robertdi8 wrote:Hey Ken,
I think your experience regarding "forgiveness" is really what I was trying to imply in my post above. When I described the changes in the "automatic pre-infusion", they were not the only brew characteristics that changed. I too have found the need for more frequent adjustments to my shot routine.
My vendor has just installed the larger diameter gicleur...I wonder if this will exacerbate these chnages?
I think I'll get him to throw in a standard 0.6 just in case.
As well, he put on the new longer hose so i too look forward to the "sounds of silence"!!!!!
Dave.

I'm no longer happy with the variation in extractions I'm getting from the machine. Initially I didn't notice it, but I've had occasion recently to pull a lot of back to back shots, and the kind of inconsistency I'm getting is just too obvious to miss. I did not have these problems with either of my old Cimbali Juniors. In the case of the vibe machine, one has the general "forgiveness" of a vibe pump with its slow pressure ramp up. In the case of my old rotary machine, it had a delay timer with line pressure preinfusion mod installed, also providing forgiveness.

So, I'm going to do an Eric S gigler/filter mod, and see how that comes out.

ken

erics wrote:For those who have replaced (or had replaced) the 0.60 mm gicleur with the 0.80 mm gicleur, the pressure presented to the coffee during a typical shot is now anywhere from 0.50 bar to 0.70 bar higher than it was in the original configuration.

Hmmm...using a double basket, I typically observe flow rates of 1ml/sec to 1.5ml/sec after the initial wetting period. According to your graph, this would suggest a pressure difference of 0.2 or 0.3 bar between the two gicleurs.

erics wrote:Then, perhaps, a little recalibration is in order.
From - WBC Procedure for Measurement of Brewing Water Temperature
...75 ml of water is collected in an elapsed time of 25 seconds...

1. IIRC, I had a conversation with Greg or Barry about that, because the figure seemed quite high. I don't remember what their explanation was, but it made sense -- for the WBC.

2. The WBC procedure is a conservative standard for testing machines, but it doesn't necessarily reflect actual water flow rate.

3. IIRC, from watching the flowmeter display on my Silvia, the flow rate for the first few seconds is about 3-5ml/sec, but soon settles down to a third or a quarter of that. It's the latter flow rate that is relevant if one wants to tweak brew pressure.

4. I agree that your graph is very handy for picking out whatever flow rate is appropriate.

5. Before Chris T comes in and says so, it is of course recommended that one tweak brew pressure by taste, using the numbers as only a rough guide.

erics wrote:As far as the weight of water in the coffee, that one I can personally confirm over many samples

A typical double shot for me might work out as follows:
16g dry coffee weight
31g spent puck weight
26g espresso weight

3g of dry matter is extracted into the cup (~19% yield). That leaves 13g of dry matter behind, so 31-13=
18g of water absorbed by the puck.

26g espresso minus 3g of coffee solids means 23g of water are in the beverage.

Perhaps 10g of water are lost to drain when the 3-way opens (this figure varies greatly depending on the machine, basket, dose, etc).

18g + 23g +10g= ~51g of water used to pull a double shot to my style, on my machine. YMMV.

No, the flow rate on my replacement GS/3 hasn't changed . But today I had occaision to open the group cap after almost nine months of use, and here's what I found:



The dreaded verdigris. The TL30 tube was pretty-much encased in the stuff all the way back to the 3-way, and you can see a thick ring of it at the margin with the top cap. Also note the loose bits of verdigris lying on the floor of the group. The verdigris was quite hard. Some grains came off the ring when scraped with a fingernail, but nothing came off the TL30 tube. It was really baked on.

I don't know what the black specks on the floor of the group are. They were very soft and just smeared and disappeared when rubbed between my fingers, almost like charcoal. I do have a carbon filter after my cation filter, but during installation I flushed it thoroughly before allowing water from it to enter the GS/3. I think the black specks must be something else.

Note the fragment of paper gasket at the lower left. The gasket was dry and baked on to the group and underside of the cap. I had to pry most of it off in small pieces. Luckily, I had a couple of replacement paper gaskets on hand for just such a contingency.

Here's the underside of the group cap:



The central plating is in excellent shape. The dark spot in the middle is just a smudge and came right off. Same for the dark spots in the upper right center. Around the perimiter you can see where the ring of verdigris shown in the first photo met the top cap. What you can't see is that the chrome plating has flaked off in several places just below the rubber gasket (not visible in the photo), right where the ring formed. It didn't look serious enough to buff off the rest of the chrome. Note that the portion of the group cap with the screw holes is actually shiny bright chrome, not brass-colored. That's a reflection in the photo. That surface was intact with no damage to the chrome. The group cap cleaned up very nicely.

The bottom piece of the group had some coffee and dirt in the recesses, but was otherwise OK and cleaned up easily.

Here's a shot showing how the verdigris went all the way back to the 3-way valve, where it was thickly caked:



You can see a few loose bits of verdigris. I think these came off when I removed the group cap.

Given the extent of the deposits, I decided to remove the TL30 tube and try to clean it and the group as thoroughly as possible. It wasn't all that hard getting the tube out. I removed the solenoid valve so many times in my original GS/3 that I could have done that in my sleep. I did mark the nut on the drain tube so that I wouldn't overtighten it and break the flare (like I did on my original GS/3 .) The only tricky part to getting the tube out is to push the nut that attaches the tube to the dispersion block out the solenoid hole while the straight portion of the tube is in the hole. There's no way it'll go through if it's at the curved portion or at the flare. Similarly, when putting it back in you have to put the curved part in first and then push the nut through while the long straight part of the tube is in the hole. I pulled the nut forward with an allen wrench. It's a little tricky, but no big deal.

Here's the tube caked with verdigris:



I examined the gicleur and was glad to see it did not appear to be obstructed:



Sorry for the blurry photo, but you get the idea. I did notice a couple of flags of teflon tape hanging loose at the base of the gicleur and removed them. Seems to me those could detach and get caught in the gicleur. The flow rate fluctuations on my first GS/3 seemed to be caused by something intermittent like teflon tape blocking the gicleur to different degrees at different times.

It took a long time to clean the TL30 tube. I soaked it in vinegar for about an hour, then lightly scrubbed it with the scrubbie side of a sponge. Then I took a real Scrubbie to it. I also tried copper polish, but that didn't do much. In the end, I had to use very fine sandpaper to take of the most stubborn deposits. Here's the result:



Now, you might think there are still a lot of dark spots left from the verdigris. But those are actually shiny plated areas. This photo shows the largest such area:



I'm at a complete loss to explain why those areas look plated. Any ideas?

Here's the group after cleaning:



Here's the group with the cleaned TL30 installed:



Lookin' pretty nice. And a closeup:



The view back to the solenoid valve:



Finally, debris left in the vinegar bath. It wouldn't have been good to have this stuff swirling around the boiler:



After re-assembly, the flow rate was about the same as before, 420-450 ml/min.

Until today, I used only cation-softened water in my GS/3 (5-micron well filter -> 2-micron sediment filters -> profession cation softener -> carbon filter.) Today I installed an Everpure Claris system for evaluation. Now that I know how easy it is to pop the group cap, I'll report on conditions inside the group over the coming months. Based on Nicholas's experience, I'm not optimistic that the Claris formulation will be any better in terms of verdigris formation. But it'll be interesting to compare.

Does anyone know if LM has made any changes to the TL30 tube or group cap plating to make them more resistant to deposit formation / flaking?