I'm highly skeptical that reasonable applications of WDT can cause a meaningful segregation or demixing based on size. I have a lot of experience with mixing of powders as powder metallurgist and ceramic engineer, fields where understanding mix homogeneity is critical. Coffee has none of the hallmarks of a material likely to exhibit rapid demixing. It is low bulk density, irregularly shaped, no density disparity between materials and of a fine enough size where electrostatic cohesion plays a strong role. It is also tacky from the coffee oils which limits fines migration. A few large clunkers may indeed move, but I doubt that is the primary impact of the WDT technique.

Phil, is that a spent puck? I'm not arguing for or against the speculation of fines migration prior to pulling the shot after WDT, but if that is a spent puck, does the water passing through the puck move fines towards the bottom of the basket?

Spitz.me wrote:Phil, is that a spent puck? I'm not arguing for or against the speculation of fines migration prior to pulling the shot after WDT, but if that is a spent puck, does the water passing through the puck move fines towards the bottom of the basket?

Spent puck, and yea, the water surely "washes" the fines away to make the larger particles more obvious.
But flipping the puck over and gently breaking it to see particles through the middle and on the bottom doesnt reveal any other obvious large particles, and certainly not in the amount found on the surface.

It's a fair observation though, and thanks for bringing it up!

kmills wrote:I'm highly skeptical that reasonable applications of WDT can cause a meaningful segregation or demixing based on size. I have a lot of experience with mixing of powders as powder metallurgist and ceramic engineer, fields where understanding mix homogeneity is critical. Coffee has none of the hallmarks of a material likely to exhibit rapid demixing. It is low bulk density, irregularly shaped, no density disparity between materials and of a fine enough size where electrostatic cohesion plays a strong role. It is also tacky from the coffee oils which limits fines migration. A few large clunkers may indeed move, but I doubt that is the primary impact of the WDT technique.

Thanks for your input and expertise!

What might you call "meaningful segregation"? Does having that number of large particles sitting on top of the puck count, when I see no others through the thickness? (obviously visual inspection in this case isnt sufficient, but it's all I have right now)

I had thought about static forces, and adhesion due to moisture\oil as well, as we are dealing with very small\light particles where electrostatic\adhesion forces can begin and do overcome gravitational ones. I'm not saying there's "zero" static in my grinds, but the zass 169dg grinder I used here has never exhibited any static at all. Super fluffy since I first got it ~4yrs ago.

I wont even begin to speculate about the adhesive forces from oils\moisture.

I do think though that its curious I would see such a large number of large particles at the surface of my puck if my claims werent at least plausible.

An interesting experiment might be to grind a single dose at two settings and do WDT\shake\tap\settle and then try to scoop\shave off a little at a time onto separate plates for comparison. This could be easily done using one of those plunger\syringe style measuring cups typically used for sticky liquids. Just imagine the measuring cup is the PF basket. If I had one I would do it.

Lastly, I would agree with you about it not being the primary impact of WDT.
It does, however, seem that it does occur. I also think it could also be argued, that with user intent, one COULD make this particle distribution claim a primary impact of "WDT" by stirring more, tapping to settle, stirring again, tapping, repeat etc.

I also seem to recall Jim S. writing something here to this effect, that by stirring and tapping, his goal, or maybe he was saying, a general goal, of this process was to move the fines lower in the puck to help block the PF basket holes to aid in controlling flow rate. I hope I'm not misquoting or misremembering, but I'm fairly certain it was him and appx what he typed.

I've been seeing this large particle size ontop of the puck with this zass and a kyocera grinder for as long as I can remember noticing....but now I think we're getting Off topic a bit. I just didnt like being called wrong and being claimed to only be speculating, when in my experience, it has been observed and true.

By meaningful I am saying I don't think that segregation, under typical scenarios, would be a contributor to failure/success of the mechanics of extraction (channeling etc.) or to flavor in the cup. Your picture is compelling but I wonder if it has to do with the small burrs hand grinders you are using. I haven't noticed such clunkers with my Vario, but I haven't been looking too closely. I also think that top of a puck undergoes some pretty violent things as the solenoid opens. Do you see clunkers in the as-tamped surface? This topic is worth discussing but may be diluting the interesting discussion specific to the Perger technique.

Getting back to Matt's performance; his claim was that a nutating tamp can increase the puck density compared to a straight down tamp, at least when the distribution of particle sizes is tight. This is believable. A straight down tamp will never turn



but a nutating one can.

kmills wrote:By meaningful I am saying I don't think that segregation, under typical scenarios, would be a contributor to failure/success of the mechanics of extraction (channeling etc.) or to flavor in the cup. Your picture is compelling but I wonder if it has to do with the small burrs hand grinders you are using. I haven't noticed such clunkers with my Vario, but I haven't been looking too closely. I also think that top of a puck undergoes some pretty violent things as the solenoid opens. Do you see clunkers in the as-tamped surface? This topic is worth discussing but may be diluting the interesting discussion specific to the Perger technique.

gotcha. I think thats a fair definition of meaningful, and I think I mostly agree with it, especially when we're considering grinds from better grinders than my small conical zass at work, where they have a tighter bell curve for particle distrubtion.

The puck was out of a gaggia factory, so no solenoid.
I very very very (very!) slowly raise the lever till I barely hear a hiss of flow and hold it there for at least 5s and then continue raising very slowly until flow sounds stop. I am at altitude, so the first bit of water is surely flash boiling, but as water is added, its probably cooled enough where its liquid water being added very slowly.

As to the as tamped question, the big clunkers are there as well, but they arent as clearly visible as post extraction thanks to the washing of the grinds. Worth noting, I feel that these big clunkers, as we're calling them, come from the very first and last turns of the burrs when there isnt a full complement of coffee grinds to help support and center the burrs. This is visible proven if I inspect the first grinds in the drawer as I see a large number of large particles. If I were to dump the entire drawer onto a plate, I think most of the other grinds would be really very even. (for what its worth, this hand grinder has made some of the tastiest extractions I've had, even in comparison to my pharos)

agreed on the discussion
Maybe if one of the mods trips over this discussion thread they can break it out?
Until then I will hold off on more commenting.

GlennV wrote:Getting back to Matt's performance, his claim was that a nutating tamp can increase puck density when the distribution of particle sizes is tighter. This is believable. A straight down tamp will never turn

<image>

but a nutating one can.

This illustrates distribution to remove air pockets, not tamping to compress the puck denser. Grind particles are highly irregular -- you are compressing something that doesn't look like a sphere but like barbed wire. According to Petracchi, even shaking sieves cannot reliably separate the grinds.

There is one and only one point to all the distribution and tampng, to makes sure the puck stays sealed and crack free while it soaks under increasing pump pressure (which is why smaller gicleurs make it easier). Once the puck is soaked, the barbed wire bits starts dissolving, the fines can migrate, and the puck turns into a proper aggregate for the rest of the shot.

My hypothesis would go something like this:

1. The Aurelia has really small gicleurs, making lungo and cafe crema shots workable possibility
2. While the usual pro techniques of finger swipes, Stockfleth's or roating the basket while dosing work for espresso grind, a coarser grind may need some extra help in distribution in line with Glenn's diagram. Nutation is fast, and more effective at levelling than these usual moves, although not as effective as WDT
3. Therefore I think Matt's technique may have limited usefulness in terms of grnder, machine and coffee.

I'll try it with the steel Vario and Bunn on the Strega, where if it has the best chance of working; but I'm not holding my breath

The example was extreme; but I would still argue that, no matter how macroscopically even the initial loose distribution of coffee, a straight down tamp can easily lead to small triangles (in x-section) of uncompressed grinds, with the load being carried through particles to the side. It seems entirely believable that moving the coffee around, by nutating, could mitigate this.

I didn't invent nutating, another alt.coffee guy called North Sullivan did. Bit I'm the second person to use it, and gave it it's name (he called it ship's mast tamping). I've been doing it for eleven years. Believe me, it does not slow done the flow compared to other competent methods.

I plan on visiting St Ali South in July... a bit too far away for our burning questions but will be happy to report back!