Cannot find the obstruction in HX/thermosiphon (Bezzera Mitica)

Equipment doesn't work? Troubleshooting? If you're handy, members can help.

#1: Post by BeMitica »


I need help figuring out where the clot lies: the thermosiphon of my Bezzera Mitica is still not working after descaling (the group remains lukewarm even after an hour of preheating). I explored a little bit the innards of the machine... here is where I am at:

I had never descaled in 5 years and a mushroom inspection revealed that it was needed! So I performed a HX descaling (with a commercial solution of Sulfamic acid) following the good advice I found on this forum, pushing from the tank all the way to the group.

I also inspected individual parts as much as I could: I detached all three hoses entering the HX:

Nr. 26 is the hose bringing the hot water from HX to the group
Nr. 29 is where the water is pushed by the pump into the HX
Nr. 25 is the hose going back to HX from the group, when thermosiphon works

These seemed clear after descaling. I even succeededin pushing a plastic wire all the way through the copper hoses 25 and 26.

Nr. 10, the bend connecting the HX to the 26 hose to the group, was still white with scale. So I scrubbed it and pushed some more descaling agent through it. Again ran some descaling from the tank to the group.

At the end of all this, there is no progress at all with the thermosiphon :( :(

It's true that I still haven't seen what is lurking in the HX itself, or in the connectors between the hoses and the HX. But I took apart as much as I could: I am not able to, and would be scared to, access the HX itself. Not easy to get tools inbetween all these hoses.

I hope you will have an insight what is going on there... and how I should proceed.

Thank you very much ...


PS: A final observation, for what it is worth:

After I draw some water through the group, it seems that sometimes the thermosiphon is working: the group is not just hot from the drawing, it remains hot. Or so it seems, sometimes's hard to judge. But at start-up, it never heats up beyond lukewarm


#2: Post by Bluenoser »

Sometimes it can get air-locked.. did you lift lever and leave for about 20-30 seconds to see if you can purge any air?

BeMitica (original poster)

#3: Post by BeMitica (original poster) » replying to Bluenoser »

I lifted the lever to just before pump contact. Then I get a slow flow of water out of the group, and after a while I hear a gargling like air moving through some hose.

Later I drew some shots and I repeated the experiment: same result (slow flow, then gargling).


#4: Post by Bluenoser »

I'd engage the pump for about 20 seconds to see if you can purge the air that way.. If not, look to see if some fitting got corrupted and air is entering.. I've not got a lot of experience with fixing these issues.. others might be much more helpful.

BeMitica (original poster)

#5: Post by BeMitica (original poster) » replying to Bluenoser »

Thank you Bluenoser. Unfortunately no improvement.

In fact I don't think the gargling air is the cause of the problem. The gargling only occurs when I drain the group by lifting the lever without engaging the pump, so it seems to me that it's just air getting through to replace the water that I drain from one (or both) of the hoses connecting the group to the HX.

But my real problem remains this:

As far as I can tell, the thermosiphon still doesn't circulate at all, and the group gets hot only from pulling shots.

Any help or insight very much appreciated... I was really hoping that by clearing the two hoses I would solve it (Nr 25 and 26 on the picture of the original post) but no luck :(



#6: Post by NicoNYC »

Hi Nicolas,

If I had to guess, there's scale inside the heat exchanger itself, likely concentrated on the lower side. That is where there is the most water, at its hottest temperature in the hx system.

Can you try disconnecting your thermosyphon pipes at the grouphead, then turn the pump on and with your finger block one tube then the other and see if you have flow from both. Obviously this should only be done with the heating element disconnected, or right as the machine has come on. This will tell you if there is in fact an obstruction in the system.
LMWDP #718

BeMitica (original poster)

#7: Post by BeMitica (original poster) »

So, I have put my finger down the mushroom hole :P Thank you NicoNYC for the idea!
then turn the pump on and with your finger block one tube then the other and see if you have flow from both.
When I close the upper pipe (from HX to group head) and run the pump, water comes at full flow through the lower pipe.

When I close the lower pipe... I have to press quite a bit harder to close it. I feel it on my finger, and the manometer says the pump gets at least 0.5 bar (it was zero when closing the other pipe).

The water still comes at full flow, but there is definitely something going on.

This seems consistent with the visual inspection, where I found scale on the bend connecting the HX to the upper pipe (in red in the picture in the original post above).

So now the question is, how to do I continue from here? I found it challenging to unscrew the the various fittings close to the HX, and reaching the HX itself seems beyond my abilities. I was OK getting the pipes loose,but going in further becomes tricky for my skill level...

BeMitica (original poster)

#8: Post by BeMitica (original poster) »

The only idea I have right now sounds a bit iffy:

I could put descaling in the tank, close the clear pipe (the lower one) with a finger, and make the pump push the descaling through the upper pipe...

Obviously a gloved hand is needed here. But maybe there is a better idea?

Team HB

#9: Post by ira »

Empty the machine of water, put descaling solution in the tank and let the machine fill, pull some shots so it gets everywhere. Depending on the instructions let the machine heat up and sit for a bit and then purge all the solution out and flus with water till it's gone. Or take off the suspect pipes, put them in a jar with solution arranged so that the bubbles from the solution rise up the pipes and pull fresh solution from the bottom and wait till the bubbles stop. In my case on a very badly scaled commercial machine it took many many hours to get the worst pipes clean.


BeMitica (original poster)

#10: Post by BeMitica (original poster) »

Thank you everybody for the good advice! Finally it's all solved, up and running with perfect thermosiphon action.

Following NicoNYC's answer I decided to force some descaling through the upper pipe connecting the HX to the group:
- disconnected heating element,
- closed the lower pipe group->HX with a gloved finger,
- descaling mix in the tank, pulling shots with still finger on the lower pipe:

After doing this for quite a while (let rest, pull a shot, let it rest, etc) the upper pipe was getting unobstructed flow from the pump. So I was pretty sure it was solved. Next morning, again no thermosiphon but I assume that I just stalled the cycle with air, so I took out the brew valve, scrubbed, lubricated, assembled, pulled a shot and put to bed for the next night.
Now: everything is restored to former thermosiphon glory!

My only regret: I did not manage to detach the bend connecting the HX to the upper pipe to the group. Since the pipe itself was clear, I guess that the scale was either in this bend or in the HX itself, either way out of my direct reach. So I had to run the descaling from the tank through the pump, which I would prefer not to do on a regular basis...

Any ideas here?

Another side comment: after reading a detailed thread on stalled thermosiphon on this forum I thought I might want to exchange the brew valve or at least make sure it was in the right way if worn by the cam (it's been 5 years...). But I did not see any wear at all:

Except perhaps the marks in the gasket ? But I don't know if it looks different new. Does this look all clear to you guys?