Hard tamping and channeling - Page 2

Beginner and pro baristas share tips and tricks for making espresso.
Posts: 264
Joined: 2 years ago

#11: Post by LittleCoffee »

The OP cites a 58mm no name vs 58.5mm normcore. Could this be a factor?

Posts: 152
Joined: 3 years ago

#12: Post by kye »

I agree Mike-R - and Lances indication that double-tamping reduces the variation in density between tamps (which might potentially be of different pressure) doesn't seem to be that significant in the data.

However, going back to your comment about cracking in the puck caused by the flex in the bottom of the basket, it might be that a hard tamp followed by a soft-tamp might be the winning formula, as the second soft tamp might be enough to compress the top part of the puck pushing the cracks closed but not so much as to flex the bottom so much that new cracks open up in the grounds when the pressure is released.
I guess it's something easily tested at home, to see if it has major impacts anyway.

Posts: 289
Joined: 14 years ago

#13: Post by Mike-R »

You might be right about a second soft tamp healing any cracks. As you say, it would require testing to be sure.

That being said, my ideas about cracks forming due to basket flex would also require testing to confirm. I think it should be pretty straight forward to test by pulling shots with two different methods of support: first with the portafilter supported in a portafilter stand, and second with the basket supported on the bottom by a flat surface. If the first method of support results in much more frequent channeling than the second, then probably basket flex is the culprit.

On a related subject, you might find it interesting that I have experimented with 9 bar pre-infusion (flow control valve wide open) without channeling issues. That compacts the puck with over 600 lb of force. I have to grind a little coarser to avoid choking the flow, but it does work. The results are interesting, but not something I do very often.

Posts: 131
Joined: 3 years ago

#14: Post by crwper »

Author of the Quantitative Cafe post here... There are a couple of main results in the tamping article:
  • Tamping with more force results in greater compression of the puck.
  • Repeated tamping also results in greater compression of the puck.
Interestingly, tamping for longer (a single long hold) doesn't seem to affect compression--in an experiment with a homemade oedometer, the puck initially compressed very quickly (as quickly as I could apply force), and then over the next 90 seconds there was no significant additional compression (measured using a dial indicator).

It's common in material science to see more compression with a greater applied force--nothing really special happening there, I think.

But the compression with additional tamps is interesting to me. It suggests, I think, that the material first compresses, then when the force is removed it relaxes into a slightly different configuration, so that when force is applied again, it's able to achieve more compression than it did the first time--maybe the smaller particles adjust to fill gaps between the larger particles. The effect may be similar to the "densification" that Lance mentions in his blind shaker video.

Maybe another way to look at this is that the puck is going to go through at least two compression cycles--one when we tamp, and the other when pressure is applied by the machine. If we tamp only once, we're still on the "steep" part of the curve in this plot:

So the second compression (when the machine applies pressure) has more effect on the arrangement of the particles in the puck. But if we tamp a second time, we might be putting those particles into a more stable configuration, so that when pressure is applied a second time, it has a more repeatable effect.

I haven't actually measured the effect on EY or consistency of 1 tamp vs. 2 tamps, so this should be taken with a grain of salt--for the most part, this is speculation based on the data gathered in the tamping post.
★ Helpful

Posts: 289
Joined: 14 years ago

#15: Post by Mike-R »

Hi Michael. Good to hear from you. By the way, I found your post on Quantitative Cafe to be very clear, but appreciate your explaining here as well.

As I mentioned earlier in this thread, each 1 bar of pre-infusion pressure exerts the equivalent of about 32 kg (70 lb) of tamping force (on the dry portion of the puck). In terms of your graph, this would be somewhat equivalent to a first tamp at a low force, then tamping a second time at a much greater force.

I'm guessing by the trends in your graph that the density of an initial tamp followed by a much harder tamp won't be much different than if the first tamp is skipped. If you have tested this or test this in the future, I would be interested to know the result. But now that I think about it, I'm not sure how practical it would be to test a tamp with 70 lbs or more of force.

Posts: 2936
Joined: 13 years ago

#16: Post by malling »

I can't possibly think of a reason a 70lbs tamp would be a very healthy thing to do, you would need a machine to tamp that amount some of the pro purqpress can handle 30kg/66lbs so it would be the best way to test it, at least from a health perspective, my main issue would be consistency in applying such desired force.

If there is a difference between one hard (60lbs/27kg tamp), a dual tamp with the equivalent of 60lbs/27kg tamp force as in quoted article, one lighter tamp followed by a 60lbs/27kg tamp.

Gladly see this with EY% measures on all but I suspect there actually won't be a

If one can achieve the same compression with dual tamp, is definitely more advisable than doing a single very hard tamp.

User avatar
Team HB
Posts: 6940
Joined: 19 years ago

#17: Post by Jeff »

There are those that hypothesize that basket flex under tamping forces may impact the "seal" of the puck to the basket.

Though I haven't seen any strong evidence of this effect, I have seen baskets flex under load and even take on a permanent set. Not so much with second-generation baskets, but definitely with some of the recent "fancy" baskets.

Posts: 3929
Joined: 12 years ago

#18: Post by jpender »

If the basket were flexing that much with a human tamping wouldn't simply applying pressure from the machine be as bad or worse?

Posts: 131
Joined: 3 years ago

#19: Post by crwper »

Mike-R wrote:As I mentioned earlier in this thread, each 1 bar of pre-infusion pressure exerts the equivalent of about 32 kg (70 lb) of tamping force (on the dry portion of the puck).
I've seen this idea mentioned before, but I think it may be a significant simplification of what's happening--someone let me know if gaps in my fluid dynamics knowledge are showing here.

I believe what we need to do is to look at the pressure drop from the top to the bottom of the puck. At the top of the puck, let's suppose we have 9 bar, and at the bottom 0 bar (obviously it's not 0 bar absolute pressure, but all we need here are relative pressures). Assuming the puck provides a uniform resistance, that pressure will drop slowly from the top of the puck to the bottom.

If we want to ask about how much compression is applied to a particular part of the puck, we need to look at the difference in pressure between two "layers". If we choose the top and the bottom, we can say that 9 bar of pressure is dropped from the top to the bottom of the puck, but this isn't the same thing as applying 534 lb of pressure to the top of the puck.

When the pressure is applied by a fluid to a packed bed in this way, I think it is applied gradually. The particles at the top of the puck only feel a very small pressure--the water is basically just flowing past them. But that small pressure is also exerted by those particles on the layer below them, so the next layer experiences a slightly higher force. By the time we get to the bottom of the puck, I believe we would still have a very small pressure applied by the fluid flowing through the puck, but about 534 lb applied across the bottom layer by the particles just above it.

In contrast, if we apply a 534 lb tamping force to the top of the puck, this force is applied directly to the particles on the top layer, and they apply the full force to the particles below them, etc. So each layer feels this 534 lb force, and will tend to yield until internal forces equal the applied force.

So I would guess that if we tamped with 534 lb, we would have a very (and more or less uniformly) compressed puck top to bottom, but if a fluid drops 9 bar of pressure from the top to the bottom of the puck, as would happen in a traditional shot, the top layer would essentially be uncompressed, but the bottom layer would be very compressed.

If that's true, then the main effect of tamping before the shot may be to give those top layers structural integrity.

Posts: 152
Joined: 3 years ago

#20: Post by kye »

I have zero background in fluid dynamics, but in thinking this through (and drafting a reply several times over and changing it as I thought more deeply each time) I have come to the conclusion that I shouldn't speculate what is happening, but also that if there is un-evenness of the puck in any way that might cause the water to move through the puck wetting it faster in that area than others, then that may cause other effects that might in turn lead to a self-reinforcing pattern (similar to the way that channels make themselves larger over time during the extraction).
This would mean that un-evenness in the puck starts setting up the circumstances for an un-even extraction and channeling even before the puck is wet.
This is definitely something to think more about, and more importantly, to see if there are experiments we could do to improve this part of puck-prep.