Modifying thermosiphon flow rate on vintage Faema Levers

A haven dedicated to manual espresso machine aficionados.
User avatar
dominico
Team HB
Posts: 2006
Joined: 9 years ago

#1: Post by dominico »

Open thermosiphon lever designs are famous for being very temp stable and for not needing flushes between shots.
They also tend to lack user friendly controls for controlling that temperature: they will stay rock solid at one temperature, but you don't really get to choose what that temperature is in a very convenient way.
Many people will not care about this modification as the "factory setting" will be in the general ballpark for most espresso blends. There are those however that either change between a lot of coffees frequently or for some reason their factory setting just runs a bit too hot or cold for them.

Some good ways to control the shot temperature on the fly and without necessarily modifying your machine can be found here: Commercial Lever Temperature Management Techniques I split that discussion out to make it more applicable to commercial levers in general and to allow this thread to focus on the thermosiphon modifications.

Modifying the Thermosiphon Circuit
The thermosiphon circuit transports heat to each group from the boiler and the constant flow combined with the thermal mass of the group keeps the temperature very stable.
The idea behind this modification is that, since each group has its own thermosiphon circuit, I can modify the flow rate of just one thermosiphon get get a different resting temperature of that group compared to the other.

Each group in my Faema President has what's called a "thermosiphon metering rod"


Located here (top down)


I previously wrote about these here Zodiaco Group Thermosiphon Limiter Configuration
when I was first experimenting with them a while ago. I found the orientation has a noticeable effect on the group temp but others were not able to repeat the results and also told me that the orientation would do little to help you control the thermosiphon flow. It was suggested however (I forget by who) that a thicker metering rod would probably have a more pronounced and repeatable effect on the temperature.

This time I measured group temp, always measuring in the same location.
My running temp range for that thermoprobe location was 151-153.


Over the last few weeks I have experimented wit "thickening" the metering rod using teflon tape to one of my groups.


That group now runs at a range of 147 - 149


Results
I now have a proven "cool group" and "hot group", my temperature management now is just choosing which group I want to the shot on.
https://bit.ly/3N1bhPR
Il caffè è un piacere, se non è buono che piacere è?

chappcc
Posts: 270
Joined: 9 years ago

#2: Post by chappcc »

Very interesting post. I have a Salvatore Compact Lever and have attached a self adhesive temperature indicator on the exterior of the group head. It generally idles around 77C (171F); after several shots it can get up to 85C (185F). I never really questioned whether this is too hot or cold; however, after seeing your temperature readings, I am now concerned this may be too hot. I don't note too much acidity which would be characteristic of too high an extraction temperature. I am now tempted to try a needle valve in the return circuit to cool the grouphead. This would be a simple modification as an experiment or permanent addition. My p-stat operates between 1.0 and 1.25 bar. After each shot, I use a small flush to clear grounds and avoid stalling the thermosiphon.
I would be interested to see where other Bosco heads idle.

User avatar
naked-portafilter
Posts: 698
Joined: 10 years ago

#3: Post by naked-portafilter »

"Cool" :-) experiment on that amazing machine. You have my all attention.

Gabor

User avatar
dominico (original poster)
Team HB
Posts: 2006
Joined: 9 years ago

#4: Post by dominico (original poster) »

Just a note that I am using the thermometer attached to the outside of the group as a relative measurement. That the thermometer read 153 before modification doesn't really mean much, it means more that there is a measurable cooling affect with the modification to the thermosiphon metering rod. The next step here to get precise temp measurements would be to use a Schomer / Scace style portafilter thermocouple, which I will do for the fun of it, but to my tongue I know that I can now pull consecutive shots on the modified group without having to lock a portafilter to make it too hot.
chappcc wrote: I am now tempted to try a needle valve in the return circuit to cool the grouphead. This would be a simple modification as an experiment or permanent addition. My p-stat operates between 1.0 and 1.25 bar. After each shot, I use a small flush to clear grounds and avoid stalling the thermosiphon.
I would be interested to see where other Bosco heads idle.
The needle valve would be a much more elegant solution for this, and it would be cool if you wanted to do a thread on your modification. Also FWIW my boiler runs at .8-1 bar.
https://bit.ly/3N1bhPR
Il caffè è un piacere, se non è buono che piacere è?

chappcc
Posts: 270
Joined: 9 years ago

#5: Post by chappcc »

dominico wrote:The needle valve would be a much more elegant solution for this, and it would be cool if you wanted to do a thread on your modification. Also FWIW my boiler runs at .8-1 bar.
I installed the needle valve in the return loop of the thermosyphon and as expected the range of control is very wide. Without the needle valve, the idle temperature is around 77C and can rise to 85C after repeated shots. With the needle valve at ~50% the idle temp is 66C. I will need to take some time finding the best setting to optimize extraction for my taste.
One confusing result is the top port of the grouphead is the cold leg of the thermosyphon. On my E61 machine. the top port was the hot leg, which is what I was expecting for the Bosco grouphead. The temperarure difference across the thermosyphon is about 12C.

User avatar
dominico (original poster)
Team HB
Posts: 2006
Joined: 9 years ago

#6: Post by dominico (original poster) »

Looking forward to pics! :-)
https://bit.ly/3N1bhPR
Il caffè è un piacere, se non è buono che piacere è?

User avatar
vberch
Posts: 596
Joined: 14 years ago

#7: Post by vberch »

Great job, Dominick!

chappcc
Posts: 270
Joined: 9 years ago

#8: Post by chappcc »

dominico wrote:Looking forward to pics! :-)
Here are some pictures:

Needle Valve in the hot leg before insulating:

Cold and hot leg grouphead connections and thermocouple attachment:

Cold and hot leg boiler connections:

I also have a thermocouple attached to the face of the grouphead on the portafilter collar.

I have the needle valve set to ~ 50%. Without the needle valve, grouphead idle was between 77C-85C depending on how long the machine was unused.

After restarting the machine this morning I got the following temperatures for the machine in idle after ~2hours:
  • T hot leg = 106C
    T cold leg = 82.5C
    T grouphead = 69C
This is a delta T across the thermosyphon of 23.5C

After my first shot, the thermosyphon stalled (both hot and cold legs were around 79C with the grouphead at 70C), so I did a brief flush to clear the screen which restarts the thermosyphon.

After ~30minutes the machine is idling at the following:
  • T hot leg = 104.5C
    T cold leg = 87C
    T grouped= 70C
My previous post about the top leg of the grouphead being cooler than the bottom leg may have been due to air in the thermosyphon loop after I added the needle valve. Now the thermosyphon appears to be working as expected.

I now need to find an "optimum" setting for the espresso I am using.

User avatar
JohnB.
Supporter ♡
Posts: 6579
Joined: 16 years ago

#9: Post by JohnB. »

chappcc wrote: My p-stat operates between 1.0 and 1.25 bar. After each shot, I use a small flush to clear grounds and avoid stalling the thermosiphon.
I would be interested to see where other Bosco heads idle.
Keeping in mind that my Sorrento is a hybrid dipper I did do extensive Scace testing earlier this year. After an idle period shot water temp peaked at 198°-199°F. With a 2 second (42-45ml) flush before you start the shot prep shot water temp peaked at 201°F. Each second added to the flush increases peak temp approximately 1°F.

Kees recommended a 45ml flush for his Idrocompresso after an idle period to bring the group up to temp. He also offered a needle valve option in place of the fixed jet in the thermosyphon.
LMWDP 267

chappcc
Posts: 270
Joined: 9 years ago

#10: Post by chappcc »

JohnB. wrote: Kees recommended a 45ml flush for his Idrocompresso after an idle period to bring the group up to temp. He also offered a needle valve option in place of the fixed jet in the thermosyphon.
I would not expect the grouphead to heat up very much with just 45ml flush. In my case just after a shot (~30ml), the surface temp of the grouphead increases by 3C.

Post Reply