I've not been able to use my new LM GS/3 this week, due to a blocked Gicleur. A week ago I returned home from a month long trip to France, and shortly after firing up my GS/3, which I purchased a couple of months ago, I noticed that the flow rate was greatly diminished, and in fact continued to diminish to the point where the machine would barely produce a water flow of ~200ml/min, when normal flow would be roughly 350-450ml/minute. I had not experienced any such problems during my initial month of ownership. The flow was so reduced that it was hard to flush off the group screen in between shots. Shot timing became just plain "weird," with the first drops of espresso starting to flow off the bottomless PF after 10-12 seconds instead of about 6 seconds previously, and the flow then seemed to speed up rapidly, much more rapidly than before once the first few drops appeared.

These problems were similar to those reported by Dick Green in this thread: La Marzocco GS3 pressure ramp takes longer

Subsequent conversations with LM USA in Seattle confirmed that the problems I was experiencing were due to a blocked gicleur, and in fact I was told yesterday that after Chris' Coffee disassembled Dick's machine, they found the same problem, a clogged gicleur.

For those unfamiliar with this part, a gicleur is basically a short section of pipe in the water intake path feeding the grouphead, that has a very constricted diameter with the intention being to restrict flow and hence effect both the water distribution and flow out of an espresso machine's group head. A gicleur can have various effects on shot flow and distribution, giving what in essence is pre-infusion, among other things. Since a gicleur is by definition a finely constricted piece of pipe, it is (understandably) a prime location for water path obstruction in an espresso machine's group head.

La Marzocco's prescribed remedy for this condition is to disassemble the group head, to unscrew the feed tube from both the group head and the solenoid ends, to dislodge the offending particle, and then to reassemble using new group head seals/gaskets in order to prevent leakage. I don't have the needed replacement gaskets/seals yet, so this sort of repair will have to wait. Another option, suggested to me by Michael Teahan in LA (espresso parts guru and master technician/machine designer for Brasilia in the past) would be to use a fine wire from the solenoid side to try to dislodge the particle without having to disassemble the grouphead. I intend to try Michael's suggestion when I come back from a planned trip to San Diego just before Thanksgiving. I don't have access to the right kind of wire yet and I'm leaving on my trip shortly so the repair will have to await my return. If I'm unable to succeed using Michael's suggestion than I'll try the standard procedure to free the blockage. I can post more in this thread in 10 days or so giving the results of my repair efforts.

Michael told me that what I was describing, what happened to my machine and to Dick's, is what they call, in the industry, "new machine syndrome." This is basically explained by the fact that when espresso machines are manufactured there is a likelihood of small bits of grunge (metal shavings, etc.) from the manufacturing process ending up inside the boiler or tubing of an espresso machine. If you are unlucky, these tiny bits of debris can become impacted in critical areas such as gicleurs, causing problems requiring repairs.

As soon as this problem became severe, and it became obvious that I would not be able to use the GS/3 this week, I went down in the basement and carried my old vibratory pump driven (circa 1995) Cimbali Junior back up into the kitchen and fired it up. Although I also have a newer rotary version of this machine, the vibe is lighter and more easily carried and set up than is the rotary, so I chose the vibe. Fortunately, with my kitchen remodel of a year ago, I have two separate locations in my kitchen which can support an espresso machine. The GS/3, turned off, occupies one, and now the Cimbali Jr. Vibe occupies the other.

Some may want to know what have I observed in the quality of my espresso beverages after going from a vibe and a rotary Cimbali Jr., with much older designs, to the LM GS/3, and then back to the oldest of the Cimbali Juniors. In all honesty, the Cimbalis made very good espressos, as (did) the GS/3. The GS/3 is easier to use from the standpoint of temperature management. In my hands as an experienced user of the Cimbalis and someone who has not had trouble using the GS/3 before this current problem, I have not noticed any change in beverage quality between the GS/3 and the Cimbalis, including the oldest Cimbali, which was manufactured 14 years ago.

As I have posted a number of times before, espresso machines are fun to play with and newer ones with gizmos are even more fun. However, if one is simply interested in espresso quality, you are kidding yourself if you think that a move from an already very good machine (for example, a well made E-61 machine, or better) to a more expensive one, is going to revolutionize the results in your espresso cup. The improvements that I have made that manifest themselves in the cup, have been with using better raw materials (e.g. COFFEE), and with better grinders. The espresso machine itself is a distant third to these, beyond a decent level.

Finally, I'd like to thank several people who have been very helpful to me in dealing with this blocked gicleur problem with the GS/3, in no particular order, Bill Crossland, Michael Teahan, and Roger from La Marzocco USA.

ken

EDIT: 12/6/2009

Please read my post made today on page 4; after two attempts with the guitar string it was necessary to disassemble the group head in order to resolve the clogged gicleur in my machine.

Ken, very interesting write up - Im sure there is a difference if you are entertaining party of coffee-hungry guests though


I think you should try to take apart the group (pending that doesnt void your waranty or anything) and dislodge the offending particle. Your machine is pretty new, so it should be fine with the gaskets that are already there (and if it does leak, you can always order the seals after the fact since you have the cimbali to hold you over in the mean time

I would think that you might be able to dislodge the offending particle with a blast of compressed air if you had a small compressor or maybe using one of these Air Dusters:
http://www.microbarn.com/details.aspx?rid=100054

JohnB. wrote:I would think that you might be able to dislodge the offending particle with a blast of compressed air if you had a small compressor or maybe using one of these Air Dusters:
http://www.microbarn.com/details.aspx?rid=100054

Compressed air didn't work for me, at least not consistently. I have a small air compressor here, which produces quite a strong blast of air. The first time I tried to clear the gicleur, the flow rate did move from about 360 ml/min to 420 ml/min. But later the flow rate fell to 360 ml/min again, and compressed air did not affect the flow rate. At one point, I replaced the solenoid valve and the flow rate returned to 420 ml/min without using compressed air. It did the same once when I simply drained and refilled the boiler. It's possible that whatever was clogging the gicleur was partly attached to it, and only getting bent or blown out of the way sometimes by whatever troubleshooting I happened to be doing.

The nature of my clog may have been different from Ken's, so it's possible compressed air would work for him. But Ken's flow rate has declined steadily, which was not the case with my GS/3, and his has dropped to a much lower level than mine ever did. My feeling is that whatever is causing Ken's clog is wedged in the gicleur a lot more firmly that whatever was clogging mine. I think the guitar string idea is a good thing to try before disassembling the group head. That said, if it's successful the debris will be pushed into the group head water supply, and hopefully into the boiler and out the OPV. If not, it might clog the gicleur again. Disassembling the group head would be a more reliable way to make sure the debris is cleaned out of the machine.

You would have to open up the line somewhere behind the gicleur so that the particle was blown completely out but if it is wedged in there you would end up having to removing it physically. The Vivaldi Minis had a similar problem as the factory was not cleaning out the sealed steam boiler properly before shipping. It was never an issue with the S1 as both covers have removable covers & evidently were cleaned well before assembly. It's unfortunate that the companies don't take the time to thoroughly clean these boilers before they leave the factory.

Compressed air is unlikely to work due to the design of relevant parts:



The gicleur is part # F.2.003, which can be seen to come off of part TL30, which is attached to the solenoid at one end and the bottom of the group head at the other. In order for the compressed air to work, without disassembling the machine, the air would have to enter through the hole in the underside of the group head through which the water flows and the screen screw is attached. It would then have to go back to the solenoid and then come forward. Assuming a good seal for the incoming air, which would be problematic, the air would have to go to the solenoid and come back out through the bottom part of part TL30, then go through the gicleur with sufficient force to push out the offending blockage. I think this is very unlikely since the air would probably find an easier path out through the solenoid.

Michael Teahan's suggestion was to use a very fine but rigid wire to push the offending particle out from the solenoid side. He suggested using "high e" electric guitar string wire, and trying this before giving up and disassembling the group head.

In operation of the GS/3, brew boiler water is sucked in through the gicleur, goes to the junction of the two pipes in the part labeled TL30, then (depending on the solenoid valve position) either is injected through the long part of the TL30 part, through the hole in the grouphead, to make espresso, or is evacuated through the solenoid, as at the point where the shot flow is terminated.

I'm assuming that when the machine is turned off, that the solenoid valve position is such that if you tried to force WATER, under pressure (with a bulb syringe or other approach), through the group screen hole opening, that the water would just end up in the drip tray. I guess it is possible that if you supplied current to the solenoid in the right way, you might be able to then force some water through the screen hole and have it come out through the path to the gicleur. Of course, you might end up really screwing things up, frying the brain board or electrocuting yourself in the process!! Using the guitar string, if it worked, would seem to be the simpler and safer approach

Once the particle is dislodged, the question would be, where does it go? From looking at this diagram, above, and with a quick glance at my machine, it appears to me that there are 3 possibilities:

(1) It stays more or less where it fell out, in the "neck" of the group, which would put it at greatest risk of getting sucked back into the gicleur in the not too distant future;

(2) It gets washed into the boiler, proper, and than has little risk of ending up close enough to the gicleur that it will get sucked in once again, causing a blockage; this would make it easier to flush out through the expansion valve port;

(3) It goes "forward" into the group head, near the water outlet from the other end of part # TL30; this would create a situation intermediate between #s 1 & 2, where immediate reblockage would be less likely, but could happen again after a period of time.

Given the time, difficulty, and effort involved in disassembling the grouphead, as opposed to simply removing the solenoid (which you would have to do in any event, whichever tact you took) and trying to push the particle through with a fine wire, I think that Michael's suggestion of using the guitar string deserves at least a try. Assuming that one could dislodge the particle, and then flushed out the boiler a few times (perhaps heating it up also to encourage this particle to move around and hopefully end up in the boiler proper), I think the odds of a reblockage would be real, but probably not anywhere near 50%.

ken

Now I finally know why they called it a 'banjo tube" on the old LM models

I didn't realize the water path was that convoluted. Pulling the solenoid & trying to fish/push it out definitely looks like your best "easy" option.

Ken Fox wrote:Michael Teahan...suggested using a "high e" electric guitar string

another_jim wrote:Now I finally know why they called it a 'banjo tube" on the old LM models

That's why machine makers talk about "tuning" their espresso machines.

But seriously....

1. I've always thought that this is why LM sold the Starbucks Lineas with 2mm or larger gicleurs (or no gicleurs) -- maintenance is much simpler without them.
2. You should have bought the Speedster. Its gicleur takes five minutes to remove and clean, then you're up and running again.