Can flow control make the world a better place?

Need help with equipment usage or want to share your latest discovery?
User avatar
Ursego

#1: Post by Ursego »

In one video, I heard that flow control can correct grind size errors. In another video, there was information that flow control can be used to improve the taste of beans roasted a long time ago. I'm a total newbie to flow control, and I'm wondering - how? What profiles to use and how do they work?

Thank you!


User avatar
Peppersass
Supporter ❤

#2: Post by Peppersass »

Those so-called benefits are not the reason flow control has become popular. It's really a consequence of a trend toward light roasts that began a little over 10 years ago.

Flow control is primarily useful for pulling shots from hard-to-extract light roasts. Typically, when light roasts are pulled using the machine's stock preinfusion and flow rate, they underextract and taste sour. You can try to grind finer, reduce the dose or pull longer to get more extraction, but with many light and very light roasts this doesn't work or results in a diluted shot. And there's a point when the grind gets so fine that it chokes the machine and you get less extraction or none at all.

With flow control, you can slow the preinfusion to the point where you can grind ultra-fine without choking the machine. The long, slow preinfusion loosens the puck, allowing water to penetrate. It also increases the contact time between the water and the grounds. Depending on the machine, you can also gradually reduce the flow rate after peak pressure has been reached in order to prevent the shot from flowing too fast as the puck loosens even more. This increases contact time and improves extraction.

IMHO, flow control has little or no benefit for medium-to-dark roasts, though some have said that some medium roasts extract better with some extension of the preinfusion time.

PIXIllate
Supporter ♡

#3: Post by PIXIllate »

I've found all roast levels benefit from the flow control technique I use. The variable is how long the hold stage is, or if there is a hold stage at all with darker roasts.

User avatar
Ursego (original poster)

#4: Post by Ursego (original poster) »

Peppersass wrote:IMHO, flow control has little or no benefit for medium-to-dark roasts
In fact, what you wrote before, is applicable to dark roast:
Peppersass wrote:you can also gradually reduce the flow rate after peak pressure has been reached in order to prevent the shot from flowing too fast as the puck loosens even more.
I have read (that was very interesting to me!) that, during the brewing time (especially when extraction is nearing its end), the coffee tablet degrades and becomes friable (because the more time passes, the more substances have already been washed out of it). As the constant pressure of 9 bar continues to act on the puck (which is no longer able to resist the flow as effectively as in the beginning), the tendency to form channels increases. Therefore, there is a recommendation to reduce the pressure during the extraction (the "spring lever profile").

This advice (as well as the Slayer shots for light roast) applies to any beans - even ones that were roasted recently and have the correct grind size.

User avatar
Jeff
Team HB

#5: Post by Jeff »

Ask a lever-machine aficionado and you'll hear that E61-style flow management is just an ugly hack for what lever machines have been doing well for decades. Controlled and controllable puck soak ("PI"), declining pressure profile during extraction. With a manual lever there is also the ability to compensate for a slightly "off" grind size during the shot. As there is more work done with temperature profiling, it may turn out that the descending-temperature profile that is gaining favor for light-roast espresso is another "yeah. my lever does that" thing.

My experience agrees with both of the above posts. Light-roast espresso benefits from a controlled soak of some duration. Classic espresso blends often end up with more of what I find unpleasant about them in the cup if you extend the soak. There's a continuum and a lot will depend on roast quality, not just level.

Slayer-style shots are another topic. I acknowledge that Slayer was instrumental in the creation of the concept of a flow-managed, pump-driven machine for the comparatively dark-roasted coffees that cafes were using at the time (2009-ish). Those coffees are probably still popular in cafes, but are far from the light-roast, SO coffees that are being pursued today by many enthusiasts. Slayer-style shots aren't in favor among the light-roast enthusiasts that I converse with or read experience of. Slowly infusing water into the puck means that parts of it have begin to extract long before others.

User avatar
Peppersass
Supporter ❤

#6: Post by Peppersass »

Jeff wrote:Slayer-style shots aren't in favor among the light-roast enthusiasts that I converse with or read experience of. Slowly infusing water into the puck means that parts of it have begin to extract long before others.
It depends on what you mean by Slayer-style shots. The modifications I made to my GS/3 AV are partly modeled on the Slayer design. I replaced the stock rotary pump with a gear pump and installed a needle valve with a solenoid bypass valve in the cold water input path.

That said, my mods are a bit more flexible than the Slayer design, and my preferred shot method isn't exactly the same as the Slayer. Once I had the mods in place, I used the flow rate recommended for the Slayer: 50ml-60ml in 30 seconds, or about 1.6ml-2ml per second. Quite slow. But unlike the Slayer, I can change the gear pump speed at any time during the shot. I found that although the gear pump, which doesn't have a bypass valve, naturally reduced the flow rate after max pressure had been reached, shots were improved if I reduced the gear pump speed even more during the back half of the shot, effectively increasing contact time.

Some time ago, I read a post by a Decent user that postulated what Jeff is saying, that the very slow preinfusion rate of the Slayer causes uneven extraction in the puck. So, I increased my preinfusion flow rate to what the post recommended, which was 3ml-4ml per second. I adjusted the grind accordingly. The result was a definite improvement. I can do a no-flow soak if I want to, but haven't had the need.

So, I'm effectively doing a Slayer shot with a slightly coarser grind, faster preinfusion and gradually declining pressure/flow during the back end of the shot. But I feel like the overall methodology still follows the key elements of the Slayer design.

User avatar
another_jim
Team HB

#7: Post by another_jim »

Jeff wrote:Ask a lever-machine aficionado and you'll hear that E61-style flow management is just an ugly hack for what lever machines have been doing well for decades.
I switched from lever to hack. Mostly, it makes no difference. But you can get higher volumes if needed; you've got the 2.5 bar steam boiler for fast cappas; and it's much less effort. So for us non-purist hacks; it's a pretty good change.

Currently enjoying 6 gram singles from a medium roasted blend I'm playing with to reverse engineer Italian bar blends; as well as two light roast naturals, a Guji and the Elida Catuai. Flow control works well for both to get sweet little espresso morsels. At this level of balance and extraction, the lighter and darker flavors are just pleasant changes of nuance, rather than anything earth shaking, exactly how I like it. It's very long and good way from wrestling with overstuffed 18 gram baskets.
Jim Schulman

PIXIllate
Supporter ♡

#8: Post by PIXIllate »

Yeah, it all comes down to getting different machine types to act like a lever.

I actually think it was me who suggested to Dick that an increase in flow rate based on a pucks maximum ability to absorb water at a rate of ~3-4ml/sec would be better than a "Slayer" flow rate.
La Marzocco Linea Mini gicleur - does it really matter if you switch to 0.6mm?

Of course this information came out of the Decent user dataset. This was also the basis for my e61 flow control technique.

User avatar
Jeff
Team HB

#9: Post by Jeff »

Jim's experience hints at a few things that seem to often get lost

* There are several "tasty points" for a given bean and roast

* People's tastes are different, my "tasty points" may not be yours

* There are many different kinds of gear that can get you to many of those, each with its own advantages and disadvantages

If you stick to "18 g in, 36 g out, in 25 seconds" you might be close to one of those tasty points. Close, not spot on. Unless you're willing to explore outside of that immediate neighborhood, you're unlikely to find any of the others -- Jim is at less than half the "right" dose for his tasty point. To get to a different "local maxima", you often need to go through some poorer shots along the way.



Above image licensed under Creative Commons Attribution-ShareAlike 3.0 License -- http://mathonline.wikidot.com/local-max ... and-minima

PIXIllate
Supporter ♡

#10: Post by PIXIllate »

I think I hit a local Maxima this morning with the Tim Wendelboe Karinga Kenyan! :D

18/33g in 42 seconds. Juicy acidity, sweetness, dark berries and a nice but not overwhelming floral underpinning.