I made special brass boiler caps that fit on the 2nd and 3rd generation La Pavoni levers several years ago. I use one of these with a wooden handle for normal runs.

I also made one for monitoring pressure and another for monitoring temperature. The temperature monitor has a long brass tube attached to it that goes below the water level in the boiler so I can read water temperature when I insert a thermometer probe. The pressure monitor has a large easily read pressure gauge, not one of the tiny ones that can fit on top of the sight glass. From time to time I do pressure or temperature monitoring during an espresso making run.

The pressure behaviour reservoirs on 2nd generation groups (1974 or so to 2000) with their steam heated groups is interesting. Data here is with a 1999 Europiccola with pressurestat. With these groups once the machine is bled, "live" steam keeps pouring into the top of the group and its condensation releases heat to keep this area at boiler temperature.

Pressures are measured in Mpa. Multiply by 10 to get bars.

Turn on machine: 0.00 Mpa

Machine warmed up, pressurestat turns off: 0.081 Mpa

Machine is now ready for bleeding. Raise handle up to just before steam and water would exit group. Pressure jump occurs: 0.123 MPa

With handle still up release a bit of steam and water, open steamer valve. Keep handle near point of releasing water through group. Pressure plunges to 0.035 MPa Allow steam to flow for 5 to 10 seconds, while flowing raise handle again to release a few drops of water. Close valve, lower handle.

In a minute or so the pressurestat will turn off, with the pressure back up near 0.081 MPa. Wait three minutes: Pressure oscillates between 0.078 and 0.081 MPa as pressurestat cycles.

Raise handle for the shot. WITH A PROPERLY BLED 2nd GENERATION GROUP THERE SHOULD BE NO INCREASE IN PRESSURE. Indeed there was no detectable pressure change as I raised the handle, attached the portafilter, and made the shot. It remained right between the 0.078 and 0.081 MPa the whole while.

I continued to watch the pressure after I turned it off. As it cooled the gauge pressure became negative. When the pressure fell to -0.049 MPa the handle flew up by itself and pressure rose back to 0.00 MPa. (The gauge I use also reads negative gauge pressures.)

When I put on the temperature cap and measure it the temperature normally reads between 99 and 102 C when the pressurestat turns off. After the bleed when there no longer is air in the boiler it rises to about 117C. When turned off the pressure simply gradually falls.

So raising the handle bleeds any air from the group and dip tube? I always assumed that the steam valve bleeding air from the boiler was sufficient to get a stable boiler temperature. So it would be interesting to compare temp after a group and boiler bleed vs just a boiler bleed.

I did a quick test with a very full boiler (little trapped air) and after a reasonable amount of steam(and presumably air) was released from the pressure relief (note: I don't have a pressure-stat), temp went to 116C and there were no boiler temperature gains from any type of bleeding.

Group temperature is heavily impacted by boiler level due to conduction between the boiler water and the group. Lower level decreases the rate of heat transfer.

Bleeding the boiler requires a bit of careful work to get it completely purged. Raising the lever displaces most of the air above the piston back into the boiler and opening the steam valve releases this steam-air mix to atmosphere. I normally vent at this point until a good plume of water vapor is observed. Then close valve and lower lever and allow boiler to heat to shutoff. Again raise lever and open steam valve until a good water vapor plume is observed, close valve and lower lever. At this loaded PF is attached and lever raised to pr infuse the coffee and displace the air and gases from the puck. Very small down lever moves will force this gas back into the boiler, after several of these moves teh lever wil go solid and the shot can be pulled. This works for darker roasts as it keeps the group at a slightly lower temperature.

Exactly my observations on my pressure gauge rpavlis.

I normally use your original water bubble method and use multiple pumps of the lever to ensure complete purge, after which there is an audible change in boiling intensity and a very rapid spike in group temperature.

I always like to start an espresso session with the boiler sight gauge near full, because then it takes less bleeding to get the group and boiler purged. The pressure spike that occurs when the handle is raised at the start of the bleeding operation will be much smaller with a boiler that is not nearly full. To my mind, being absolutely consistent is the route to make completely consistent espresso. That consistency must include getting the system into the same state with each session. Without consistency one can get unexplained absolutely wonderful and unexplained terrible shots. With consistency one can do the things that make the wonderful shots and not have the terrible ones!

rpavlis wrote:... "live" steam keeps pouring into the top of the group and its condensation releases heat to keep this area at boiler temperature.

OldNuc wrote:Group temperature is heavily impacted by boiler level due to conduction between the boiler water and the group. Lower level decreases the rate of heat transfer.

Do you have data for these apparently contradictory statements?

They are not contradictory. The steam or hot water heating at the initial purge is 1 point in time and the conduction heating is further down the time line. As the steam-water mix initially enters the top of the group it decreases temperature as it gives up energy to the group, temperature decreases slightly. The temperature will recover back to the saturation temperature for the boiler by conduction if no shots are pulled. This is why after the first shot turning off the machine tends to prevent the slow overheating that occurs if left on.

What Robert says above is absolutely correct, be consistent. If you are consistent with the basic operating routine then the results will be consistent. The thermodynamics are complex and really only a curiosity item if you develop a consistent routine.

I did a test with a low boiler level and my conclusion (again, for a dual-switch, pressure relief model) is to let it hiss on high heat for about a minute, then turn it down to low heat. With that done, bleeding and boiler level don't matter.

As always, how hot the group gets before you pull a shot is important. 76C works for me.