I'm surprised by the difference in flow rates, and that the gicleur shots actually flowed faster. I guess when the puck soaks at lower pressure, the particles swell and lock before the fines can get to the bottom.

Randy G. wrote:You have been on the forefront of such research for a good, long time and a lot of us are enjoying better espresso because of it...

As have you, Randy!

another_jim wrote:I'm surprised by the difference in flow rates, and that the gicleur shots actually flowed faster.

I'm surprised that you're surprised. We were talking about this on alt.coffee back in 2004. I didn't have the fancy graphs back then, though.

another_jim wrote:I guess when the puck soaks at lower pressure, the particles swell and lock before the fines can get to the bottom.

I think that's correct, at least for a good portion of the fines. If fewer get to the bottom, do fewer of the finest fines make it through the holes and into the cup, reducing mouthfeel?

Randy G. wrote:Thanks for the hard work, Andy. You have been on the forefront of such research for a good, long time and a lot of us are enjoying better espresso because of it...

Indeed, thanks Andy for sharing your results, despite that trying to understand them hurts my head.

AndyS wrote:Interestingly, this initial slowdown makes the shot run far faster once full pressure is reached (see graph). It also means that baristas use a significantly finer grind on gicleur-equipped machines as compared to non-gicleur machines. How that affects the resulting shot is a mystery.

You aren't kidding it's a mystery. I've been scratching my head about the importance of slowing the ramp up in pressure ever since I evaluated the Elektra A3. I documented my findings in Pressure profiles, preinfusion and the forgiveness factor:


Comparisons of the pressure profile of three espresso machines

Everything I read and intuition told me that the two gently curved pressure curves would belong to more forgiving espresso machines that produced superior shots. For this particular lineup, neither was true. Since then I've given up attempting to predict in-cup results with measurements like brew pressure or temperature profiles. They're certainly meaningful metrics to manipulate when using the same gear as you and Greg have done, but across different designs, other (unknown) factors come into play, rendering blanket statements like "preinfusion is a good thing" specious. It's got to be qualified, e.g., modifying espresso machine X by adding preinfusion reduces the likelihood of channeling and thus increases the number of superior espressos.

HB wrote:They're certainly meaningful metrics to manipulate when using the same gear as you and Greg have done, but across different designs, other (unknown) factors come into play, rendering blanket statements like "preinfusion is a good thing" specious.

Thought I'd chime in here in support of Andy's observations. I recently added a 0.6 mm gicleur to my Techno, reducing the water debit from about 230 ml in ten seconds down to about 70 ml. The Techno has the sort of preinfusion in which the pump is run for about half a second, then switched off for about a second, then switched back on again. It also, as it happens, has it on for two of the switches and off for the third, making easy A-B comparisons possible. At least with this machine, with the gicleur, on the two coffees I tried (Moka Kadir and Klatch WBC), using 14 g in an 18 g basket for about a five-second dwell with a bottomless portafilter (dwell time without preinfusion), I saw a difference. The shots with preinfusion had a slight bitterness or harshness on entry that disappeared if I didn't use the preinfusion. The difference was small but absolutely repeatable. Thought I'd throw it into the mix.

Best,
David

A few posts back, I said, in part, this:

erics wrote:Andy - Another idea would be to video a scale readout during the brew process and graph the data much as you did with the Silvia flat-liners you did previously.

Well, I emailed the mfg of the scale I use - happens to be a rather inexpensive ($15) one.

This was my question to them:

scalemodel=ibalance 601
request=RS 232 question - at what frequency can I obtain
data from the this port?, i.e. 1/s, 2/s, 4/s

Interested in using scale to measure flow (weight) of liquid water into a cup at AVERAGE flowrates of 3.0 ml per second over a, say, 45 second period. The flow through my device intentionally varies over this 45 second period and I am deserious of graphing flowrate as a function of time for the 45 seconds. Can this be accomplished - on your end?

This was their response: After checking with our engineer,the continue output is 4/S and you can set the frequency using the manual.

Just so there's no misinterpretation, the scale I was asking about costs ~ $150 and has an RS232 port however even my much less costly scale has essentially the same display update, as visually tested.

DavidMLewis wrote:The shots with preinfusion had a slight bitterness or harshness on entry that disappeared if I didn't use the preinfusion. The difference was small but absolutely repeatable. Thought I'd throw it into the mix.

David, if I understand you correctly, you're NOT saying that the gicleur produced the bitterness, but that the on-off-on preinfusion cycle produced the bitterness. Yes?

Any further thoughts on how the gicleur affected things?

I must admit to being blissfully unaware and now mildly fascinated by the restricted vs unrestricted flows.

I find fuller bodied shots coming from restricted flow - but is this down to a finer grind producing even more ultrafines that make their way into the cup altering the tactile balance? (not sure if this is a stupid question or not?)

King Seven wrote:I find fuller bodied shots coming from restricted flow - but is this down to a finer grind producing even more ultrafines that make their way into the cup altering the tactile balance?

Thanks for the very interesting observation. If I were given to wild speculation (which of course, I'm NOT ), I would speculate that EVERYTHING might be "fuller" in a restricted shot, because you have the ability to extract more with the slightly finer grind. But I could be wrong!

HB wrote:Everything I read and intuition told me that the two gently curved pressure curves would belong to more forgiving espresso machines that produced superior shots. For this particular lineup, neither was true. Since then I've given up attempting to predict in-cup results with measurements like brew pressure or temperature profiles.

Agreed, predictions based on graphs and numbers doesn't work.

FWIW, "preinfusion" and "gently curved" are relative. A four second rampup to full pressure isn't as gentle as a six second rampup (!), and there are many preinfusion types and variations.

AndyS wrote:FWIW, "preinfusion" and "gently curved" are relative. A four second rampup to full pressure isn't as gentle as a six second rampup (!), and there are many preinfusion types and variations.

I'm not sure how much difference gicleurs versus E61 pressure cylinders versus line pressure, no pump preinfusions make. But I think there might be a simple water volume difference between groups. E61s in all their versions have a lot of volume between the 3 way release point and the top of the puck, whereas the Silvia and little Elektra groups do not. I don't know about other groups. A classic E61 has around 10 seconds of dwell time; but it could be a lot of that is just taken up filling the various nooks and crannies in the group; and that the actual puck-absorbing-water time is about the same as in other groups.