for my Bezzera Magica, an E61 HX,
for the last 4 months, I am recording with a datalogger (2 analog inputs) the temperature profiles for the sessions;
with K type probes in different points on the thermosyphon pipes and on the group;
no scace, no styrofoam;
lately, I have an Eric's E61 adapter with K thermocouple, keeping records for group temp profile and for temperature on the lower thermosyphon pipe, then on the upper (closer to the GH);
current settings are for pstat: 1.07-1.19 bar, autofill probe 1 mm up from the lowest position;

the problem: there is a consistent double hump profile in the first 10-12 sec of the extraction;
first hump corresponds to the interval when the pump is running and there is no coffee drop on the bottomless PF; then, when coffee appears, the temp rises again (do not know if preinfusion is over) and the second hump finishes after ~12 sec from the start of the shot:



I wonder if the short E61 preinfusion (5-6 sec) could be an explanation, and if the preinfusion would be longer wouldn't be a single hump like usual? is this preinfusion duration adjustable somehow or influenced from other settings?
or is just a kind of malfunction in another area?
BTW, with a maximum of 97C in long idle, I thought there is a restrictor on the upper tsyph pipe, but now I doubt it;

I had an initial temperature bump that was caused by steam (a very fast and effective carrier of heat). No idea if that would apply here.

It might be useful to consider your curve the result of several overlayed curves. Ie, steam, cool metal heating up, a HX cooling down as cool water enters and heating up as heat is conducted from the main boiler, sensor response time delays, water sitting in the lines, etc.

No flush?

You've got two major things changing the temp - the HX is one and the thermal mass of the group is the other. When the too-hot water enters the group, it heats up the group a bit, but without continuing to flow (which would feed more heat into the group head), it cools. Then, once it does flow, you've got colder water entering the HX, bringing down the temp again.

What do the measurements look like if you do a brief flush (1-2oz) before locking the pf?