Cimbali Jr S/1 boiler level question

Need help with equipment usage or want to share your latest discovery?
User avatar
jknotzke

#1: Post by jknotzke »

Hello

This is in continuation of this thread: Problem with Sight Glass Water Level On Cimbali Jr

I posted a follow up question to @civ privately and he kindly replied with the thoughtful request that I post his reply and my follow up questions via a post.

To set the stage, I own a 1999 Cimbali JR S/1. I have owned this machine for 20 years. This is a pour over machine who's boiler is primed manually. There is no auto-fill on the boiler. You have to push a button which turns on the pump which pumps water into the boiler.

The level of the boiler is determined by a site glass. In this site glass is water and a small plastic ball. This indicates the level that is currently in the boiler:



As @civ pointed out to me 8 years ago, this is done using the notion of communicating vessels. The bottom of this site glass is connected to the bottom of the boiler. The top of this site glass is connected to the top of the boiler. I can post a side photo if it helps.

This machine has been completely refurbished. Everything except the switches and frame were replaced as new. This was done three months ago. Since then, I have been using rpavlis water. Therefore, scale is not an issue.

The question is this. When I turn on my machine and the element turns as, as the machine heats up, the ball rises.. It will often rise to the point where it simply disappears. (It's difficult to see in the photo I posted but there is a MAX line 2/3 the way up the glass written on the stainless steel.) The ball goes WAY past the max level. No pump was ever turned on during this phase. Only the element was activated.

What I have noticed thus far is this:

1) When I turn off the machine, the ball and the water level where the ball floats will eventually drop and stay at that level until I turn back on the machine.
2) When the machine is hot, if I pull a shot, the ball will often drop down to a level that matches the level it was at when cool. But I can't say this is the case 100% of the time. But, there is a pattern.
3) If turn on the steam wand (which is connected just before the false pressure valve which is connected to the site glass, the ball will fly upwards and remain. If I also blip the false pressure valve releasing steam, the same will happen.

Not sure if 3) is relevant but I figured I might as well let you know what I have experienced. There have been times, where blipping the false pressure valve causes the ball to fly up and then quickly drop down to what I think is the correct level.. But again, this happens rarely, but it has happened.

My question is this: what is causing the ball to rise when the machine is heated and what is causing it to drop when cooled and why does the ball drop often when pulling an espresso ?

Thanks !

J

Aida Battle: Indigo Reserve from world renowned Finca Kilimanjaro in El Salvador
Sponsored by Aida Battle
User avatar
civ

#2: Post by civ »

Hello:

Glad to see you back. ;^D
jknotzke wrote: ... a 1999 Cimbali JR S/1.
... pour over machine ...
... no auto-fill on the boiler
... completely refurbished.
... except the switches and frame ...
... using rpavlis water.
... scale is not an issue.
Right.
I believe this Cimbali is identical to the Faema E98.
I have just sent Dan a *pdf of the S1 manual in case you are needing one.
Get it here: /downloads/ ... ompact.pdf (Thanks Dan!)
jknotzke wrote: ... turn on my machine ...
... as the machine heats up, the ball rises.
... will often rise to the point where it simply disappears.
The Cimbali S1 (essential information here) does not have an autofill but does have a "Level control device", as seen in this circuit diagramme and labeled Pe.



I have never used or seen the innards of an S1, so I don't know how this device works but it is there to do just that: keep the level under control.

So my first guess is that it is not working properly.

From the looks of it, pressing Pi (fill button) activates Pv (pump) and Evc (boiler fill valve) and energises what would seen to be a contact working in or with Pe, but I cannot say more without seeing the part.
jknotzke wrote: 1) When I turn off the machine, the ball and the water level where the ball floats will eventually drop ...
This is normal.
jknotzke wrote: 2) ... machine is hot, if I pull a shot, the ball will often drop down to a level that matches the level it was at when cool.
... But I can't say this is the case 100% of the time.
Does not seem at all normal.
jknotzke wrote: 3) If turn on the steam wand ...
... the ball will fly upwards and remain.
I'd expect that turning on the steam wand and releasing pressure woud cause movement in water level.
But to fly upwards ...
jknotzke wrote: ... blip the false pressure valve releasing steam, the same will happen.
You are essentially doing the same thing and getting the same effect.
jknotzke wrote: ... if 3) is relevant ...
Everything is relevant.

You don't say if you refurbished the machine yourself or sent it to a shop.
If to a shop, have you asked about this?

If the boiler level is showing a drop when pulling a shot, it would seem to indicate that pump pressure is having some effect on it.
This should not happen under any circumstance.

This is an HX machine and the brew circuit has no connection to the boiler.
The only thing that happens is the transfer of heat from the boiler to the brew water in the brew circuit, through the HX tube inside the boiler.

Like this:



Follow the red line from the first circle after the pump to the second one at the brew head.

Would a crack in the HX tube cause a drop in water level?
I think that, if anything, it would cause it to go up, not down.
ie: the pump would be sending water into the boiler through a crack in the HX tube as the pump pressure would be >1.5 bar.

I've run short of ideas and not having the machine in front of me does not help.
For now I'll leave this to those with more experience in HX circuits.

Maybe Stefano has something to add.

Cheers,

CIV

User avatar
jknotzke (original poster)

#3: Post by jknotzke (original poster) »

civ wrote: I believe this Cimbali is identical to the Faema E98.
I have just sent Dan a *pdf of the S1 manual in case you are needing one.
Get it here: /downloads/ ... ompact.pdf (Thanks Dan!)


The Cimbali S1 (essential information here) does not have an autofill but does have a "Level control device", as seen in this circuit diagramme and labeled Pe.
Yes, the Faema E98 looks to be identical to the S/1. Thanks for the PDF. I had never seen that before.

The Pe that is labeled in that diagram is basically a pressure switch that is in the control box. It's connected to a silicone tube that is connected to a brass bell that is lowered into the reservoir. It's what tells you if the reserver is low on water or doesn't have enough water to function and it cuts the pump. You can see that here:



(labelled Pe bottom right of the diagram).

It's the same sort of pressure switch you find in dishwashers or washing machines. As water in the reservoir fills, it pushes air through the tube which flips a membrane/switch.. I have been trying to find a replacement from Cimbali but I have come to the conclusion I am going to have to just grab one for a dishwasher.... It doesn't look like Cimbali carries this part anymore. But that part was probably never theirs to begin with..

I have replaced almost everything on this machine at least once in the 20years I have owned it. I didn't do the complete refurbish because I don't have the room in my condo to completely take apart an espresso machine.. my wife would have had a fit.. not to mention the fear I had of not being able to find all the parts. For instance, I couldn't find a boiler..

The technician that did refurb this machine has been doing this for about 40 years.. He made his living fixing commercial machines in coffee shops and got fed up with the stress of anxious coffee shop owners and trying to find parking.. He told me that the Cimbali JR has one small flaw and it has to do with the placement of the pipe coming out of the boiler that leads to the site glass. He claims there is another machine that has this flaw.. I can't remember which.. He told me that as it heats, the ball rises and one way to get a proper reading of how much water is in the boiler is to run the hot water tap and then run the pump to introduce colder water into the boiler and this will cause the ball to drop.. I never tested this theory..

To be honest, I still don't quite understand how the placement of the tube coming out of the boiler leading to the site glass is the cause of this. I took this theory with a grain of salt.

My theory is this must have something to do with pressure and/or temperature. I can say for certain that turning off the machine will cause the ball to drop and stay at that level until I turn it back on. So clearly, this phenomenon is somehow related to temperature and or pressure. When I run the pump to make an espresso, cold water does pass through the HX which does introduce cold water to the system.. The ball does often drop when I do this and then gradually climbs again.

But I haven't gotten much further than this in my theory. Cimbali stopped using this method a long time ago. They switched to autofill with a water level sensor directly in the boiler..

Anyhow, if anyone has any idea as to what is going on, I'd LOVE to hear your theory. It's not so much in attempts to "fix" this problem, but mostly to understand the physics of what's going on. I've been Googling for years and years trying to determine what is causing this with no luck.

J

User avatar
civ

#4: Post by civ »

Hello:
jknotzke wrote: ... Faema E98 looks to be identical to the S/1.
Cimbali took over Faema some time in the last quarter (?) of the 20th. century.
Many Faema E98s started to be sold fitted with parts having the Cimbali label.
jknotzke wrote: Thanks for the PDF.
You're welcome.
jknotzke wrote: The Pe that is labeled in that diagram is basically a pressure switch ...
I see ...

Then that's one problem to look at.
It may not be working properly.
Reacting slowly, not at all, at times, maybe filled with whatever it can get filled with after many years of use.
jknotzke wrote: ... pushes air through the tube which flips a membrane/switch.
Take the switch out, see if you can take it apart and give it a good cleaning.
ie: brass bell, silicone (?) hose and pressure switch.
Check the rubber hose is not losing pressure somewhere along it's length.

Chances are it is all it needs. 8^)
You can then test it by very gently blowing through the silicone hose and see if it clicks/makes contact.
Use a multimeter to check the switch.
jknotzke wrote: ... Cimbali JR has one small flaw and it has to do with the placement of the pipe coming out of the boiler ...
... as it heats, the ball rises and one way to get a proper reading ...
Hmm ...
I seriously doubt that.
We would have surely heard about it here at HB.

The trick there is for both the glass pipe and the ball to be squeaky clean.
A lot of crud, barely visible, sicks to the inside of the sight glass and the ball.

It's combersome to clean up, so very few techs do it.
Not to mention, obviously, the deposits we spoke about in your original thread.
jknotzke wrote: ... this phenomenon is somehow related to temperature and or pressure.
What I know for sure is that the brew circuit and the boiler are totally isolated from each other.
The only contact between them is by heat transmision through the copper wall of the HX cylinder which is immersed in the boiler.
No pressure is transmitted at any point.

Now ...
This holds true as long as the boiler intake Lucifer valve (Evc) maintains a good seal while closed.

If the Lucifer's valve seal is dirty or worn, it may at some point let some of the pressure/water in the boiler out through the over pressure valve (Vp) at the moment it opens.

It is the only place boiler pressure can get through, to as the pump is generating ~9.0 bar and the puck is also resisting ~9.0 bar.
This will happen when pulling a shot and a brief opening of the OVP will show up in the sight glass.

Take apart the Lucifer valve and see that it gets a good thorough cleaning then test it again.

Edit:
I remembered this post of mine from 2019 about a leaking Lucifer valve.
Your Cimbali S1 probably has the same proprietary Cimbali model.

Lucifer boiler fill valve leak

Keep us posted.

Best,

CIV

User avatar
jknotzke (original poster)

#5: Post by jknotzke (original poster) »

Thanks for the reply.

I will take a look at the Lucifer.. It was replaced during the refurb.

This morning I turned on the machine, and for three hours, the ball did not move. Then it started to creep up. Made an espresso, it did not drop. So ya, it's an enigma wrapped in a mystery !

The pressure switch for the reservoir is also acting wonky. I have to blow and suck on the bell to get the switch to work correctly. If not, it always indicates that the reservoir is low on water when it's full to the brim.. This seems to happen while the machine is off. I used to be able to pull the bell out and drop it back in and this would cause enough air to be pushed into the tube causing the switch to activate.. But that trick doesn't appear to always work now. It requires a little more motivation.. I haven't been able to find a replacement from Cimbali. So, I'll probably have to get one for a dishwasher and hook up a multimeter to figure out which contact does what.. I am sure that replacement part won't come with a manual as it will be configured to work with whatever dishwasher it was intended for..

I'll keep digging on this floating ball mystery.. It should be such a simple problem to solve for as the physics aren't that complicated.. But hells bells if I can figure it out.

Thanks again for the reply!

J

User avatar
civ

#6: Post by civ »

Hello:
jknotzke wrote: Thanks ...
You're welcome.
jknotzke wrote: I will take a look at the Lucifer.
It's a bit of a bitch to remove, but it is the only way to give it a proper inspection/cleaning
jknotzke wrote: Made an espresso, it did not drop.
Try this:

Put a blind basket in the PF and pull a shot.
When you press to pull the shot the pump will start to ramp up and as it gets to ~9.0 bar you will see the ball in the sight glass go down.
Stop the shot immediately after you see that happen, boiler water is being pumped through the OVP.

Lacking a blind basket, cut a circular piece of some pliable plastic bag material and cover the botom of the basket before tamping in some (ground) stale beans.

You are seeing the ball going down but without being able to reproduce it reliably because not all your shots (grind+tamp) cause the pump to ramp up past the OPV's set point, which should not be over 9.0 bar.

ie: the valve leaks but does not let water pass every time you pull a shot.
This is because it happens only if the OVP valve opens and that depends on the grind + tamp + water temperature + combination.
And when it does oopen, it is just for a very small fraction of the duration of the shot.
ie: 1 or 2s out of 30.

I can't see any other way (ie: a leaking valve) that could be justify what you are seeing.
Run the test above and let us know how you fared.
jknotzke wrote: ... pressure switch for the reservoir is also acting wonky.
... have to blow and suck on the bell ...
As expected.
It is probably gummed up, never cleaned up.

But It should not be too hard to take it apart.
And the mechanism should be easy enough to clean up.

Invite your friends Prudence and Patience to have a look at it.
Take your time, hurry will get you nowhere.
Post pictures here at HB and everyone will benefit from what you discover.
jknotzke wrote: ... requires a little more motivation.
Hmm ...
No.
Motivation will definitely end up killing kill it.
It is a cheaply made pressure switch which has to be able to work with a difference of maybe 20cm water pressure.
jknotzke wrote: ... keep digging on this floating ball mystery.
My money is on a dirty/poorly sealing Lucifer valve.
But it is your machine. 8^D

Cheers,

CIV

User avatar
jknotzke (original poster)

#7: Post by jknotzke (original poster) »

Hi

I'll give your test a try tomorrow.

One thing that may be of interest, today, after shutting off the machine and letting it cool to the point where there was no pressure in the boiler, the ball did not drop.

I removed the false pressure valve and pushed a screw driver into the site glass and pushed down on the ball and the ball and water level quickly dropped
to what I consider the correct level of water in the boiler.

So maybe there's an air bubble forming ?

I'll report back tomorrow with the test you suggested.

Oh and I had a look at the pressure switch:



And it looks to me like it's riveted shut. I don't think I'll be able to take it apart.. Unless you know of a way..


J

ECM Manufacture: @ecmespresso #weliveespresso
Sponsored by ECM Manufacture
User avatar
civ

#8: Post by civ »

Hello:
jknotzke wrote: ... test a try tomorrow.
Good.
That will tells us a lot/everything.
jknotzke wrote: ... no pressure in the boiler, the ball did not drop.
... removed the false pressure valve ...
... screw driver into the site glass ...
... pushed down on the ball
... water level quickly dropped ...
The ball is getting stuck and (amazingly enough) in such a manner that it creates a seal in the glass tube.
Since the negative pressure is very small, it does not pull the ball down because it is a relatively tight seal.
Eventually, the seal will break.

Interesting.
I think you may have solved one of the problems.

Try to get the ball out, use a straw/tube and suck on it while you pull it out.
It is very light.

Then see if you can sand all around it with some fine emery

Before putting it back in, clean the inside of the tube with a thin wooden dowel with the end wrapped in some soft towel and alcohol.
It should not be a tight fit inside the glass and be well strapped to the dowel.
You do not want to leave the towel inside the tube.
jknotzke wrote: ... an air bubble forming ?
In a way.

Think a soft drink straw filled with liquid but you seal the drinking end with your tongue and pull it out.
Little or no liquid will come out unless you take your tongue off the end.
Same effect, actually depends on the relationship between the height of the liquid column and the section of the tube.
jknotzke wrote: ... a look at the pressure switch:
Try to get it out and post a photo of the other side.
I'd like to see what is on the other end of those rivets and if there are any part number markings we can go by.

I'm off to roast my fortnight's supply of beans. 8^D

Edit:Clean up the sight glass/ball act before you run any test.
It just dawned on me that the only control you have over boiler level is the fill button.
Along with what you see in the sight glass.
ie: if the ball sticks, you won't be taking the finger off the button because you think the level is at a place where you need to fill.

That would solve the ball moving up effect.
The ball moving down is another matter.

Best,

CIV

User avatar
jknotzke (original poster)

#9: Post by jknotzke (original poster) »

Just a quick update:
  • I took out the ball (the best method was to use a funnel over the site glass and flood it with water which popped the ball out into the funnel)
  • I took a dishrag, soaked it in alcohol and pushed it into the site glass a few times over.
  • I took the ball and ran it over a file all over
  • I looked under the pressure switch and there's nothing but bare metal. I have ordered a dishwasher 2 level pressure switch
    In the meantime, I created a jumper bypassing the pressure switch and disconnected the orange light. It works fine.
The machine has been on for 4hrs. The ball did rise a little and I was going to backflush to perform the test you suggested but by the time I got setup, the ball dropped by itself..

I will come back when I have more news..

User avatar
civ

#10: Post by civ »

Hello:
jknotzke wrote: ... update:
... took out the ball ...
... funnel over the site glass and flood it ...
Very ingenious! 8^D
jknotzke wrote: ... dishrag, soaked it in alcohol ...
... a few times over.
Did you see any residue in the rag after that?
jknotzke wrote: ... took the ball and ran it over a file all over
Did you notice that some material was actually being filed off?
You could also try rolling it for a while beween two sheets of 200 grit sandpaper.
jknotzke wrote: ... under the pressure switch and there's nothing but bare metal.
I see ...
Any markings indicating brand, part #, etc.?
jknotzke wrote: ... ordered a dishwasher 2 level pressure switch ...
jknotzke wrote: ... created a jumper bypassing the pressure switch and disconnected the orange light.
... works fine.
Should work as long as you keep an eye on the tank level.
This is the only use for the switch.
ie: to make sure there is water in the tank and fill it if there is not.
jknotzke wrote: ... has been on for 4hrs.
... ball did rise a little ...
... going to backflush to perform the test ...
... ball dropped by itself..
The test was to see if the Lucifer valve leaks when the OVP opens during a shot.
ie: the ball (water level in the boiler) drops during a shot.
Using a blind basket mimics a shot that would make the OVP open.
Sorry if I was not clear.

I should have suggested to try the test <i>without</i> the ball inside the sight glass and see if the water level drops.

Cheers,

CIV