Incentive for and considerations on PID ing VBM Domobar

[Hofer]

There was a substantial post at alt.coffee on PIDing E61 group non-HX machines.
After PIDing a new non-HX Domobar, I feel that some points need further clarification and discussion over.
The first is an incentive for PIDing. There was a [rather controversial] info at a.c. on a [rather inferior] thermal stability of Gaggia Classics and even Silvias. But it looks like a consensus exists that E61 group machines, both HX and non-HX, are very thermally stable due to their massive 4kg head.
I should confess that I had very strong intention to PID the Domobar in any case. But after making preliminary time-temperature measurements I had a very strong "case" to PID it.
A thermocouple from under the steaming button-type thermostat on the top of the boiler was connected to the designated controller Fuji PYX4. The following table shows time-temperature data measured in a thermal equilibrium state after the machine being idle for several hours.

Temp Time
97.2 0 ON Heating
97.0 18 MIN 45 sec
109.5 45 OFF
121.7 1:40 MAX Cooling down
120.0 2:25 18sec/C 11min20sec
118 3:07
116 3:43
114 4:20
112 4:57
110 5:35
108 6:20 25sec/C
106 7:10
104 8:02
102 9:08 35sec/C
100 10:17
99 11:02
98 11:40
97.4 12:05 ON

Summing the table up: 45 sec heating, 11 min 20 sec (!!!) cooling down, on-the-boiler temperature surf 97.0-121.7C, about 25K (!!!).
Several measurements were made for the extraction water temperature at different temp-surf points. Can't find the log this moment but the max-min difference was more than about 8K.
The original thermostat is with a bulb temp sensor, model Prodigy TR86: the same type as used in new Isomac Zaffiros .
The above data contradicts the measurements brought in Zaffiro's review at: http://www.coffeegeek.com/proreviews/de ... erformance
"Zaffiro has about a 6-8F (2.5C) range tolerance in its thermostat; that is, the top of the cycle may be 200F (93.4C) (which you can adjust! A first for a single boiler/dual purpose boiler machine!) but the thermostat won't activate the boiler element until it reads 194F (90C)."
The review states cycling time of 67 sec. It is a heating time, I suppose. But then 11 minutes the machine is cooling down. At "the top of the cycle" (when the lights are going off) the water temperature is 93.4C (as for coffeegeek). But next 55 sec (45sec-1min40sec from commencing the cycle) the temperature on the boiler raises for about 12K (109.5-121.7C), and brewing temperature for about 4K accordingly, with a speed of about 1K/10sec. It means, as for my PID-biased understanding, substantially non-consistent results with a "blind" surfing and very strong incentive for PIDing the machine.

The second point that IMHO worth discussion is finding the right place for a thermocouple.
In the Bionic Zaffiro (see ttp://home.earthlink.net/~mrfuss/)
"Thermal probe installed through a hole drilled in the top of the boiler, reaching just above the midpoint between the E-61 thermosyphon inlet and outlet and on the opposite side of the boiler from the reservoir inlet."
Sunnyfield posted to a.c. that he had located ("unfortunately"?) TC in the well of the boiler of his Domobar.
But is there any difference? I think the TC placing only has an importance if one supposes that PIDing "on the fly" is possible and/or desirable. Speaking of the possibility/ability for brew water temperature regulation during extraction, following little "experiment" has been performed. The coffee lever and steam button (heater) were operated the same time with commencing the extraction. Observed: the on- the- boiler PV temperature went down during extraction. My conclusion: the thermal inertia of the boiler [not speaking about the inertia of the group] makes brewing water temperature regulating "on the fly" an utopia. I didn't find taste difference for coffees with or w/o constant heating during extraction, but the above method could be taken as "the best fit" for regulating to the constant brewing temperature. (Next step could be a mod of installing separate heat accelerator).
Assuming that regulating "on the fly" is not possible, the ultimate goal for PIDing is getting the consistent starting point in terms of brewing water temperature. Now, is getting on-the-boiler temperatures of 112+-0.4C satisfying? Does it make much sense to move TC to the right/to the left/closer to inlet/inside the boiler? Anyway the machine should be idle for at least 30 min to be heated to reach sufficient thermal equilibrium. It doesn't matter the optimum 112C would be gained in 3 or 5 minutes, matters to which extent the temperature will be stable after about half an hour. That is why I have chosen to self-tune the fuzzy when the machine is thoroughly heated and the temperature is oscillating near the optimum point. These P,I,D allow m y temperature be in the very narrow band of +-0.5C. And my temperature is on-the-boiler upper don't-remember-to the right or to the left- and not inside the boiler- because I assume it doesn't matter. Even more, I think that on the boiler's surface temp measuring moderates temp fluctuation caused by water layers movement up and down and to the left and to the right.

These are the points I would gladly discuss with you, knowledgeable guys.

(also posted at alt.coffee today)