole) on the top edge of the cylinder. The water runs down the cylinder channel to exit into the cylinder just below lower piston gasket but above the front face of the piston.
This forces the water to run around between the cylinder wall and piston wall , filling the chamber evenly from the whole of the piston perimeter. Much like a toilet pan really. Wetting the puck from the outer perimeter to centre.
With the lower piston gasket facing forwards, and critically positioned on the inlet it allows me to do very small pumps to set up preinfusion in the puck; thus only one pull and no channeling for 1oz of liquid gold.

180degrees wrote:With the lower piston gasket facing forwards, and critically positioned on the inlet it allows me to do very small pumps to set up preinfusion in the puck; thus only one pull and no channeling...

What lies behind the word "thus" here is exactly what we're trying to tease out, only with a full pull, not a series of minimalist preinfusion pulls.

Let's walk through your scenario:

Piston full upstroke: lower piston seal is positioned so that water can enter the cylinder via the square notches.
Piston minimalist downstroke (just a couple of mm for preinfusion): a little water is forced through the filter.
Piston minimalist upstroke: with this operation, if the lower piston seal were perfectly sealed, it would create a small vacuum. And maybe it does. And maybe it's not disruptive enough to cause any problems because the coffee in the filter at this point is a deformable soggy medium, not a hard cake, that can be resettled with downward pressure and water-flow after it has experienced an insignificant up-suction.

What we need is a pyrex portafilter and basket

it would create a small vacuum

After the intial upstroke there's plenty of air caught inside the cylinder, partly due to the inflow of water going down the side walls of the cylinder, across the face of the puck and rising with the air trapped on top.
Especially noticeable on the first shot. The puck sits with 3/4 oz of water on top of it. The downstroke compresses the air, which applies pressure to the water forcing some of it into the puck. The minimalist upstroke just uncompress's the air, the weight of the water protecting the puck. At the end of the full downstroke there is always some bubbles from the compressed air, again especially noticeable on the first shot.
After the first shot the amount of air is always a lot less, and progressively so the more shots pulled. I would propose that the first shot, under pressure properly fulls the top of the piston which is now down and full of water and it is that water flowing lifted up by the piston,under pressure that is being forced into the cylinder inlet thru the cutouts on top of the cylinder wall flowing across the cylinder inlet and refilling the bore.
Why is the first shot so airy and subsequent shots less ? ---- because any air left in the outer wall of the cylinder is forced past the cylinder gasket on the second shot upstroke and out past the screen gasket effectively bleeding the head. We have a 2 stroke espresso machine.
Sorry it's a bit wordy, should really do a flow diagram.
Cheers

Very very interesting and I think if I understand it correctly, very different than the Ponte Vecchio design. But I am not sure I understand it correctly.

180degrees wrote about one square hole and a channel.

Timo wrote about square notches at the top of the cylinder.

Are they discussing the same thing?

Sorry about my thickheadedness. But does that protrusion on the left side of the cylinder contain the aforementioned square hole and the channel which allows the water to exit through another hole that is below the piston on its upstroke? Or is it entering the via the notches as Timo suspects?

Thanks
Geoffrey

hmm, quite different than my Europiccola group. It just has a small inlet hole near the top of the brass cylinder. No plastic.

And, back to the OP's question, maybe its the Gaggia Mini. I'll try to get a volume measurement soon.

BTW, I've been taking two or more pulls on the Ponte Vecchio and Caravel since I started using them (5000 replicates or so), and I really think the concerns about disturbing the puck during a second pull are overrated. At least the way I prepare my baskets, I've rarely seen any evidence of fractures on the puck. Over time I've developed a more slow and gentle approach to raising (manual) or lowering (spring) the lever, but even early on when I did it faster it didn't seem to cause problems.

Not to say I wouldn't want a nice single-group commercial lever on the counter. Currently, however, about HALF the space (and energy) needed for it is being taken up by BOTH the Ponte Vecchio AND MiniGaggia, so I've a happy combination of 45 and 58 mm groups, with steam on demand and precise control over shot temperature.

Cheers,
PC

180degrees wrote: ... After the intial upstroke there's plenty of air caught inside the cylinder, partly due to the inflow of water going down the side walls of the cylinder, across the face of the puck and rising with the air trapped on top. ...

Yes, some air remains in the cylinder.

Here's my understanding of the dynamic. It assumes, arguendo, that all seals are sealing 100%.

UPSTROKE
When the piston has risen to where its face (lower seal) is above the water inlet, much of the air inside the cylinder gets displaced by the inrush of water-under-boiler-pressure; the displaced air is forced out through the coffee in the filter, which is very porous at that point. Until the piston's downstroke closes the inlet, the boiler and cylinder form a continuum, and any air that does not get forced out of the cylinder gets mixed with the incoming water and compressed a little by boiler pressure. The pressure in the continuum will drop as some pressurized steam escapes through the filter, at least until the puck becomes wet and swells and effectively slows or even blocks egress. How much or how little pressure is in the chamber will depend on the p-stat setting and whether the heating element has kicked back on during the pull; other ancillary factors such as grind and dose and freshness and basket geometry will determine how much can be pushed through the coffee-in-filter by boiler pressure alone.

DOWNSTROKE
The downstroke compresses any air remaining in the cylinder as its contents are pushed through the coffee.
Some air remains above the saturated puck. The piston is in its lowest position and the bottom seal traps the air between the puck and the bottom of the seal. Residual pressure from the piston's compression may cause bubbling at the spouts (or on the bottom of the filter, if naked PF), the only egress . A small amount of air remains above the puck nonetheless; how much will depend on how full the basket is and on how much or how little air was forced out of the cylinder by boiler pressure during the upstroke.

SECOND UPSTROKE
When the piston rises, a slight vacuum is created. Whether the vacuum is sufficient to rupture the puck, and the severity of such rupture if it should occur, will depend on the amount of air that remains trapped above the puck, on the speed/force of the upstroke, and on the degree to which that trapped air remains compressed or has reached equilibrium with the atmospheric pressure on the outside of the filter.

PRACTICAL HYPOTHESIS
An overdosed heavily compacted basket is more likely to be ruptured by a second upstroke than a moderately dosed lightly tamped basket because the latter has a greater cushion of (stretchable) air above the puck, which mitigates the vacuum-effect. If the portafilter would sneeze, chances are good that a (slow) second pull wouldn't disrupt the puck.

I agree with peacecup on this issue: it's not that big a deal. I always do double strokes on my Sama and PV (and Caravel for that matter). Also I am not that careful about the second stroke. I never really have a 'ruptured puck'. Sometimes the top of the puck is a little dimpled- or cratered.

Also I generally fill the basket as much as I can- and certainly on the PV if I tried to quickly remove the PF immediately after the second pull- I would get a 'sneeze'.

Not wanting to hi-jack this topic maybe we should move to a new topic on this one sysop ?
--- but I wouldn't disagree with Timo either
But is it just me that has more air in the first shot ? if not then,
Presumably this extra air is coming from the inner chamber of the group head and as Timo says being ejected to the cylinder by boiler pressure as more shots are pulled.
Would be good to have a technique that would empty the air pre the first shot, it certainly reduces the volume and quality of pull.