Hello, happy weekend and thanks for looking!

I recently descaled my very early model GS3 (SN<100) and since then it seems to behave differently:

1. If I use the machine in full power mode the thermal switch on the main on/off switch trips (not the wall outlet fuse, the one on the back of the GS3). This must be because the GS3 is drawing more current. Unfortunately I don't have a multimeter handy here and can't check for shorts on the heat elements. However, when I switch to half power mode both boilers get to temperature and everything operates normally.

2. A day after the descaling I noticed the steam boiler was over filled (water dispensing from the steam wand and at one point the OPV). I disassembled the steam boiler fill solenoid and inspected the water level sensor. The both seemed OK. So I contributed this to lose debris from the descaling and ran some water more through the tea wand and also emptied the boilers. Since running the water through I have not noticed the boiler being overfilled anymore.

3. Before the descaling when heating the machine the brew boiler pressure would get to about 12bar (set by the pressure relive valve). But I've noticed this doesn't happen anymore. When I brew espresso I can still get to the 9 bar and sustain the pressure during the brew. I have been looking at the hydraulics to see if this could happen by a leak from the brew boiler to the steam boiler (also would explain 2). But unless the HX is damaged this seems impossible. And if the Hx was leaking I would expect impact on the brew pressure and and the overfilling of the steam boiler to continue which is not.

4. The AV function seems to be slightly inconsistent: there has been a couple of times where I think less water was dispensed when I used the programmed settings (with coffee in the group head). I checked the amount of water dispensed without any coffee in the group head and it seemed to be quite consistent. My understanding is that the AV sets the volume regardless of the group head pressure, is that correct? So I expect very consistent volume unless tue pressure relief valve gets engaged which I checked and i didn't see any dripping during the brew cycle at 9 bar. I have visually inspected everything under pressure and there is no indication of leakage.


EDIT: I can't remember if it was always like this but when I dispense hot water from the tea wand with pump on the brew pressure goes upto 9 bar. I think this is normal but thought I'd check!



Thanks so much for looking and any ideas are super appreciated.

Cheers,
Mohsen

Hey Mohsen,

1 is certainly a loose connection tion on your thermal switch. You need to unplug the machine, slide out the electronics cassette to inspect the thermal switch terminals. One or more will likely be burnt/brown color vs clean and blue/white/yellow/whatever like the others. REPLACE the burnt spade connector on the wire side and clean the terminals on the thermal switch and you should be good to go.

2 is worrisome. Do you have the extra check valve between the tea water mixing block and the steam boiler? I can't remember.

3 and 4 both sound like you have air in the coffee boiler. Just bleed the air out of the group by cracking the bleed screw under the plastic group cap.

Cheers!

- Jake

I did a series of tests by turning one boiler off (lowering the temp to min) and then the other and indeed it looks like when the machine stabilizes, the pressure in the steam and brew boiler settle on the same value. I.e when I turned the steam boiler off both boilers settled at 0. When I turn the steam boiler on and set the temp on the brew boiler to 80 both boilers settle at around 1.5 bar.

So seems like there is some sort of a path between the 2 boilers. I can't imagine it's through the HX because then achieving 9 bar when brewing espresso should be impossible.

I was wondering if this could be caused by the malfunctioning check-valves. Especially the one between the steam boiler and the mixing valve. This is an older machine with just one check-valve in that line and I'll be upgrading that to the double valves soon. I was wondering if a failure on that check-valve could cause this?

Thank you!

Thanks so much Jake.

Perfect! I'll check the witch for 1!

On 2, indeed I have only one check valve in the line. I'll be adding the second soon. I was thinking that could be related as well.

On 3 and 4. I did bleed the air out and replace the copper washer I'll repeat and report!

Thanks so much again for all the help.

Cheers,

Jake, correct me if I'm wrong, but running tea water with the pump shouldn't affect the coffee boiler pressure. With the autofill solenoid and tea water solenoid open, water coming in from the pump will take the path of least resistance, i.e., into the steam boiler and out the tea wand. Pressure in the coffee boiler shouldn't rise unless there's resistance in the steam boiler or tea wand path.

If I'm right, it could be that the check valves or solenoid valves into and/or out of the steam boiler aren't opening fully or are clogged with debris. If the check valve at the mixing valve is contaminated, that could also cause it to not fully close, which would cause the steam boiler to overfill.

[N.B.: The rise in coffee boiler pressure wouldn't be caused by a leaky HX. I think that would have the opposite effect -- i.e., a drop in coffee boiler pressure.]

It is hard to say. I think it depends on the mixing valve position.

A wide open cold water flow would not build much pressure in the coffee boiler, but if you have the temperature set higher, more pressure will build against the needle valve in the mixing block and you will see higher boiler pressure.

Agreed on the HX relationship. A leaky HX will generally only result in coffee boiler water (high pressure) leaking into the steam boiler, causing an overfilled boiler. Given the prevalence of other contributing factors on GS/3, this is unlikely, even with the earlier tests showing a correlation between pressures.

The HX transfers some steam boiler pressure to the coffee boiler even in the absence of leaks. If the coffee boiler is 100% saturated and the check valves are functioning, the water in the HX can heat enough to trigger the expansion valve, even when the coffee boiler itself is still room temperature. Varying amounts of air in the coffee boiler will reduce this effect to the point of no pressure generated if there is enough air, though the amount of air required to create a specific pressure in the coffee boiler is a big question mark for me. My guess is that there is enough of an air pocket to cause some relationship between the boilers that looks like a leak.

Please let us know how it turns out after addressing the thermal switch and getting all the bubbles out of the group. Bleed the group cold, too. As the water in the HX will boil if the steam boiler is up to temp and this makes bleeding the group really difficult. It also makes you think you have a leak because steam bubbles will keep coming out the bleed screw.

Cheers!

- Jake

Hi Jake,

So after trouble shooting all options I kind of was convinced the HX is the culprit. So I went ahead and disassembled the hydraulics - your reseal post was awesome. Check valves seem to be ok but the HX is indeed leaking:



It may have been covered up by the scaling and then exposed when I descaled. I'll be checking the heating element next. This does explain the water and pressure issues but not the thermal switch tripping. PS I checked the switch and connectors and it all seemed fine. So I think there is more current being drawn...

Cheers,
Mohsen

Yowza!

Great job in the troubleshooting.

Thanks for all the inputs. You guys were awesome in pointing out the root-cause. I also learned a lot about how the hydraulics work on the gs3
When going through your resealing post it looks like your HX was not made out of copper. Do you recall that by any chance?

When searching online I've seen different pictures. Some looks like stainless and some copper. I have yet to find a reputable online store in the US with the part. May just reach out to LM.

I've also been fantasizing the idea of cutting of the copper piping and silver weld a stainless steel pipe instead to fix this for good. I understand copper is more conductive but I don't think that'd matter in this case. Any ideas?

Cheers,
Mohsen

I think they replaced all the copper with stainless. If it is less conductive you can open the mixing valve on the HX a bit. I am not sure it is adjustable. Never took it apart.

Since the HX is just a copper tube run at rather low temps - perhaps it can be silver soldered shut (with no lead solder).maybe some gas repair shop can do it.