One thing that is important when thinking about these machines is that the temperature of the boiler liquid water, and water vapour above it is the same, normally perhaps 116-117 degrees. The bleeding procedures clear air out of the space in the boiler above the liquid water, and also the top of the group (in 1974-2000 group models). Steam constantly condenses on surfaces that are colder releasing the heat of vaporisation until the surface reaches the temperature of the rest of the system. If one know the temperature at any point in the system, one knows the temperature of all liquid and vapour throughout the boiler and the part of the group that is connected by channel to the reservoir. (Knowing the pressure enables one to calculate this without a thermometer. That is why pressure gauges are useful.)

Until the group is bled, its top is not hot because air trapped in prevents the entry of steam through the little steam channel, but instantly after the bleeding operation it is HOT.

Probably the temperature determining location on the 1974-2000 group that is most reliable to determine when to pull the shot is on the flare of the group's bell. There is no hot live steam behind it and one is measuring how much heat has been conducted into the part of the group where most of the water is just before the shot is made.

In reality the bottom part of the group is heated by (1) conduction from the top part of the group through the piston and top group walls. (2) conduction through the flange that connects the group to the boiler. (3) Hot water that enters just before pulling the shot that is used to pull the shot.

(the 1961-1973 groups and post 2001 groups do not have the steam passage that heats the top of the group to boiler temperature. They are heated by conduction through the flange, and (mostly) by bringing hot water around the cylinder liner. They are thus less inclined to over heat. This is certainly desirable when pulling many shots, but whether good for making one or two is highly debatable!!!)

The good thing about measuring the temperature of water as it leaves the portafilter is that you know EXACTLY what happened. With three inputs of energy to the group, measuring temperatures at any other point may be misleading and may not be quite right.

The disadvantage of measuring temperature as it leaves is that you know AFTER THE FACT that the water was too hot or cold. But you know this exactly. You can tell you need corrective action in the future.

rpavlis wrote:(the 1961-1973 groups and post 2001 groups do not have the steam passage that heats the top of the group to boiler temperature. They are heated by conduction through the flange, and (mostly) by bringing hot water around the cylinder liner. They are thus less inclined to over heat. This is certainly desirable when pulling many shots, but whether good for making one or two is highly debatable!!!)

Why might the 1974-2000 models with their steam heated groups be superior to those with cylinder liners for making one or two shots?

The live steam entering the top of the group of the 1974-2000 group gets it hot quickly. After purging the top of the group by raising the handle whilst steam is exiting either the steam wand (pressurestat models) or relief valve(two switch models) one only has to wait until enough heat conducts downward enough to make the shot. (About three minutes.)

Thanks for clarifying that. I wasn't certain if you were referring to the speed of preparation or the taste of the shots. I find it takes seven minutes from room temperature to be at the point of pulling my first shot (1999 pre-millenium), during which time I grind the beans, prepare the puck and heat the cup. I hadn't though about the possibility that it would take longer to heat a millennium model to the desired group temperature.

Your explanation about the passage of heat from the top of my premillenium to the brew chamber makes perfect sense and is consistent with the way the brew chamber temperature seems to lag behind the temperature on the surface of the group just above the bell flare. Thank you for sharing your expertise.
Buckley

I would like to add that I have had good success with measuring brew temperature through thermocouple wire. I was able to get my hands on very thin wire, which I bend over the PF basket, leaving the exposed tip to rest on top and in the middle of a prepared puck. The normal seal of the group gasket to the basket rim is achieved without leakage, due to the super thin, two wire setup. I theorize about the grounds being at room temperature, and the absorption of the heat of the incoming water, and other temperature factors, but I feel that within a few seconds, there is stabilization. I am using an accurate (calibrated) mulitmeter at the other end.
To me, all of the typing about temperature and pressure is secondary, as it only points towards the goal of knowing what the water temperature is that is hitting the puck. I suppose that once good settings are known through my puck method, they can be repeated through other measuring devices, and I could remove the basket probe forever. But, unless you are measuring the water temperature, the correlation to pressure and to the heat of other parts of the metal machine, are subject to offset, lag and tolerance.

While I love the machined brass spout with the thermometer probe inserted, the measurment is post saturation by being well below the PF, so there is some loss of heat in that setup. How much? I dunno.

I did something similar to Robert, but far less elegant a while back with my La Pavoni Pro. I have a naked portafilter, so I taped a bare wire K thermocouple to the pf handle so that the K's tip was about 1/16" below the center of the double basket and hopefully in the coffee stream. Then I used my phidget and Artisan to record the pull. I now use a bare tc to measure the group temp, so I guess it might be interesting to do again, but this time with 2 lines going to Artisan.

regards,
Rich