Micky_G wrote:What a fantastic and epic thread! And to think, I started it all by simply existing!

In all seriousness, I'm extremely proud of my younger brother for taking up the challenge of creating a consistent and amazing espresso and for taking the time to share it the way he has, in the fashion he has. Sadly, his exposure to my burr grinder (which I still have, though the terrible krupps espresso machine is long gone) was such a tiny and insignificant thing to me, that I completely missed the opportunity to really engage with him on how amazing a thing extracting coffee from ground up, nearly burnt seeds is.

I'm hoping to find some inspiration and will be on the lookout for an affordable machine in the not too distant future.

Learning lots here, keep the discussion going all!

Hey, you had to more than just exist; you also had to appreciate really mediocre coffee!

Thanks for taking the time to read my ramblings and chime in! Documenting the journey and the continued learning has been so much fun. There's nothing more rewarding than realizing just how much I don't know in front of a whole group of folks whose own knowledge and experiences outnumber my own at least a hundred to 1. It's a blast

AssafL wrote: A single, double. triple all need specific volumes of water and given a particle size - specific amount of time for the water to permeate and saturate the puck. The key is to avoid leaving a dry section at the bottom (that will either seal the puck or spritz). The dry part will - at the very least - skew the extraction yield - or altogether ruin the pull.

AssafL wrote: One simply cannot treat different basket sizes as a "simple" parameter - the basket size and dose work in concert with grind setting, PI volume, PI time, & pump pressure, to ensure proper extraction (EY). It all has to change and the main reason is pre-infusion.

This is a compelling argument. Playing the devil's advocate, I would ask about all the excellent shots pulled on a stock Linea EE/AV or other instant-on machine, but I think we'll get there further down.

AssafL wrote: So the following observations may be true:
1. Fixed (or limited range) "volume" PI machines - all the e61, most gicleur'd pump machines, etc. will have an optimum dose (or a narrow range of dose) where the results will be optimal.

I think a more precise observation may be that for any given dose on such machines, there is a very narrow range of grind to yield acceptable results and there is likely an optimal dose where these machines are more forgiving and easier to extract well over a wider range of grinds. This optimal dose is likely the one to yield best results, but many coffees extract differently, and it may be that the sweet spot for a given bean/roast exists outside the sweet spot for the machine, making it a specifically taxing chore to dial in these machines under such circumstances. Better grinders, more consistent prep, etc... all maximize the barista's ability to work within the operating window of the machine and help them find the overlap between bean and machine. I surmise that this overlap exists more often than not, but it may be a very narrow sliver of the Venn diagram of extraction, and is not likely to give you the best unless, as you say, you nail the dose... I don't know though. All of this makes it sound like it's really challenging to get great results and if that were the case, you'd think there would be a whole website dedicated to folks coming together and trying to figure out how to get consistently great espresso out of their equipment...

AssafL wrote: It is therefore an abuse of an e61 machine to use it with a 21gr dose. Unless one fits a finer gicleur, modulates a volumetric pump or otherwise changes the working point to the system to dump a lot of water - slowly - on the puck - before the PI chamber pressurizes.

Not sure I would go that far unless we also demonize pulling doubles with an E61. I'm not enough of a history buff to know, but weren't the Italians still pulling singles in the heyday of the E61? I'm not sure how to accurately measure the volume of water delivered to the puck at low pressure since I dont have my E61 any more. One could do a fairly decent job estimating it by examining pump curves at the lifting pressure of PI valve and noting the water debit of the group at that pressure, but the chamber being pre-gicleur is problematic for bench top calculations. We can calculate how long it takes to fill the chamber, but the volume through the gicleur is dynamic, and dependant on the puck. To your point, a 7g single only needs 14g of water to saturate, but needs a significantly tighter grind to generate 9 bar back pressure, so 14g of water into a 7g single puck might take less time than 30g into a 15g double, but I don't know it doesn't actually take longer. Furthermore, isn't some PI better than none?

As an aside, this whole concept may help explain why the VST "The Single" 41mm single dose basket is so successful. The reduced exposed surface area lessens the compressive force of the water while also increasing the puck resistance in the process, allowing a grind that is more in line with doubles...

AssafL wrote: That is not to say that mistakes don't sometime come out tastier (espresso is all about myopically focusing on localized maxima in flavor). If one over-extracts due to a channeling in a dry area of the puck (insufficient PI) one should really dose less, or grind coarser and fully percolate at the preferred extraction.

I think this summarizes our collective thesis statement quite well. "The best extraction made without [complete] preinfusion is the one that achieves locally even extraction throughout the shot, at the likely expense of being extracted less than would be ideal." That is to say that grinding coarser to avoid choking the shot is a compromise that is not needed when one fully saturates the puck prior to pressurizing. One can still get consistently good and even great extractions, but they are leaving the very best the bean (some beans, at the very least) has to offer on the table by making this compromise.

AssafL wrote: NB2 - 58mm. Larger portafilters have more force applied to them so it is clear why home machines prefer smaller filters (smaller pumps and cheaper bayonets). But why do larger portafilters (usually) perform better? My guess is more uniform PI given the fingering flow you show above. In fact, having pulled many times on a la Pavoni (48mm basket), i'd reflect that fingering flow in the puck makes getting PI right difficult.

That's an interesting viewpoint/observation. I would have imagined that with the lower forces on the smaller baskets that compressive choking would be less of a concern, and as such, full pressure could be reached somewhat earlier than with a wider, shorter puck. In fact, a larger H dimension in our filter cake analogy leads to a lower differential pressure across each representative slice of the puck and could arguably result in a more even extraction through the depth of the puck. I believe Dalla Corte subscribes to this notion. It would be interesting determine why it is, exactly, that 58mm is the commercial standard. I suspect it is a silly reason that holds zero scientific rationale when put under any real scrutiny. As to why they perform better, I think it's a numbers game. More 58mm portafilters, more folks experienced with them, more data suggesting better results?

Cheers!

- Jake

ira wrote:As a wild guess I would assume that one should be able to get an identical shot from essentially any diameter basket if the puck thickness is the same and the pressure profile and flow per unit area is the same. Which means same taste but smaller shot for smaller diameter basket if all else stays the same.

Ira

This got me thinking about "flow per unit area" being the constant and I looked around a bit to see what I could find. Note my correction to the above post: (IMS makes "The Single" and it takes a 58mm tamper. I was thinking of the VST/LM single basket that takes a 41mm tamper...)

Jake_G wrote:...the VST "The Single" 41mm single dose basket

As it turns out, 41mm is exactly half the area of a 58mm basket. 58mm double, 41mm single. But what about the poor triple? We don't have access to a 71mm basket, but that's what it would take to get a triple with the same puck height as a 58mm double. We could play the game a little differently and claim that a 58mm triple is "the standard", which gives us a 47mm double (47.35, to be exact ) and a teeny, tiny 33.5mm single.

So, the question remains. Would these make identical shots?

I strongly suspect that the cooled off shots, brewed to the same brew ratio would be nearly identical. However, thanks to the genius of Schulman et al, and the judicious application of the scientific method found here, we have strong evidence that the result in the cup, at the time of the extraction could be wildly different...
While this study was produced to identify the difference different baskets produce in taste and technique, there are some interesting parallels to the idea of uniform puck height. The violet diagonal line running from the lower left, up and to the right represents a constant brew ratio of 1:1.5. The clustered lines running up and to the left represent families of baskets with similar geometry. For all shots (there were 6 prepared per basket in each family) the beans, grind and 25 second shot length were all held constant. As the dose for each basket was increased, the brew ratio moved from normale to ristretto.

There are a few things that stand out to me on the graph above, but probably the one thing most relevant to Ira's comment is the three lines (green, black and blue) graphed for the double baskets. All 3 are 58mm baskets on top and have 43mm hole patterns, all three have the exact same beans, ground exactly the same. Yet, if you draw a vertical line on the chart around 15g dose, you get a yield varying all the way from 14g with the MF double baskets (green) to 26g with the LM double baskets (blue), with the Faema baskets splitting the difference at just over 20g. Same grind, same dose, drastically different output. What gives?

Not having the baskets around to take measurements on, I see two possible scenarios that explain this:

1) It is the hole design of the basket that impacts the flow, and my earlier statement would be wrong.

Jake_G wrote: If all the flow can travel through one 0.5mm gicleur without more than a half a bar pressure drop, how could there be any appreciable pressure drop at all across the many similarly sized holes in the bottom of a basket?

Dead wrong. I struggle with this. If feels like the at the flow rates of extraction, the basket hole pattern itself can't impact the overall resistance of the puck appreciably, but I've been very wrong before...

2) It is less a function of hole size and more a function of basket shape, which changes the puck height required to achieve a given dose that impacts the flow of a given basket. This could be the case, but I don't know without measuring baskets that I don't have.

However, I do have a stock "102" 14g "precision" Rancilio basket as well as a ridgeless 14g basket out of an old Gaggia machine, which seems to have Faema-like hole sizes and spacing. Doing some quick and dirty measurements, it turns out that while these two baskets are the same height, the Gaggia basket has more capacity. The Rancilio basket tapers just after the ridge down to the 43mm hole pattern at the bottom, giving up a fair amount of capacity in the process. The Gaggia basket is straight-walled until a large radius curves in to the hole pattern.

The result of this is that the height a 13.5g puck would be considerably different in each of these baskets, with the Gaggia basket losing a few milimeters in puck height to the narrower taper on the Rancilio. As such, given the same grind and dose, the Gaggia would pour faster than the Rancilio and one would be tempted to tighten the grind, which would change the result in the cup. However, if one instead increased the dose to attain the same puck thickness and increased the yield accordingly, I think (and plan to try this and verify!) that you would get a larger but otherwise identical espresso out of the Gaggia basket compared to what comes out of the Rancilio.

If the above theory proves correct, to truly test this theory of puck height uniformity giving us identical results, I need to somehow mount a La Pavoni 49mm basket into my portafilter and then compare a 58mm triple to a 49mm double with doses adjusted to give the same puck height. I'm thinking a backflush disk bored out to fit should be able to support an LP basket with a flattened rim and an o ring to seal... I should then be able to seamlessly go between a 21g 58mm triple and a 15-ish g 49mm double with no adjustment whatsoever other than yield....

One last note is how exceptionally well the graph posted above shows the sensitivity to dose of the VST baskets. The most stable basket would be the LM single, which gives nearly the same output in 25 seconds, regardless of swings in dose. The LM double basket, while flowing more than the Faema or MicroFine baskets, seems also to be the least sensitive to dose of the double baskets. The VST baskets, however, are just nuts with respect to changes in dose drastically changing the volume in the cup for a 25 second pour. Not sure why or if I really care, but it is interesting and certainly supports the claims of the masses with respect to the baskets being intolerant of dosing outside their rated volume.

Cheers!

- Jake

AssafL wrote:Fingering flows?

It took me longer than it should have do digest this response.
Strangely, it was the post by John regarding nutation and how it makes "a better puck" that got the wheels turning in a cohesive fashion.

My practice has been so far to wait until the entirety of the basket is saturated before engaging the pump and my results have been consistent and better than before, but let's face it. The results of a moron with an old, once expensive machine with no gicleur and no expansion valve and no idea what good coffee tastes like aren't that hard to improve upon. But I digress...

Decent offers a truly visionary way of exploring puck integrity by utilizing a flow meter and real time at the puck pressure to give you information regarding the progression of the shot and the degradation of the puck. I surmise that what Matt was playing with was using a constant flow rate and attempting to build as much back pressure as possible without fracturing the puck. Poor prep (or less than perfect, really) can lead to fingering flows, where streams of extraction progress more rapidly than others. I think of this a little like quantum chaos theory meets plinko. Every obstacle the water encounters on its way through the puck leads to a direction change. Water takes the path of least resistance, and once a path, or several have been established, solids are stripped from those paths and they become the preferred paths through the puck. The result with our typical machines that provide a constant pressure above the gicleur is that the flow increases as the fingered flow paths widen. Since pressure drop across the gicleur is proportional to flow, we would also see a pressure drop at the puck, if we were measuring it.

Decent provides a different lense for this thought experiment by actively adjusting pump duty to achieve and maintain a constant flow profile. As a result, one can objectively quantify the integrity of the puck by applying a low and fixed flow rate and identifying the peak back pressure generated (I suspect there is a practical limit here as to what is desireable...) and the rate of erosion (this should be minimized. Period.) While I haven't seen the charts aside from what is floating around, I strongly suspect that too high a peak pressure results in either a choked machine or a fractured puck and a prompt drop in pressure as channels open up and vent the pressure... even extraction throughout the entire puck should be represented by the chart that has a highest local maximum at the point of saturation coupled with the slowest decline in pressure as the shot progresses, given a grind level that produces your desired yield in the desired shot time. It can be debatable what grind, dose, yield and time are desired, but I can't think of a good argument for not maximizing the area under the pressure curve for those parameters.

Why?

Because the slowest puck erosion is (I think by definition) the most even extraction. Fingering flow paths should lead to decreased back pressure and pockets of over-extracted coffee where the fingers developed blended with a fair amount of under extracted coffee everywhere else. I dare say that when Perger and Rao say that nutating creates a better puck, the result will almost certainly be better in the cup, all other things being equal.

I tried nutating the last couple days and it does seem to yield a more even bleed through but changed the characteristics of the pour in a way that will require more work to figure out whether it's better at its best with it without nutating. One thing for sure is that I can fit an extra gram in the basket when nutating, possibly more. This is annoying because I had everything figured out and now I have yet another rabbit hole to trot down before I feel like messing everything up by throwing pressure/flow profiling into the mix.

**sigh**

So,

Tying that back to my question of how 3 different baskets with the same bean, ground the same way, at the same dose could pour so differently, I think "fingering flows" could certainly be a contributor of a higher yield shot. The question is why?

AssafL wrote: Perhaps there is another: headspace.

Good point. The volume between the gicleur and the puck divided by the water debit should give you a rough estimate of the time it takes to fill the headspace. Longer time, better saturation before back pressure is generated. But would those total time be the same if you factored in the delayed time to fill the larger headspace coupled with the increased rate of the pour? Basically, would we expect the high yield puck to have a small headspace and got to work on brewing a shot faster and put more in the cup, or was it the basket with extra headspace that had better PI and flowed faster that had the higher yield? While Jim took notes and commented on the general effects of hole size and pattern on dwell time and pour rate, there isn't quite enough detail in the basket discussion to concretely answer those questions; nor does it give us any insight into whether or not flow fingering (ew...) could be the culprit...

What I can (and shall) do is dose my typical 17.X dose into my triple basket and see if there is more dwell before drops appear, and if there is any change in the bleed through pattern. Unfortunately, my triple basket also has nearly perfectly straight sides, so the puck will be thinner and it has more holes over a larger area, so there won't be a good way to quantify the root cause of any results, but even bleed through should mean less fingering and that should equal a better extraction. At a minimum, I can adjust my dose and yield so that puck thickness is maintained across my 14g, 18g and 21g baskets and scale the yield accordingly without carrying the grind to see what happens. It might be interesting.

More to come...