No announcement yet.

Strega - PIDS and temperature profiling.......

  • Filter
  • Time
  • Show
Clear All
new posts

  • Strega - PIDS and temperature profiling.......


    I thoughts there is value in documenting some of the modifications I have done on my Strega as other Strega owners might find it informative.

    I decided to install a PID on the group of my Strega – it was relatively straight forward, and could be done in a way that had minimal impact on the looks of my machine, but added flexibility and functionality. Pretty much all the parts I purchased from Auber. Nothing new here – quite a few people have done it. The standard wide band temperature switch was replaced with a pt100 temperature sensor.

    Perhaps of more interest, I also decided whilst I was there to add a PID to the boiler. Yes, I know the consensus is that there is little value in doing so, but I figured since I was doing work on the machine, the additional costs were relatively modest and it was an interesting challenge. It would give me the flexibility to readily change boiler temperature with a few button pushes in a way that is completely reproducible, unlike adjusting the pressurestat. And it would get rid of the constant clicking of the pressurestat.

    It wasn’t until I really started looking at it carefully that I realised that adding the PID to the boiler was a deal more challenging than the group PID The main challenges were where to place the temperature probe, and how to preserve the safety devices built into the Strega (e.g. low water cut out in boiler or tank, over temperature protection etc.). The group PID has those mechanisms in the heating circuit, so the PID-driven SSR simply replaced the temperature switch and preserved the protections. For the boiler PID it was more difficult as the protections were not in the main heating circuit. I ended up using the safety protected circuit to power the PID – the PID is only powered when the none of the safety mechanisms are activated, achieving the same protections as in the group PID circuit but through a different mechanism.

    I thought the value for others might come from the impact on the temperature profiles of changes to both the boiler and group temperatures. I did about 5 repeats of various combinations of group and boiler temperature and temporally aligned the highest temperature to generate average temperature profiles for each temperature setting. The temperatures were measured with a thermocouple in a home made portafilter basket that has resin replacing the coffee to mimic the normal dead space. Flow was adjusted with a needle valve to try to replicate a flow rate to about the same as seen in a slow shot.

    I have provided a couple of different graphs of the data.

    Temperatures 1 contains three profiles where I held the group temperature at 86, and three different boiler temperatures were used. As people have speculated, the impact of boiler temperature on the brew temperature is modest, but it does have an effect.

    Temperatures 2 contains some graphs where I varied both the boiler temperature and the group temperature.

    So what have I learnt:
    • Both group and boiler temperature influence the actual brew temperature.
    • As expected, the impact of the boiler temperature is less, but a change of boiler temperature of 5C creates about a 1.5C change in the initial brew temperature.
    • The greater the difference between group and boiler temperature, the greater the temperature decline during the shot. The group PID doesn’t usually cut in, as the group temperature is mostly above that set by the group PID for the duration of the shot, although for longer shots you will see that in some circumstances the group temperature is perhaps down to the level that the group PID might kick in.

    From a PID perspective, I am please I have done both boiler and group. It gives easy, reproducible control over both variables, and that can changed very quickly. In particular, if you are less concerned about steam performance, the boiler PID would enable quite a large influence of boiler temperature on brew temperature. I roast my own coffee, and pretty much each batch is different (bean origin and level of roast), and so the ability to easily adjust parameters is useful.

    I have also included a photo of the machine, and the location of the PIDs. I have sought to preserve the looks, and so the PIDs are hidden under the water tray. Since the water tray is so easy to remove on the Strega, they are very easy to adjust.

    One project remains – incorporate a control for the pre-infusion so I can adjust pre-infusion pressure. I could just use a dimmer, but that would mean a knob somewhere on the machine, and limited ability to reproduce the pressure setting. So I have got a remote controlled triac from China which will give me the ability to have a number of adjustable presets, operated by a small hand held remote. I just have to install it, which is a little more complex that it would be with the dimmer knob as I need to isolate both sides of the vibe pump from the wiring for the fill function. Think I have it sorted, just need to do the wiring and install it. I also have a Bezzera 40mm brew gauge which I might install on the right side of the machine to match that on the left - haven't decided on that yet.

    I hope my fellow Strega owners find this interesting.


    Attached Files

  • #2
    That's brilliant Dave, thanks for sharing! I've done similar (PID boiler & group head) on my Brugnetti Aurora and haven't looked back.

    The boiler PID isn't critical for temperature precision as you noted, as a typical 0.2 bar band translates to only about 3 deg C (perhaps about 1 deg influence on the brew temp based on your graphs). But more importantly it got rid of the Sirai clicking noise for me. PID is also much easier to adjust than having to open up the case to turn the screw on the pstat.

    I found the mod provide an extremely precise temperature performance not possible with a stock lever, and my shots have improved to be more consistent all-around (no longer influenced by ambient or warm up). This was noticeable during the winter, especially when compared to the warm summer (I have a temp strip on the group and it can idle as much as 5 deg C lower in winter). Some would argue the massive metal mass of the group head should provide a stable thermal behaviour, yes, but it doesn't guarantee the group to idle at the right temperature.

    You probably have thought of this too, if you ever need to pull back-to-back shots, you can lower the boiler temp even more, and increase the grouphead temp. This way the group would run more like a saturated group. I did that for a while when experimenting, no problem with immediate consecutive shots and the group head temperature wouldn't even budge when ran that way.



    • #3
      I agree with the temperature stability - that has greatly improved, more so due to the Group PID.

      Unlike the E61 HX machines which you could count on to gradually overheat if left, with the 10C deadband of the stock group temperature sensor it was hard to know whether it was too hot or possibly too cold. Flushing just added to the uncertainty.

      Now, I never need to think about a flush - just prep and brew. It does take a little while for teh group to stabilise after a shot, but about the same amount of time as preparing the next shot.