My Carimali Eta beta (single boiler HX with PID) has been working fine up until now. Now it heats up fine; normal pressure and temperature. (1.1 bar, 100 degrees C) Then as I use it to draw shots and steam, the steam pressure just keeps on going down. After about 3 cups it's down to about 0.4 bar.

The puzzle is that the temperature stays at normal. I removed the thermocouple and put in a temperature probe, and it confirms that the temperature is still at 100-101 degrees C. By the time I had pulled off all the panels to get to the boiler to confirm the temperature, the pressure had dropped to nothing. So the PID is correctly turning off the heating element because the temperature is in range, (and the temp probe confirms the temperature in the probe pipe is correct) but there is no pressure. If I let it all cool down and start again, it will work for one pressure cycle again.

Can anyone shed any light on this?

The only thing I can think of that has changed from when it was working until now, is that scale could have built up in the boiler, but it seems such a large step change. The boiler still appears to be filling correctly , and I can draw hot water, steam, etc. (Just that the steam pressure dissipates)

vacuum breaker is stuck closed. remove, clean and replace; or fit a new one.

Hi Paul, I have stripped down the valve and it looks fine; the rubber looks good, and I have cleaned off the very small deposits on the valve seat. I have cycled up the machine again, and checked that the valve is open when cool, and it closes as the boiler heats up. As the temperature rises, the valve closes, and I can't hear or see any leaking steam. Boiler pressure builds up to 1.1 bar, but still drops off over about 40-60 minutes. I did some further testing, and the temperature I am getting out of the brew head is lower than it should be, at about 84-86 degrees C.

Does anyone else have any ideas I can try?

Haven't been here for a while so late reply. Hope you've sorted the problem out by now.

There can only be 2 ways for the pressure to drop I would think. A leak or the heater isn't on. If there was a leak I would expect there to be evidence as it would take a lot of steam/water to battle against the heater and the pump would keep refilling.

If the heater is switching on ok when the machine is cold, coming up to temperature/pressure, switching off and not switching back on the heater element itself is probably ok but the pressure stat isn't. It may may be able to switch when cold but not when hot or there's some blockage in the pipework which is temperature sensitive. If the stat's ok could possibly be the electronics I suppose for the same reason. Swapping the stat would be easy enough to try.

Please explain exactly what you are doing. 1.1 bar pressure is a temperature of 122C in the boiler. Normally, boilers are controlled by a pressure stat. You mention a PID that is measuring 100C but don't mention what it is measuring -- HX water, boiler steam etc.

What you describe, on the surface, makes no sense. If you have a PID set to 100C to control the boiler with a probe in the boiler, the pressure would never get above 0 bar unless it is false pressure due to a jammed vacuum breaker. So please be a lot more specific.

The solution had a D'Oh component to it. I had used non-thermocouple wire to extend the thermocouples to the PID units. What was puzzling was that it would apparently work correctly for the first 20 minutes or so. I think what was happening was that the readings to the PID were changing as the whole unit and the wires heated up. The PID readings were no longer relating to the temperature inside the boiler where the thermocouple junction was, but probably to where the joins to the extension were.

Removing the extensions cured the problem.

Oh well, live and learn.

At least I'm back to great coffee again; peace is restored to the household.

That would sure do it. Thermocouples work on voltage feedback. Change the wire, change the resistance, measurement goes out the window. Simply using TC wire is not quite good enough, you need to use the correct wire for the TC type you are using. You cannot extend a T type with K wire, if I am not mistaken.

I would not gave guessed that to be the answer. Wow. So I guess when you extended the wire you ended up reading the sum of three separate junctions. I wonder if that could be useful in another application where you were expecting it.

CRCasey wrote:So I guess when you extended the wire you ended up reading the sum of three separate junctions. I wonder if that could be useful in another application where you were expecting it.

Not unless you just happen to know the temperature at the junctions with the extension wires (for instance, in an ice slurry). Otherwise, the extra junction adds an unknown variable.

Jim