Peppersass wrote:While waiting for the new board, I decided to inspect the steam boiler level probe to see if there might be any damage that would explain why the GS/3 overfilled the steam boiler. Here's what the probe looked like when it came out of the steam boiler:



There are some mineral deposits near the tip, but not a lot. What's interesting is that the probe is cut off. It looks like cutters were used to snip the probe:



Roger at CC confirmed that this is done at the factory to set the proper water level.

Here's what the probe looked like after I cleaned it with distilled white vinegar and water:



And here's the end of the tip after cleaning:



I'm not completely sure how the probe works, but I think the GS/3 simply measures the resistance between the probe and the boiler body, which is at ground potential. This would be infinite resistance when water containing conductive minerals isn't connecting the probe and boiler (i.e., the level is lower than the probe), and some finite amount of resistance would be detected when conductive water connects the two.

I tested the probe by filling a conductive metal cup with water. Resistance between the probe and cup was infinite with the probe out of the cup, of course. Resistance measured just under 2 megohms when I plunged the probe into the cup. So the probe appears to work. My only question is whether the resistance when covered with water is low enough for the GS/3 to detect. Anybody know if 2 megohms is about the right range?

The probe is pretty simple, just a piece of metal really, so unless it was heavily coated with deposits that altered the resistance, which doesn't appear to be the case, I don't see how it can malfunction. It must have been either a loose connection or perhaps a problem with my logic board (maybe water had accumulated in the brain box over time from the faulty vacuum breaker.)

I'm somewhat concerned about the base of the probe. There was a ring of partly disintegrated blackened material around the base of the threads. I cleaned this off with a utility knife, which exposed a hard, white plaster-like material that appears to fill a recess in the nut:



It's not clear whether the copper washer will seal properly now. I won't know until I get a working logic board. Meanwhile, does anyone know what the story is on the material in or on the nut at the base of the threads? I see there are fiber washers available at espressoparts for the steam level probe. Is the white material just the remnant of a baked fiber washer, or is it some sort of insulator/sealer? Should I replace the level probe?

While I was at it, I also pulled the brew boiler temperature probe. I have seen a little more fluctuation in the brew temperature than there was when the machine was new, and after measuring with a Scace I found that the temperature offset was about a degree too high. This made me wonder if mineral deposits on the temperature probe might be throwing off the measurements. Here's what the probe looked like when it came out of the machine:



The probe itself looks better than I thought it would. The threads were a bit messy, though:



Note the white area at the base of the threads. This is Teflon tape, which I later pulled off after cleaning the threads. This is how the probe looked after cleaning:



You can see a few small deposits remain. I was concerned about damaging the surface of the probe and decided not to scrub them off with a Scotchbrite pad.

That flattened area at the base of the threads, where the Teflon tape was, concerned me. When I told LM that I wanted to pull the probes for inspection, they told me not to use Teflon tape when reinstalling the probes -- it shouldn't be necessary. But that flattened area looked like it would allow water to seep past the copper washer. Sure enough, that's what happened when I reinstalled the probe and filled the boiler at line pressure (about 3 BAR.) The only way I was able to get a good seal was to put a couple of wraps of Teflon tape around the base of the threads, filling that flat area. Later, Roger at CC told me they always use teflon tape on the probes. Anyway, it seems to be holding OK at 3 BAR. We'll see what happens at 9 BAR when I get a working logic board.

I can understand why LM recommends against using teflon tape. If you aren't careful when wrapping the threads, bits of teflon tape can get into the water path, and then you're going to get a clogged gicleur or a stuck check valve. I'm always very careful to tape low on the threads and avoid letting the tape stick into openings.

For those who are interested, I also took the opportunity while the machine was down to open the group cap to see whether more of the dreaded "verdigris" had accumulated in the year since I last opened the cap. The answer is yes. A complete description of that adventure is posted here: Clogged Gicleur With Particular Reference to a La Marzocco GS/3.