Most useful visual input for effective flow control?

Beginner and pro baristas share tips and tricks for making espresso.
boren

#1: Post by boren »

Is it the bar value on the grouphead pressure gauge? The position of the flow control handle (based on testing actual flow rate in various positions)? How fast / disrupted the flow actually looks?

Thanks!

PIXIllate
Supporter ♡

#2: Post by PIXIllate »

I'm currently trying to tweak my latest flow profile. Once I get it to where I'm happy with both light and medium roast I was thinking of shooting a video to show how easy it is once you turn the theory and mechanics into repeatable measured steps.

Basically there are a few stages those being pre-infusion, pause/zero flow/Bloom, main extraction, and potentially tapering off. The idea I'm working with is that it's optimal to give the puck all the water it needs to fully saturate as quickly as possible and then give it some time to soak/Bloom before it hits the primary compression point at ~4bars, at which point you are no longer able to "pre-infuse" as the puck is compressed.

My current steps look like this:
-set FC to a position that gives ~7.5ml/sec and run for the first 4 seconds
-move FC to the 4ml/sec position until the pressure builds to 2-3bar
-close the FC and wait until ~20 seconds from the beginning of the shot
-you should have ~2g in your cup by this point if you've dialed in you grind correctly
-open the FC back up to the 7.5ml/sec position for the main extraction
- you can taper back the flow but most of the time I don't bother

Hope this helps. Someday I'll get around to shooting that video.

boren (original poster)

#3: Post by boren (original poster) »

PIXIllate wrote: -move FC to the 4ml/sec position until the pressure builds to 2-3bar
If you measured the flow rates, is the bar reading actually that important? And is it any different on the grouphead pressure gauge compared to the one on the body of the machine?

The reason I'm asking is because I bought a flow control device but haven't installed it yet. I would prefer to skip installing the grouphead pressure gauge and just keep the grouphead thermometer (which I like a lot). If knowing which flow rates correspond with which position is sufficient, then it's probably not a huge compromise to just install the flow control without the pressure gauge (and maybe refer to the one that's already built into the machine as reference).
-close the FC and wait until ~20 seconds from the beginning of the shot
Do you keep the pump working during this time?

User avatar
BaristaBoy E61

#4: Post by BaristaBoy E61 »

I don't have a flow control device on our E61 but it is direct plumbed and we do have line level preinfusion. I would say that the most important visual is what you see at the bottom of a naked portafilter.
"You didn't buy an Espresso Machine - You bought a Chemistry Set!"

User avatar
Peppersass
Supporter ❤

#5: Post by Peppersass »

Chris's instructions make a lot of sense, though I don't do the fast saturation and bloom. I do a modified Slayer shot on my highly modified GS/3 AV. I set the preinfusion flow rate to about 3 ml/sec. With a very fine grind, the puck gets saturated in 5-10 seconds and pressure begins to rise. Peak pressure occurs in 15-20 seconds. Once the peak is reached, I wait until the bottom of the basket is covered with coffee or until the first drop falls into the cup. Then I restore the flow to the GS/3's full flow rate. Once the stream gets going, I reduce the flow rate to keep the stream flowing at a steady 20-30 ml/min through the end of the shot (that's for a single.)

I use the flow meter in the GS/3 to monitor the flow rate with an external Arduino microprocessor. That allows me to set the preinfusion rate and more precisely control the flow in the back half of the shot. However, the next best thing is to use a graduated cup measure to determine the flow rate at each setting of your flow control device.

PIXIllate
Supporter ♡

#6: Post by PIXIllate »

boren wrote:If you measured the flow rates, is the bar reading actually that important? And is it any different on the grouphead pressure gauge compared to the one on the body of the machine?

The reason I'm asking is because I bought a flow control device but haven't installed it yet. I would prefer to skip installing the grouphead pressure gauge and just keep the grouphead thermometer (which I like a lot). If knowing which flow rates correspond with which position is sufficient, then it's probably not a huge compromise to just install the flow control without the pressure gauge (and maybe refer to the one that's already built into the machine as reference).

Do you keep the pump working during this time?
Unfortunately the pressure gauge at the group is pretty much essential in my opinion.The one on the machine doesn't measure pressure fast enough to allow you to respond in real time or where it matters, which is at the water/coffee interface.

Pressure is an outcome (or result) of flow. Think of it this way, you've created a high integrity puck of finely ground coffee and locked the portafilter in place. In doing so you've built a (temporarily) finite amount of space. As you start the pump and force water into that space initially the pressure does not rise as the water first fills the headspace above the puck and then is absorbed by the coffee itself (which can hold about 2-2.5 times its dry weight).

Once the coffee begins to reach saturation, that's when pressure starts to build. One of the things we learned from the Decent Espresso machines is that a puck can absorb water at a maximum of about 3-4ml/sec. In a perfect world you want all of the coffee to get wet (which begins extraction) as quickly as possible. The amount of pressure applied (I'm aiming for 2-3 bars) at this pre-infusion stage speeds up getting the bottom of the puck saturated so it reaches a state of equilibrium with the top of the puck in the shortest amount of time possible, thus avoiding a top to bottom unven extraction gradient.

So, I put a bunch of water (7.5 x 4 =30g) into the system right away and then move to a lower 4ml/sec flow so I have better control over hitting that 2-3 bar target for pre-infusion pressure to force a quick saturation of the puck. I do this usually for ~2-4 seconds which adds an additional ~8-16g of water, which means I now have enough water in the system to fully saturate the ~18g puck.

My next goal is to keep the pressure below 4 bars until I get to ~20 seconds on the shot clock. Usually I don't have to do anything once I close the flow control as the pressure hold is already setup in the previous steps. This allows for full wetting and reorganization of the puck before it compresses and no longer can swell and fill in gaps in the bed. When you have the grind dialed in this should give you about 1-2 g in your cup just as you hit that 20 second point. Then you can open the flow back up and let it hit your peak pressure. Mine is usually 8-8.5 bars.

As you can see, an accurate, fast pressure gauge is preety critical for this approach.

I have a dual boiler so there's no need for a temperature gauge on the group. With your HX machine it's a tougher choice. If it's repeatable enough that you use the same flushing routine over and over and get reliable temperatures then maybe you could do without it. I've heard some people use a T junction to have both. This is one of the reasons I never considered a HX machine.

Dick's approach is another valid one. He and I have discussed the relative merits of each before. But at that point you mastered the techniques and now taste is your guide.

boren (original poster)

#7: Post by boren (original poster) »

My machine isn't HX, it's has a single boiler and a PID, but temperature stability is not where I want it to be. For this reason my next machine will likely not have an E61 group head.

As for pressure, if the puck is not 100% saturated but only 80% (before starting the main extraction), do you expect that to result in a difference that can actually be detected, e.g. in a triangle taste test? Would overshooting and starting the main extraction a couple of seconds too late negatively impact the cup?

PIXIllate
Supporter ♡

#8: Post by PIXIllate replying to boren »

If you have a single boiler PID and don't switch into steam mode you should be able to achieve excellent temperature repeatability. E61's are very predictable and a declining shot temperature is not necessarily a bad thing.

Yes, I detect large differences both by taste and with a refractometer when the puck is not fully and evenly soaked and extracted. I don't see any way around having a group pressure gauge if you really want to build a reliable, repeatable flow controlled shot profile.

I have ~30 years of blind tasting experience with wine and frankly coffee is a pretty simple beverage compared to that. It's exponentially more difficult to prepare properly though. Your experience may be different than mine and not everyone is the same kind of taster that I am. Keep in mind I'm always aiming to get the maximum out of everything. Not everyone enjoys approaching things in that manner.

User avatar
SteveRhinehart

#9: Post by SteveRhinehart »

While I know and can verify that my flow control valve changes water flow into the group, in practice I can see that it does not change the reading on the group manometer except for a very small window near zero flow. Visually, the stream out of the bottom of my basket will change over the course of the shot, but I know for certain that I can't consistently spot the difference between 2.5 g/s and 1.5 g/s.

So in the spirit of the original question, here's what I use for logging flow control:
1. The valve position. At one point I recorded the unladen water flow rate across the range but in practice I'm not so much concerned with that as I am with output flow. For me, 9 o'clock is fully closed and 12 o'clock is fully open. I assume that it is more or less a continuous change in flow from closed to open, and on that I assumption I tweak the valve according to what I would like the flow rate to do.

2. Lunar 2021 flow indicator. Just for at-a-glance flow measurements when I'm looking at my mass yield in the cup. I can check that it is either increasing, decreasing, or holding steady according to what I was intending with the last change of flow rate. The dots do correlate with a numerical flow rate, but I am typically only glancing at the scale, too quickly to really determine what the exact g/s might be.

3. Smart Espresso Profiler app. I don't use it often, but the flow graphs on the SEP app are really useful for getting a hang of flow adjustments. I don't really like keeping my phone out while I make coffee though so it doesn't get used as much as the above.
PIXIllate wrote:I have ~30 years of blind tasting experience with wine and frankly coffee is a pretty simple beverage compared to that.
Total aside - I'd always heard coffee is more complex than wine, with like 3X or more identifiable flavor compounds. I'm an expert taster in neither, so I won't claim to know heads or tails either way.

boren (original poster)

#10: Post by boren (original poster) »

SteveRhinehart wrote:While I know and can verify that my flow control valve changes water flow into the group, in practice I can see that it does not change the reading on the group manometer except for a very small window near zero flow.
That was also my experience with a friend's flow control unit, but I just assumed there may be something wrong with his grouphead pressure gauge. If that's the normal behavior, then how does one even make good use of the gauge when every tiny movement of the handle make such a big difference?

The method you're using with the Acaia scale makes sense. Do you think the flow rate display of the (much cheaper) Timemore Black Mirror Nano be usable for this?