Yesterday I replaced the brew boiler pressure gauge on my GS/3, and thought others might benefit from a description of how to do it (or how not to do it ).

I'm not exactly sure when my brew boiler gauge failed, but at one point I remember needing to lower the pump pressure quite a bit after doing a drain/refill. I didn't give it a second thought (though I should have) until a subsequent drain/refill session when I noticed the gauge wasn't getting all the way to zero when I drained the boiler, even after shutting off the water:



With the water on, the gauge was reading really high, around 5-6 BAR. It should have been 2.5-3 BAR. I wasn't sure how linear the error was, but probably should have realized that the necessity to change pressure probably meant it was off by at least as much at the middle and high end. Unfortunately, I didn't record the reading before changing the pressure.

I continued to use the machine with the defective gauge set to 9 BAR until I could measure the pressure at the group head. I have a homemade PF gauge, but the flow is controlled with a needle valve and it's hard to get accurate readings. Then I saw Psyd's modification of his Scace I to include a pressure gauge, and decided to follow suit (the Scace flow meter is much more accurate and reliable than my needle valve.) It took a while to get all the parts, but eventually I was able to confirm a 2 BAR error at the group head. This means I was brewing at about 7 BAR instead of 9 BAR. I don't think this made a huge difference, but I wanted the machine to operate to spec. I used the modified Scace to set the pressure to about 8.5 BAR at the group, which resulted in a reading of about 11-11.5 BAR at the boiler (if that math confuses you, remember that there's about a 1-BAR drop in pressure between the group and boiler due to the gicleur.) I also adjusted the OPV up a couple of BAR, to 14, so it would open at 12 BAR.

When I called Chris Coffee about the problem, I got an uncharacteristic response from what I think was a junior tech. He said I had to make sure the water was off. I told him I did that. He said I had to drain the boiler. I told him I did that. He then said I had to disconnect the gauge tube from the boiler. I said, "Huh?" He gave me some sort of mumbo jumbo about maybe the tube was blocked, despite my telling him that the gauge was reading high, not low. Trapped air in the tube? Maybe, but not likely. Besides, what would I be able to do about it? That tube is very thin, very long, coiled, and soldered at the gauge end. I wouldn't be able to push a blockage out anyway. Besides, I think it was either Greg or Eric who posted here that what I was seeing is a common failure mode for this type of pressure gauge. The exercise of removing the tube hardly seemed necessary and was easier said than done. I decided to leave the battle for another day and lived with the problem a little longer.

When my OCD finally caught up (and I realized the machine only has a few more months under warranty), I called Roger, the Service Manager at Chris Coffee. As is always the case, Roger (and most of his people) are first-rate knowledgeable service techs who are totally customer oriented. Roger immediately agreed to get me a replacement gauge. It arrived a week or so later (probably had to be ordered from LM), and I did the replacement. All I asked Roger about doing the replacement was if I should remove the front panel (the piece with the display and buttons.) Roger said yes. I probably should have asked more questions!

Of course, the brew boiler must be cooled, drained and the water shut off before any attempt is made to remove the gauge. The big problem is getting to the 16mm nut at the connection of the copper tube to the brew boiler, and getting to the 14mm nut behind the gauge which holds it to the frame. I went through every metric wrench I could find, most of which were too long, and several other tools, most of which were too large. For the brew boiler nut, I ended up alternating between a 16mm and 17mm wrench and trying very hard not to strip the nut. Finally, with great difficulty, I managed to get the nuts off and remove the gauge.

I was very concerned about damaging the delicate copper tube of the new gauge, either at the soldered-on fitting at the boiler end or at the solder joint at the gauge. The tube must be bent into a fairly sharp turn at those points in order to clear the body of the boiler (at the gauge end) and for the fitting to mate with the opening into the brew boiler. Also, the tube is considerably longer than it needs to be. LM had already coiled one end of it, but, unlike the original tube, they didn't coil it at the boiler end. I had to do that. I did all these bends and coils rather nervously, but they all worked out OK.

It wasn't hard to get the nut started at the boiler end, but it was really hard to get the nut at the gauge end started. After finally getting them going, I ran into considerable difficulty tightening them fully, just has I did when I tried to loosen them. Finally, the nuts were tight and I tried to replace the front panel. Uh-oh. The gauges didn't fit into the holes in the panel very well -- they were quite offset from their original positions. I remove the panel and played around with the brew gauge position (again, much easier said than done.) After several iterations, it still wasn't right. And when I did get it right, the mounting holes in the front panel didn't line up.

It was about this time that I notice a small bolt that connects the gauge support bracket to the GS/3 frame:



The bolt is just above the center of the brew boiler.

Duh! I thought the bracket was welded to the frame and wasn't movable. All I had to do was remove that bolt and I would be able to lift the bracket and gauges out of the way enough to get full access to the mounting nuts. Plus, loosening the bolt allowed the bracket to be repositioned so that the gauges fit perfectly in the front panel holes when its screw holes were aligned.

It's easy to get a 10mm socket in between the nut and gauge tubing if you push the tubing slightly out of the way. I used a short-handle ratchet (couple of inches), which is advisable.

It's a good thing I discovered the bolt. When I got the machine back together and turned on the water supply, the connection at the brew boiler leaked. That nut has to be really cranked down, even at 2.5 BAR pressure. Obviously, it has to withstand a lot more than that, so it took every bit of leverage I could muster, and a proper 16mm wrench, to torque the nut down. That wouldn't have been possible without loosening and moving the bracket.

Everything went back together, and after refill I was delighted to see the new gauge at 2.5 BAR with just the supply water on. After the boiler heated up, the gauge showed 9 BAR when brewing, which is exactly where I hoped it would be when I set the old gauge to 11 BAR (as stated earlier, pressure drops by one BAR between the boiler and puck due to the gicleur, but rises about a BAR due to resistance of the puck, or the Scace flow restrictor. The two cancel out and the brew pressure works out to about 9 BAR at the puck when you set the brew gauge to 9 BAR -- convenient!)

Another job well done! As usual, I figured out the right way to do it after the job was nearly completed .

Thank you for detailing your experience, Dick, along with detailed explanations on how to do this replacement.

One question I have, however, is how much shorter will be the lifespan of parts such as this pressure gauge, as a result of doing what may be unnecessary maintenance on the machine. You have stated in previous posts that you drain your boilers frequently, apparently in response to instructions you have read in the manual. I have responded in previous threads that I felt this was unnecessary, that the only thing that was really important was to be sure that minerals were not building up in the boiler(s) over time. This would be much more of a concern in the steam boiler than in the brew boiler, for obvious reasons.

It is also well-known here that there have been problems with some GS/3s related to gicleur clogging which appears to be at least correlated with corrosion on the copper parts in the brew boiler (the infamous TL-30 part). No one seems to really understand this process very well, but we do know that the machines that were the most effected by it were ones that sat for a year or so with their boilers drained but still damp, in a warehouse in Seattle. Removing the water from the boiler, even for a short period of time, might contribute to this process. We simply do not know.

I do flush a lot of water through my brew boiler weekly, and effectively replace virtually all the water in the steam boiler at least 2x per month. I believe that is sufficient. The boiler (and pressure gauge) remain always pressurized with my regimen.

I had a discussion about draining the boilers of an LM with Michael Teahan, an experienced former espresso machine designer (for Brasilia) and tech with decades of experience, who now owns a parts business with Angelo M. in L.A. Michael knows more about espresso machines than just about anyone else I've ever met. He told me he'd only drain the boilers in an LM if it was going to sit unused for 6 or more months, assuming that boiler safe water was being used.

So I do wonder if this early failure could be in part related to overly-fastidious machine maintenance in a low volume home setting.

This is the sort of example I would cite when people talk about how they take their commercial grinders apart every few weeks to clean out the grommous. I am not the originator of this expression but I do like to quote it: if it ain't broke, don't fix it.

Anyway, good luck and hopefully this will be the last time you will have to do this.

ken

Ken Fox wrote:One question I have, however, is how much shorter will be the lifespan of parts such as this pressure gauge, as a result of doing what may be unnecessary maintenance on the machine.

Ken, I think your question could be answered in part by finding out what causes pressure gauges like the ones used on the GS/3 to fail in the particular way mine did, which is apparently a common failure mode -- i.e., not going to zero under zero pressure, but still capable of movement throughout the rest of the range. If the failure can be caused by mineral debris deposits or other foreign matter in the water, then there's a possibility that repeated boiler draining might increase the chance of failure. But if the failure has to do with a defective part or manufacturing error, then it was just bad luck that mine failed (or perhaps it was the quality of the gauge selected by LM for this particular application.)

Maybe one of the fluid dynamics experts among us can answer the question.

Note, however, that if deposits or debris can be the cause, it's hard to say whether frequent draining of the brew boiler increases the chances of it happening. We really don't know if the green verdigris or other debris found in the group head can form in a short time when the TL30 tube is exposed to air, or if it takes a long time. We also don't know if draining helps to rid the boiler of the stuff, or if it disturbs the deposits making it more likely that they'll find their way to small orifices like the gicleur, pressure gauge, check valves, etc.

What we need is the Fantastic Voyage team to take their tiny ship inside the brew boiler to explore the possibilities.

Thanks for the detailed posting, Dick! I'm definitely filing this one away in the unlikely need for future reference.

Peppersass wrote:Ken, I think your question could be answered in part by finding out what causes pressure gauges like the ones used on the GS/3 to fail in the particular way mine did, which is apparently a common failure mode -- i.e., not going to zero under zero pressure, but still capable of movement throughout the rest of the range. If the failure can be caused by mineral debris deposits or other foreign matter in the water, then there's a possibility that repeated boiler draining might increase the chance of failure. But if the failure has to do with a defective part or manufacturing error, then it was just bad luck that mine failed (or perhaps it was the quality of the gauge selected by LM for this particular application.)

Maybe one of the fluid dynamics experts among us can answer the question.

These gauges are presumably different in design than the pressurestat gauges one finds on other machines, which probably have pressurized steam in them normally rather than hot water at the full pressure of the brew boiler. I did have a couple of the pressurestat gauges freeze up on me, over the years, in my Cimbalis, and it was always after the machine had been depressurized (turned off, drained), when I tried to refill and reuse the machine after a period of non-use.

Although I tried various maneuvers to get them working again, they never worked and needed to be tossed.

ken

Ken Fox wrote:One question I have, however, is how much shorter will be the lifespan of parts such as this pressure gauge, as a result of doing what may be unnecessary maintenance on the machine. You have stated in previous posts that you drain your boilers frequently, apparently in response to instructions you have read in the manual. I have responded in previous threads that I felt this was unnecessary, that the only thing that was really important was to be sure that minerals were not building up in the boiler(s) over time.

+1 (Same goes for descaling).

[I agree with Ken twice a year. This probably exhausts my 2010 quota.]

At the risk of following Marshall into the deep abyss of agreeing with Ken :

I think what Ken says about not draining the boilers on a regular basis makes some sense. If you run a fair amount of water through both boilers on a daily basis, which I do, then it's hard to see what draining the boilers will accomplish. Even if you don't run a lot of water through, you can always do that once a month instead of draining the boilers.

Given Ken's story about frozen pressure stats and his earlier observation about cation-softened water leaving a whitish powder residue behind, combined with my own observations of that powder residue in various places, and my experience of it blocking my steam boiler vacuum breaker, my suspicions are growing that cation-softened water with high levels of alkalinity (and/or perhaps other disolved solids) leaves a powder residue behind when it evaporates, and that residue can foul parts in one's espresso machine (pleaese excuse the run-on sentence!) The powder doesn't precipitate out of hot/boiling water like the scale due to hardness, and doesn't build up on parts the same way, but it seems to be rather less than benign. It also appears that it's important to minimize opportunities for such cation-softened water to evaporate inside the machine.

While it seems unlikely to me that the very short time it takes to drain my boilers is sufficient for water to evaporate, it's always possible that it is sufficient. Another possibiliy is that evaporation on some parts is hastened if they're still hot. This suggests that if you're going to drain the boilers, it may be best to allow the machine to cool completely.

But, bottom line, it seems prudent to drain the boilers only when necessary for repairs. (And Ken, I don't disassembly my grinder on a regular basis, either. )

Marshall wrote:+1 (Same goes for descaling).

[I agree with Ken twice a year. This probably exhausts my 2010 quota.]

Hi Marshall,

We do need to examine the above statement and to put it into its proper context.

On numerous prior occasions, you have stated rather clearly (and vociferously) that you don't do any work on your espresso machines, and you advise other individuals to do likewise. For all I know, you don't even replace worn group gaskets. No problem with that, not everyone is mechanically inclined or interested in working on mechanical things. I have extensive experience in both working on and modifying espresso machines; so be it.

Another oft-quoted theme of yours has been to recommend certain water treatments that have been promoted in a book, one of whose authors is the principal in a water treatment company that sells related apparatus. Although we did not agree (at all) on this topic, you were unable to get much support from other posters, either, for these views in at least a couple of recent prior HB threads, if memory serves.

I'm an experienced home roaster, for good or ill, and you buy your coffee already roasted.

We have our own opinions on the direction of high end coffee, and cafes, but as opinions they are like a**holes, everyone has one.

So, my friend, there may be less here than meets the eye.

ken

Ken Fox wrote:On numerous prior occasions, you have stated rather clearly (and vociferously) that you don't do any work on your espresso machines, and you advise other individuals to do likewise.

I have only been advising owners of GS/3's that they have left the "you're on your own" category we have all experienced with previous machines. Professional service is widely available for La Marzoccos for those who don't have endless hours to spend tinkering on a work bench or who don't enjoy the repair process.

Ken Fox wrote:For all I know, you don't even replace worn group gaskets.

Group gaskets wear out?

Ken Fox wrote:Another oft-quoted theme of yours has been to recommend certain water treatments that have been promoted in a book, one of whose authors is the principal in a water treatment company that sells related apparatus. Although we did not agree (at all) on this topic, you were unable to get much support from other posters, either, for these views in at least a couple of recent prior HB threads, if memory serves.

Ummm, my purpose was to bring to the attention of consumers willing to learn, that the most respected thinking in water treatment for coffee avoids cation systems. I have never seen a top-end coffee bar that uses them, and I do look. There have been several warnings here recently about the salt residues left behind by cation. If some people have a religious devotion to treating their coffee with salt, I will not stand in their way (or look to H-B posters as the best authorities on the subject.)