Interesting observation: No preinfusion = considerably cleaner taste

[MOSFET]

At least on my machine. I am using a primitive form of preinfusion where a time delay relay delays pump turn-on, adjustable from 0.1 to 10 seconds. So unregulated mains pressure is used to preinfuse. I have 6 bars of mains pressure, so the pressure is high during preinfusion and coffee sometimes starts to drip out before the pump engages. When I turn off the preinfusion a lot of bitterness and muddy flavor goes away. So I'll either not preinfuse or add a regulator. But in my opinion preinfusion may help prevent channeling and probably does not improve flavor otherwise. Now that I have lots of tamping practice, a good tamper, and a good basket, I don't get channeling anymore, so I don't consider the preinfusion necessary. I'm sure this has been discussed but I haven't followed much discussion on preinfusion.

Keith

[another_jim]

I've always had that suspicion, judging by the differences in E61 versus short dwell machine shots.

However, the effect may be blend dependent. When Ken and I did the comparison on his Junior (2.5 bar, 5 second preinfusion), we could not find a fall off in taste in the DP Harar and DP heavy blend we used. There was a hint of this with the WP heavy blend, but it tasted so lousy no matter how it was done, that differences were hard to taste, and we soon abandoned using it.

[Ken Fox]

At least on my machine. I am using a primitive form of preinfusion where a time delay relay delays pump turn-on, adjustable from 0.1 to 10 seconds. So unregulated mains pressure is used to preinfuse. I have 6 bars of mains pressure, so the pressure is high during preinfusion and coffee sometimes starts to drip out before the pump engages. When I turn off the preinfusion a lot of bitterness and muddy flavor goes away. So I'll either not preinfuse or add a regulator. But in my opinion preinfusion may help prevent channeling and probably does not improve flavor otherwise. Now that I have lots of tamping practice, a good tamper, and a good basket, I don't get channeling anymore, so I don't consider the preinfusion necessary. I'm sure this has been discussed but I haven't followed much discussion on preinfusion.

Keith

In the testing that Jim S. and I did, with my Cimbali Jr. D machine, we found that you needed 3 bar, at least, to preinfuse the entire puck thoroughly during a 6 second period. Anything less we tested failed to accomplish this goal.

Being as "typical" regulated pump pressure is around 9 bar, I would argue that 6 bar for preinfusion isn't "preinfusion" at all, but "infusion." You may be demonstrating that 6 bar, at least with your machine, your beans and your taste --- isn't a very good espresso infusion pressure, something that many of us would guess if asked, even though we probably haven't tried it.

My impression is that 6 bar preceding 9 bar does not accomplish much. Also, if your machine incorporates some degree of preinfusion in its design, the benefit of adding a delay timer is probably very limited.

ken
p.s. I would also argue that a moderate amount of barista skills may prevent channeling on some machines, and may be less successful on others. In a rotary machine which has a near instantaneous pressure ramp up, such as mine, moderate skills are not going to cut it consistently, although great skills likely would.

[cannonfodder]

Interesting. I get the exact opposite on my Faema. I find my preinfusion shots better than my no infusion shots. I have a similar setup, a delay on make 0-10 second relay. I had extractions problems without it. My machine has zero preinfusion and was just hammering the puck leading to side channeling or channel jets. I have no channeling or uneven extraction problems on my other two machines.

I run a LM ridgeless double basket with a 17g dose and a Reg convex tamper. I am running a regulated mains line (half inch) at 3 bar for 5.5 seconds. That infuses the entire puck and pin head sized drops are just emerging from the portafilter when the pump engages (9 bar). My channeling issues went away and the quality of the shot improved.

[another_jim]

I run a LM ridgeless double basket with a 17g dose and a Reg convex tamper. I am running a regulated mains line (half inch) at 3 bar for 5.5 seconds. That infuses the entire puck and pin head sized drops are just emerging from the portafilter when the pump engages (9 bar). My channeling issues went away and the quality of the shot improved.

One way to distinguish the indubitable improvement in percolation one gets with preinfusion from the perhaps deleterious effect it has on the taste is to try the experiment with singles.

A single basket is basically a chopped off cone (some have a step in them). This means there's a huge triangular wedge of grounds over every part of the filter diameter, and side channeling is basically impossible. If the no preinfuse singles taste equal or better, then the improvement from using it is due to better percolation, rather than a better dissolving of the coffee into the water.

I added the needle valve preinfuse to the Elektra mainly because doubles were very hard to level well enough to get perfect bottomless shots with the 1.5 second dwell time. Extending it out to 4.5 seconds hasn't hurt the singles, and has benefited the doubles.

[pdx]

My machine allows line pressure preinfusion (4.75 bar at my machine) & has a .6mm gliceur. I've not actually seen benefit from preinfusing unless I really flub the tamp somehow. I get better shots from well-built pucks with full pressure (mitigated by the .6mm jet.) My impression is that an eased pressure ramp-up is more important than a sustained low pressure exposure before infusion. I would restrict flow before modifying pump behavior.

[cannonfodder]

Funny you should mention singles Jim. The tamper road show arrived at my home last night. To reduce the over caffeinated effects I dug around and found the single basket from my Isomac and popped it into the Faema. More tamper testing with lower caffeine intake with singles. With the pressure and preinfuse set the same, I am not getting as dynamic a cup from the single basket. In fact it is coming across as all body and little flavor range. Almost like a SO and not a blend. With 30% Harar I should be getting much more fruit and acidity.

Prior to putting in the pump delay preinfuse, I had resolved the channeling issues. This machine is much less forgiving than anything I have worked on (Isomac, Gaggia, and La Marzocco). I had to really concentrate on my skills to get good shots. The adjustable relay made the forgiveness factor better. Whether my cup quality increase was due to better percolation or preinfusion can be debated forever. However on my machine, with my modest barista skills, I got a better cup.

I was going to replace the group jet as pdx suggests. I however have not been able to find a group jet for this machine. It was Spanish manufactured for Faema and appears to be of 'unusual' design. When I rebuilt the machine I used parts from two different Faema models, but the group jets are unique.

[MOSFET]

Funny you should mention singles Jim. The tamper road show arrived at my home last night. To reduce the over caffeinated effects I dug around and found the single basket from my Isomac and popped it into the Faema. More tamper testing with lower caffeine intake with singles. With the pressure and preinfuse set the same, I am not getting as dynamic a cup from the single basket. In fact it is coming across as all body and little flavor range. Almost like a SO and not a blend. With 30% Harar I should be getting much more fruit and acidity.

I wonder if the overall temperature has shifted up simply due to there being less thermal mass (smaller puck). Maybe if you shave off a degree the fruit will come back.

Keith

[another_jim]

Funny you should mention singles Jim. The tamper road show arrived at my home last night. To reduce the over caffeinated effects I dug around and found the single basket from my Isomac and popped it into the Faema. More tamper testing with lower caffeine intake with singles. With the pressure and preinfuse set the same, I am not getting as dynamic a cup from the single basket. In fact it is coming across as all body and little flavor range. Almost like a SO and not a blend. With 30% Harar I should be getting much more fruit and acidity.

I get a milder shot from a single, but not like you describe. Given the Faema style single basket, I use around 11 grams in it versus 17.5 in the double; so perhaps what I'm doing aren't really singles

Prior to putting in the pump delay preinfuse, I had resolved the channeling issues. This machine is much less forgiving than anything I have worked on (Isomac, Gaggia, and La Marzocco). I had to really concentrate on my skills to get good shots. The adjustable relay made the forgiveness factor better. Whether my cup quality increase was due to better percolation or preinfusion can be debated forever. However on my machine, with my modest barista skills, I got a better cup.

I don't know how these non-stop/e67 groups are drilled. Micheal Teahan thinks the sloped water channel to the bell in the E61 would be a problem unless one has a fairly narrow jet (0.5mm) and the long preinfusion it uses. The E61 dispersion block is also very small (the size of a quarter) rather than covering the whole puck. So the wetting is bound to start out uneven. If the later Faema groups also use this, one would need to slow down the initial water flow.

[Jepy]

To me, pre-infusion to soak the puck was never the goal, even though I have experimented with doing this at different pressures and grinds. Because I like the flavor from triple baskets (coffee dependent 16-18 grams loaded), there is quite a space between the dispersion screen and packed coffee, so trying to respect Pascal's law, easing(currently set at just under 2 bar) the beginning of the shot just enough to fill this space before applying more pressure.
The size gicleur I use is just big enough to not force a lag time of full pressure of more than 1/2 second(the.6's have no faster then a 2 second ramp on my machine). The reason for this is I had found some coffees IMO to be better with a faster ramp up after filling this space. Still though, I find my favorite blends to have a ramp of around 2-4 seconds.
I've found an interesting pressure profile that I've been running with great results, it goes like this:
4.5 seconds @1.95 bar,
3 second ramp to 8.3 bar-hold for 6 seconds,
4 second ramp down to 4.15 bar-hold 2 seconds,
2 second ramp back to 8.3 bar,
20 second ramp down to 6.9, I usually pull it just before it gets to 6.9
The puck is very soggy, and spongy, which may go against what you might think as being correct.
The body and mouth feel is amazing, with even the liquid portion having a thicker mousse like viscosity. My theory is that on the first ramp down of this profile, the coffee puck expands letting the water contact a little more of the coffee, so that on the next stage of pressure, a little more flavor is allowed to be extracted. The final slowly decreasing pressure at the end may help in making these shots creamier by leaving some of the bad behind.
Sound crazy?

BTW, the coffees I've run at this profile. Light house Roasters espresso blend, Stumptown Hairbender, Ecco cafe Brasil Cachoeira Yellow Bourbon

Andy S,if you're out there, try this on your set up, let me know what you find......